Looking to boost your learning?

Ensure you’re on track to excel in your studies with trusted

A-level study and revision resources from Hodder Education.

My Revision Student Guides

Notes Consolidate knowledge and
strengthen exam technique with
There are hundreds of My Revision
our convenient and content-
Notes across a wide range of
packed range of Student Guides.
subjects. My Revision Notes
Benefit from clear topic-by-
will set you on track to achieve
topic course coverage alongside
your best with structured topic
challenging sample questions and
summaries to test your knowledge
answers that will keep you on
and understanding of key content,
track for success.
accompanied by a mixture
of activities and exam-style
questions that will help you to
perfect your written answers.

Also available as eTextbooks Also available as eTextbooks

www.hoddereducation.co.uk/mrn www.hoddereducation.co.uk/studentguides

Student Magazines
Workbooks Ideal for developing independent
learning skills, our Student
Maximise your potential with
Magazines provide in-depth
these write-in workbooks written
information surrounding subject
by experienced authors to help
content, with topical articles
you to practise and apply what
and expert exam advice to help
you have learned during your
deepen your understanding and
course. Build confidence and
put knowledge into context.
independent learning skills
through a blend of focused course
guidance and varied practical
activities.

Answers to workbook
questions available online
at hodderplus.co.uk Also available as eMagazines
www.hoddereducation.co.uk/studentworkbooks www.hoddereducation.co.uk/magazines

For more information about all our revision titles

and series across a wide range of subjects, visit:
www.hoddereducation.co.uk/revision

IFC catalogue ads.indd 1 29/10/2020 14:40

325487_01_MRN_AQA_CS_001-033.indd Page 1 6/29/21 9:39 PM f-0116 /103/HO02206/work/indd

AQA
A-level

COMPUTER
SCIENCE
THIRD EDITION

Mark Clarkson
325487_01_MRN_AQA_CS_001-033.indd Page 2 6/29/21 9:39 PM f-0116 /103/HO02206/work/indd

Every effort has been made to trace all copyright holders, but if any have been inadvertently
overlooked, the Publishers will be pleased to make the necessary arrangements at the first
opportunity.
Although every effort has been made to ensure that website addresses are correct at time of
going to press, Hodder Education cannot be held responsible for the content of any website
mentioned in this book. It is sometimes possible to find a relocated web page by typing in
the address of the home page for a website in the URL window of your browser.
Hachette UK’s policy is to use papers that are natural, renewable and recyclable products
and made from wood grown in well-managed forests and other controlled sources. The
logging and manufacturing processes are expected to conform to the environmental
regulations of the country of origin.
Orders: please contact Hachette UK Distribution, Hely Hutchinson Centre, Milton Road,
Didcot, Oxfordshire, OX11 7HH. Telephone: +44 (0)1235 827827.
Email education@hachette.co.uk. Lines are open from 9 a.m. to 5 p.m., Monday to Friday.
You can also order through our website: www.hoddereducation.co.uk
ISBN: 978 1 3983 2548 7
© Mark Clarkson 2021
First edition published in 2016. This edition published in 2021 by
Hodder Education,
An Hachette UK Company
Carmelite House
50 Victoria Embankment
London EC4Y 0DZ
www.hoddereducation.co.uk
Impression number 10 9 8 7 6 5 4 3 2 1
Year 2025 2024 2023 2022 2021
All rights reserved. Apart from any use permitted under UK copyright law, no part of this
publication may be reproduced or transmitted in any form or by any means, electronic
or mechanical, including photocopying and recording, or held within any information
storage and retrieval system, without permission in writing from the publisher or under
licence from the Copyright Licensing Agency Limited. Further details of such licences (for
reprographic reproduction) may be obtained from the Copyright Licensing Agency Limited,
www.cla.co.uk
Cover photo © Maksym Yemelyanov - stock.adobe.com
Illustrations by Aptara, Inc.
Typeset in India by Aptara, Inc.
Printed in Spain
A catalogue record for this title is available from the British Library.
325487_01_MRN_AQA_CS_001-033.indd Page 3 6/29/21 9:40 PM f-0116 /103/HO02206/work/indd

Get the most from this book

Get the most from this book

Everyone has to decide his or her own revision You can also keep track of your revision by ticking
strategy, but it is essential to review your work, off each topic heading in the book. You may find
learn it and test your understanding. These Revision it helpful to add your own notes as you work
Notes will help you to do that in a planned way, through each topic.
325487_01_MRN_AQA_CS_001-033.indd Page 5 6/12/21 7:38 PM f-0236-new /103/HO02206/work/indd

topic by topic. Use this book as the cornerstone My Revision Planner 7 Computer organisation and architecture
158 Internal hardware components
162 The stored program concept

of your revision and don’t hesitate to write in it —

My Revision Planner

My Revision Planner
162 Structure and role of the processor and its components
Introduction 173 External hardware devices
6 Command words
8 Consequences of using computers
1 Fundamentals of programming 182 Moral, ethical, legal and cultural issues and opportunities
9 Programming
9 Fundamentals of communication and networking

personalise your notes and check your progress by

22 Programming paradigms
188 Communication
23 Object-oriented programming
190 Networking
2 Fundamentals of data structures 195 The internet
34 Data structures and abstract data types 201 The Transmission Control Protocol/Internet Protocol
39 Queues 10 Fundamentals of databases

ticking off each section as you revise.

41 Stacks 214 Conceptual models and entity relationship modelling
43 Graphs 216 Relational databases
45 Trees 217 Database design and normalisation techniques
47 Hash tables 222 Client-server databases
48 Dictionaries 11 Big Data
48 Vectors 224 Big data
3 Fundamentals of algorithms 12 Functional programming
53 Graph traversal 227 Functional programming paradigm
57 Tree traversal 230 Writing functional programs
60 Reverse Polish notation 13 Systematic approach to problem solving

Tick to track your progress

62 Searching algorithms 235 Aspects of software development
65 Sorting algorithms
241 Now test yourself answers
67 Optimisation algorithms
4 Theory of computation 253 Exam practice answers
71 Abstraction and automation
83 Regular languages
89 Context-free languages
91 Classification of algorithms
96 A model of computation

Use the revision planner on pages 4–7 to plan your

5 Fundamentals of data representation
103 Number systems
104 Number bases
108 Units of information
109 Binary number system

revision, topic by topic. Tick each box when you

121 Information coding systems
125 Representing images, sound and other data
6 Fundamentals of computer systems
139 Hardware and software
141 Classification of programming languages

have:
143 Types of program translator
145 Logic gates
152 Boolean algebra

4 5

✚ revised and understood a topic Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_02_MRN_AQA_CS_034-052.indd Page 34-35 6/12/21 7:35 PM f-0236-new

My Revision Notes AQA A-level Computer Science Third Edition

/103/HO02206/work/indd

✚ tested yourself 2 Fundamentals of data structures

This code will step through the items in the array n times, where n is the size
of the array. This code would work for all arrays regardless of their size. This
is far more efficient than stepping through several variables one at a time, as
this would require a separate line of code for each variable and the program
Making links
One-dimensional arrays
can be an effective way

✚ practised the exam questions and gone online

2 Fundamentals of data structures

would need to be edited every time the number of items changed. of representing a vector.
Vectors are used in many

Data structures and abstract Multi-dimensional arrays games and physical models,
and are discussed in more

data types
A two-dimensional array can be written down as a table. detail towards the end of
A two-dimensional array can be thought of as an array of one-dimensional this chapter.
arrays.

to check your answers and complete the quick

Data structures
By identifying which row is being addressed using one array index, a
Data structures are designed to hold several items of data at a time. This particular value can be addressed within that row using a second array index.
Data structure The
makes it easier to manage, find, update and process data than to work with Numbers [0]
concept of storing multiple
individual variables. values together with a [0] [1] [2] [3]
For instance, finding the biggest value from ten separate variables would single identifier. [0] 1 3 5 7 Numbers [0][3]

quizzes
require a separate selection statement to interrogate each one, and additional Array A data structure for
code would be needed each time the number of variables increased. Numbers [1]
holding values in a table. Numbers
Using a data structure called an array means that it is possible to step through [0] [1] [2] [3]
[0] [1] [2] [3]
each item using a loop, reducing the amount of code and making the code [1] 11 13 17 19
[0] 1 3 5 7
much more flexible . Exam tip
[1] 11 13 17 19 Numbers [2]
Arrays can also be defined
Single- and multi-dimensional arrays [2] 23 29 31 37 [0] [1] [2] [3]
using the first index to find
[3] 41 43 47 51 [2] 23 29 31 37
(or equivalent) Numbers [2][2]
the column and the second
Numbers [3]
index to find the item in that
An array is a fixed-size collection of values, all of the same data type. column. Always read the
Index A value indicating the [0] [1] [2] [3] question carefully as any
Single-dimensional arrays position of a value with in [3] 41 43 47 51 ambiguities should be made
A single-dimensional array can be written down as a single row of a table. an array or list. Numbers [3][1] clear.

Each individual item in the array is referred to using the identifier (or List A data structure similar Figure 2.1 A two-dimensional array called Numbers
variable name) of the array, followed by the index in brackets (usually square to an array, commonly used Making links
In the example above the contents of Numbers[0][3] can be found by isolating
in Python in place of an
brackets). Each index position can hold data. row 0 and then finding the item in position 3. One common use for a
array.
two-dimensional array is
Index [0] [1] [2] [3] [4] A two-dimensional array is useful because it can store more complex data; for
storing a matrix. Matrices
example, the scores for different students or values for different days.
Data 1 3 5 7 11 Exam tip are often used in complex
It is possible to find the size of a two-dimensional array using the Length maths, and can also be used
In most languages (including
function (exact syntax varies by programming language). to represent a graph data
Note C#, Delphi/Pascal, Java &
structure. This is explored
VB.Net) the array index For an array with indexes defined as [row][column], to find the number of
Arrays are always declared as a fixed size and cannot be changed later in the program. further later in this chapter.
starts at 0, but be very rows, find the number of one-dimensional arrays the table can be split into:
In an array, all data must be of the same data type. careful when reading
NumberOfRows ← Length(ArrayName)
Lists are very similar to arrays, however it is possible to alter the size of a list while the exam questions as some
Matrix A rectangular, two-
program is running and it is possible to store data of differing types in the same list. languages and some To find the number of columns, find the length of a row in the table: dimensional collection of
scenarios will have the array
In some languages (notably including Python) it is much more common to use a list, NumberOfColumns ← Length(ArrayName[0]) values.
index starting at 1.
whereas in others it is much more common to use an array. Nested A selection
To interrogate a two-dimensional array it is necessary to use nested loops.
You are not expected or required to show your understanding of the difference statement or loop inside
FOR i ← 0 to Length(Numbers) // Step through each row another selection statement
between the two, and you should use whichever version is most suitable for your
chosen programming language. FOR j ← 0 to Length(Numbers[i]) // Step through each or loop.
value in that row
Arrays are an effective solution for storing several values because a FOR loop OUTPUT Numbers[i][j] Exam tip
can be used to iterate (or loop) over an array.
ENDFOR Make sure you are able to
In this example Numbers is the name of an array and Length(Numbers) hand trace a program such
ENDFOR as the one above to fully
returns the size of the array (the number of items is contains).
Using the principle above, it is possible to think of a three-dimensional array understand which counter
FOR i ← 0 TO Length(Numbers) variable is addressing the
as an array of two-dimensional arrays.
OUTPUT Numbers[i] row or column index of the
34 table. 35
ENDFOR

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads My Revision Notes AQA A-level Computer Science Third Edition

Features to help you succeed

Exam tips Exam skills
Expert tips are given throughout the book to These summaries highlight how to specific skills
help you polish your exam technique in order to identified or applicable in that chapter can be
maximise your chances in the exam. applied to your exam answers.

Now test yourself Revision activities

These short, knowledge-based questions provide These activities will help you to understand each
the first step in testing your learning. Answers are topic in an interactive way.
available online at www.hoddereducation.co.uk/
myrevisionnotesdownloads Exam practice
Practice exam questions are provided for each topic.
Definitions and key words
Use them to consolidate your revision and practise
Clear, concise definitions of essential key terms are your exam skills.
provided where they first appear.
Summaries
Key words from the specification are highlighted in
bold throughout the book. The summaries provide a quick-check bullet list for
each topic.
Making links
Online
This feature identifies specific connections between
topics and tells you how revising these will aid your Go online to check your answers to the
exam answers. exam questions and try out the extra quick
quizzes at www.hoddereducation.co.uk/
myrevisionnotesdownloads 3

My Revision Notes AQA A-level Computer Science Third Edition

325487_01_MRN_AQA_CS_001-033.indd Page 4 6/29/21 9:40 PM f-0116 /103/HO02206/work/indd

My Revision Planner
My Revision Planner

Introduction
6 Command words
1 Fundamentals of programming
9 Programming
22 Programming paradigms
23 Object-oriented programming

2 Fundamentals of data structures

34 Data structures and abstract data types
39 Queues
41 Stacks
43 Graphs
45 Trees
47 Hash tables
48 Dictionaries
48 Vectors
3 Fundamentals of algorithms
53 Graph traversal
57 Tree traversal
60 Reverse Polish notation
62 Searching algorithms
65 Sorting algorithms
67 Optimisation algorithms
4 Theory of computation
71 Abstraction and automation
83 Regular languages
89 Context-free languages
91 Classification of algorithms
96 A model of computation
5 Fundamentals of data representation
103 Number systems
104 Number bases
108 Units of information
109 Binary number system
121 Information coding systems
125 Representing images, sound and other data
6 Fundamentals of computer systems
139 Hardware and software
141 Classification of programming languages
143 Types of program translator
145 Logic gates
152 Boolean algebra

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_01_MRN_AQA_CS_001-033.indd Page 5 6/29/21 9:40 PM f-0116 /103/HO02206/work/indd

7 Computer organisation and architecture

158 Internal hardware components
162 The stored program concept

My Revision Planner
162 Structure and role of the processor and its components
173 External hardware devices
8 Consequences of using computers
182 Moral, ethical, legal and cultural issues and opportunities
9 Fundamentals of communication and networking
188 Communication
190 Networking
195 The internet
201 The Transmission Control Protocol/Internet Protocol
10 Fundamentals of databases
214 Conceptual models and entity relationship modelling
216 Relational databases
217 Database design and normalisation techniques
222 Client-server databases
11 Big Data
224 Big data
12 Functional programming
227 Functional programming paradigm
230 Writing functional programs
13 Systematic approach to problem solving
235 Aspects of software development

241 Glossary

My Revision Notes AQA A-level Computer Science Third Edition

325487_01_MRN_AQA_CS_001-033.indd Page 6 6/29/21 9:40 PM f-0116 /103/HO02206/work/indd

Introduction
Introduction

As a student of computer science, it is important that you understand three

things in relation to the A-level examinations.
✚ Assessment objectives
✚ Command words
✚ Exams and NEA

Assessment objectives
✚ AO1 You should be able to demonstrate knowledge and understanding of
the principles and concepts of computer science, including abstraction, logic,
algorithms and data representation.
✚ AO2 You must be able to apply your knowledge and understanding of the
principles and concepts of computer science, including to analyse problems in
computation terms.
✚ AO3 You must be able to design, program and evaluate computer systems
that solve problems, making reasoned judgements about these and presenting
conclusions.

Command words
Familiarity with the relevant command words is important. It helps you to
avoid wasting time in the exam room (for example, trying to evaluate when
there is no requirement for it). The most frequently used command words
used for the A-level papers are listed here.
✚ Calculate… requires you to work out the value of something. A correct
final answer, to the required degree of accuracy and with the correct units,
will score full marks. Working is usually required, and correct working can
score marks. (AO2)
✚ Compare… requires you to identify similarities and differences between
ideas, technologies, or approaches. (AO1)
✚ Create… requires you to write program code that solves a problem. Even if
you struggle to get the syntax correct, marks are awarded for evidence of
the approach taken as well as for working code. (AO3)
✚ Define… requires you to specify the meaning of a technical term in order
to show that you understand what it means. (AO1)
✚ Describe… requires you to set out the characteristics of a device or a
computing concept. These can be very short questions worth 1 mark, or
long-answer questions worth up to 12 marks. (AO1)
✚ Discuss… requires you to present the key points. These questions are
typically of medium length, worth 4-6 marks. You should aim to present as
many points as you can and, where appropriate, provide balance between
advantages and disadvantages. (AO1)
✚ Draw… requires you to produce a diagram. These are usually technical
diagrams such an E-R diagram for a database, or a logic circuit diagram. (AO2)
✚ Explain… requires you to provide purposes or reasons. These questions
are used to assess your knowledge of a topic and can sometimes be used
within a specific context. If a question is asked in relation to a particular
scenario then you should always make sure you link to the scenario in
your answer. (AO1, AO2)
✚ Express… requires you to convert an input into a particular format.
Examples might include encrypting a message or calculating a value. (AO2)
✚ Modify… requires you to take an existing section of program code and add
to or edit it. This is commonly asked as part of Section D in Component 1,
involving changes to the skeleton program. (AO3)
✚ State… requires you to express some knowledge in clear terms. These questions
6
are usually short answer questions and are usually worth 1 mark. (AO1)

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_01_MRN_AQA_CS_001-033.indd Page 7 6/29/21 9:40 PM f-0116 /103/HO02206/work/indd

✚ Suggest… requires you to present a suitable case or solution. (AO1, AO2)

✚ Test… requires that you run the code that you have written in order to
check that it functions as it should. This is usually combined with a write/
create/modify question in component 1 and screenshots of the testing

Introduction
process should be included. (AO3)
✚ Write… requires you to create a program or a Boolean expression to solve
a specific problem. These questions can be quite lengthy and partially
complete answers are usually worth a significant proportion of the marks.
(AO2, AO3)

Exams and NEA

The AQA A-level in Computer Science is assessed using three components.
There are two exams, both taken at the end of the course, and one piece
of non-examination assessment (NEA) which is based on a programming
project.
Component 1 Component 2 Component 3
On-screen exam Written exam Programming project (NEA)
2.5 hours 2.5 hours Internally assessed
40% weighting 40% weighting 20% weighting

Component 1
Component 1 is an on-screen exam, largely focused on programming
and computation. In advance of the exam, you and your teachers will
be provided with a skeleton program and a small pack of background
information.
The exam will be taken using a programming language you have studied
throughout the course, and will be pre-selected by your teacher. The available
languages are:
✚ C# ✚ Python
✚ Java ✚ VB.net
✚ Pascal/Delphi
The skeleton program is a working program that is functional but could
be improved. Typical examples include text-based role-playing games or
simulations.
For A-level students the pre-release information is available from September
of that academic year (typically Year 13).
The pre-release information should be given to you by your teacher, but they
may choose not to give the information out straight away, depending on how
and when they plan to deliver the content.
The A-level exam is split into four sections:
1 Section A: Questions about programming and computation; for example,
finite state machines, standard algorithms, trace tables, Turing machines and
computation logic.
2 Section B: A programming problem (unrelated to the skeleton program);
for example, writing a program to convert between binary and denary
numbers.
3 Section C: Questions about the skeleton program; for example, identifying
specific variables or programming constructs, hierarchy charts, class
diagrams and explaining the purpose of specific elements.
4 Section D: Improving the skeleton code; for example, adding an extra menu
option, improving exception handling, adding new functionality.

Component 1 tests programming ability as well as theoretical knowledge

covered in Chapters 1–4, and the skills covered in Chapter 13.
7

My Revision Notes AQA A-level Computer Science Third Edition

325487_01_MRN_AQA_CS_001-033.indd Page 8 6/29/21 9:40 PM f-0116 /103/HO02206/work/indd

Programming syntax
Some questions in Component 1 will include algorithms written using AQA
pseudo-code. It is important to be able to read and understand this pseudo-
code and, in some cases, to write program code in your chosen programming
Introduction

language based on the pseudo-code algorithms provided.

This book is not intended to teach you how to program using one specific
language and uses pseudo-code similar to that used by AQA. Any code that
is not labelled as being from a particular language is instead written in AQA
pseudo-code.

Component 2
Component 2 is a traditional-style written paper, largely focused on the more
theoretical components of this course.
The topics covered in this paper test subject content from Chapters 5–12, and
typically include:
✚ data representation ✚ networking
✚ computer systems ✚ databases and Big Data
✚ computer hardware ✚ functional programming.
✚ consequences of computing
As this is a written paper, no practical programming activities are assessed,
though some practical elements such as calculations, data conversions and
trace tables (especially for assembly language and instruction sets) are likely
to appear.
This paper will generally include short answer questions (1–2 marks) that will
assess your knowledge, questions that will assess your ability to apply your
knowledge (3–4 marks), and a small number of longer-answer questions that
require detailed discussion (6–12 marks).
For these longer-answer discussion questions, credit is awarded for
identifying, discussing and evaluating potential issues.
✚ Marks are awarded for identifying relevant knowledge; for example,
suggesting appropriate input devices for collecting data or identifying the
methods by which wireless data transmissions can be intercepted.
✚ To reach the top mark bands it is important to follow a line of reasoning, using
your knowledge to write in connected sentences in a way that makes sense
and relates to the context of the question. Explaining how each point links to
the scenario and adding as much technical detail and vocabulary as you can
makes it more likely that you will score well on this type of question.
✚ Always make sure you back up any arguments or suggestions you make
with facts, logical arguments and technical details, as unsubstantiated
statements don’t demonstrate your understanding.

Component 3
Component 3 is a non-examination assessment (NEA) component, based on
completing a programming project.
The programming project is extremely open-ended, and it is up to you to
identify a real-world problem and then work through each phase of the
systems development lifecycle in order to solve it.
It is beyond the scope of this book to go into detail in terms of completing
the programming project, but typical examples include online booking and
scheduling systems, computer games with a simple AI component, animal
population simulations, and so on. There is no definitive list of expected or
excluded projects and your teacher will be able to provide with much more
specific guidance.

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_01_MRN_AQA_CS_001-033.indd Page 9 6/29/21 9:40 PM f-0116 /103/HO02206/work/indd

1 Fundamentals of programming

Programming
Data types
It is important to declare variables using the correct data type. This will make
sure that memory is not wasted, and that the program is able to process the
data correctly.
Different data types are processed in different ways; for example, adding two
strings produces a different result to adding two integers.
"123" + "456" = "123456"

123 + 456 = 579

The main data types are:
Data type Suitable for Examples
integer whole numbers 7, 9, 163
real/float fractional numbers 7.3, 9.6, 0.5
Boolean True/False True, False
character a single letter or other character 'b', 'W', '@'
string text "word", "EC4Y 0DZ"
date/time a date, time, or combination 14:18, 17/10/2004
pointer a memory address #bf4e3a67
record a collection of fields RECORD username, password
array a table of data [1,3,5,7,11,13,17]

Different data types use different amounts of memory. If a numeric variable

will always hold a whole number, it will use less memory stored as an integer
than it will if stored as a real or float.

Note
All programming languages deal with data types slightly differently.
It is important to note that Python does not support the character data type, using
only a string to store text of any length. Python also does not support the array data
type. The closest alternative is a list data type.

Most of the data types are quite straightforward. Exam tip

✚ Records are more complex as they allow the programmer to group
There are many additional
together attributes of different types; for example, a user record may have
data types, some quite
attributes for Username, Password and Last_Login. specific (for example, for
✚ An array is a table-like structure. Each field in an array must be of the whole numbers Java has a
same data type (e.g. an array of integers, or an array of strings). choice of byte, short, int or
long). The mark schemes
Making links
allow for any appropriate
Arrays are an important data structure that are used to store sets of data together. answer though only those
The use of arrays can increase the efficiency of programs when processing lots of listed above are strictly
data. Arrays are covered in detail in the chapter on data structures. required.

My Revision Notes AQA A-level Computer Science Third Edition

325487_01_MRN_AQA_CS_001-033.indd Page 10 6/29/21 9:40 PM f-0116 /103/HO02206/work/indd

Built-in and user-defined data types Exam tip

Built-in data types are those already defined by a programming language.
User-defined data types are
The exact list varies from language to language, but most of those in the table
mostly like to appear in the
above are built-in to most programming languages.
1 Fundamentals of programming

skeleton code. Look for any

It is possible for a user to define their own data types, combining more data types made up of other
than one value per variable. The use of user-defined data types means that variables in the skeleton
program code can be simplified, with attributes being grouped together. This code and make sure
can be carried out in various ways in different languages. A good example that you are comfortable
might be a user data type that is made up of a username and a password. working with them in the
syntax used in your chosen
Now test yourself language.

1 What data type is most suitable for storing a whole number?

2 What is the difference between a character and a string?
3 What sort of data would be stored in a pointer?
4 Suggest two fields that could be included in a user-defined record for recording
high scores on a game.

Answers available online

Programming concepts
For each of the programming concepts you should be familiar with both the
Pseudo-code A format
pseudo-code used by AQA and the syntax used in your own programming
for program code that
language.
is not specific to one
programming language.
Declaration Used extensively in
Variable declaration is the process of creating a variable. Component 1. Pseudo-code
is useful for describing an
In languages which are strongly typed (for example, C#, Pascal/Delphi, Java algorithm that could be
and VB.net) the data type of the variable is stated, followed by the variable’s coded in one of several
identifier. In these languages it is possible to declare a variable without different languages.
initially assigning a value to it.
Syntax The strict rules
✚ int Age; (C# and Java)
and structures used within
✚ var Age: integer; (Delphi and Pascal)
a specific programming
✚ Dim Age As Integer (VB.net)
language. You will only be
In languages which are weakly typed (for example, Python) a value must be assessed in the syntax of
assigned to a new variable, and no data type is declared as the data type can one programming language
be changed during the course of the program. for Component 1.
✚ Age = 18 (Python) Declaration The creation
Constants are declared in a similar way, but with a key word that indicates of a variable in memory.
the variable cannot be changed. Because the value of a constant cannot be
changed later, they must be declared with an initial value:
Exam tip
✚ const int MAX = 3; (C#)
✚ Const MAX = 3; (Pascal/Delphi) In Component 1 you will
✚ final int MIN = 3; (Java) be presented with pseudo-
✚ Const MAX As Integer = 3 (VB.net) code algorithms to read,
and to turn into program
Python does not support the use of constants and therefore it is not possible code, but you will not be
to prevent the accidental assignment of a new value to a constant. expected to write answers
using pseudo-code.
Making links
Practical programming
It is important to consider where a variable is declared. A variable declared inside an questions are intended to
IF statement, loop or subroutine can only be used within that section of the code be coded using the specific
and will be destroyed once that section has ended. This issue is dealt with more in the programming language you
section on local and global variables. have been entered for.

10

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_01_MRN_AQA_CS_001-033.indd Page 11 6/29/21 9:40 PM f-0116 /103/HO02206/work/indd

Assignment Assignment Changing the

Changing the value of a variable is called assignment. value of a variable stored in
✚ Assignment in AQA pseudo-code often uses a left arrow (for example, memory.
Score ← 12).

1 Fundamentals of programming
✚ Assignment in most programming languages uses an equals sign (for
example, Score = 12)

Now test yourself

5 What is meant by the term declaration?

6 Describe one difference between the declaration of a variable and the declaration
of a constant.
7 What symbol is used to indicate assignment in AQA pseudo-code?
8 What symbol is used to indicate assignment in most programming languages?

Answers available online

Sequencing
Program code is executed in the order in which it appears. For example, given
the code:
x ← 5
OUTPUT x
x ← x + 1
The value 5 will be output before the value of x is increased.

Selection Selection A program

All but the simplest computer programs need to include decisions. A simple structure that makes a
decision of whether to execute a particular block of code is controlled by a decision about which block
selection statement, often implemented using an IF statement. of code to execute, typically
implemented using an IF
IF score > 90 THEN statement.
OUTPUT "Top marks!" SWITCH An alternative
ELSE IF score > 60 THEN selection structure to an IF
statement that is slightly
OUTPUT "Good job." quicker to execute if used
ELSE with exact values rather
than a range.
OUTPUT "Try again."
END IF
Exam tips
There are no limits to how many ELSE IF statements can be used, but with
a large number it is possible to use a SWITCH statement instead. A SWITCH Remember that IF and
statement is slightly faster to execute if used with single values (rather than ELSE IF commands need
checking for a range) a conditional statement
with a Boolean (True/
SWITCH (grade) False) answer. The ELSE
CASE "A*" THEN command doesn’t as it will
catch all other possibilities.
OUTPUT "Top marks!"
You are not required to
BREAK use a SWITCH statement
and can score full marks
CASE "A" THEN
using IF-ELSE IF-
OUTPUT "Very good job." ELSE. The skeleton code or
pseudo-code may include a
BREAK
SWITCH statement so you
… should make sure you are
familiar with the concept. 11
END SWITCH

My Revision Notes AQA A-level Computer Science Third Edition

325487_01_MRN_AQA_CS_001-033.indd Page 12 6/29/21 9:40 PM f-0116 /103/HO02206/work/indd

Nested selection
Nested IF statements occur when an IF statement is placed inside an IF
Nested Placing a
statement.
programming structure
IF animal = "dog" THEN inside another programming
1 Fundamentals of programming

structure.
IF target = "sheep" THEN
OUTPUT "Dog chases sheep"
Revision activity
ELSE
✚ Write a program in your
OUTPUT "Dog wags tail"
chosen programming
END IF language that uses an
IF statement to allow
END IF the user to choose one
of four options.
Now test yourself
✚ Re-write the program
9 What programming concept is implemented using an IF statement? to use a SWITCH
10 Which part of an IF statement does not need a condition? statement.
11 What alternative to an IF statement is sometimes used if there are several
possible options? Exam tips
12 What is meant by a nested IF statement?
Be very careful with the
Answers available online start and end conditions
when creating FOR loops.
While a pseudo-code
Iteration algorithm may explicitly
When a block of code needs to be repeated, this is referred to as iteration. start and end at given
values, the implementation
Definite iteration for a FOR loop may be
Definite iteration, or count-controlled iteration, refers to the use a FOR loop, less clear (for example,
where the number of times to repeat is known. Even if the number of times to in Python for i in
repeat isn’t always the same, if it is known at the start of the loop then a FOR range(1,5) would stop
loop should be used. at i = 4 and in Java for
(int i = 1; i < 5;
You should be familiar with the syntax of a FOR loop in your own i++) would also stop at
programming language as well as the pseudo-code you might see in an exam: i = 4).
FOR i ← 1 TO 5 Although the specification
refers to definite and
OUTPUT "This is step " + i
indefinite iteration, question
ENDFOR papers typically use the
terms count-controlled and
Indefinite iteration condition-controlled iteration.
Indefinite iteration or condition-controlled iteration refers to a loop where
the number of iterations is not known. A typical example might be a
Iteration The repetition of
validation loop, asking a user to enter a valid input and repeating while a process or block of code.
their answer is invalid.
Definite iteration,
Indefinite loops can be further split into those that assess the condition at the or count-controlled
start of the loop (a WHILE loop) and those that assess the condition at the end iteration Iterating a fixed
of the loop (a DO-WHILE loop). number of times (also
In AQA pseudo-code: known as a count-controlled
loop, implemented as a
A WHILE loop will have the condition at the top of the loop, allowing the program to FOR loop).
bypass the code inside completely. For example: Indefinite iteration, or
value ← 5 condition-controlled
iteration Iterating until a
WHILE value < 100 DO
condition is met (also known
value ← value * 2 as a condition-controlled
ENDWHILE loop, implemented in AQA
pseudo-code as a WHILE
12 loop or DO-WHILE loop).

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_01_MRN_AQA_CS_001-033.indd Page 13 6/29/21 9:40 PM f-0116 /103/HO02206/work/indd

A DO-WHILE loop will have the condition at the bottom of the loop, forcing the program Note
to pass through the code at least once. For example:
Python does not support
DO
the use of a DO-WHILE

1 Fundamentals of programming
OUTPUT "Enter shoe size:" loop, but students are still
expected to be familiar with
ShoeSize ← INPUT
the concept. One solution
WHILE ShoeSize > 12 to this problem is to copy
and paste the first iteration
of the code before a WHILE
Nested iteration loop. Another is to create
Nested iteration means having a loop within a loop. This is used in a number
a REPEAT … UNTIL
of applications, including when working with 2D arrays, and in a number of structure.
standard algorithms including the bubble sort.
FOR i ← 1 TO 3
FOR j ← 1 TO 5 Remember
OUTPUT "Outer loop = " + i + " | Inner loop = " + j Remember that if you know
how many times to loop you
ENDFOR
should use a FOR loop. If
ENDFOR you don’t know how many
times then use a WHILE or
Now test yourself DO-WHILE loop.
13 What type of loop is also called a count-controlled loop? Remember that the inner
loop is completed multiple
14 When should you use a condition-controlled loop?
times for each step around
15 Describe the difference between a WHILE loop and a DO-WHILE loop. the outer loop. Selection
16 Suggest two possible uses for nested iteration. and iteration statements
can be nested inside each
Answers available online other, potentially many
layers deep.
Revision activity

✚ Write a program that uses iteration to print out the 12 times table.
Debug The process of
✚ Write a program that uses nested iteration to print out all of the times tables from 1
identifying and removing
to 12.
errors from program code.
Self-documenting
Meaningful identifier names code Program code that
It is important to choose identifiers for variables (and subroutines) that tell other uses naming conventions
programmers something about the purpose of that variable (or subroutine). and programming
✚ Var1, Var2 and Var3 are very poor choices as they make it very hard to conventions that help other
read the code and to debug if there are any errors. programmers to understand
✚ UserName, UserAge and DateUserLastLoggedIn are much more the code without needing
meaningful and are key to writing self-documenting code. to read additional
documentation.
It is not usually possible to use spaces in identifier names. Common strategies
to aid readability include the use of CamelCase, Kebab-Case or Snake_Case.
Now test yourself
Exam tip
17 Describe two
In Component 1, Section B and Section D, it is very important to use any identifier
advantages of using
names exactly as they are provided in the question. If the identifier name is not given
meaningful identifier
explicitly then remember that making it easier for the examiner to understand the
names.
code makes it more likely you will pick up the marks.
18 Explain the term self-
documenting code.
Note
Answers available
The use of meaningful identifier names to produce self-documenting code can have a online
significant impact on the marks available in the NEA.
13

My Revision Notes AQA A-level Computer Science Third Edition

325487_01_MRN_AQA_CS_001-033.indd Page 14 6/29/21 9:40 PM f-0116 /103/HO02206/work/indd

Subroutines
Information on subroutines can be found in the section Purpose of subroutines.

Arithmetic operations
1 Fundamentals of programming

There are several key operators you should make sure you are familiar with.
Simple arithmetic operators include those for addition (+), subtraction (-), Exam tip
multiplication (*) and division (/).
DIV and MOD are common
It is important to understand the different types of division operation that can
terms for integer division
be carried out: and modulo. Make sure you
✚ Real or float division will result in a real or float answer; for example, are comfortable with how
5 / 2 = 2.5 they function.
✚ Integer division will strip any fractional part of the answer, effectively
rounding down; for example, 5 DIV 2 = 2
✚ The modulo operator will find the modulus – the remainder of an integer Operators Symbols used
division, as a whole number; for example, 5 MOD 2 = 1 to indicate a function.

DIV and MOD are useful for converting between different number systems, Modulus The remainder of
the division of one number
including conversion between units of time, imperial measurements, and
by another.
between numbers using different bases (for example, denary and hexadecimal).
Exponentiation The
Exponentiation refers to powers; for example, 2^3 = 2*2*2 = 8
raising of one number to the
Rounding can be carried to a given number of decimal places or significant power of another.
figures, generally using a function; for example, round(3.14,1) = 3.1,
Rounding Reducing the
round(3.16,1) = 3.2 number of digits used to
Additional functions can be used that always round up or always round down; represent a value while
for example, roundup(3.142,1) = 3.2 maintaining its approximate
value.
Rounding down has the same effect as truncation.
Truncation Removing any
Exam tip value after a certain number
of decimal places.
Some questions in Component 1 will assess your understanding of programming
principles in general and may involve having to read and show understanding of
pseudo-code. Other questions will require you to write your own program code in your
chosen programming language, so it is important to make sure you are familiar with
the specific operators and functions that are used in that programming language.

Now test yourself Revision activity

19 What is the difference between float division and integer division? Write a program that uses
DIV and MOD to convert a
20 What is the value of 7.86 when rounded to one decimal place?
given number of hours into
21 What is the value of 7.86 when truncated to one decimal place? a number of days.
Answers available online

Relational operations
Relational operators are used in comparisons, typically in selection
statements and condition-controlled loops. A relational operation, or
comparison operation, will always return either True or False.
The relational operators are as follows:
= or == Equal to
!= or <> or ≠ Not equal to
> Greater than
>= Greater than or equal to
< Less than
14 <= Less than or equal to

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_01_MRN_AQA_CS_001-033.indd Page 15 6/29/21 9:40 PM f-0116 /103/HO02206/work/indd

Note
Different programming languages have different syntax rules for the equal to
comparison operator.

1 Fundamentals of programming
AQA pseudo-code and VB.net use a single equals (=).
C#, Pascal/Delphi, Java and Python use a double equals (==).

Now test yourself Making links

State whether each of these is True or False when x ← 60. C# and Java are not able to
22 x != 60 make use of the standard
relational operators when
23 x < 60
comparing strings. String
24 x >= 60 handling operations are
25 x = 60 discussed on page 16.

Answers available online

Boolean operations
Boolean operations can be used to invert the logic of a conditional statement,
or to combine two or more conditions together.
NOT Will invert the logic (that is, a condition that returns True will become False).
AND Both conditions must be True.
OR Either one condition must be True, or both.
XOR One condition must be True and the other False.

Making links
The exam for Component 2 assesses understanding of Boolean logic in much more
detail, including questions on Boolean algebra. The same basic principles apply
whether combining logical statements in a practical programming setting or solving
Boolean equations. For a more in-depth examination of Boolean logic see Chapter 6.

Now test yourself Revision activity

Where a = 50 and b = 100, would each overall condition be True or False? Write a program to calculate
the output of simple logic
26 NOT (a = b)
circuits for NOT, AND, OR
27 a < 100 AND b > 100 and XOR.
28 a > b OR b = 2*a
29 a < b XOR b = 100

Answers available online

Variables and constants

Concept of a variable
Computer programs are essentially sequences of instructions for how to Memory The location
process data. This data must be stored in memory in order to be accessed and where instructions and data
the computer program will save and retrieve data as needed. are stored in a computer.
Variables are made up of four components: Identifier A technical term
✚ A memory address – the location of the variable in memory. for the name of a variable.
✚ An identifier – a name used to identify the variable in the program code.
Data type A definition of
✚ A data type – a definition of what type of data can be stored in that
what type of data can be
variable.
stored in that variable. 15
✚ A value – the actual value of the variable at that moment in time.

My Revision Notes AQA A-level Computer Science Third Edition

325487_01_MRN_AQA_CS_001-033.indd Page 16 6/29/21 9:40 PM f-0116 /103/HO02206/work/indd

Variables are used to store values that are used in a computer program. The
Exam tip
value of a variable can be changed while the program is running (such as a
score, a running total, a user’s name). Component 1, Section C will
often start with asking you
A named constant is a variable whose value cannot be changed while the
1 Fundamentals of programming

to ‘state the identifier for…’

program is running. Examples might include constants used in calculations
some variable or subroutine
(such as Pi) or user-defined values that should be used throughout the
in the skeleton code.
program (such as a maximum number of turns or a maximum size).
For those questions you
It is a common convention to use all caps in the identifier for a constant; for must only write down the
example, MAX_SIZE or MIN_VALUE and to declare constants at the start of a identifier, and not the whole
program. line of code.

Named constants are useful because:

Named constant A
✚ they make it easier to update the program if the value of the constant
variable whose value
needs to be changed
cannot be changed while
✚ it reduces the need to use absolute values (that is, it is better to use a
the program is running.
constant in the condition for a loop than a hard-coded number)
✚ it stops the value being changed accidentally
✚ it makes code more readable. Exam tip
Remembering two or
Now test yourself three standard advantages
30 What four components make up a variable? for each aspect of
programming (such as
31 What is the meaning of the term identifier for a variable?
named constants) will help
32 What is the main difference between a variable and named constant? you answer those questions
33 What are the advantages of using a named constant? quickly and accurately in the
exam.
Answers available online

String handling operations

Working with strings is an important part of programming, and it is very
important to be familiar with the specific syntax used in different languages.
Comparing strings is handled using different syntax in different languages.
✚ In AQA pseudo-code and VB.net, a single equals (=) is used to test for
equality
✚ In Pascal/Delphi and Python, a double equals (==) is used to test for equality.
✚ In C# and Java a function must be used; for example, if (StringOne.
equals(StringTwo)) (C# and Java).
There are several other string handling operations with which you should be
familiar. The exact syntax will vary significantly depending on your chosen
Concatenate Join two
programming language and it is important to make sure you are confident
(or more) strings together;
in carrying out all of these operations in that language. Some examples are
for example, ABC + DEF =
included below, in a variety of different languages, in order to demonstrate
ABCDEF.
each operation.
The key operations are as follows.
Key operator Example
Finding length of a string len(string) or string.length()
Finding a character at a given position string.charAt(3) or string[3]
Retrieving a substring string[0:6] or string.Substring(0,6)
Concatenating two or more strings stringOne + stringTwo or
concat(stringOne,stringTwo)
Finding the numeric code for a character ord('D') or (int)'D'
Finding the character from a given numeric code char(68) or chr(68)
Converting strings into different formats, and vice-versa converting a string into an integer, float or date/time format

16

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_01_MRN_AQA_CS_001-033.indd Page 17 6/29/21 9:40 PM f-0116 /103/HO02206/work/indd

Exam tips Substring A series of

characters taken from a
Substring methods in different languages use different parameters.
string.
✚ Python and Java require the start and end positions.

1 Fundamentals of programming
✚ C#, Delphi and VB.net require the start position and the length.
Make sure you read any pseudo-code questions that involve substrings carefully,
and make sure you are comfortable programming with substrings using the language
chosen for your exam.
Some languages treat a string as an array of characters and can use an index array to
refer to a specific character; for example, string[3]. Some languages require the
use of functions such as charAt(int).
Make sure you are familiar with all of the operations above in your own language in
advance of the Component 1 exam.

Now test yourself

34 Explain what is meant by concatenation.

35 Which string handling operation should be used to extract a person’s first name
from the string FullName?
36 State and describe two other string handling operations.

Answers available online

Revision activity

✚ Write a program that will ask for a string and display each character’s
numeric code.
✚ Write a program that will ask for a series of numeric codes and convert them
into a single string which is displayed.

Random number generation

A large number of programs rely on random number generation.
Most languages have the ability to generate a random float between 0 and 1;
for example, in Delphi:
float: Random;
Generating a random number in a specific range can then be achieved using
multiplication; for example, to find a random float between 0 and 3 using VB.net:
Rnd() * 3
Most languages have a function for finding a random integer; for example, in
Python:
random.randint(1,3)

Exam tip
In most years there is some random number generation included in the skeleton code.
If you forget the correct syntax during the exam then make sure you know where to
find a working example from the skeleton code to help you.

Revision activity
✚ Write a program that will ask for a minimum and maximum number and will
generate a float between those two values.
✚ Write a program like that one above that will generate an integer between those
two values.
17

My Revision Notes AQA A-level Computer Science Third Edition

325487_01_MRN_AQA_CS_001-033.indd Page 18 6/29/21 9:40 PM f-0116 /103/HO02206/work/indd

Exception handling
Exception handling is used to deal with situations where a program may
Exam tip
crash. The program should try to execute a block of code and then catch the
1 Fundamentals of programming

error and deal with it appropriately. Make sure you can

explain the purpose of any
Although the exact syntax may vary between languages, the basic construct
exception handling found
typically reads: in the skeleton code as this
try may be a question asked in
Section C.
{
strInput = input("Enter a number")
Making links
numInput = int(strInput)
Using exception handling
} in your NEA programming
project is a good way
catch (Exception)
to demonstrate good
{ programming techniques.
print("Error, did not recognise a number")
numInput = -1 Revision activity

} Write a program that asks

users to enter a 2-digit
Now test yourself number as a string (such as
‘12’), convert it to an integer,
37 What does the initialism RNG typically refer to? and then double it. Use
38 What runtime error might occur in a program that asks a user to enter numeric exception handling to catch
values? errors such as a non-digit
39 What programming structure should be used to deal with these types of errors? entry (such as ‘twelve’).

Answers available online

Purpose of subroutines
A fundamental aspect of improving the efficiency of program code is the use
of subroutines. Exam tip

A subroutine allows a programmer to take a section of program code that Make sure you are able
performs a specific task and move it out of line of the rest of the program. to explain the difference
between a function and a
This means that the programmer can call the subroutine in order to run that
procedure, and that you can
block of code at any point.
identify which subroutines
Subroutines are called using the subroutine’s identifier, followed by are which in the skeleton
any arguments that must be passed in parentheses; for example, code. This is a common
DisplayGreeting(name). topic in section C.

Subroutines that do not require any parameters must still be called using
parentheses; for example, DisplayDate(). Subroutine A named block
of code designed to carry
The advantages of subroutines are:
out one specific task.
✚ the subroutine can be called multiple times without needing to duplicate
Call The process of running
the code
a subroutine by stating
✚ changes to the subroutine only need to be made once its identifier, followed by
✚ it is easier to read the code any required arguments in
✚ it is easier to debug the code if there is a problem parentheses.
✚ subroutines can be re-used in other programs Pass The transfer of a
✚ the job of writing a program can be split, with each programmer tackling value, or the value of a
their own subroutines. variable, to a subroutine.

Making links
Using meaningful, self-documenting identifiers for subroutines is important to make
18 your code more readable. It is essential to do this as part of your NEA.

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_01_MRN_AQA_CS_001-033.indd Page 19 6/29/21 9:40 PM f-0116 /103/HO02206/work/indd

Subroutines are broken down into two types: functions and procedures.
Function A subroutine that
A function is a subroutine that returns a value once it has finished executing. returns a value.
A typical use of a function is to carry out a calculation and return the result. Procedure A subroutine
that does not return a value.

1 Fundamentals of programming
A procedure is a subroutine that does not return a value. A typical use of a
Module A file that contains
procedure is to display some data and/or prompts to the user.
one or more subroutines
Subroutines can be grouped together in a file to form a module or library. that can be imported and
These subroutines can then be re-used in other programs. used in another program.
Library A collection of
It is common to import libraries that have already been written to help solve modules that provide
problems; for example, in Java import java.util.* or, in Python, import related functionality.
random.

Now test yourself

40 State three advantages for using subroutines.

41 Explain why it is important for subroutines to have meaningful identifiers.
42 What is the name for a subroutine that doesn’t return a value?
43 What is the name for a subroutine that does return a value?

Answers available online

Parameters
The parameters of a subroutine are the variables that must be passed to a Parameters The variables
subroutine when it is called. This is declared when the subroutine is written. that a subroutine needs in
For example, in the code Procedure DisplayTemperature(int Temp, order for the subroutine
bool Celsius), the subroutine called DisplayTemperature needs to be passed to run.
an integer value and a Boolean value. Arguments The actual
values that are passed to
When a subroutine is called, the values must be passed as arguments; for the subroutine at runtime.
example, DisplayTemperature(20,True).

Returning a value
Functions must always return a value at the end of their execution. This is Return To pass a value or the
carried out with a return statement. contents of a variable back
For example: to the place in the program
where the function was called.
Function Double(int StartVal)
return 2*StartVal Now test yourself
End function 44 Describe what is meant
The value that is returned is passed back to the part of the program that by the term parameter.
called the function. 45 Explain the purpose of a
return statement.
It is possible for a subroutine to have several different return statements – for
example, within a selection structure – but the subroutine will stop once a Answers available
value has been returned. online

Local variables
When a subroutine is called, any variables passed as parameters and any Local variables Variables
variables declared within that subroutine are referred to as local variables. that are declared within a
These variables can only be accessed within that subroutine and will be subroutine and can only be
destroyed once the subroutine has finished executing. This is referred to as accessed by the subroutine
the scope of the variable. during the execution of that
subroutine.
Scope The visibility of variables
(either local or global).
19

My Revision Notes AQA A-level Computer Science Third Edition

325487_01_MRN_AQA_CS_001-033.indd Page 20 6/29/21 9:40 PM f-0116 /103/HO02206/work/indd

The advantages of local variables are:

✚ uses less memory as local variables are destroyed once the subroutine has
finished executing
✚ less likely to accidentally change the value of a variable somewhere else in
1 Fundamentals of programming

the program
✚ easier to debug
✚ variable identifiers can be re-used in separate subroutines Modular Independent of
✚ subroutines can be more easily re-used (subroutines are modular). other subroutines.

Knowing that local variables are destroyed once that section of code has
finished executing, it is important to note that variables declared within a
selection statement or iteration structure will also be destroyed at the end of
that section of code. It is therefore very important to choose carefully where
in the program a variable will be declared.

Global variables
Global variables are variables that are declared in the main program and can
Global variables Variables
be read or altered in any subroutine.
that can be accessed from
Accessing global variables from within subroutines reduces the need for any subroutine.
passing parameters and using return statements but should generally be
avoided where possible.
Making links
Global variables are, however, useful for named constants.
Try to limit your use of
Now test yourself global variables when
working on your NEA
46 Explain what is meant by the scope of a variable. programming project. Using
47 Explain three advantages of using local variables. a modular structure for your
subroutines will increase
48 Explain three disadvantages of using global variables.
the range of marks you are
49 Describe a situation where it would be appropriate to use a global variable able to access.
Answers available online

Stack frames
When a subroutine is called a stack frame is created. This stack frame
contains: Stack frame The collection
✚ the return address – where to return to in the program once the subroutine of data associated with a
has finished executing subroutine call.
Call stack A data structure
✚ parameters – the variables to which data was passed when the subroutine
that stores the stack frames
was called
for each active subroutine
✚ local variables – any variables declared within that subroutine.
while the program is running.
Newly called subroutines are added to the top of the call stack and the top
stack frame is removed once that subroutine has been completed.
Exam tip
Making links Make sure you can recall
Stacks are a complex data structure with a variety of uses in programming. Stacks and the three things stored
other complex data structures are explored further in Chapter 2. in a stack frame as this
is a common question in
Section A.
Now test yourself

50 What three things are stored in a stack frame?

Note
51 When is a new stack frame created? If a program becomes stuck
52 Where are stack frames stored? in a loop that continually calls
one or more subroutines then
Answers available online eventually the call stack will
become too full, triggering a
20 stack overflow error.

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_01_MRN_AQA_CS_001-033.indd Page 21 6/29/21 9:40 PM f-0116 /103/HO02206/work/indd

Recursive techniques
Some algorithms are best solved by solving smaller and smaller instances
Recursion The process of
of the same problem. To achieve this, a function must call itself repeatedly –
a function repeatedly calling

1 Fundamentals of programming
this is known as recursion. One example of recursion is used in calculating a itself.
factorial.
A factorial is calculated by multiplying a number by all of the integers less
than itself. The ‘!’ is used as the mathematical symbol for ‘factorial’. For
example:
5! = 5 × 4 × 3 × 2 × 1
This can be simplified as 5 × 4!
4! = 4 × 3!
3! = 3 × 2!
2! = 2 × 1!
1! = 1
A recursive subroutine will continue to call itself (known as the general case) General case A case in
until it reaches a decision that returns a value without calling itself (known as which a recursive function is
the base case or terminating case). called and must call itself.
Base case The case in
The pseudo-code algorithm for a factorial calculator might read: which a recursive function
Function Factorial (int n) terminates and does not call
itself.
IF n <= 1 THEN
RETURN 1
Exam tips
ELSE
Following a recursive
RETURN n * Factorial (n-1) algorithm can be very tricky
and it is important to get
END IF
lots of practice using trace
End Function tables to step through
recursive algorithms.
If n = 1 or less then the function cannot attempt to break the task down
any further, and should stop calling itself recursively, returning the result of Section B questions can
1 (1! = 1). This is the base case. require the use of recursion
to achieve full marks,
In all other cases the function will call itself. This is the general case. though a non-recursive
It is possible to have more than one general case and more than one base case. solution will always be
possible.
Now test yourself

53 What is meant by a recursive function? Making links

54 Explain what is meant by the base case in a recursive algorithm. While Sections B and D
55 Explain what is meant by the general case of an algorithm. may assess your ability to
write program code using
56 Using the pseudo-code for a factorial calculator, above, state the maximum
recursive techniques,
number of stack frames on the stack when calculating Factorial(4).
Sections A and C may
Answers available online require you to demonstrate
your ability to hand trace a
recursive algorithm. Trace
Revision activity tables are covered in more
detail in Chapter 4.
✚ Use the pseudo-code provided to program and test a working factorial calculator.
✚ Use a debugger to step through the factorial calculator and inspect the stack frames.
✚ Create a program that uses recursion to display the Fibonacci sequence
(in which the previous two numbers are added to get the next one; for example,
1,1,2,3,5,8,13…).

21

My Revision Notes AQA A-level Computer Science Third Edition

325487_01_MRN_AQA_CS_001-033.indd Page 22 6/29/21 9:40 PM f-0116 /103/HO02206/work/indd

Programming paradigms
Programming paradigms
1 Fundamentals of programming

A programming paradigm is a way to classify a programming style or

Paradigm A particular style
approach, based on its features and its approach. For Component 1 you are
or approach to designing a
expected to have studied:
solution to a problem.
✚ procedural-oriented programming
Procedural-oriented
✚ object-oriented programming.
programming An
Each paradigm is discussed in more detail below, and it is expected that you approach to solving a
will have had practice programming using both paradigms throughout your problem using subroutines
preparation for the exam. to tackle smaller sub-
problems.
Exam tip Object-oriented
programming An
Study the skeleton program carefully to help predict what kinds of questions you might
approach to solving a
be asked. If the skeleton program uses object-oriented techniques then it is more likely
problem using a model
that any questions on this topic will appear in Section C and will refer to the skeleton
based on real-world objects
code and pre-release scenario. If the skeleton code doesn’t use object-oriented
that are designed to interact
techniques then it is more likely that questions on procedural-oriented programming
in a realistic way.
will appear in Section C, and Section A will include some more generic questions on
object-oriented programming.

Now test yourself

57 What is the meaning of the term programming paradigm?

58 What are the two programming paradigms that you should know?
59 Which paradigm uses a model based on real-world objects?

Answers available online

Procedural-oriented programming
Procedural-oriented programming is designed to allow programmers to use a
Making links
structured, top-down approach to solve a given problem.
For more on subroutines
The program designer uses decomposition to break the problem down into
see the section Purpose of
increasingly small sub-problems, each of which can then be solved using a
subroutines above on
subroutine (either a function or a procedure).
page 18.
The main program will then be constructed by calling subroutines which will,
in turn, call other subroutines in order to solve the original problem. Top-down approach
A method of planning
Data can be passed to subroutines and values can be returned from them,
solutions that starts with
allowing the different parts of the program to interact with each other.
the big picture and breaks
This type of computer program is generally simpler to understand and the it down into smaller sub-
subroutines can be re-used at different points in the program without needing problems.
to copy it. This type of program is very modular and can easily be updated by Decomposition A
changing one subroutine. method of solving a larger
problem by breaking it up
into smaller and smaller
Hierarchy charts problems until each
Procedural-oriented programs can be represented using hierarchy charts. problem can’t be broken
Simple hierarchy charts show the relationship between subroutines, such down any further.
as the chart shown in Figure 1.1, with lines used to connect subroutines Hierarchy charts A
wherever one subroutine calls another. diagram that shows which
subroutines call which other
subroutines. More complex
versions will also show what
22 data is passed and returned.

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_01_MRN_AQA_CS_001-033.indd Page 23 6/29/21 9:40 PM f-0116 /103/HO02206/work/indd

Exam tip
Program
Questions involving
hierarchy charts almost

1 Fundamentals of programming
exclusively appear in
Section C, referring to the
Initialise Input Process Output
skeleton program, and
usually involve a ‘fill the
gaps’ style of question.
Make your own hierarchy
Capture
Enter data Validate charts when studying
form
the skeleton code to help
you understand how the
subroutines fit together and
Figure 1.1 A simple hierarchy chart
see past papers for example
of this style of question.
Now test yourself

60 How is decomposition used when designing solutions using procedural-oriented

programming?
61 In a hierarchy chart, what is the meaning of a line connecting two subroutines?

Answers available online

Revision activity
✚ Open a complex program you have been working on in your lessons and create a
hierarchy chart to show the relationships between each subroutine.
✚ Consider the steps involved in a two-player game such as Rock-Paper-Scissors
or Noughts & Crosses. Create a hierarchy chart to show how the game could be
broken down into subroutines and how those subroutines would be related.

Object-oriented programming
Classes
Object-oriented programming (often referred to as OOP) uses a different
Class The definition of the
approach to programming. attributes and methods of a
The programmer thinks about real-world objects and creates a class to group of similar objects.
describe the attributes and methods for that type of object. For instance, in Attributes The properties
Figure 1.2, the class called Customer has: that an object of that type
✚ the attributes: Name, Address, and Date of birth has, implemented using
✚ the methods: Edit customer and Delete customer. variables.
Methods Processes or
Customer Account actions that an object
of that type can do,
Name
Account number implemented using
Address Attributes subroutines.
Balance
Date of birth
Edit customer Check balance
Methods
Delete customer Add interest
Figure 1.2 Classes containing attributes and methods

Attributes are implemented using variables to hold the relevant data.

Methods are implemented using subroutines to carry out each action.

23

My Revision Notes AQA A-level Computer Science Third Edition

325487_01_MRN_AQA_CS_001-033.indd Page 24 6/29/21 9:40 PM f-0116 /103/HO02206/work/indd

Objects Object A specific

When the program is run, objects are instantiated (created) based on the class instance of a class.
definitions. Once instantiated, the object is stored in a variable. Instantiation The
process of creating an
1 Fundamentals of programming

object based on its class

Note
definition.
The object is stored in memory and the variable stores the memory address of that Constructor A special
object so that it can be accessed. method (with the same
name as the class) that is
called when the object is
When an object is created, a special method called the constructor is
instantiated.
called, and this will initialise the object, typically setting the attributes and
sometimes calling other methods.
Each instance of an object will have its own values for each attribute, but
all instances of that object will be able to carry out the same methods. For
instance, in Figure 1.3:
✚ the object Account1 is an instance of the class Account, with Account
number 12345 and a balance of £7.63
✚ Account2 is a different instance of the class Account, with a different
account number and balance.
All objects of this type have the same attributes, however the value of the
attributes can vary from object to object.

Account1 Account2 Account3

Account number: Account number: Account number:
12345 73526 92649
Balance +7.63 Balance -3,107.12 Balance +11,836.01
Check balance Check balance Check balance
Add interest Add interest Add interest
Figure 1.3 Three unique instances of an Account object

Now test yourself

62 What two features are defined in a class?

63 Describe the relationship between an object and its class.
64 What method is called when an object is instantiated?
65 When two objects of the same class are instantiated, describe the features of each
object which will definitely be the same and which might be different.

Answers available online

Encapsulation
Deciding how to structure and organise classes can be difficult, but the main
Encapsulation The
rule is to group together objects with common characteristics (attributes) and
concept of grouping similar
behaviours (methods). Keeping these features together in one class is called
attributes, data, and
encapsulation.
methods together in one
Encapsulation is useful because it means that program code is more modular. object.
This means that the code is easier to debug, can be re-used more easily and Information hiding
teams of programmers can work on individual classes without needing to Controlling access to data
know how other classes are programmed – they only need to know what stored within an object.
methods can be accessed.
Encapsulation is also helpful as it allows for information hiding, in which the
data stored in the attributes can be kept within that object and access to that
data can be controlled.

24

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_01_MRN_AQA_CS_001-033.indd Page 25 6/29/21 9:40 PM f-0116 /103/HO02206/work/indd

Access specifiers Access specifier (or

Each attribute (variable) and method (subroutine) within a class has an access modifier) A
access specifier (also commonly known as an access modifier) which controls keyword that sets the
whether other objects are able to directly interact with it. accessibility of an attribute

1 Fundamentals of programming
or method.
Access specifier Symbol Attributes Methods
Public An access specifier
public + can be viewed or updated can be called by any that allows that attribute or
by any object of any class object of any class method to be accessed by
private - cannot be viewed or cannot be called by any any other object.
updated by any other other object, regardless Private An access specifier
object, regardless of class of class that protects that attribute
or method from being
protected # can only be viewed or can only be called by this
accessed by any other
updated by this object object or another object
object.
or another object of that of that class, or a subclass
Protected An access
class, or a subclass
specifier that protects that
The convention is to declare attributes as private so that other objects cannot attribute or method from
directly interact or affect the values that are stored. This make it less likely being accessed by other
objects unless they are
that a class written by another programmer could adversely affect the overall
instances of that class or a
program.
subclass.
To allow access to those data, a class should include getters and setters – Getter A function used
public methods that allow other objects to ask an object to return the value of to return the value of an
a specific attribute, or that ask an object to update a value. attribute to another object.
One example might be a game character that has a score and a number of Setter A procedure used to
lives which are both set to private, and uses public methods to allow other allow another object to set
objects to interact with those values: or update the value of an
attribute.
Character = Class
Private:
Revision activity
Score: Int
✚ Using pseudo-code,
Lives: Int describe the class
Public: attributes and methods
required for a virtual pet.
Function GetScore() ✚ Design and build
Function GetLives() an object-oriented
program that uses a
Procedure AddPoints() Calculator class
Procedure LoseALife() with methods such as
Press0, Press1,
Procedure GainExtraLife() PressPlus and
End Class PressEquals.

Now test yourself

66 Identify three things that are grouped together using encapsulation.

67 Suggest two advantages of using encapsulation.
68 State the name, symbol and meaning for the three main access specifiers.
69 Which access specifier should be used, by default, for attributes? Inheritance The idea
that one class can use the
70 Explain how it is possible for private attributes or methods to be accessed by other
attributes and methods
objects.
defined in another class.
Answers available online Subclass (or child class)
A class that inherits the
attributes and methods
Inheritance from another class.
Inheritance describes an ‘is a’ relationship between classes. For example, a Base class (or parent
horse is a type of animal. class) A class whose
attributes and methods are
To implement this, subclasses (or child classes) inherit attributes and inherited by another class. 25
methods from a base class (or parent class).

My Revision Notes AQA A-level Computer Science Third Edition

325487_01_MRN_AQA_CS_001-033.indd Page 26 6/29/21 9:40 PM f-0116 /103/HO02206/work/indd

This means that the same code can be re-used without needing to be copied,
Revision activity
and if the code needs to be changed then it only needs to be changed in the
base class, reducing the risk of errors and making it easier to debug. ✚ Design a base
class, Room, and
In class diagrams, inheritance is showing using a hollow arrow, which always
1 Fundamentals of programming

subclasses Kitchen,
points towards the base class.
LivingRoom and
Bedroom. Consider
Account
the attributes that could
be inherited and those
Current Mortgage
which must be declared
in the subclasses.
✚ Design classes that
Figure 1.4 An inheritance diagram
could be used to model
Figure 1.4 shows a class structure in which two types of account (Current and various forms of public
Mortgage) are subclasses will inherit the attributes and methods from the transport, including
Account base class. buses, taxis and trains.

Exam tip
Exam tip
In AQA Pseudo-code, subclasses are declared by adding the base class name in
parentheses; for example: The terms subclass
Dog = Class (Animal) and child class are
It is rare for the exam to ask students to describe a class using pseudo-code. However, interchangeable and both
when this is the case, then language-specific syntax is accepted; for example: mean the same thing.
Dog = Class extends Animal The terms base class
and parent class are also
interchangeable. It is helpful
Subclasses will have additional properties and methods specific to that type
to either use the terms
of object (for example, a current account might have a withdrawal limit or the
subclass and base class
ability to allow an overdraft, whereas a mortgage account might have a fixed
OR child class and parent
end-date and the ability to allow a payment holiday).
class for consistency, but
Both subclasses will inherit the attributes and methods from the base class both sets of terminology are
(that is, they both have an account number and a balance and both have acceptable.
methods for checking the balance and adding interest).

Now test yourself

71 Suggest equivalent terms for child class and parent class.

72 What symbol is used to show an inheritance relationship on a class diagram, and Aggregation The idea that
which direction does it point? one class can be related
to another using a ‘has a’
73 Suggest two advantages of using inheritance.
relationship.
Answers available online Container class (or
container object) A class
or object that contains, or is
Aggregation made up of, other objects.
Aggregation describes a ‘has a’ relationship between classes; for example, a Associated class (or
building has a room. associated object) A class
In the implementation of aggregation, one container object contains a or object that is used to
reference to an associated object as an attribute (variable). make up another object.
Association aggregation
There are two types of aggregation: A type of aggregation
✚ association aggregation where the container object
✚ inheritance aggregation. can be destroyed without
destroying its associated
Association aggregation objects.
In association aggregation the two types of object are associated with each Composition aggregation
other, but not dependent on each other. A type of aggregation
If the container object is removed or destroyed, the associated objects are where, if the container
undamaged. In a real-world example, employers and managers in a business object is destroyed, its
all have a job role. A business that is changing the way it works may decide associated objects are also
a certain role is no longer needed. However removing the job role does not destroyed.
26 remove the employees and managers – they are still part of the business.

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_01_MRN_AQA_CS_001-033.indd Page 27 6/29/21 9:40 PM f-0116 /103/HO02206/work/indd

In class diagrams, association aggregation is shown using a hollow diamond,

which always points towards the container class.

Job Role

1 Fundamentals of programming
Manager Employee

Figure 1.5 Class diagram showing association aggregation

Composition aggregation
Composition aggregation describes a more dependant relationship in which
the container object is directly made up (composed) of the associated objects.
If the container object is removed or destroyed then so are the associated
objects. In our example the container object Workforce is made up of
managers and employers. Removing Workforce from the model altogether
results in the removal of the employees and managers.
In class diagrams, composition aggregation is shown using a filled diamond,
which always points towards the container class.

Workforce

Manager Employee

Figure 1.6 Class diagram showing composition aggregation

Now test yourself

74 What is the difference between inheritance and aggregation?

75 Describe the difference between association aggregation and composition
aggregation.
76 State which symbols are used to show each type of aggregation on a class diagram.
77 In a computer model featuring a Room class, a Window class and a Door class,
which class is the container class?

Answers available online

Polymorphism
Polymorphism means ‘many forms’. It covers the potential scenario of having Polymorphism Literally
two or more methods with the same name. ‘many forms’ – the ability for
✚ A method to ring(PhoneNumber) would involve dialling that specific two methods with the same
number. name to carry out their
✚ A method to ring(Name) would involve looking up the phone number for actions using different code.
the person you want to ring and then ringing that number. Overriding A method in
a subclass re-defining a
Both methods have the same name, but different parameters. By looking
method inherited from a
at the data type of the value that is passed, the object can identify which
base class.
method to use.
Having methods with the same names but different parameters means that
the same basic goal can be achieved using different steps, or methods, and is
Exam tip
a common feature in object-oriented programming. Look carefully through
the skeleton code for any
Overriding examples of polymorphism.
If there are any then this is a
Overriding is the situation where a base class has a method and a subclass
likely question to appear in
has a method of the same name, but with different steps. The subclass Section C.
method overrides (takes priority over) the method from the base class.
For example, an Account class can have a method CloseAccount that
transfers the remaining balance to the customer.
27

My Revision Notes AQA A-level Computer Science Third Edition

325487_01_MRN_AQA_CS_001-033.indd Page 28 6/29/21 9:40 PM f-0116 /103/HO02206/work/indd

A Mortgage class is a subclass and inherits the CloseAccount method, but

in this case the CloseAccount method needs to be re-written so that the
account can only be closed if the mortgage has been paid off.
1 Fundamentals of programming

Now test yourself

78 Explain what is meant by the term polymorphism.

79 Explain what is meant by the term overriding.
80 How can a computer identify which method to run if there are two methods with
identical identifiers in that class?

Answers available online

Revision activity
✚ Create flashcards for each piece of key vocabulary in this chapter.
✚ Examine the skeleton program for your Component 1 exam, or a previous skeleton
program, and identify as many examples of instantiation, inheritance, aggregation,
polymorphism and overriding as possible.

Why object-oriented programming is used

There are several advantages for choosing object-oriented programming over
procedural-oriented programming.
✚ OOP more closely represents the real world, which is made up of objects of
various types.
✚ Program code is easier to debug as each class is modular.
✚ Classes can be re-used in other programs.
✚ Code can be re-used through inheritance and aggregation.

Object-oriented design principles

There are three main design principles you are expected to be familiar with:
✚ encapsulate what varies
✚ favour composition over inheritance
✚ program to interfaces, not implementation.
Encapsulate what varies
This means that in the design stage a programmer should try to identify the
features of a model that are likely to change in the future and encapsulate
those features.
One example is a method designed to display the number of wheels that a
vehicle has, using a selection statement:
IF Vehicle.Type = "bicycle" THEN
OUTPUT "Vehicle has 2 wheels"
ELSE IF Vehicle.Type = "car" THEN
OUTPUT "Vehicle has 4 wheels"
ELSE IF Vehicle.Type = "boat" THEN
OUTPUT "Vehicle has 0 wheels"
Making links
ENDIF
Object-oriented design
A better design would use encapsulation to store the number of wheels as principles go alongside
part of a vehicle class: software development
principles, covered in
OUTPUT "Vehicle has " + Vehicle.NumberOfWheels + " wheels"
Chapter 13. Clear definition
This means that if the program is changed to add a tricycle vehicle type, there and analysis of a problem
is no need to change the existing code. will directly inform the
object-oriented code
The use of inheritance can also make it much easier to encapsulate features
design.
28 or behaviours that vary, using the technique of overriding within subclasses.

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_01_MRN_AQA_CS_001-033.indd Page 29 6/29/21 9:40 PM f-0116 /103/HO02206/work/indd

Favour composition over inheritance

Composition aggregation and association aggregation provide a more flexible
relationship between classes than inheritance.
This is advantageous because each class can be tested independently of other

1 Fundamentals of programming
classes, which is not true for subclasses.
Changing a base class can cause unexpected side-effects in subclasses,
so using composition rather than inheritance means it is safer to update
program code in the future.
A particular problem with inheritance is that a subclass can’t inherit from two
different classes at once. This means that if a new behaviour is common to
two different classes then it may not be easier to re-use code without copying
it into both classes.

Program to interfaces, not implementation

Interfaces are a special concept which can be used to group similar Interface A collection of
behaviours together and make them available to unrelated classes that have subroutines that can be
access to that interface. implemented by unrelated
classes.
For example, a Dog class might have access to an interface called Annoy,
which will call the methods Bark and ChewSlippers. A Person class might
also have access to the interface called Annoy, and in this class it will call the
methods TalkLoudly and PickNose.
By grouping both sets of behaviours into one interface, a programmer can
make either type of object ‘annoy’ without concerning themselves with the
actual implementation of how each unrelated class will carry out that process.
This technique uses a mixture of encapsulation (grouping similar behaviours
together) and abstraction (hiding the complexity of what is happening).

Now test yourself

81 Describe three advantages of using object-oriented programming.

82 Explain why composition is preferable to inheritance.
83 Which design principle involves the grouping together of similar behaviours with
unrelated classes?
84 Which design principle involves the grouping together attributes or methods that
are likely to be changed in the future?

Answers available online

Writing object-oriented programs

It is important to be familiar with the syntax of your chosen programming
language.
For each of the following, these techniques may be present in the skeleton
program and you may be required to make use of them, but you are unlikely
to be asked direct questions about them.
Abstract method A
Abstract methods are declared but contain no program code. These are
method that has no
sometimes used in base classes to create the requirement to override a
program code and is
method. An example may be a Speak method in an Animal class that must
intended to be overridden.
be overridden for each of the Dog, Cat and Parrot subclasses.
Virtual method A method
Virtual methods are methods that may be overridden. In some languages that may be overridden
(Java and Python) all methods are virtual by default. In other languages (C#, (this is the default in many
Delphi/Pascal and Visual Basic) methods must be declared as virtual in the languages).
base class if they are to be overridden. Static method A method
within a class that can be
Static methods are methods that can be called within a class without the need
called without the need
to create an object of that class. This means less memory usage as the program
to create an object of that
doesn’t need to create the object in order to be able to run those methods and class. 29
should be used when any object should be able to make use of those methods.

My Revision Notes AQA A-level Computer Science Third Edition

325487_01_MRN_AQA_CS_001-033.indd Page 30 6/29/21 9:40 PM f-0116 /103/HO02206/work/indd

Exam tip Exam tip

Study the skeleton program carefully and make sure you can identify any examples It is easy to forget the
of each concept and technique mentioned in this chapter. It is likely that a question exact meaning of each
1 Fundamentals of programming

on that concept will appear in Section C and it is possible you will be expected to use relationship symbol. Make
those techniques when writing your own program code for Section D. sure you remember the
following.
✚ Inheritance is shown
Class diagrams using a hollow arrow,
The design of classes can be described using a class diagram, using arrows which always points
and diamonds to describe the relationships towards the base class.
✚ Association aggregation
Simple class diagrams just show the names of classes and the relationships
is shown using a hollow
between them.
diamond, which always
More detailed class diagrams show the attributes and methods within each points towards the
class as well. container class.
✚ Composition
These diagrams show the name of the class in the top section, attributes in aggregation is shown
the middle section and methods in the bottom section, along with access using a filled diamond,
specifiers. which always points
towards the container
Base class or parent class
class.
Account
– AccountNumber: String
– OpeningDate: Date
– CurrentBalance: Currency
– InterestRate: Real
+ GetAccountNumber()
+ GetCurrentBalance()
# AddInterest()
+ SetInterestRate

Revision activity
✚ Create a class diagram
Current Mortgage
that could be used
– PaymentType: String to model a house,
– Overdraft: Boolean – EndDate: Date including a House
Subclasses or container class and
+ SetPaymentType() + GetEndDate() associated classes that
child classes
+ SetOverdraft() + SetEndDate() include Kitchen,
LivingRoom,
+ GetOverdraft() Bedroom and
Garden. Consider the
Figure 1.7 A class diagram showing the design for different types of account
relationship between
Note the way that attributes are described using Identifier: Data Type. each class.
✚ Examine the skeleton
Now test yourself program for your
Component 1 exam,
85 What type of method can be called even if an object of that class has not been or a previous skeleton
instantiated? program, and create a
86 What key word means that a method can be overridden? detailed class diagram
87 What type of method must be overridden? to show attributes and
methods in each class as
88 Identify six things should be included in a detailed class diagram.
well as the relationships
Answers available online between them.

30

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_01_MRN_AQA_CS_001-033.indd Page 31 6/29/21 9:40 PM f-0116 /103/HO02206/work/indd

Summary

Programming ✚ In object-oriented programming the programmer

✚ Each variable has an address in memory, an identifier, designs a program around the attributes and

1 Fundamentals of programming
a data type and a value. It is important to choose a behaviours of real-world objects
meaningful identifier ✚ A class contains the attributes (variables) and methods
✚ Named constants are variables whose values cannot (subroutines) associated with that type of object and
be changed while the program is running functions as a blueprint for how objects of that class
✚ Selection is used to decide which block of code to will behave
execute and is implemented using an IF or SWITCH ✚ Objects are instances of a specific class and
statement instantiated by calling their constructor
✚ Iteration is used to repeat a block of code and either ✚ The process of grouping objects with common
uses a count-controlled (FOR) loop if the number of attributes and behaviours together is called
repetitions is known or a condition-controlled (WHILE encapsulation
or DO-WHILE) loop if the number of repetitions is not ✚ Access to an object’s attributes and methods is
known controlled using access specifiers
✚ Arithmetic operators include basic arithmetic, plus ✚ It is common for all attributes to be declared as private,
rounding, truncation and the use of DIV and MOD and for setter and getter methods to be declared as
✚ Relational operators can be combined with Boolean public in order to allow other objects to interact in a
operators to create more complex conditions controlled manner
✚ String handling operations include the basic skills of ✚ Inheritance describes an ‘is a’ relationship. A subclass
finding the length, addressing specific positions and will inherit the attributes and methods of a base class
concatenation (joining) of strings, as well as extracting ✚ Aggregation describes a ‘has a’ relationship. A
substrings, converting to and from the numeric container class will be linked to one or more associated
character codes and converting data types classes
✚ It is important to be able to use random number ✚ In association aggregation, if the container class
generation and exception handling in your is destroyed then the associated classes will be
programming untouched
✚ Exception handling is an important tool when a block of ✚ In composition aggregation, if the container class
code has a chance of failure (for example, opening a file is destroyed then the associated classes will be
that may not exist, or processing user inputs that may destroyed as well.
be in the wrong format) ✚ Polymorphism is the term for using having two or more
✚ Subroutines allow for code to be re-used. Parameters methods with the same identifier in the same class,
describe the data that must be passed to a subroutine but using different parameters to allow the program to
when it is called decide which method to run
✚ Functions are subroutines that return a value. ✚ Overriding is the term for a subclass having a method
Procedures are subroutines that do not return a value with the same identifier as a method in the base class;
✚ Local variables are preferable to global variables the method in the subclass will always take precedence
in most cases as they use less memory, make ✚ There are three design principles to remember:
subroutines more modular and make programs easier ‘Encapsulate what varies’, ‘Favour composition
to read and debug. over inheritance’ and ‘Program to interfaces, not
✚ Stack frames contain the return address, parameters, implementation’
and local variables of a subroutine call while it is ✚ You should not be asked about abstract, virtual and
running static methods but these may appear in the skeleton
✚ A recursive function is a function that calls itself. It code
must include at least one base case, a point at which ✚ Class diagrams should always show the name of the
the function will stop calling itself class, the attributes for that class (including data types
and access modifiers) and the methods for that class
Programming paradigms (including access modifiers)
✚ There are two main programming paradigms to study – ✚ Inheritance is shown using a hollow arrow that points
procedural-oriented programming and object-oriented towards the base class
programming ✚ Association aggregation is shown using a hollow diamond
✚ In procedural-oriented programming the main problem that points towards the container class
is broken down into smaller sub-problems, each of ✚ Composition aggregation is shown using a filled diamond
which is solved using a subroutine that points towards the container class
✚ Hierarchy charts are used to show which subroutines
call other subroutines

31

My Revision Notes AQA A-level Computer Science Third Edition

325487_01_MRN_AQA_CS_001-033.indd Page 32 6/29/21 9:40 PM f-0116 /103/HO02206/work/indd

Exam practice

1 Dave has been asked to write a program as part of a project looking at rainfall. At the end of a week the user of the
program will enter the total rainfall (measured in mm) for each one of the last seven days as a number with one
1 Fundamentals of programming

decimal place. The program should calculate and then output:

✚ The amount of rain that fell on the day with the least rainfall.
✚ The amount of rain that fell on the day with the most rainfall.
✚ The total amount of rainfall for that week.
✚ The mean average amount of rainfall for that week.
a) An iterative structure should be used in the program described above. Explain whether a count-controlled or
condition-controlled iteration structure would be most suitable, and why. [2]
b) Write a program using your chosen programming language that will solve the problem described above. [10]
c) Show the output of the program when the following values are input. [1]
2.2 7.6 1.4 3.1 2.7 1.5 3.2
2 In the following table, the first column contains pseudo-code for an algorithm that will check to see if a piece of text is
a palindrome (a word or phrase that reads the same forwards or backwards).
The second column contains pseudo-code for a recursive algorithm that will carry out the check.
The subroutine Length() will return the number of characters in a given string.
The subroutine Substring(start,end) will return the string starting from position start and
ending at position end. For example "Message".Substring(2,5) will return "ssag".

FUNCTION Palindrome()
OUTPUT "Enter a word or phrase"
Phrase ← INPUT
Result ← PCheck(Phrase)
IF Result = TRUE THEN
OUTPUT "This is a palindrome"
ELSE
OUTPUT "This is not a palindrome"
ENDIF
ENDFUNCTION

FUNCTION PCheck(Phrase)
NumChars = Length(Phrase)
IF NumChars < 2 THEN
RETURN TRUE
ELSEIF Phrase[0] = Phrase[NumChars-1] THEN
RETURN PCheck(Phrase.Substring(1,NumChars-2))
ELSE
RETURN FALSE
ENDFUNCTION

a) Explain what is meant by a recursive algorithm. [1]

b) Explain what is meant by a base case, and describe one base case from the subroutine PCheck. [2]
c) Write a program in your chosen language that will ask the user to enter a word or phrase and check if it is a
palindrome without using recursion. [7]
d) Show the output of the program when the word “RACECAR” is entered. [1]
e) Show the output of the program when the word “BEAR” is entered. [1]

32

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_01_MRN_AQA_CS_001-033.indd Page 33 6/29/21 9:40 PM f-0116 /103/HO02206/work/indd

3 This class diagram is a partial representation of the relationships between some of the classes in a program for a
garage.
Vehicle

1 Fundamentals of programming
- Registration: String
- Colour: String
- New: Boolean
- Owner: Owner
+ GetDetails()
# ChangeColour()
+ ChangeOwner()

Person Car Van

- Name: String - NumberOfDoors: Integer - WheelBase: String
- Email: String
+ GetDetails() + GetDetails()
+ GetDetails() # ChangeColour()
+ UpdateEmail()

a) State the identifier of a base class. [1]

b) State the identifier of a subclass. [1]
c) State the access specifier for the attribute Registration. [1]
d) State the identifier of a protected method. [1]
e) The method ChangeColour appears in both the Vehicle class and the Van class. State the name of
the concept that allows this to happen. [1]
f) Identify the relationship between the Car class and the Vehicle class. [1]
g) Identify the relationship between the Person class and the Vehicle class. [1]
4 a) Explain the difference between a public method and a protected method. [2]
b) Give two reasons why most attributes in object-oriented programming are declared as private and
methods such as GetDetails or UpdateScore are used to access them. [2]
c) Explain the meaning of the phrase favour composition over inheritance. [2]
5 A program is used to store and process the details of students in a college. The Student class contains several
attributes and methods, including two methods named Student.
When a new student is added to the program then one of these methods must be run.
PRIVATE PROCEDURE Student ()
// Some code
ENDPROCEDURE
PROCEDURE Student (Name, DoB, Email)
// Some code
ENDPROCEDURE
a) State the process for creating a new object based on its class. [1]
b) State the name for a method that must be run when an object of that class is created. [1]
c) State the name for the concept that allows there to be two methods with identical identifiers in one class. [1]
d) Explain how it is possible for the computer program to decide which method to run when the method
Student is called. [3]

33

My Revision Notes AQA A-level Computer Science Third Edition

325487_02_MRN_AQA_CS_034-052.indd Page 34 6/29/21 10:01 PM f-0116 /103/HO02206/work/indd

2 Fundamentals of data structures

Data structures and abstract

data types
Data structures
Data structures are designed to hold several items of data at a time. This
Data structure The
makes it easier to manage, find, update and process data than to work with
concept of storing multiple
individual variables. values together with a
For instance, finding the biggest value from ten separate variables would single identifier.
require a separate selection statement to interrogate each one, and additional Array A data structure for
code would be needed each time the number of variables increased. holding values in a table.
Using a data structure called an array means that it is possible to step through
each item using a loop, reducing the amount of code and making the code
much more flexible .

Single- and multi-dimensional arrays

(or equivalent)
An array is a fixed-size collection of values, all of the same data type.
Index A value indicating the
Single-dimensional arrays position of a value with in
A single-dimensional array can be written down as a single row of a table. an array or list.

Each individual item in the array is referred to using the identifier (or List A data structure similar
variable name) of the array, followed by the index in brackets (usually square to an array, commonly used
in Python in place of an
brackets). Each index position can hold data.
array.
Index [0] [1] [2] [3] [4]
Data 1 3 5 7 11 Exam tip
In most languages (including
Note C#, Delphi/Pascal, Java &
VB.Net) the array index
Arrays are always declared as a fixed size and cannot be changed later in the program. starts at 0, but be very
In an array, all data must be of the same data type. careful when reading
Lists are very similar to arrays, however it is possible to alter the size of a list while the exam questions as some
program is running and it is possible to store data of differing types in the same list. languages and some
scenarios will have the array
In some languages (notably including Python) it is much more common to use a list, index starting at 1.
whereas in others it is much more common to use an array.
You are not expected or required to show your understanding of the difference
between the two, and you should use whichever version is most suitable for your
chosen programming language.

Arrays are an effective solution for storing several values because a FOR loop
can be used to iterate (or loop) over an array.
In this example Numbers is the name of an array and Length(Numbers)
returns the size of the array (the number of items is contains).
FOR i ← 0 TO Length(Numbers)
OUTPUT Numbers[i]
34
ENDFOR

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_02_MRN_AQA_CS_034-052.indd Page 35 6/29/21 10:01 PM f-0116 /103/HO02206/work/indd

This code will step through the items in the array n times, where n is the size
Making links
of the array. This code would work for all arrays regardless of their size. This
is far more efficient than stepping through several variables one at a time, as One-dimensional arrays
this would require a separate line of code for each variable and the program can be an effective way

2 Fundamentals of data structures

would need to be edited every time the number of items changed. of representing a vector.
Vectors are used in many
games and physical models,
Multi-dimensional arrays and are discussed in more
A two-dimensional array can be written down as a table. detail towards the end of
A two-dimensional array can be thought of as an array of one-dimensional this chapter.
arrays.
By identifying which row is being addressed using one array index, a
particular value can be addressed within that row using a second array index.
Numbers [0]

[0] [1] [2] [3]

[0] 1 3 5 7 Numbers [0][3]

Numbers [1]
Numbers
[0] [1] [2] [3]
[0] [1] [2] [3]
[1] 11 13 17 19
[0] 1 3 5 7
Exam tip
[1] 11 13 17 19 Numbers [2]
[2] 23 29 31 37
Arrays can also be defined
[0] [1] [2] [3]
using the first index to find
[3] 41 43 47 51 [2] 23 29 31 37
the column and the second
Numbers [2][2] index to find the item in that
Numbers [3]
column. Always read the
[0] [1] [2] [3] question carefully as any
[3] 41 43 47 51 ambiguities should be made
Numbers [3][1] clear.
Figure 2.1 A two-dimensional array called Numbers
Making links
In the example above the contents of Numbers[0][3] can be found by isolating
row 0 and then finding the item in position 3. One common use for a
two-dimensional array is
A two-dimensional array is useful because it can store more complex data; for
storing a matrix. Matrices
example, the scores for different students or values for different days.
are often used in complex
It is possible to find the size of a two-dimensional array using the Length maths, and can also be used
function (exact syntax varies by programming language). to represent a graph data
structure. This is explored
For an array with indexes defined as [row][column], to find the number of
further later in this chapter.
rows, find the number of one-dimensional arrays the table can be split into:
NumberOfRows ← Length(ArrayName)
Matrix A rectangular, two-
To find the number of columns, find the length of a row in the table: dimensional collection of
NumberOfColumns ← Length(ArrayName[0]) values.

To interrogate a two-dimensional array it is necessary to use nested loops. Nested A selection

statement or loop inside
FOR i ← 0 to Length(Numbers) // Step through each row another selection statement
FOR j ← 0 to Length(Numbers[i]) // Step through each or loop.
value in that row
OUTPUT Numbers[i][j] Exam tip

ENDFOR Make sure you are able to

hand trace a program such
ENDFOR as the one above to fully
Using the principle above, it is possible to think of a three-dimensional array understand which counter
as an array of two-dimensional arrays. variable is addressing the
row or column index of the
table. 35

My Revision Notes AQA A-level Computer Science Third Edition

325487_02_MRN_AQA_CS_034-052.indd Page 36 6/29/21 10:01 PM f-0116 /103/HO02206/work/indd

[0] [1] [2] [3]

Numbers [0] [0] 1 3 5 7
[1] 11 13 17 19 Numbers [0][1][3]
2 Fundamentals of data structures

[2] 23 29 31 37
[3] 41 43 47 51

[0] [1] [2] [3]

Numbers
Numbers [1] [0] 1 3 5 7
[0] [1] [2] [3]
[0] [1] [2] [3] [1] 11 13 17 19
[0] [1] [2] [3]
[0] 11[0] 33[1] 55[2] 77[3]
[0] [2] 23 29 31 37
[0] 11 33 55 77
[1][0] 11 13 17 19
[1] 11 13 17 19 [3] 41 43 47 51
[1] 11 13 17 19
[2][1] 23 11 29 13 31 17 37 19
[2] 23 29 31 37
[2] 23 29 31 37
[3][2] 41 23 43 29 47 31 51 37 [0] [1] [2] [3]
[3] 41 43 47 51
[3] 41 43 47 51 Numbers [2]
[3] 41 43 47 51 [0] 1 3 5 7
[1] 11 13 17 19
Numbers [2][1][2]
[2] 23 29 31 37
[3] 41 43 47 51

[0] [1] [2] [3]

Numbers [3] [0] 1 3 5 7
[1] 11 13 17 19
[2] 23 29 31 37
Numbers [3][3][1]
[3] 41 43 47 51

Figure 2.2 A three-dimensional array represented as an array of two-dimensional

arrays

Similarly, an array of n-dimensions is possible, where n is any positive integer.

Exam tip
While questions on two-dimensional arrays are quite common, three-dimensional
arrays are generally limited to potential skeleton programs and NEA programming
projects. Examine the skeleton code carefully for any examples of multi-dimensional
arrays.

Now test yourself

1 Explain the purpose of an index in an array

2 Describe two advantages for using an array of integers instead of several integer
variables.
3 True or false? All arrays start counting from 0.
4 What data structure is typically used instead of an array in Python?
5 True or false? It is not possible to have an array of more than four dimensions.
6 How many FOR loops need to be nested in order to interrogate a two-dimensional array?

Answers available online

Revision activity
✚ Create a two-dimensional array called MovieRatings. Populate the array with
film review ratings, with each row representing one review website and each
column representing one film.
✚ Create a menu and write program code that will allow the user to:
✚ find the min, max and mean average review score for each film
✚ find the min, max and mean average review score from each website
✚ find a specific review rating given the name of the film and the review website.
✚ Create a 2-dimensional array to store the game state for a game of noughts and
36 crosses.

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_02_MRN_AQA_CS_034-052.indd Page 37 6/29/21 10:01 PM f-0116 /103/HO02206/work/indd

Fields, records and files

A record holds groups of related items together, using fields. An example
might be a user record, containing fields to store the first name, last name Record A collection of

2 Fundamentals of data structures

and username for people logging onto a website. related data.
Field A category of data
First name Last name Username within a record or group of
Dave Smith Davo263 records.

Nicola Tandy NicTandy

Stephen Duffy DuffS17

Each item in a record can be a different data type, making a record more
complex than an array, but also more flexible.
There are many ways to implement a record, and this is handled differently in
different programming languages. Understanding and recognising records of
data and which fields are being used is the key concept.
One common concept when considering records is the importance of writing
records to a file.
Files are a way of permanently storing data which would otherwise be lost
File A persistent collection
once the program or subroutine has finished executing. of data, saved for access
Text files use a character set (such as ASCII or Unicode) to store the data as and accessible once the
text and typically use a delimiter such as a comma (,) or colon (:) to separate program or subroutine has
the individual items of data. These files are usually saved with a .txt or .csv finished running.
(comma separated value) file extension. Text file A file that uses
A saved file from a game of Noughts and Crosses might be saved like this as a text encoding (such as ASCII
.csv file, where a comma is used to separate each item of data. or Unicode) to store data.
Binary file A file that uses
X binary values to represent
each item of data.
O X

X O

-,X,-,-,O,X,-,X,O
An alternative is to save the game as a binary file. A binary file only stores
binary information, and needs a key to translate binary into particular items
of data. For instance, for the Noughts and Crosses game, using two bits per
cell, the following key could be used:

00 Empty

01 X

10 O

11 Not used Exam tip

Make sure you can identify
Then the saved game file would be stored as: the advantages and
disadvantages of using
000100 binary files versus text files.
001001 It is extremely unlikely that
you will be asked to write
000110 code to work directly with
The advantages of using a text file to store this data are that the data can be: binary files, though you
✚ read by other programs may choose to explore
✚ easily understood. this as part of your NEA
programming project. 37

My Revision Notes AQA A-level Computer Science Third Edition

325487_02_MRN_AQA_CS_034-052.indd Page 38 6/29/21 10:01 PM f-0116 /103/HO02206/work/indd

The advantages of using a binary file to store this data are:

✚ smaller file size
✚ data cannot be read by other programs (more secure).

Note
2 Fundamentals of data structures

For A-level it is only necessary to consider sequential file access. That is, reading files
from start to end, in order.

Exam tip
The syntax for writing to and reading from files varies significantly in different
programming languages. It is important to be familiar with file handling routines in
your chosen programming language and to look carefully for any file handling that
takes place in the skeleton program.

Now test yourself

7 What are the two types of file?

8 Describe one advantage and one disadvantage for using a binary file.
9 What is used to separate individual values in a .csv file?
10 Which of the following is correct: a field is made up of records, or a record is made
up of fields?

Answers available online

Revision activity
✚ Create a two-dimensional array to represent the state of a game of noughts and
crosses. Write program code that will:
✚ save the game state as a text file
✚ load a game state from a text file
✚ save the game state as a binary file
✚ load a game state as a binary file.
✚ Compare the file size of the text and binary files. Try opening both types of file in a
simple text editor (such as Notepad) and compare the contents.

Abstract data types/data structures

While most programming languages have built in data structures such as
arrays (or similar), more complex data structures are not always directly
supported.
Abstract data types can be implemented using a collection of existing tools,
typically involving the use of an array and one or more other variables. Each
one is discussed in more detail below. Abstract data type A
complex data structure in
They are referred to as abstract data types because the complexity of the data
which the complexity of
structure is hidden and a programmer can make use of them without needing
how the data is stored or
to be directly concerned with how they are implemented.
accessed is typically hidden
✚ Some data structures (including arrays) are static. That means that, once from the programmer.
declared, their size is fixed and cannot be increased or decreased.
✚ Static data structures are assigned a fixed amount of memory when Static The size of, and the
declared, making access to the values quick, however memory can be memory assigned to that
wasted if some spaces are left empty. data structure is fixed and
cannot be changed.
✚ Some data structures (including lists) are dynamic. That means that their
size is not fixed and they can grow and shrink as necessary. Dynamic The size of, and
the memory assigned to
Dynamic data structures are assigned memory as needed which uses
that data structure can
memory more efficiently, but accessing the data can be slower as the memory
change.
38 assigned to the data structure may become fragmented.

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_02_MRN_AQA_CS_034-052.indd Page 39 6/29/21 10:01 PM f-0116 /103/HO02206/work/indd

Queues
A queue data structure works in a very similar way to a queue in the real

2 Fundamentals of data structures

world.
Queue A data structure
Items are retrieved from the front of the queue and new items are added at in which items are stored
the rear of the queue. in the order in which they
The term for this type of data structure is First In, First Out (FIFO). should be accessed.

Common uses for queues include: First In, First Out (FIFO)
✚ buffering (storing data as it arrives until it can be processed) Those items placed into the
queue first will be the first
✚ simulating a card game (cards are drawn from the front and replaced at
ones to be accessed.
the back).
Pointer A value that stores
Making links an address. In the context
of queues this is usually the
Queues are a fundamental part of the breadth first search (BFS) which is explored in index of the front or rear
Chapter 3. item.

A queue can be implemented using a single-dimensional array and two

integer variables, referred to as the Front Pointer (FP) and the Rear Pointer (RP).

FP RP
[0] [1] [2] [3] [4] [5] FP 0
Dave Angelina Faaris RP 2

There are several actions that can be carried out on a queue:

Enqueue add new data to the rear of the queue Linear queue A queue in
Dequeue remove one item from the front of the queue which data is stored in a
line.
Peek (or Top) look at, but don’t remove, one item from the front of the queue
IsEmpty check to see if the queue is empty Circular queue A queue
which wraps around in a
IsFull check to see if the queue is full circle. If implemented using
an array, the last index is
The IsFull and IsEmpty actions are necessary because a program will crash
followed by the first index.
if there is an attempt to enqueue data when the queue is already full or to
dequeue data from an empty queue. Priority queue A queue
which stores a priority for
There are three types of queue: each value so that the items
✚ Linear queue with the highest priority can
✚ Circular queue be accessed first.
✚ Priority queue

Implementing a linear queue

In a linear queue a front pointer is not needed.

RP
[0] [1] [2] [3] [4] [5]
Dave Angelina Faaris Kev RP 3

Items are retrieved from the front of the queue by moving each item forward
in the array:

RP
[0] [1] [2] [3] [4] [5]
Angelina Faaris Kev RP 2

This can be inefficient, especially for large queues, as moving each item in the
array takes time. 39

My Revision Notes AQA A-level Computer Science Third Edition

325487_02_MRN_AQA_CS_034-052.indd Page 40 6/29/21 10:01 PM f-0116 /103/HO02206/work/indd

Implementing a circular queue

In a circular queue a front pointer is needed.
The front pointer is moved whenever an item is dequeued:
2 Fundamentals of data structures

FP RP
[0] [1] [2] [3] [4] [5] FP 1
Dave Angelina Faaris Kev RP 3

Note
In a circular queue there is no need to delete the value from the array, only to move
the pointers. The data that has been dequeued will eventually be overwritten when
the rear pointer comes back around.

When a pointer reaches the end of the array, it wraps around back to the start.
For example, given the following state of a circular queue:

FP RP
[0] [1] [2] [3] [4] [5] FP 2
Dave Angelina Faaris Kev RP 5

When a new item is enqueued, the rear pointer will wrap around to 0.

RP FP
[0] [1] [2] [3] [4] [5] FP 2
Aadya Dave Angelina Faaris Kev RP 0

Note
Exact implementations vary slightly, for example some implementations have the rear
pointer pointing to the first empty space rather than the last full space.

A circular queue is much more efficient that a linear queue when the queue is
large as it avoids the need to move each item each time a value is dequeued.
A circular queue is also a more complex data structure and a linear queue
may be more appropriate for a smaller queue.

Implementing a priority queue

A priority queue enqueues items to the rear of the queue, but also records a
priority level.
When an item is enqueued the new item's priority will be compared to each
item in the list, starting at the back, in order to find the correct position to
insert the item. This means that the items with the highest priority will
always be dequeued first.
In the example below, a priority queue is used to store the inputs to a
processing system on an aircraft. The bottom row represents the priority
level, where a higher number means a higher priority.

FP RP
[0] [1] [2] [3] [4] [5]
Engine Steering Calculate
Warning Input Position
5 4 3
40

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_02_MRN_AQA_CS_034-052.indd Page 41 6/29/21 10:01 PM f-0116 /103/HO02206/work/indd

A new item, Rudder Input, is added to the queue, with priority 4.

Starting at the rear of the queue, the priority is compared to Calculate
Position. Because the priority is higher, this item is moved back one space.

2 Fundamentals of data structures

The priority is then compared to Steering Input. Because the priority is the
same (or higher), the new item is inserted into the list behind it.

FP RP
[0] [1] [2] [3] [4] [5]
Engine Steering Rudder Calculate
Warning Input Input Position
5 4 4 3

Now test yourself Exam tip

11 What does the acronym FIFO stand for? Questions relating to

queues may focus on
12 What is the meaning of the term enqueue?
hand-tracing algorithms,
13 Suggest two uses for a queue data structure
discussing the
14 Which type of queue will enqueue items in a position based on their importance? circumstances in which
15 What pointers are required for a circular queue? different types of queues
16 Explain why the method IsFull is required when designing a queue data structure. are needed (Section A),
suggesting what changes
Answers available online may be needed to adapt
one type of queue into
another (Section C) or
Revision activity making changes to an
✚ Create a Queue class in your chosen programming language. Include the following existing implementation
methods. (Section D).
✚ IsFull
✚ IsEmpty
✚ Enqueue
✚ Dequeue
✚ Peek
✚ Implement a linear queue, a circular queue and a priority queue.

Stacks
A stack is a Last In, First Out (LIFO) data structure. Stack A data structure in
Much as with a stack of plates, new items are placed at the top of the stack which items are added to
and items are also removed from the top of the stack. the top and removed from
the top, much like a stack of
Common uses for stacks include: plates.
✚ reversing a list of values
Last In, First Out (LIFO)
✚ performing undo operations
Those items placed into the
✚ as a call stack for keeping track of subroutine calls.
stack most recently will be
A stack can be implemented using a single-dimensional array and a single the first ones to be accessed.
integer variable to point at the top of the stack.

[5] Top 2
[4]
[3]
[2] Change font Top
[1] Make bold
[0] Type

41

My Revision Notes AQA A-level Computer Science Third Edition

325487_02_MRN_AQA_CS_034-052.indd Page 42 6/29/21 10:01 PM f-0116 /103/HO02206/work/indd

There are several actions that can be carried out on a queue:

Making links
Push add new data to the top of the stack Stacks are a fundamental
Pop remove one item from the top of the stack part of the depth first search
2 Fundamentals of data structures

(DFS) which is explored in

Peek (or Top) look at, but don’t remove, one item from the top of the stack
Chapter 3.
IsEmpty check to see if the stack is empty
IsFull check to see if the stack is full

The IsFull and IsEmpty actions are necessary because a program will crash if
there is an attempt to push data when the stack is already full or to pop data
from an empty stack.
When a value is pushed onto the stack it is placed at the top position and the
value of the top pointer is incremented.

[5] Top 3
[4]
[3] Delete selection
[2] Change font Top
[1] Make bold
[0] Type

When a value is popped from the stack the value at the top position is
returned and the top pointer is decremented.

[5] Top 2
[4]
[3] Delete selection
[2] Change font Top
[1] Make bold
[0] Type

Now test yourself Note

17 State the acronym used to classify a stack data structure, and state the words this In a stack there is no need
stands for. to delete the value from
18 Describe two potential uses for a stack. the array, only to move the
pointer. The data that has
19 What error could occur when trying to Push an item? State how this situation could
been popped will eventually
be dealt with.
be overwritten when new
20 Describe the difference between Pop and Peek. data is pushed and the top
pointer moves back up.
Answers available online

Revision activity
✚ Create a Stack class in your chosen programming language. Include the following
methods:
✚ IsFull
✚ IsEmpty
✚ Push
✚ Pop
✚ Peek
✚ Take a program that you have written that makes use of subroutines and write down
the state of the call stack each time a new subroutine is called or a value returned.
✚ Repeat the first bullet for a recursive program that you have written.

42

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_02_MRN_AQA_CS_034-052.indd Page 43 6/29/21 10:01 PM f-0116 /103/HO02206/work/indd

Graphs
A graph is a data structure designed to represent more complex
Graph A data structure
relationships.

2 Fundamentals of data structures

designed to represent the
Typical uses for graphs include: relationships between items.
✚ route finding Node Used to represent an
✚ representing relationships in human networks (such as in social networks) item of data, drawn as a circle.
✚ computer networks Also known as a vertex.
✚ project management
Edge A line used to
✚ game theory (mathematical modelling of strategic decisions).
represent the relationship
Each graph is made up of nodes (or vertices) and edges (or arcs). between two nodes. Also
known as an arc.
There are several types of graph.

Types of graph
Unweighted and undirected graphs
Town A Town B
A simple graph (unweighted and undirected) shows how two or more items
are connected.
Figure 2.3 An unweighted,
undirected graph
Weighted graphs
A weighted graph is used where the connection between the two nodes has a Weighted graph A graph in
cost or value (such as a distance). which each edge has a value
or cost associated with it.
30
Town A Town B Directed graph A graph
some edges can only be
Figure 2.4 A weighted graph traversed in one direction,
shown with an arrow head.

Directed graphs Disconnected graph A

graph in which two or more
In a directed graph, one or more edges has an arrow which shows that this
nodes are not connected to
edge can only be traversed in one direct.
the rest of the graph.

Charles Pauline
Exam tip
Harry Make sure you can identify
the different types of
graph. Be careful with the
Dave
Jack term ‘edge’. This doesn’t
mean the outskirts of the
graph, but the connections
Figure 2.5 A directed graph
between each node.
In a directed graph, not all edges need to be one-directional.
Now test yourself
Disconnected graphs 21 Explain the following terms.
A disconnected graph is one in which two or more vertices are not connected a Weighted graph
to the rest of the graph. b Directed graph
c Disconnected graph
A B
22 State two other terms
for a line or a connector
E
in a graph.
23 What symbol is only used
C
in a directed graph?
D 24 Suggest two possible uses
for a graph data structure.
Figure 2.6 A disconnected graph Answers available
online 43

My Revision Notes AQA A-level Computer Science Third Edition

325487_02_MRN_AQA_CS_034-052.indd Page 44 6/29/21 10:01 PM f-0116 /103/HO02206/work/indd

Adjacency lists and adjacency matrices

While graphs can be an effective visual way to represent a system for
humans, to store them in a computer system uses one of two methods: an
2 Fundamentals of data structures

adjacency list or adjacency matrix

An adjacency list uses a list for each node to store details of adjacent nodes. Adjacency list A method
Vertex Adjacent vertices of representing a graph
A B by listing, for each node,
A B,C
just the nodes that are
E B A,C,E connected to it directly.
C A,B,D
C D C,E
D
E B,D

Figure 2.7 An adjacency list for an unweighted graph

For a weighted graph, this list also includes the cost of travelling along
each edge.

20 Vertex Adjacent vertices

A B 25
A B,20,C,30
30 E B A,20,C,30,E,25
30
C A,30,B,30,D,35
40
C 35 D C,35,E,40
D
E B,25,D,40

Figure 2.8 An adjacency list for a weighted graph

This is efficient for sparse graphs, where the presence of edges does not need Sparse graph A graph with
to be tested often, because sparse graphs have very few edges and so not few edges.
much storage space is needed. However reading and processing each item in
an adjacency list can be inefficient if the list is large. Adjacency matrix A
method of representing
An adjacency matrix uses a grid to store details of adjacent nodes. a graph using a grid with
For an unweighted graph the adjacency matrix uses a binary number to show values in each cell to show
the presence or lack of an edge. which nodes are connected
to each other.
A B C D E
A B Dense graph A graph with
A 0 1 1 0 0 many edges.
E B 1 0 1 0 1
C 1 1 0 1 0
C D 0 0 1 0 1
D
E 0 1 0 1 0

Figure 2.9 An adjacency matrix for an unweighted, undirected graph

This is efficient for dense graphs, where the presence of edges needs to be
tested often, because the adjacency matrix is simple to step through. However
a lot of space can be wasted if the graph has few edges.
Undirected graphs will produce an adjacency matrix with a diagonal line of
symmetry.
Directed graphs will produce an adjacency matrix without symmetry, as not
all edges can be traversed in either direction.

44

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_02_MRN_AQA_CS_034-052.indd Page 45 6/29/21 10:01 PM f-0116 /103/HO02206/work/indd

A B C D E
A B
A 0 1 0 1 0
E B 0 0 0 0 1

2 Fundamentals of data structures

C 0 0 0 0 0
C D 0 0 1 0 1
D
E 0 0 0 0 0

Figure 2.10 An adjacency matrix for a directed graph

Weighted graphs will produce an adjacency matrix that stores the cost of each
edge.
Where nodes are not adjacent the cost of traversing is infinite.

20 A B C D E
A B 25
A ∞ 20 30 ∞ ∞
30 E B 20 ∞ 30 ∞ 25
30 Exam tips
C 30 30 ∞ 35 ∞
40 Make sure you are confident
C 35 D ∞ ∞ 35 ∞ 40 recalling the pros and
D
E ∞ 25 ∞ 40 ∞ cons of adjacency lists and
adjacency matrices, as this
Figure 2.11 An adjacency matrix for a weighted graph is a common question.
The number of edges is the main consideration when deciding whether an The difference between
adjacency list or an adjacency matrix is the most appropriate data structure a graph and a tree is that
for recording the details of a graph. a tree has no cycles. This
question comes up quite
Advantages Disadvantages often.
Adjacency list Good for sparse graphs (few Poor for dense graphs as it
edges) or situations where takes longer to process each
edges don’t need to be tested value.
often as it takes up less
storage space.
Adjacency matrix Good for dense graphs (many Poor for sparse graphs as it
edges) or situations where wastes storage space.
edges do need to be tested
often as it is easier to process
each value.

Now test yourself Revision activity

25 Which representation for a graph, when written down, looks most like a grid? ✚ Create a weighted graph
26 Which representation for a graph is most suitable for a sparse graph? showing the time taken
to travel between five
27 When else would that representation be most suitable?
towns near your home.
28 When would an adjacency matrix not be symmetrical? ✚ Create an adjacency
29 True or false? An adjacency list cannot be used for a weighted graph. matrix and an adjacency
list to represent the
Answers available online graph.

Trees
A tree is a special example of a graph which has no cycles (or loops). Tree A graph which has no
cycles (it is not possible to loop
Trees are always connected (no disconnected nodes) and undirected (edges back around to a previously
can be traversed in both directions). traversed part of the tree).
A rooted tree has one node as the root, or starting point, and all of the edges Rooted tree A tree with a
tend to lead away from the root. Rooted trees are commonly used to represent rooted node, from which all
a hierarchical structure, such as the chapters in this book. They are very edges leading away from
45
commonly used as binary search trees, as described below. that root.

My Revision Notes AQA A-level Computer Science Third Edition

325487_02_MRN_AQA_CS_034-052.indd Page 46 6/29/21 10:01 PM f-0116 /103/HO02206/work/indd

A-level Computer Science 2

1

3
4
2 Fundamentals of data structures

Component 1 Component 2

5
1 2 3 4
Programming Data Structures Algorithms Computation 6

Figure 2.13 A non-rooted tree

Stacks Graphs Trees Cat

Budgie Fish
Rooted Non-rooted

Figure 2.12 A rooted tree to represent chapters and topics in this book Rabbit

A non-rooted tree doesn’t have a clear start point and appears at a glance to
Figure 2.14 A binary search
be more like a graph, though with no cycles.
tree for storing pets in
A binary tree is a rooted tree in which each node has a maximum of two alphabetical order
children.
Cat
Binary trees consist of:

Root the node that is the starting point for the tree Budgie Fish
Branch a node that comes after the root and has one or more children
Leaf a node that does not have any children
Dog Rabbit

Binary search trees are especially useful for storing data in an ordered way
Figure 2.15 A binary search
and can be quickly and easily traversed to provide an ordered list.
tree with the addition of a
When adding data to a binary search tree, each new node is added according node for Dog
to its order.
In the example in Figure 2.14, where pets are stored alphabetically, to insert Binary tree A tree in which
the value Dog we would: each node has a maximum
✚ Compare to the root: Dog comes after Cat so follow the right edge of two children.
✚ Compare to the next node: Dog comes before Fish in the alphabet so we Root A node that has no
follow the left node parent nodes.
✚ If there is no child node in that direction, then we add one:
Branch A node that has a
The order in which the nodes are written will be different depending on the parent node and at least
order in which nodes are added. However the binary search tree will always one child node.
be able to provide a correctly ordered list.
Leaf A node that has no
Note child nodes.
Binary search tree A
Binary trees are also useful for representing equations and can be traversed in such
binary tree used to store
a way as to produce equations in infix notation, Polish notation and reverse Polish
data in order so that it
notation (RPN).
can be quickly and easily
searched.

Now test yourself

30 What is the difference between a graph and a tree?

Making links
31 What is the main feature of a binary tree?
32 Describe two advantages of using a binary search tree to store data. Infix and reverse Polish
notation are explored in
Answers available online Chapter 3.
46

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_02_MRN_AQA_CS_034-052.indd Page 47 6/29/21 10:01 PM f-0116 /103/HO02206/work/indd

Hash tables
A hash table is a data structure in which values are found by looking up their
associated key by following a hashing algorithm. Revision activity

2 Fundamentals of data structures

When the data is added to the hash table the hashing algorithm is applied to ✚ Complete a binary
the value in order to generate its key. search tree containing
eight items you can see.
A simple hashing algorithm for an ID number might be to add the digits ✚ Create a new binary
together, divide by 6 and record the remainder. (This is known as modulo 6 search tree containing
division). the same eight items,
For example: but added in a different
order.
ID number 25463 ID number 34255
Sum of digits = 2 + 5 + 4 + 6 + 3 = 20 Sum of digits = 3 + 4 + 2 + 5 + 5 = 19 0
Remainder = 20 % 6 = 2 Remainder = 19 % 6 = 1 1 34255
2 25463
The key can then be used as the index for the value in the hash table.
3
In a large hash table this allows for a value to be found quickly. 4
If the program were to search for the ID number 25463 then it could be run 5
through the hashing algorithm and the index number calculated rather than
having to perform a search. This reduces the search time.
Hash table A data
A problem can occur where two values generate the same key. This is referred structure for holding data
to as a collision. that is indexed according to
its key.
In the event of a collision, a simple solution is to store the value in the next
available index. Hashing algorithm An
algorithm for calculating a
Once a hash table is more than half full there will be many collisions, and so key for each value.
it is wise to rehash the table.
Collision The situation
This involves designing a new hashing algorithm that will provide more where two different items of
possible values (for example, using modulo 20 rather than modulo 6) and re- data produce the same key.
calculating the hashes for each existing value.
Rehashing The process of
Ideally a hashing algorithm should be designed to avoid this problem, recalculating the keys for all
however rehashing might be needed if data storage requirements grow over existing data using a new
the life of a programming project. hashing algorithm.

Making links
Hashing algorithms are also used for checksums (to check the validity of a file or other
piece of data) and for encryption (hashed passwords are typically stored in databases Revision activity
rather than plaintext passwords). Checksums and other error checking processes, and ✚ Design a simple hashing
encryption, are described in Chapter 5. algorithm for storing the
names of films based
on the length of the film
Now test yourself title.
✚ Implement a hash table,
33 Describe the relationship between a value and its key in a hash table. storing the names
34 What is meant by a collision in a hash table? of films in an array
based on the hashed
35 Describe two steps that must be taken when rehashing.
keys. Display an error
36 Describe three potential uses for a hashing algorithm. message if two films
37 A hashing algorithm for an ID number is written as ‘sum of each digit MOD 6’. generate the same key
Calculate the hashes for each of the following values. and devise an improved
a 7216 hashing algorithm to
b 5891 reduce the number of
c 0275 collisions (for example,
alter the modulo
Answers available online division).
47

My Revision Notes AQA A-level Computer Science Third Edition

325487_02_MRN_AQA_CS_034-052.indd Page 48 6/29/21 10:01 PM f-0116 /103/HO02206/work/indd

Dictionaries
A dictionary is a collection of key–value pairs, in which the value is found by
Dictionary A data
looking up the key.
2 Fundamentals of data structures

structure based on key–

Keys must be unique, but values can be duplicated. value pairs.

One use for a dictionary is information retrieval, for example in frequency Frequency analysis The
analysis (storing the value of how many instances of a word, letter or other process of examining how
key appears) or as a high score table: often something occurs.
A useful tool for trying to
{"Dave" : 23, "Kev" : 37, "Angelina" : 42} break some encryption
Another use for a dictionary is lossless compression where a dictionary of methods.
values is stored and the original file is represented with just the keys, in order. Lossless compression
For instance using the following dictionary: Reducing the size of a file
without the loss of any data.
Key Value
1 ask
2 not
3 what
4 your
5 country
6 can
7 do
8 for
9 you
10 ,
11 .
We can compress the sentence:
"Ask not what your country can do for you, ask what you can
do for your country."
This sentence can be stored as:

1 2 3 4 5 6 7 8 9 10 1 3 6 7 8 4 5 12

Now test yourself

38 Describe the essential concept of a dictionary using exactly three words.

39 Identify which field is which in a dictionary designed to store data for frequency
analysis.
40 Identify which field is which in a dictionary designed to store what type of food
each dog staying at a kennels should be fed with.

Answers available online

Vectors
Vector representation
In mathematics, scalars are simple numbers that only have a size. Vectors on Scalar A single number
the other hand are numbers that have both a magnitude and a direction. used to represent or adjust
Vectors are often used to represent a position. For example, a 2D vector can be the scale of a vector.
used to can be represent a position on a 2D plane. Vector A data structure
used to represent a position
Vectors can be shown in the list representation – when it is represented (for example, a 2D vector
as a list of numbers, such as [3,5]. In this two-dimensional vector the
48 represents a position on a 2D
plane).

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_02_MRN_AQA_CS_034-052.indd Page 49 6/29/21 10:01 PM f-0116 /103/HO02206/work/indd

first number represents the horizontal position, and the second number
represents the vertical position.
The same vector can be represented visually as a position on a two-
dimensional grid, or as an arrow.

2 Fundamentals of data structures

A two-dimensional vector is known as a 2-vector. The symbol for all possible
real numbers is R. A two-dimensional vector made up of real numbers exists
Figure 2.16 A graphical
in a two-dimensional space called R2. representation of the
A function maps one set of numbers to another. In formal mathematics, a vector [3,5]
function f can map a set of numbers, S, to a co-domain R, where S represents
a set of simple numbers and R represents the set of all real numbers: Note

f:S→R A list representation is

equivalent to a 1D array
Using a function representation the position vector [3,5] could be written as: representation.
✚ 0↦3
✚ 1 ↦5
In this case S is the set of numbers {0,1}. The symbol ↦ means ‘map to’.
And as a dictionary representation this vector could be written as {0 : 3, 1
: 5} and could be thought of as ‘move along 3, then move up 5’. Exam tip
A three dimensional vector would follow the same pattern; for example, [7, Although there are
2, 5] could be written as {0 : 7, 1 : 2, 2 : 5} and thought of as ‘move numerous ways to
along 7, up 2 and back 5’. represent a vector, a list
of values is the most
A three-dimensional vector made up of real numbers is known as a 3-vector frequently used model in
over R, and exists in a three-dimensional space called R3. A four-dimensional exam papers.
vector made up of real numbers is known as a 4-vector over R and exists in a
four-dimensional space called R4.

Vector manipulation
Because vectors are not normal numbers, the have special rules when adding
and multiplying. Vector addition involves adding the equivalent values
together. For example,
[3, 5] + [7, 1] =
[3+7, 5+1] =
[10, 6]
The result of vector addition is translation; for example, taking the vector [3,
5] and moving (or translating) by [7, 1].

Vector addition Adding

two vectors together
in order to perform a
translation.
Translation Moving a
vector by adding another
Figure 2.17 Vector addition
vector to it.
Scalar-vector multiplication means multiplying each value in a vector by a Scalar-vector
single number (or scalar). multiplication Multiplying
each element of a vector
2x [3, 5] =
by a number in order to
[2x3, 2x5] = increase the scale of a
vector.
[6, 10]
Scaling Changing the scale
The result of scalar-vector multiplication is scaling; for example, taking the of a vector by multiplying
vector [3, 5] and scaling it by 2. the vector by a scalar.
49

My Revision Notes AQA A-level Computer Science Third Edition

325487_02_MRN_AQA_CS_034-052.indd Page 50 6/29/21 10:01 PM f-0116 /103/HO02206/work/indd

The convex combination of two vectors means that two vectors must be
multiplied by two scalar numbers which add up to exactly 1 and are both
positive (greater than or equal to 0).
0.2 x [3, 5] + 0.8 x [7, 1] =
2 Fundamentals of data structures

[0.6, 1.0] + [5.6, 0.8] =

[6.2, 1.8]
The result of the convex combination of 2 vectors will always be a vector that
is on the straight line between the two original vectors.
Convex combination is often written as αu + β v, where u and v are vectors
and α + β = 1 and α ≥ 0 and β ≥ 0.
Figure 2.18 Scalar-vector
The dot product, or scalar product, of two vectors is written as u • v, for multiplication
example
[3,5] • [7,1] Convex combination The
process of multiplying two
The result is found by multiplying the values within each position of the (or more) vectors by scalar
vector together and adding the results. values that are both positive
[3,5] • [7,1] = and add up to exactly 1.
Dot product A calculation
3x7 + 5x1 = used to help find the angle
21 + 5 = between two vectors.

26
Note
The result of a dot product is always a scalar, not a vector.
The result of the convex
The dot product of two vectors can be used to calculate the angle between the
combination of 3 (or more)
two vectors.
vectors will always be a
It is not necessary to know how to calculate the exact angle, but the table vector that is within the
below gives a general indication. shape described by the
original vectors.
Dot product Angle between the vectors
0 Exactly 90°
Bigger than 0 Between 0 and 90°
Less than 0 Between 90° and 180°

Note
The dot product of two vectors is used in many computer games to calculate the
angles between characters in order to check whether sentries can see an intruder or
to help non-player characters make decisions relating to route finding.

Now test yourself

41 What is the difference between a vector and a scalar?

42 Describe four different ways a vector can be represented.
43 Add the vectors [-3,2] and [6,-4].
44 State the effect of multiplying a vector by a scalar value.
45 If a convex combination of two vectors is calculated using the formula αu + βv,
what two rules dictate the possible values of α and β?
46 Describe the significance of the value of any convex combination of two vectors.
47 Describe the purpose of calculating the dot product of two vectors.

Answers available online

50

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_02_MRN_AQA_CS_034-052.indd Page 51 6/29/21 10:01 PM f-0116 /103/HO02206/work/indd

Summary
✚ Data structures provide a way to store collections of ✚ Trees are graphs that don’t include any cycles (loops)
data together rather than in individual variables ✚ A rooted tree has one vertex which is the root and all

2 Fundamentals of data structures

✚ Arrays are designed to store tables of data and can be edges flow away from that root
single-dimensional or multi-dimensional ✚ In a binary tree each vertex can have a maximum of
✚ Two-dimensional arrays can be accessed using two two children
nested FOR loops ✚ A hash table is a data structure that creates a map
✚ Text files store data using ordinary text (for example, between keys and values
ASCII or Unicode) ✚ Hash tables are used to organise data by using a
✚ Binary files store binary codes for each value hashing algorithm to generate a key
✚ Binary files are usually smaller in size but cannot be ✚ A collision occurs when two values generate the same
read by other programs hash
✚ Abstract data types are more complex data structures ✚ Collisions can be handled by rehashing – designing a
✚ Queues are a FIFO (First In, First Out) data structure new hashing algorithm and recalculating the hash for
✚ Items are enqueued at the back of the queue and each item
dequeued from the front of the queue ✚ Dictionaries are a collection of key–value pairs
✚ Common uses include buffers, card games and ✚ Dictionaries can be used for information retrieval or for
breadth first searches lossless encryption
✚ In linear queues the items are shuffled forward ✚ Vectors are quantities that have a size and a direction
when the front item is dequeued ✚ Vectors can be represented as a list of numbers, a
✚ In circular queues the rear pointer and front pointer position or arrow on a chart or as a dictionary
are moved back when items are enqueued or ✚ Adding vectors results in translation
dequeued and wrap around to the beginning of the ✚ Multiplying vectors by a (scalar) number results in
queue again scaling
✚ In priority queues each item is given a priority and ✚ The convex combination of two vectors means
items with higher priorities are dequeued first multiplying each by a number between 0 and 1, where
✚ Stacks are a LIFO (Last In, First Out) data structure both numbers add up to exactly one
✚ Items are pushed onto the top of the stack and ✚ The resulting vector from a convex combination will
popped from the top of the stack always be a point on the straight line between the two
✚ Common uses include the undo feature, call stacks original vectors
in programming and the depth first search ✚ The dot product of two vectors is found my multiplying
✚ The graph data structure uses vertices/nodes and each first term, second term, etc. and adding the
edges/arcs to represent relationships between items results together
✚ Graphs can be weighted or unweighted, directed or ✚ The dot product can be used to calculate the angle
undirected, connected or disconnected between two vectors
✚ Sparse graphs, or graphs where the existence of ✚ Vectors can be in multiple dimensions: a three-
edges doesn’t need to be checked often, should be dimensional vector is known as a 3-vector over R,
represented using an adjacency list which exists in a three-dimensional space called R3
✚ Dense graphs, or graphs where the existence of edges
does need to be checked often, should be represented
using an adjacency matrix

Exam practice

1 A programmer is writing a program to simulate a card game using a circular queue

data structure to represent a deck of 52 cards. Cards are dealt from the top of the
deck and replaced at the bottom of the deck.
a) State why a queue is the most appropriate data structure to represent the
deck of cards. [1]
b) Explain why a circular queue is the most appropriate implementation of a
queue in this situation. [2]
c) Describe one problem that could occur when a card is drawn from the
deck and suggest how this problem could be dealt with. [2]
d) A player is allowed to check the top card in the deck without taking that
card. State the name of the function in a queue which allows this to happen. [1]

51

My Revision Notes AQA A-level Computer Science Third Edition

325487_02_MRN_AQA_CS_034-052.indd Page 52 6/29/21 10:01 PM f-0116 /103/HO02206/work/indd

2 This graph represents flight routes between five cities.

A B
2 Fundamentals of data structures

C D

a) Tick one box that best describes this graph. [1]

Directed
Unweighted
Tree
Disconnected
b) Copy and complete the adjacency list for this graph. [2]

A
B
C
D
E

c) Copy and complete the adjacency matrix for this graph. [2]

A B C D E
A 0 1
B
C
D
E

d) Describe two situations where an adjacency matrix would be more

appropriate than an adjacency list. [2]
3 In a computer game the position of a security guard is described by the
vector, u, [3,5] and is moving in the direction described by vector, v, [–4,4].
The player is standing at the position described by the vector, w, [1,4].
a) Calculate the resulting position of the security guard by adding the vectors
u and v. [1]
b) Calculate the vector, x, between the security guard’s original position
(u) and the player’s position (w). [1]
c) Calculate the result of 3w. [1]
d) Calculate the result of 0.7u + 0.3w. [2]
e) Explain the significance of the result of part c. [1]
f) Explain why the calculation 0.7u + 0.4w would not produce a result
with the same significance. [1]
g) Calculate the result of v • x. [2]
h) Explain the meaning of the dot product and why finding the dot product
of v and x could be useful as part of the game mechanics. [2]

52

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_03_MRN_AQA_CS_053-070.indd Page 53 6/29/21 10:00 PM f-0116 /103/HO02206/work/indd

3 Fundamentals of algorithms
Several standard algorithms are covered in this chapter.
Algorithm A sequence
For each algorithm, you are expected to be familiar with: of instructions that are
✚ the purpose of the algorithm followed in order to solve a
✚ the general steps involved in the algorithm problem.
✚ the program code for the algorithm Hand-trace Also known
✚ how to hand-trace the algorithm. as dry run. Without using
You are not expected to memorise the program code for each algorithm, a computer, completing a
although you should be able to recognise each algorithm if you see it. table to record the values
of each variable as the
Making links program is executed line-
by-line.
Hand-tracing algorithms is an important concept that appears in every Component 1
exam. More detail on how to tackle trace table problems is covered in Chapter 4.
When you are confident with the principles of hand-tracing algorithms it is advised to
practise with each of the examples in this chapter.

Graph-traversal
Graph traversal algorithms are designed to find a route from one node of a
Making links
graph to another.
For more on graphs, see
There are two main approaches to this problem, the breadth first search (BFS)
Chapter 2.
and depth first search (DFS).
In each of the examples below, the algorithm is being used to find a route
from node A to node F.

Breadth first search

In the breadth first search, the algorithm:
Graph traversal Inspecting
✚ removes the first node in the queue
the items stored in a graph
✚ inspects this node by adding all adjacent nodes to the queue data structure.
✚ marks the node as visited.
Breadth first search
It repeats this sequence until the required node is found, or all nodes have (BFS) Searching a graph by
been visited. checking every node one
We start by adding the first node (node A in this example) to a queue. step from the starting point,
then every node two steps
F H away, etc.

E G Depth first search (DFS)

C Searching a graph by
heading in one direction
D Front Rear until you reach a dead end,
B
then stepping back one
A node at a time.
A

In each step, the item at the front of the queue is removed, and each adjacent
node is added to the queue.
In this first step, we remove the first item, A, from the queue and inspect it.
The adjacent nodes are B, C and D so these are added to the queue and A is
marked as visited:

53

My Revision Notes AQA A-level Computer Science Third Edition

325487_03_MRN_AQA_CS_053-070.indd Page 54 6/29/21 10:00 PM f-0116 /103/HO02206/work/indd

F H

E G
C
3 Fundamentals of algorithms

D Front Rear
B
B C D
A

Visited: A
In the next iteration the first item (node B) is removed from the queue and
inspected. The only adjacent node to B is C, which is already in the queue. As
there are no new adjacent nodes no items are added to the queue. Node B is
marked as visited.

F H

E G
C

D Front Rear
B
C D
A

Visited: A, B
Again, the first item in the queue is removed and inspected. In this case it is
node C. Node C is adjacent to nodes E and G which are added to the queue.
Node D is also adjacent to node C but is already in the queue. Node C is
marked as visited.

F H

E G
C

D Front Rear
B
D E G
A

Visited: A, B, C
The first item in the queue, node D, is removed and inspected. There are
no new adjacent nodes to node D, and so the queue is unchanged. Node D is
marked as visited.

F H

E G
C

D Front Rear
B
E G
A

Visited: A, B, C, D
The first item in the queue, node E, is removed and inspected. Node E is
inspected next, which has an adjacent node F (the target). F is added to the
queue and E is marked as visited.

54

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_03_MRN_AQA_CS_053-070.indd Page 55 6/29/21 10:00 PM f-0116 /103/HO02206/work/indd

F H

E G
C

3 Fundamentals of algorithms
D Front Rear
B
G F
A

Visited: A, B, C, D, E
By the end of the algorithm the nodes have been visited in the order
[A,B,C,D,E,G,F,H], however by keeping track of the parent for each node (which
node was being interrogated when that node was found) it is possible to
identify the shortest route – in this case [A, C, E, F].
This algorithm will always find the shortest path through an unweighted
graph, however it may take many steps to do so in a large graph.
Revision activity
A queue is the most appropriate data structure to use as each newly
Using a section of the
discovered node is added to the rear of the queue and nodes are inspected in
London Underground
the order in which they are found (FIFO).
Walking Tube Map (you
can find a PDF of this on
Now test yourself
the Transport for London
1 What data structure should be used in a breadth first search? website, tfl.gov.uk), carry
out a breadth first search
2 State one potential use for a breadth first search.
to find the shortest route
Answers available online between two stations.

Depth first search

In the depth first search a very simplistic description is to traverse the graph Note
in one direction until a dead end is found, at which point the algorithm
backtracks until it finds an alternative route it has not tried. This is very There is more than one way
similar to the approach most people take in a maze. to implement the depth first
search, and the following
In each iteration the top item in the stack is removed and inspected for description covers one
adjacent nodes which have not yet been visited: specific implementation.
✚ if they exist, then these adjacent nodes are added to the top of the stack You may come across
✚ if they don’t exist, then the top item in the stack is removed. slightly different steps,
In this case the starting node (node A) is added to a stack. however the overall
principle remains the same.
Top F H Top

E G
C

D
B

Bottom A A Bottom

When node A is removed from the stack and inspected, the nodes B, C and D
are added.

Top F H Top

E G
C
D
D
B C

Bottom A B Bottom 55

My Revision Notes AQA A-level Computer Science Third Edition

325487_03_MRN_AQA_CS_053-070.indd Page 56 6/29/21 10:00 PM f-0116 /103/HO02206/work/indd

The node at the top of the stack is always inspected first. Node D is not
adjacent to any new nodes, and so the stack is unchanged.

Top F H Top
3 Fundamentals of algorithms

E G
C

D
C B C

B Bottom A B Bottom

Inspecting node C means that nodes E and G are added to the stack.

Top F H Top

E G
C
G
D
B E

B Bottom A B Bottom

Inspecting node G leads on to node H, which is added to the stack.

Top F H Top

E G
C
H
D
E B E

B Bottom A B Bottom

Node H is inspected but does not have any new adjacent nodes, and so the
stack is unchanged.

Top F H Top

E G
C

D
E B E

B Bottom A B Bottom

The next node to be inspected is node E, effectively back-tracking from the

dead end at node H. Node F is now discovered and added to the stack.

Top F H Top

E G
C

D
B F

B Bottom A B Bottom

At this point the node has been identified and the route take is described as
[A, D, C, E, F].
A stack is the most appropriate data structure to use as each newly
discovered node is inspected immediately (LIFO) and the route to this point
56 can be stepped through in reverse (popped) to find a new path.

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_03_MRN_AQA_CS_053-070.indd Page 57 6/29/21 10:00 PM f-0116 /103/HO02206/work/indd

This algorithm can be effective at finding a route through a maze more

quickly than the BFS because not every route must be explored, though the
route may not be the optimal route.

3 Fundamentals of algorithms
Algorithm Data Structure to use Useful for
Breadth First Search (BFS) Queue Finding the shortest path between two nodes
Depth First Search (DFS) Stack Navigating a maze

Now test yourself Making links

3 What data structure should be used in a depth first search? It is important to be familiar
4 State one potential use for a depth first search. with the workings of the
algorithms covered in this
Answers available online chapter, and how to hand-
trace the algorithm. This
chapter focuses on the
Revision activity principles of the algorithms.
Using a section of the London Underground Walking Tube Map, carry out a depth first Hand tracing technique is
search to find the shortest route between two stations. covered in Trace tables in
Chapter 4 and the pseudo-
code for the breadth first
Exam tip search and depth first search
is included in that chapter.
You are not expected to memorise the exact syntax for the BFS and DFS algorithms,
though you should be familiar with the principles of the algorithm and be able to
describe which algorithm might be chosen in different circumstances.
It is also important to be able to hand-trace both the BFS and DFS algorithms. The
pseudo-code algorithm would be provided for this, as well as a trace table to fill in.

Tree traversal
Tree traversal is carried out using one of three recursive algorithms, all of
which look similar.
✚ Pre-order tree traversal
✚ In-order tree traversal
✚ Post-order tree traversal
The difference between the algorithms is in when the node’s value is output.
Making links
// In-Order Traversal
For more on trees see
FUNCTION Traverse (node)
Chapter 2.
OUTPUT node
IF node.left THEN Tree traversal Inspecting
Traverse(node.left) the items stored in a tree
data structure.
IF node.right THEN
Pre-order tree traversal
Traverse(node.right)
Inspecting or displaying the
ENDFUNCTION value of each node before
// In-Order Traversal moving on to other nodes.

FUNCTION Traverse (node) In-order tree traversal

Inspecting or displaying the
IF node.left THEN value of each node after
Traverse(node.left) checking the node on the
left but before checking the
OUTPUT node
node on the right.
IF node.right THEN
Post-order tree traversal
Traverse(node.right) Inspecting or displaying the
value of each node after
ENDFUNCTION 57
checking for all child nodes.

My Revision Notes AQA A-level Computer Science Third Edition

325487_03_MRN_AQA_CS_053-070.indd Page 58 6/29/21 10:00 PM f-0116 /103/HO02206/work/indd

// Post-Order Traversal
FUNCTION Traverse (node)
IF node.left THEN
3 Fundamentals of algorithms

Traverse(node.left)
IF node.right THEN
Traverse(node.right)
OUTPUT node
ENDFUNCTION

Pre-order tree traversal

In pre-order tree traversal the value of the node is always output first, before
checking for nodes on the left, and then for nodes on the right.
Because the tree traversal algorithms use a recursive subroutine, each of the
left nodes is inspected until there are no more left nodes to be found. At this
point the right nodes are inspected.
If there are no nodes left to inspect then the subroutine returns to the
previous node.
The order in which nodes are examined can be described using the following
technique.

Pre Post A
In

B C

D E F

Figure 3.1 Pre-order tree traversal

In Figure 3.1, the green outline shows the order in which the nodes are
traversed, starting with Node A and then checking the left node each time
until a dead end is reached. The subroutine then goes back a stage and checks
the right-hand node.
Using pre-order tree traversal the value of the node is output as soon as that
node is first inspected. This can be shown by placing a horizontal bar to the
left each time the green outline meets a node.
The output of pre-order tree traversal can be found by following the outline
and writing down the value of the node each time it crosses a purple line.
With this tree the output would be A, B, D, E, C, F.
The pre-order tree traversal is used when copying a tree, as each node value
will be immediately copied to the new tree as soon as it is found, meaning
that it is impossible to copy node E, for example, before node B has already
been copied.

Making links
A binary search tree is a data structure used to store data that can be quickly
searched. The binary search algorithm is explored on page 66.

58

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_03_MRN_AQA_CS_053-070.indd Page 59 6/29/21 10:00 PM f-0116 /103/HO02206/work/indd

In-order tree traversal

For in-order tree traversal the value of the node is output after the left-hand
nodes are inspected, but before the right-hand node is inspected.

3 Fundamentals of algorithms
This can be represented by using a vertical bar underneath each node.
Whenever the outline path crosses a vertical bar, the value from that node
should be output.

Pre Post A
In

B C

D E F

Figure 3.2 In-order tree traversal

The output from the tree in Figure 3.2 would be D, B, E, A, C, F.

In a binary search tree the nodes are arranged so that an in-order tree
traversal would output the nodes in the correct order.

Post-order tree traversal

For post-order tree traversal the value of the node is output after both the left
AND the right nodes have been inspected.
To represent this horizontal bar on the right of each node indicates the point
in the subroutine where the value can be output. Each time the green outline
passes a brown bar, that value has been processed.

Pre Post A
In

B C

D E F

Figure 3.3 Post-order tree traversal

The output from the tree in Figure 3.3 would be D, E, B, F, C, A.

Post-order tree traversal is useful for emptying a tree, as nodes with children
cannot be destroyed until the child nodes have already been destroyed.
Post-order tree traversal is also used for converting from infix notation to
postfix notation.

Making links
Infix notation is how people typically write mathematical equations. Postfix notation
is an alternative format that is easier and faster for a computer to process. Both
notations are described in more detail later in this chapter.

59

My Revision Notes AQA A-level Computer Science Third Edition

325487_03_MRN_AQA_CS_053-070.indd Page 60 6/29/21 10:00 PM f-0116 /103/HO02206/work/indd

Each method of traversal has a different effect on the order in which the items
are displayed.

Uses Mathematical expressions

3 Fundamentals of algorithms

Pre-order Copying a tree Prefix

In-order Outputting the contents of a binary Infix

search tree in ascending order
Post-order Emptying a tree Postfix (reverse Polish notation)
Converting from infix to postfix
notation (see next section)

Now test yourself

5 Which tree traversal algorithm will print the value of the root node as the
last line?
6 Which tree traversal algorithm would be most appropriate for copying a tree?
7 Describe two possible uses for post-order tree traversal.
8 Which tree traversal algorithm can be used to output a list in ascending order?

Answers available online

Revision activity
Practise creating a binary search tree by adding the names of people, pets, sports
teams, and so on.
✚ Try using all three tree traversal algorithms to copy the tree, and identify which
method works best.
✚ Try using all three tree traversal algorithms to empty the tree, and identify which
method works best.
✚ Try using all three tree traversal algorithms to display the values in the list, and
identify which is in the correct, sorted order.

Reverse Polish notation

Mathematical expressions can be expressed using a number of different
Infix notation A method
notations.
of writing mathematical or
Typically, people are used to writing mathematical expressions using infix logical expressions with
notation. Here the operations are written in between their operands; for operators placed between
example, (2 – 5) x 7. operands; for example, 2 + 2.

Postfix, or reverse Polish notation (RPN) looks slightly different because the Postfix / Reverse Polish
operations are written after their operands; for example, 2 5 – 7 x. notation A method of
writing mathematical or
Reverse Polish notation is useful for processing because:
logical expressions in which
✚ there is no need for parentheses/brackets the operators appear after
✚ the processing can be carried out using a stack. their operands; for example,
When processing RPN using a stack, each value is added to the stack until 2 2 +.
an operator is found. At this point two items are popped from the stack, the Operation A function to be
answer calculated and then pushed back onto the stack. applied; for example, +, -,
AND, OR.

Making links Operand A value or object

that is to be operated on.
Because RPN can be executed on a stack it is often used in interpreters; for example,
when dealing with bytecode and Postscript. Bytecode is used in some languages as
a partial step in translation between source code and object code. This is explored in
60 Bytecode in Chapter 6.

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_03_MRN_AQA_CS_053-070.indd Page 61 6/29/21 10:00 PM f-0116 /103/HO02206/work/indd

Worked example
In RPN, the following calculation 2 5 – 7 x is performed, using a stack, as follows:

3 Fundamentals of algorithms
2 5 – 7 x

2 5 Operator Result 7 Operator Result

pushed pushed found. Pop pushed pushed found. Pop pushed
values and values and
perform perform
calculation. calculation.

5 7

2 2 –3 –3 –21

Transforming between infix and postfix/reverse Polish notation can be carried out by inspection (that is, by looking at the
expression and working out the answer in your head).
The transformation can also be carried out using a tree.

3 x

2 5

Figure 3.4 A binary tree representing a mathematical expression

Using in-order tree traversal on Figure 3.4 gives the result 3 + 2 x 5
Remembering to apply BIDMAS rules to infix notation, this gives 3 + 10 = 13
Using post-order tree traversal on Figure 3.4 gives the result 3 2 5 x +
When the first operator is found this will be applied to the two preceding values, giving 3 10 + = 13

Revision activity Exam tip

✚ Practise writing infix notation expressions using a binary tree and then using post- Make sure you are
order tree traversal to convert them into RPN. confident with converting
✚ Draw out and step-through the process of using a stack to evaluate, or solve, RPN between infix notation and
expressions. reverse Polish notation
by inspection as well as
understanding how trees
Now test yourself can be used to perform the
translation algorithmically.
9 Transform each of the following infix expressions into reverse Polish notation.
a 3 + 7
b 12 – 3 + 9
c 4 x 12 + 6
d (7 – 3) ÷ 2
10 Transform each of the following reverse Polish expressions into infix notation.
a 12 6 –
b 15 2 x 6 +
c 3 7 2 x –
11 Give two advantages for using reverse Polish notation.

Answers available online

61

My Revision Notes AQA A-level Computer Science Third Edition

325487_03_MRN_AQA_CS_053-070.indd Page 62 6/29/21 10:00 PM f-0116 /103/HO02206/work/indd

Searching algorithms
Linear search
3 Fundamentals of algorithms

The linear search is a very inefficient but simple algorithm for finding an item
Linear search A searching
in an array of items.
algorithm in which each item
The basic aim of the algorithm is to check each item in an array, one-by-one, is checked one-by-one.
from start to finish.
FOR i ← 0 to Length(Items)
IF Items[i] = Target
OUTPUT "Found"
ENDIF
ENDFOR
The linear search is very simple to program and can be effective for very
small lists. It is also useful if the list is unordered, however it is extremely
inefficient for large lists.

Making links Exam tip

The efficiency of algorithms is an important consideration when deciding how to solve You are expected to be
a problem. This is measured using time-wise complexity and space-wise complexity, very familiar with the linear
both of which are discussed in Comparing algorithms in Chapter 4. search and you should
be confident to program
Time-wise complexity is a measure of how efficient an algorithm is in terms one from scratch in the
of time. It is worth trying to remember the time-wise complexity for each of Component 1 exam.
the algorithms in the next sections.

Algorithm Time-wise Complexity

Linear search O(n)

Binary search
The binary search is an example of a ‘divide and conquer’ approach to
Binary search A searching
searching.
algorithm in which the
Given a sorted array of items the algorithm will check the middle item. middle item is checked and
Because the list is sorted, this means that half of the array can then be half of the list is discarded.
discarded, halving the possible number of values to search each time.
The implementation of a binary search uses pointers to mark the Start, End
and Midpoint of the array, Mid, calculating the midpoint by adding the Start
and End values and halving them.
Mid = (Start + End) / 2

Start Mid End Start 0

[0] [1] [2] [3] [4] [5] [6] End 6

Mid 3

62

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_03_MRN_AQA_CS_053-070.indd Page 63 6/29/21 10:00 PM f-0116 /103/HO02206/work/indd

Where there are an even number of items to choose from the algorithm can
be programmed to use the item on the left, or the item on the right. Using
integer division will mean that the item on the left is chosen in this case.
Mid = (Start + End) // 2

3 Fundamentals of algorithms
Start Mid End Start 0

[0] [1] [2] [3] [4] [5] End 5

Mid 2

It is important to remember that the midpoint value should be discarded if it

Integer division A division
does not match, and so the full implementation may look like this:
operation in which any
Start ← 0 fractional part is discarded,
equivalent to rounding down
End ← Length(Items) – 1
to the nearest whole number.
Mid ← (Start + End) // 2
WHILE Start <= End AND Items[Mid] != Target
IF Items[Mid] < Target THEN
Start ← Mid + 1
ELSE
End ← Mid – 1
ENDIF
Mid ← (Start + End) // 2
ENDWHILE
IF Items[Mid] = Target THEN
OUTPUT "Item found"
ELSE
OUTPUT "Item not found" Making links
ENDIF Time-wise complexity is
explored in more detail in
The binary search is considered to be very efficient. It is generally much
Comparing algorithms in
quicker than the linear search for large, sorted lists. However, it cannot be
Chapter 4.
carried out on an unsorted list.

Algorithm Time-wise Complexity Exam tip

Binary search O(log n) You are expected to be
very familiar with the binary
search and you should
Now test yourself be confident to program
one from scratch in the
12 Give one disadvantage of using a linear search.
Component 1 exam.
13 Suggest two situations where a linear search would be an appropriate choice.
14 Suggest one advantage of using a binary search instead of a linear search.
15 Suggest one disadvantage of using a binary search instead of a linear search.
16 Explain why a binary search is more efficient than a linear search.

Answers available online

Revision activity

✚ Practise programming the linear search to find an item from a list of celebrity names.
✚ Practise programming a binary search from scratch.
✚ Hand trace the binary search with some sample values and then use your
programming platform’s debugging tool to check your accuracy. 63

My Revision Notes AQA A-level Computer Science Third Edition

325487_03_MRN_AQA_CS_053-070.indd Page 64 6/29/21 10:00 PM f-0116 /103/HO02206/work/indd

Binary tree search

A binary tree search uses the same principles as a binary search to find an
Binary tree search A
item held in a binary tree.
searching algorithm in which
3 Fundamentals of algorithms

Because each item is added to a tree in order, a recursive algorithm can a binary tree is traversed.
be used to travel through the tree, moving through the left or right node Binary tree A tree in which
depending on the target value. each node can have no more
An example implementation might look like this: than two child nodes, each
placed to the left or right of
FUNCTION BinaryTreeSearch (Node) the preceding node so that
IF Node.Value = Target THEN the tree is always in order.

Return True
Cat
ELSEIF Node.Value > Target AND Exists(Node.Left) THEN
Return BinaryTreeSearch(Node.Left)
Budgie Fish
ELSEIF Node.Value < Target AND Exists(Node.Right) THEN
Return BinaryTreeSearch(Node.Right)
Rabbit
ELSE
Figure 3.5 A binary tree
Return False
ENDIF
ENDFUNCTION
And a trace table might look like this:
Target Call Node Value returned
Dog 1 Cat

2 Fish Making links

3 Dog True Time-wise complexity is
explored in more detail in
2 Fish True
Comparing algorithms in
1 Cat True Chapter 4.
Hand-tracing algorithms
Algorithm Time-wise Complexity is an important concept
Binary tree search O(log n) that appears in every
Component 1 exam. More
detail on how to tackle trace
Binary search trees are an efficient method of searching because they can
table problems is covered in
remove large sections of the possible values at each step, much like the more
Trace tables in Chapter 4.
traditional binary search.

Now test yourself

17 State one advantage of using a binary search tree to store data and then a binary
tree search to retrieve data instead of:
a an array and a linear search
b an array and a binary search.

Answers available online

Revision activity
Create a binary search tree using any data you like (such as film names or book titles)
and hand trace the steps in searching for specific items.

64

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_03_MRN_AQA_CS_053-070.indd Page 65 6/29/21 10:00 PM f-0116 /103/HO02206/work/indd

Sorting algorithms
Bubble sort

3 Fundamentals of algorithms
The bubble sort is a relatively simple, but very inefficient sorting algorithm. Bubble sort A sorting
algorithm in which pairs of
In the first pass, each item is compared to its immediate neighbour and the
items are sorted, causing the
value are swapped if necessary. In this way, the highest value bubbles to the
largest item to bubble up to
end of the list each time.
the top of the list.
If there are n items in the list, there are (n-1) comparisons in each pass, so in
Pass Travelling through
a list of six items there are five comparisons.
a list from start to finish
[0] [1] [2] [3] [4] [5] exactly once.

9 7 3 12 4 8 9 and 7 are compared and swapped

7 9 3 12 4 8 9 and 3 are compared and swapped

7 3 9 12 4 8 9 and 12 are compared and not swapped

7 3 9 12 4 8 12 and 4 are compared and swapped

7 3 9 4 12 8 12 and 8 are compared and swapped

7 3 9 4 8 12 The first pass is completed

The same algorithm is then repeated (n-1) times; that is, there are (n-1)
passes because, in the worst case scenario (where the last number in the list
is the smallest value), the value will be moved down one place each time.
Exam tip
FOR i ← 0 to n – 1 //This loop is for each pass
You are expected to be
// Complete one pass: very familiar with the
FOR j ← 0 to n – 1 //This loop is for each comparison bubble sort and you should
be confident to program
IF Items[j] > Items[j+1] THEN one from scratch in the
Temp ← Items[j] Component 1 exam.

Items[j] ← Items[j+1]
Items[j+1] ← Temp
ENDIF
ENDFOR
ENDFOR
There are two main methods to improve the efficiency of the bubble sort.
1 Record the number of swaps completed in each pass. If 0 swaps were made
then the list must be in order and the algorithm can be stopped.
2 For each pass (outer loop), reduce the number of comparisons by 1. This is
because the highest value is guaranteed to have bubbled to the end of the
list each time, and so on the second pass it is not necessary to make the
final comparison. On the third pass it is not necessary to make the final
two comparisons, and so on.
For a large list, or a list that is nearly sorted, these improvements can have a
dramatic impact on the running time of the algorithm.
The bubble sort is still considered to be very inefficient in most cases and is
rarely used in industry. Making links
The bubble sort can be suitable for a list that is nearly sorted. Time-wise complexity is
explored in more detail in
Algorithm Time-wise Complexity
Comparing algorithms in
Bubble sort O(n2) Chapter 4.
65

My Revision Notes AQA A-level Computer Science Third Edition

325487_03_MRN_AQA_CS_053-070.indd Page 66 6/29/21 10:00 PM f-0116 /103/HO02206/work/indd

Now test yourself

18 State the number of comparisons needed to run through one pass of the bubble
sort for a list of 10 items.
3 Fundamentals of algorithms

19 Suggest two methods for improving the efficiency of the standard bubble sort.

Answers available online

Revision activity

✚ Take a physical collection (such as CDs, DVDs, books) and practise carrying out a
bubble sort in real life.
✚ Practise programming the basic bubble sort from scratch.
✚ Improve your basic implementation by reducing the number of steps in the inner
loop by 1 each time and by using a flag to check whether any swaps were made
during that pass.
✚ Hand-trace the bubble sort with some sample values and then use your
programming platform’s debugging tool to check your accuracy.

Merge sort
The merge sort is significantly more efficient than the bubble sort, though it is Merge sort A sorting
also significantly more complex. algorithm in which items are
✚ A list is broken in half, and in half again until we have a series of single split into single items and
elements. then merged into sorted
✚ The elements are placed into pairs and each pair is sorted. pairs, fours, eights and so on.
✚ Each group of two pairs is merged into a sorted group of four.
✚ Each group of two fours is merged into a sorted group of eight.
✚ Repeat as necessary until the list is complete.
3 7 12 9 4 8 6 2

3 7 12 9 4 8 6 2

3 7 12 9 4 8 6 2

3 7 12 9 4 8 6 2

3 7 9 12 4 8 2 6

3 7 9 12 2 4 6 8

2 3 4 6 7 8 9 12

Making links Recursive An algorithm that

uses a sub-routine which
The nature of this solution, repeatedly breaking down and then merging lists, means calls itself.
that a recursive solution is an ideal way of tackling this problem. Recursion is
introduced in Chapter 1.

When merging two lists, only the top two values need to be compared each
time as each list is already guaranteed to be in order.
Making links
This reduces the overall number of steps required and means that the
algorithm is much more efficient than the bubble sort in most cases. Time-wise complexity is
explored in more detail in
The merge sort also has a fixed number of steps, no matter how ordered or
Comparing algorithms in
66 unordered the list may be. As a result, some lists that are nearly in order can
Chapter 4.
be sorted more quickly using a bubble sort, though this is rare.

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_03_MRN_AQA_CS_053-070.indd Page 67 6/29/21 10:00 PM f-0116 /103/HO02206/work/indd

Algorithm Time-wise Complexity

Merge sort O(n log n)

3 Fundamentals of algorithms
Note
Though not covered in this specification, there are many other sorting algorithms.
These include the quick sort, heap sort, insertion sort and shell sort. An investigation
into the efficiency of different sorting algorithms could make for an interesting NEA
programming project.

Exam tip
You may be asked to trace the merge sort, or to demonstrate your understanding
of how the merge sort works. You are unlikely to be asked to program a merge sort
from scratch so focus your revision on understanding the algorithm rather than
programming it from memory.

Algorithm Time-wise Complexity

Linear search O(v)

Binary search O(log n)

Binary tree search O(log n)

Bubble sort O(n2)

Merge sort O(n log n)

Now test yourself Making links

20 Give two situations in which the merge sort would be more suitable than the As discussed, time-wise
bubble sort. complexity is a measure of
how efficient an algorithm
21 Give one situation in which the bubble sort would be more suitable than the
is in terms of time and is
merge sort.
explored in more detail in
22 Which sorting algorithm always has a fixed number of steps? Comparing algorithms in
Chapter 4. However, it is
Answers available online
worth trying to remember
the time-wise complexity
Revision activity for each of these
algorithms:
Take a physical collection (such as CDs, DVDs, books) and practise carrying out a
merge sort in real life.

Optimisation algorithms
Dijkstra’s shortest path algorithm
Dijkstra’s shortest path algorithm is a graph traversal algorithm that will
Shortest path The route
find the shortest route from a given starting node to every other node on the between two nodes on a
graph. It is very efficient for route finding across weighted graphs, though a graph that incurs the least
breadth first search is more efficient for unweighted graphs. cost. In an unweighted graph
The trace table for Dijkstra’s algorithm provides the shortest overall distance the cost for each path can be
to each node in the graph and the previous node to travel from. This means assumed to be 1.
that the quickest path to different nodes from the starting node can be found
quickly, without the need to carry out the algorithm again.

67

My Revision Notes AQA A-level Computer Science Third Edition

325487_03_MRN_AQA_CS_053-070.indd Page 68 6/29/21 10:00 PM f-0116 /103/HO02206/work/indd

Initially the distance from the start node to each node is set to ∞ (infinity).
Then the algorithm works as follows:
✚ For every node attached to the current node:
✚ add the distance from this closest unvisited node to the current node,
3 Fundamentals of algorithms

and add it to the distance from the current node to the start node – this
is the distance of the unvisited node to the start node
✚ if this calculated distance is less than the current distance between the
start node and unvisited node, the value is updated – this is how the
shortest route is calculated
✚ note the current node as the previous ‘parent’ node for the unvisited node
✚ repeat until all nodes attached to the current node have distances
recorded.
✚ Mark the current node as visited.
✚ Select a new current node, which is the closest unvisited node, and repeat
all the steps above.
✚ Repeat until the target is reached.

Worked example

2 7
A B F
2

5 4 G
4

4
C 6
D 3 E

Figure 3.6 An example graph

Applying Dijkstra’s algorithm to Figure 3.6, starting from Node A, would produce the following results:

A B C D E F G
Shortest distance 0 2 5 11 14 9 11
Previous node - A A C C B F

Once Dijkstra’s algorithm has produced a result, the shortest route from node A to any other node can be found by
working backwards, using the ‘previous node’.
✚ The shortest route to G ends by moving to node G from node F.
✚ Node F is reached from node B.
✚ Node B is reached from node A.
Therefore, the shortest path is A–B–F–G.
Exam tip
It is not necessary to
Making links
remember each and
Hand-tracing algorithms is an important concept that is assessed in every Component 1 every step in Dijkstra’s
exam. More detail on how to tackle trace table problems is covered in Chapter 4. shortest path algorithm,
When you are confident with the principles of hand-tracing algorithms it is advised to but understanding how the
practise with each of the examples in this chapter. algorithm works makes it
simpler to complete a trace
table. You should make
Now test yourself sure you are confident
identifying the shortest path
23 Explain the purpose of Dijkstra’s algorithm. to a given node once the
24 State what two values must be stored for each node following the application of trace is complete.
Dijkstra’s algorithm.

Answers available online

68

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_03_MRN_AQA_CS_053-070.indd Page 69 6/29/21 10:00 PM f-0116 /103/HO02206/work/indd

Revision activity
Using a section of the London Underground Walking Tube Map, carry out Dijkstra's
shortest path algorithm to find the shortest route from one station to all other stations

3 Fundamentals of algorithms
in that section.

Summary
✚ Graph traversal algorithms are used to find a path from ✚ RPN is used in interpreters based on a stack; for
one node to another in a graph example, Postscript and bytecode
✚ The breadth first search (BFS) scans all of the nodes ✚ The linear search steps through each item from a
one step away from the start point, then all the nodes list in a line, and is very inefficient, with a time-wise
two steps away, and so on complexity O(n)
✚ The breadth first search uses a queue data structure ✚ The binary search checks the middle item from a list and
and is used to find the shortest path between two discards half of the values each step, making it much
nodes of an unweighted graph more efficient, with a time-wise complexity O(log n)
✚ The depth first search (DFS) continues in one direction ✚ The binary search can only be applied to an ordered list
until it reaches a dead end and then steps backwards ✚ The binary tree search discards an entire branch each
until it finds a new, unused path step, making the search quick and efficient, with a
✚ The depth first search uses a stack data structure and time-wise complexity O(log n)
is used to find a route through a maze ✚ The bubble sort involves swapping pairs of values so
✚ Tree traversal algorithms are used to interrogate each that the largest value will bubble up to the end each
node in a tree pass
✚ The three main tree traversal algorithms look very ✚ The bubble sort can be improved by reducing the
similar, with only the position of the print statement number of comparisons in each step by 1 and by
moving halting if no swaps are made in a full pass
✚ Pre-order tree traversal is used for copying a tree ✚ The bubble sort has a time-wise complexity O(n2)
✚ In-order tree traversal is used to output a sorted list of ✚ The merge sort involves splitting a list into individual
values from a binary search tree and for outputting an items and then continually merging them into sorted
infix notation expression lists of 2, 4, 8, … items, comparing only the front-most
✚ Post-order tree traversal is used for emptying a tree items in each case
and for outputting a reverse Polish notation expression ✚ The merge sort is much more efficient than the
✚ Operators are terms such +, −, AND, OR bubble sort for large, unsorted lists, with a time-wise
✚ Operands are the values on which operators act, such complexity O(n log n)
as numbers (for example, 3, 81.2) or variables (for ✚ Dijkstra’s shortest path algorithm is used to find the
example, x) shortest route from one node to any other node in that
✚ Infix expressions place the operator inbetween its graph
operands ✚ Dijkstra’s algorithm is used for route finding across
✚ Reverse Polish notation (RPN) or outfix expressions weighted graphs
place the operator after its operands
✚ RPN expressions do not need to use brackets and can
be evaluated using a stack

Exam practice

1 This graph represents train routes between different towns, and the distance
between each station in miles. Dave lives in town A and wants to travel to town G.

2 7
A B F
2

5 4 G
4

4
C 6
D 3 E

a) Dave has the choice of using a breadth first search or a depth first search.
State the name of the graph traversal algorithm that would be most suitable to
find the shortest path between two nodes. [1]
b) State what data structure would be most suitable to be used in the
algorithm identified in Question 1a), justifying your choice. [2]
69

My Revision Notes AQA A-level Computer Science Third Edition

325487_03_MRN_AQA_CS_053-070.indd Page 70 6/29/21 10:00 PM f-0116 /103/HO02206/work/indd

c) Give a suitable use for the algorithm that was not chosen in Question part a). [1]
d) Dave is advised to used Dijsktra’s shortest path algorithm since he travels by
train a lot. Explain why using Dijkstra’s algorithm will save time overall. [2]
2 This array, Cars, contains the names of car manufacturers.
3 Fundamentals of algorithms

[0] [1] [2] [3] [4] [5]

AMG BMW Corvette Ferrari McLaren Porsche

Below is a pseudo-code algorithm for a search algorithm.

Start ← 0
End ← Length(Cars) – 1
Mid ← (Start + End) // 2
WHILE Start <= End AND Cars[Mid] != Target
IF Cars[Mid] < Target THEN
Start ← Mid + 1
ELSE
End ← Mid – 1
ENDIF
Mid ← (Start + End) // 2
ENDWHILE
IF Cars[Mid] = Target THEN
OUTPUT "Car found"
ELSE
OUTPUT "Car not found"
ENDIF
a) Copy and complete the trace table below by following the given algorithm. [3]

Target Start End Mid Cars[Mid] Output

Porsche 0 5

b) Copy and complete the trace table below by following the given algorithm. [4]

Target Start End Mid Cars[Mid] Output

Aston Martin 0 5

c) State the name for the algorithm shown. [1]

d) State one feature of the list that makes it possible to use this algorithm. [1]
e) Describe one situation in which a linear search would be more appropriate. [1]
3 This list of city names has been stored in an unsorted array.
Newcastle Berlin Paris Barcelona London Amsterdam
Maariyah has been asked to sort the list of names in alphabetical order and
has access to a bubble sort algorithm and a merge sort algorithm.
a) State which of the two available algorithms is typically most efficient. [1]
b) Describe the situation where the other algorithm would be suitable. [1]
c) Maariyah decides to use the bubble sort.
Explain one method she could use to improve its efficiency and how it
would help. [2]
d) Show the results of the bubble sort algorithm on the list above after one pass. [2]
70

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_04_MRN_AQA_CS_071-102.indd Page 71 6/29/21 9:59 PM f-0116 /103/HO02206/work/indd

4 Theory of computation

Abstraction and automation

Problem-solving
The ability to solve logic problems is an important part of being a good
Logic problem A puzzle
programmer, and logic puzzles can take many forms.
which is intended to
Exam tip be solved using logical
reasoning.
Any form of logic puzzle could be used in an exam paper and past examples include
murder mystery puzzles, riddles and even sudoku style puzzles. Syllogism An argument
that uses logical reasoning.

Logical syllogisms are a test of logical understanding in which two statements

are made and logical deductions must be made. The logical deductions can be
true, false, or uncertain. To help avoid any real-world bias the statements are
usually nonsensical.

Worked example
Statements:
✚ All elephants are green.
✚ Some elephants are big. Exam tip
Proposed deduction: Try looking online for
✚ All big things are green. puzzles, riddles and logic
problems to help prepare
Response:
for this topic, which is often
The deduction is uncertain (or does not follow) because we can’t know if there are covered in the first question
other big things that are not elephants. in Component 1.

Now test yourself

1 All sweets are blue. All sweets are small. Which of these statements follows?
Revision activity
A All small things are blue.
B All blue things are sweets. Search the internet for logic
C Some blue things are small. puzzles such as:
✚ sudokus
2 A farmer is travelling with a wolf, a goat and a cabbage. If left alone the wolf will eat
✚ logic problems
the goat and the goat will eat the cabbage. The farmer must cross a river in a boat,
✚ syllogisms
but can only take one companion at a time. How can the farmer get all three across?
and practise solving them.
Answers available online

Following and writing algorithms

Understanding algorithms
An algorithm is a sequence of steps that can be followed in order to complete
Algorithm A sequence of
a task. An algorithm always terminates (in other words, it stops).
steps that should be carried
Algorithms can be represented using pseudo-code, which is a form of out in order.
program code but is not written in any particular programming language.
Pseudo-code A simplified
Pseudo-code is intended to be read and understood by programmers, who can notation similar to program
then translate the intentions of the program designer into a working program code, used in program
in a specific language. design.
This is particularly relevant to the Component 1 exam as students in different 71
centres will be programming using different programming languages.

My Revision Notes AQA A-level Computer Science Third Edition

325487_04_MRN_AQA_CS_071-102.indd Page 72 6/29/21 9:59 PM f-0116 /103/HO02206/work/indd

Pseudo-code doesn’t always have strict rules about the exact syntax that is
used, although AQA questions are written using a consistent format.
It is important to be familiar with reading and understanding each of the
following programming constructs in pseudo-code:
4 Theory of computation

Sequence The concept that the instructions are carried out in the order that Sequence Executing
they are written. instructions in the order
Assignment The concept that a value, pointer, or the result of a process can be they are written.
assigned to a variable; for example: Assignment Changing the
Total ← Total + ExtraValue value stored in a variable.
Selection The concept that a decision over whether to execute a particular Selection Using conditions
part of the program can be made based on a condition; for example: to decide whether to
execute part of a program.
IF Total < 100 THEN
Iteration Repeating or
…
looping through a section
ENDIF of code.
Iteration That concept that a section of code can be repeated; for example:
FOR i ← 0 to 5
…
ENDFOR

It is important that you can read an algorithm represented in pseudo-code. It

Exam tip
is possible that you will be given a pseudo-code algorithm in Section B which
you will be asked to implement in your chosen programming language, If you are asked to
particularly if you are sitting the AS-level examination. However, at A-level it implement a pseudo-
is more common that you will be given an explanation of an algorithm rather code algorithm it is very
than pseudo-code, which makes this task more challenging. important to use the
variable identifiers given in
Revision activity the program, and to make
sure that any prompts are
Look through past papers for examples of pseudo-code and implement these in your
written exactly as they
chosen programming language.
appear in the exam paper.

Trace tables
One way in which your ability to read and understand algorithms is assessed
Trace table A table for
is using a trace table to hand-trace an algorithm.
recording the values of
In these questions a pseudo-code algorithm is provided and the task is to fill variables as a program runs.
in a table showing how the values of the variables in the program change.
Hand-trace Simulating the
The key advice for completing trace tables is: execution of an algorithm
✚ use your finger to record where you are up to in the algorithm at all times without running it, using a
✚ write the values of the variables as they change from left to right, moving trace table to record the
to a new line when necessary state of the program after
✚ only write down a value when that variable changes. each instruction.

For example, hand-tracing the algorithm below using the following list of
values:
List

[0] [1] [2] [3] [4]

"Apple" "Banana" "Grape" "Orange" "Pear"

Target ← Banana
Start ← 0

72 End ← Length(Items) – 1

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_04_MRN_AQA_CS_071-102.indd Page 73 6/29/21 9:59 PM f-0116 /103/HO02206/work/indd

Mid ← (Start + End) // 2

WHILE Start <= End AND Items[Mid] != Target
IF Items[Mid] < Target THEN

4 Theory of computation
Start ← Mid + 1
ELSE
End ← Mid – 1
ENDIF
Mid ← (Start + End) // 2
ENDWHILE

Target Start End Mid Items[Mid]

"Banana" 0 4 2 "Grape"
1 0 "Apple"
1 1 "Banana"

Because you should always read left to right through the table, the final
change to the value of Start takes place on a new line, and it is not necessary
to replace the value of End.
It is also common to be asked to identify the purpose of an algorithm.
More complex algorithms lead to more complex trace tables and it is common
for some data to be filled in, and those boxes greyed out, in order to help
students check that their progress through the trace table is correct.
More complex trace tables often refer to the standard algorithms discussed in
Chapter 3. Common examples include the breadth first search and depth first
search.

Breadth first traversal

Figure 4.1 is an example of a graph. The pseudo-code algorithm for a breadth
B
first traversal of the graph is shown underneath. D
PROCEDURE Breadth(Start, Target)
A
Queue.Enqueue(Start) E
C
Discovered[Start] ← True
Complete ← False
DO F
Current ← Queue.Deqeue()
Figure 4.1
FOR each node Temp adjacent to Current DO
IF Discovered[Temp] = False AND
Complete = False THEN
Queue.Enqueue(Temp)
Discovered[Temp] ← True
Parent[Temp] ← Current
IF Temp = Target THEN
Complete ← True
ENDIF
ENDIF
ENDFOR
WHILE Queue Not Empty AND Complete = False
73
ENDPROCEDURE

My Revision Notes AQA A-level Computer Science Third Edition

325487_04_MRN_AQA_CS_071-102.indd Page 74 6/29/21 9:59 PM f-0116 /103/HO02206/work/indd

The task is to trace the algorithm to show how the graph is traversed when
given the call Breadth(A,E).

Note
4 Theory of computation

This particular algorithm is designed for an unweighted graph.

Making links
For a full and detailed explanation of how to complete this trace table see:
www.hoddereducation.co.uk/myrevisionnotesdownloads

The full trace table is shown below.

Queue Start Target Discovered Complete Current Temp Parent

Front Back A B C D E F A B C D E F
A C F F F F F F - - - - - -
A T F
- A
B
B T A
F
BF T A
F B
A
C
FC T B
T

Working backwards from node C we can see that the parent node is node B,
and from node B the parent node is node A. Therefore the route found is
A→ B → C.

Depth first traversal

Figure 4.2 shows another example algorithm, this time for a depth first
traversal of the same graph.

B
D

A
C E

Figure 4.2

74

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_04_MRN_AQA_CS_071-102.indd Page 75 6/29/21 9:59 PM f-0116 /103/HO02206/work/indd

PROCEDURE Depth(Current,Target)
Stack ← [Empty]
Discovered ← [False]

4 Theory of computation
Stack.Push(Start)
DO
Current ← Stack.Pop()
IF Not Discovered[Current] THEN
Discovered[Current] ← True
FOR each node Temp adjacent to Current
AND not in Stack DO
IF Not Discovered[Temp] DO
Stack.Push(Temp)
ENDIF
ENDFOR
ENDIF
WHILE Stack is Not Empty And Current != Target
ENDPROCEDURE
The task is to trace the algorithm to show how the graph is traversed when
given the call Depth(A,C).
The first three lines of the algorithm have already been completed and the Note
initial state of the program has been highlighted in grey in order to indicate
This particular algorithm is
that no changes should be made to those cells. A typical exam question
designed for an unweighted
might include the values of variables Current and Target in order to support
graph.
students in ensuring they are carrying out the algorithm correctly.

Current Target Stack Discovered Temp

Bottom Top A B C D E F
A C A F F F F F F
A -
B
BF
F B T
BA
BAE
E BA T
BAC
BACF
F BAC
C BA T
BAD
BADE

Making links
For a full and detailed explanation of how to complete this trace table see:
www.hoddereducation.co.uk/myrevisionnotesdownloads
75

My Revision Notes AQA A-level Computer Science Third Edition

325487_04_MRN_AQA_CS_071-102.indd Page 76 6/29/21 9:59 PM f-0116 /103/HO02206/work/indd

Now test yourself

3 Look at the graph (Figure 4.3) and the pseudo-code underneath.

A B C
PROCEDURE Breadth(Start, Target)
4 Theory of computation

Queue.Enqueue(Start)
Discovered[Start] ← True D E F

Complete ← False
DO
Current ← Queue.Deqeue() Figure 4.3
FOR each node Temp adjacent to Current DO
IF Discovered[Temp] = False AND Complete = False THEN
Queue.Enqueue(Temp)
Discovered[Temp] ← True
Parent[Temp] ← Current
IF Temp = Target THEN
Complete ← True
ENDIF
ENDIF
ENDFOR
WHILE Queue Not Empty AND Complete = False
ENDPROCEDURE
a Copy and complete this trace table to show the steps involved in following the breadth first search. The function
call is ShortestPath(A,E).

Queue Discovered Parent

Start Target Complete Current Temp
Front Back A B C D E F A B C D E F
A E F F F F F F - - - - - -
A
B

C D

C D E

b Copy and complete the trace table below to show the steps involved in following the depth first search. The
function call is DFS(A,E).
PROCEDURE Depth(Start,Target)
Stack <- [Empty]
Discovered <- [False]
Stack.Push(Start)
DO
Current <- Stack.Pop()
IF Not Discovered[Current] THEN
76

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_04_MRN_AQA_CS_071-102.indd Page 77 6/29/21 9:59 PM f-0116 /103/HO02206/work/indd

Discovered[Current] <- True

FOR each node Temp adjacent to Current AND not in Stack DO
IF Not Discovered[Temp] DO

4 Theory of computation
Stack.Push(Temp)
ENDIF
ENDFOR
ENDIF
WHILE Stack is Not Empty And Current != Target
ENDPROCEDURE

Current Target Stack Discovered Temp

Bottom Top A B C D E F
A E F F F F F F

Answers available online

Trace tables with recursion Recursive An algorithm

When tracing recursive algorithms it is usual to have a column for the call which calls itself.
number. Previous calls are then repeated when data is returned and new calls
to the same subroutine are given new, unique numbers.
In the example below, an algorithm has been written to traverse the binary
tree shown in Figure 4.4. Cat

An example implementation might look like this:

Budgie Fish
FUNCTION BinaryTreeSearch (Node)
IF Node.Value = Target THEN
Dog Rabbit
Return True
IF Node.Value > Target AND Exists(Node.Left) THEN Figure 4.4 A binary tree

Return BinaryTreeSearch(Node.Left)
ELSEIF Node.Value < Target AND Exists(Node.Right)
THEN
Return BinaryTreeSearch(Node.Right)
ELSE

77

My Revision Notes AQA A-level Computer Science Third Edition

325487_04_MRN_AQA_CS_071-102.indd Page 78 6/29/21 9:59 PM f-0116 /103/HO02206/work/indd

Return False
ENDIF
ENDFUNCTION
4 Theory of computation

It can be helpful to see the sequence of a recursive algorithm, as follows:

Call 1 – Node ‘Cat’ Call 2 – Node ‘Fish’ Call 3 – Node ‘Rabbit’

FUNCTION BinaryTreeSearch
(Node)
IF Node.Value = Target
THEN
Return True
IF Node.Value > Target
AND Exists(Node.Left) THEN
Return
BinaryTreeSearch(Node.Left)
ELSEIF Node.Value <
Target AND Exists(Node.Right)
THEN
Return
BinaryTreeSearch(Node.Right)
FUNCTION BinaryTreeSearch
(Node)
IF Node.Value = Target
THEN
Return True
IF Node.Value > Target
AND Exists(Node.Left) THEN
Return
BinaryTreeSearch(Node.Left)
ELSEIF Node.Value <
Target AND Exists(Node.Right)
THEN
Return
BinaryTreeSearch(Node.Right)
FUNCTION BinaryTreeSearch
(Node)
IF Node.Value = Target
THEN
Return True
IF Node.Value > Target AND
Exists(Node.Left) THEN
Return
BinaryTreeSearch(Node.Left)
ELSEIF Node.Value <
Target AND Exists(Node.Right)
THEN
Return
BinaryTreeSearch(Node.Right)
ELSE
Return False

ENDIF
ENDFUNCTION
ELSE
Return False
ENDIF
ENDFUNCTION

78

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_04_MRN_AQA_CS_071-102.indd Page 79 6/29/21 9:59 PM f-0116 /103/HO02206/work/indd

ELSE
Return False
ENDIF
ENDFUNCTION

4 Theory of computation
If the target was ‘Rabbit’ then the trace table would look like this:

Target Call Node Value returned

Dog 1 Cat
2 Fish
3 Rabbit True
2 Fish True
1 Cat True

Worked example
Here is an example of another pseudo-code algorithm and its final trace table.
FUNCTION Fact(x)
IF x < 2 THEN
RETURN 1
ELSE
RETURN x * Fact(x – 1)
ENDIF
ENDFUNCTION

Call x Return
1 4
2 3
3 2
4 1 1
3 2 2
2 3 6
1 4 24

The final result indicates that 4! = 24, which is correct.

Worked example
A more complex example might include more than one recursive call. For example:
FUNCTION Fib (int x)
IF x < 2 THEN
RETURN 1
ELSE
RETURN Fib(x-1) + Fib(x-2)
ENDIF
ENDFUNCTION
As with the non-recursive trace tables, it is common to be provided with pre-
completed values highlighted in order to help you check that you are completing the
table correctly.
79

My Revision Notes AQA A-level Computer Science Third Edition

325487_04_MRN_AQA_CS_071-102.indd Page 80 6/29/21 9:59 PM f-0116 /103/HO02206/work/indd

Call x Fib(x-1) Fib(x-2) Return

1 3
2 2
4 Theory of computation

3 1 1
2 2 1
4 0 1
2 2 1 1 2
1 3 2
5 1 1
1 3 2 1 3

A full work through of these trace tables is provided online.

Making links
For a full and detailed explanation of how to complete this trace table see:
www.hoddereducation.co.uk/myrevisionnotesdownloads
For more on recursive algorithms, including general case and base case, see Chapter 1.

Now test yourself

4 Complete the trace table below by hand-tracing the algorithm, from the FUNCTION
Fact (X) algorithm in the first Worked example above, when the call Fact(5) is made.

Call x Return
1 5
2
3
4
5
4
3
2
1

5 Complete the trace table below by hand-tracing the algorithm, from the FUNCTION
Fib (X) algorithm in the second Worked example above, when the call Fib(4) is made.

Call x Fib(x-1) Fib(x-2) Return

1 4
2
3
4
3
5
3
2
80

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_04_MRN_AQA_CS_071-102.indd Page 81 6/29/21 9:59 PM f-0116 /103/HO02206/work/indd

Call x Fib(x-1) Fib(x-2) Return

6
2

4 Theory of computation
1
7
8
7
9
7
1

Answers available online

Making links
Tree traversal algorithms are
Revision activity covered in Chapter 3
✚ Find pseudo-code algorithms for each of the key algorithms in Chapter 3 and hand- Factorials are covered later
trace them, using your own lists as inputs. in this chapter (See Maths
✚ Practise tracing recursive algorithms in particular, including tree-traversal and for understanding Big-O
factorial algorithms. notation).

Abstraction
Abstraction is a method of reducing the complexity of a problem.
Abstraction Making
This makes the problem easier to understand, and easier to solve. a problem simpler by
removing or hiding features.
There are a number of different techniques, each discussed below.

Representational abstraction Note

In order to design a solution to a real-world problem it is necessary to remove
It is important to be able
any features or details that are unnecessary.
to identify and describe
A famous example of this is the London Underground Tube Map, in which each technique in an exam
geographical features are hidden and the resulting map is not drawn to scale question, but it is also very
in order to make it easier for commuters to understand and navigate. useful to be able to use and
apply your understanding –
particularly as part of the
Generalisation NEA programming project.
Another method of abstraction involves identifying the key features or
characteristics of a problem in order to identify the type of problem.
When trying to navigate the London Underground, for example, the problem
can be identified as a graphing problem, with stations as nodes and tube lines
as edges.

Information hiding
When designing a solution it is helpful to hide any details that that do not
need to be accessed by other parts of the program.
In procedural-oriented programming this can be carried out by using local
variables which are not accessed by other subroutines.
In object-oriented programming this can be carried out by declaring attributes
and methods as private in order to hide them from other objects
81

My Revision Notes AQA A-level Computer Science Third Edition

325487_04_MRN_AQA_CS_071-102.indd Page 82 6/29/21 9:59 PM f-0116 /103/HO02206/work/indd

Procedural abstraction
In order to avoid repeatedly typing very similar code to carry out tasks that
rely on the same method over and over again, it is good practice to use a
4 Theory of computation

subroutine to abstract the process.

Procedural abstraction describes a method of ensuring that subroutines are
as generalised and re-usable as possible. For example, a subroutine called add
One To Users Score is extremely specific, however a subroutine called add
Two Numbers can be re-used in a wider range of scenarios.
Best practice in procedural abstraction is to separate the user interface
(prompts and responses) and the processing of data into separate subroutines,
and to use generic identifiers for both the subroutine names and the variables
used. For example, DisplayGreeting(Greeting) is poorly designed, whereas
DisplayMessage(Message) is more re-usable.

Functional abstraction
Functional abstraction aims to hide the complexity of a specific algorithm, or
how a part of the program works, using a subroutine. A programmer can then
call that subroutine without needing to know exactly how it functions.
A good example is a method to print a message to the screen. This is actually
a very complex process that involves re-drawing the screen with the new
data. However it is often the very first program written by new programmers
and the complexity is almost completely hidden from the programmer by
putting the code inside a procedure called print.
When a programmer wants to carry out this very complex procedure, they
can simply call print (for example, print("Hello World")) - the need to
understand the detail of how the message is printed is avoided.

Data abstraction
In data abstraction, because the details of how the data are actually stored
in a computer are hidden, it is possible to create structures that behave in a
particular way.
It is straightforward to understand how structures such as stacks, queues,
graphs and trees work at a conceptual level. However, these structures are not
really stored in the computer as stacks, queues, graphs or trees.
For instance:
✚ Implementing a circular queue can be achieved using a one-dimensional
array and two pointers.
✚ Once implemented, a programmer can then interact with this circular
queue using subroutines such as enqueue, dequeue and peek.
✚ However, the programmer does not need to be concerned with the way
that the data is actually stored within the array – they just need to know
how to use the enqueue, dequeue and peek subroutines.

Problem abstraction/reduction
Problem abstraction involves removing unnecessary detail from the problem
in order to reduce the problem to one that has already been solved.
One example might be that the owner of a chain of shops may want to visit
each of their stores in order to award a prize for the employee of the month at
each location.
By removing the unnecessary detail of the purpose of the visit, this can be
identified as an example of the travelling salesman problem, for which a
82
number of potential solutions already exist.

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_04_MRN_AQA_CS_071-102.indd Page 83 6/29/21 9:59 PM f-0116 /103/HO02206/work/indd

Decomposition
Decomposition is the process of taking a large problem and breaking it down
Decomposition Breaking
into smaller and smaller sub-problems until each sub-problem covers exactly
a problem into smaller sub-

4 Theory of computation
one task.
problems.
A programmer can then create a subroutine for each task in order to create a
complete solution.
This technique doesn’t necessarily reduce the overall level of complexity,
but is intended to provide a method of breaking the problem into more
manageable chunks.

Composition
Composition refers to the concept of combining features together.
Composition Combining
One example is using a subroutine which calls another subroutine. For parts of a solution together
example, a subroutine to start a game of cards may be composed of to create a solution made of
subroutines to shuffle the deck and then deal the cards to the players. component parts.

Another example is the concept of combining data objects. For example, a

binary tree data structure can be composed of three one-dimensional arrays;
one containing the values for each node, one containing the index for the left
branch and one containing the index for the right branch.

Automation
Automation is the overall goal of applying some or all of the abstraction
Automation Designing,
methods discussed above.
implementing and executing
In order to solve a problem, it is necessary to: a solution to solve a
✚ design an algorithm to solve the problem problem automatically.
✚ implement the data structures to store the data that must be stored
✚ implement the algorithms using program code
✚ execute the code.
By completing these four steps it is possible to automate processes and to
model real-world problems and scenarios in a meaningful and useful way.

Now test yourself

6 Describe the difference between procedural and functional abstraction.

7 State which method of abstraction is used to remove unnecessary aspects of the
problem.
8 Explain the difference between problem abstraction and generalisation.
9 What type of abstraction is used to allow the creation of stacks and queues using
an array and some pointers?
10 List the four steps involved in automation.

Answers available online

Regular languages
Finite state machines (FSM) with and
without output
A finite state machine can be used to represent potential states within a
Finite state machine
system in order to help the programmer understand the task more clearly.
(FSM) A computational
For this use, each state represents a real-world state. For instance, in a model which has a fixed
83
computer program that controls a lift the states might be: number of potential states.

My Revision Notes AQA A-level Computer Science Third Edition

325487_04_MRN_AQA_CS_071-102.indd Page 84 6/29/21 9:59 PM f-0116 /103/HO02206/work/indd

✚ stopped at a floor ✚ travelling down

✚ travelling up ✚ overloaded.
Another use for a finite state machine is to check whether a particular series
of inputs is acceptable.
4 Theory of computation

State Transition Diagrams – no output State transition diagram

Finite state machines are often represented as a state transition diagram. A diagram showing
the states, inputs and
a transitions in a FSM.
S0 S1

b, c a, c

a, b, c S4 b b

a
a, b
c
S3 S2

Figure 4.5 A state transition diagram for a finite state machine with no outputs

The state transition diagram above shows the following features.

✚ Each circle represents a state.
✚ Each arrow represents a transition.
✚ Each label represents an input.
The FSM starts at state S0 – indicated by the arrow.
State S3, with two concentric circles, is the accepting state.
State S4, with no way to transition out, is a trap state.
An acceptable input would end at S3, because this is the accepting state.
This diagram is used to check whether a series of inputs is valid or not; for
example:
✚ a is not acceptable because it ends at S1
✚ b is not acceptable because it ends at S4
✚ aa is not acceptable because it ends at S4
✚ abc is acceptable because it ends at S3
✚ abcc, abccc, abcccc, etc. are all acceptable
✚ abca is not acceptable because it ends at S4
The finite state machine can also be represented as a state transition table,
State transition table
with each row indicating the starting state, input and next state.
A table showing the states,
Part of a state transition table to represent the same finite state machine is inputs and transitions in
shown below: a FSM.

Starting State Input Next State

S0 a S1
S0 b S4
S0 c S4
S1 a S4
S1 b S2
S1 c S4
… … …

84

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_04_MRN_AQA_CS_071-102.indd Page 85 6/29/21 9:59 PM f-0116 /103/HO02206/work/indd

Now test yourself Revision activity

11 What type of state is shown with two concentric circles? Create state transition
12 Identify two methods for representing a finite state machine. diagrams for everyday

4 Theory of computation
devices, such as hairdryers
13 A finite state machine can be used to represent the states of a real-world problem.
and ovens.
Suggest one other use for an FSM.

Answers available online

State transition diagrams – with output

The finite state machines with outputs that you will see are also called Mealy
Mealy machine A finite
machines. These are similar to the previous example, except that the labels
state machine in which each
on the transitions show both an input and an output.
input has a corresponding
1|1 output as well as a transition
between states.

S1
1|0

S0 1|0 0|1

0|0
S2

0|0
Figure 4.6 A state transition diagram showing a Mealy machine

The first value in the label for a transition is the input, and the second is the
Exam tip
output.
Starting at state S0, an input of 1 will give an output of 0. It is very important to read
the question very carefully.
This finite state machine does not have an accepting state, as its purpose is In the example above,
not to validate an input, but to display an output based on its input. reading the input from right
In this particular case an input of 0110, reading from left to right, would give to left would give a different
an output of 0011. answer – the output 1100,
carrying out a left logical
The purpose of this finite state machine is to carry out a right logical shift.
shift. You need to pay
Mealy machines can also be described using a state transition table, with an careful attention to these
extra column indicating the output. details in the question.

Starting state Input Output Next state

S0 0 0 S2
S0 1 0 S1
S1 0 1 S2
S1 1 1 S1
S2 0 0 S2
S2 1 0 S1
Revision activity
Now test yourself
Create state transition
14 What name is given to a finite state machine where each input has a diagrams to demonstrate
corresponding output? binary operations such
as left shifts, right shifts,
15 Explain why finite state machines with outputs do not always have an accepting state.
and twos complement
Answers available online conversion. 85

My Revision Notes AQA A-level Computer Science Third Edition

325487_04_MRN_AQA_CS_071-102.indd Page 86 6/29/21 9:59 PM f-0116 /103/HO02206/work/indd

Maths for regular expressions

In order to better understand regular expressions, it is important to
Set theory A branch of
understand some relevant mathematical principles relating to set theory.
mathematical logic in which
4 Theory of computation

You should not be assessed specifically on this mathematical understanding, sets are collections of
but it will form part of the basis for the regular expressions topic. objects.
A set is an unordered collection of values, with no duplicates. Set comprehension A
A = {1, 2, 3, 4, 5} collection of rules to define
which values are in a set.
A set can also be described using a set of rules, known as set comprehension
or set-builder notation.
These symbols are used in set comprehension:
Making links
| ‘such that’ Set theory, and particularly
∈ ‘is a member of’ the sets of (natural
∧ logical operator for ‘and’ numbers),  (integers), 
(rational numbers) and 
≤, ≥, < > logical operators for inequalities (real numbers), are essential
to understanding the maths
Here is the set comprehension for A:
that underpins regular
A = { x | x ∈  ∧ x ≥ 1} expressions. This topic is
This translates as: A is the set of x, such that x is a member of the set of explained in more detail in
natural numbers and x is greater than or equal to 1. Chapter 5.

An empty set can be described using {} or Ø.

A compact representation of a set can be listed without needing to specify x.
A = { anbn | n > 0 ∧ n < 4 } would give the set {ab, aabb, aaabbb}

Now test yourself

16 Which one of these statements is true?

a Sets must be written in order.
b Sets cannot contain duplicate values.
c Sets cannot be empty.
17 What values would be in each of the following sets?
a A = { x | x ∈  ∧ x < 3 ∧ x > -2 }
b B = { x | x ∈  ∧ x/2 < 2 }
c C = { bna2bn | n ≥ 1 ∧ n ≤ 3 }

Answers available online

Countable set A set with

Categorising sets the same number of values
A countable set is any set with the same number of elements as some subset as a subset of the natural
of the natural numbers. For example, a set with 72 elements is countable. numbers .
Some sets are finite sets – they have a fixed number of values and their Finite set A set with a fixed
elements can be counted off in a similar way to an index in an array (for number of values.
example, 1st element, 2nd element, 3rd element, and so on).
Infinite sets A set whose
The number of values in a set is referred to as its cardinality. values will go on forever.
Some sets are infinite sets, and will carry on forever. Countably infinite sets
✚ Some infinite sets are countable sets; for example, the set of natural A set whose values will go
numbers – these can be called countably infinite sets. on forever, but can still be
✚ Some infinite sets are not countable sets; e.g., the set of real numbers. counted (for example, ).
The Cartesian product of two sets (written A x B) contains the set of ordered Cardinality The number of
pairs of values in A and B. values in a finite set.
A x B = { (a,b) | a ∈ A ∧ b ∈ B} Cartesian product The set
A = {1, 2} and B = {4, 5, 6} of ordered pairs of values
86 from both sets.
A x B = { (1, 4), (1, 5), (1, 6), (2, 4), (2, 5), (2, 6) }

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_04_MRN_AQA_CS_071-102.indd Page 87 6/29/21 9:59 PM f-0116 /103/HO02206/work/indd

A subset is made up of some elements of the main set. The main set is
Subset A selection of the
sometimes called the superset.
values found in a set.
Subsets can be described in one of these three ways:
✚ A ⊆ B means ‘A is a subset of B’.

4 Theory of computation
Exam tip
This means that everything in A is also in B. They may or may not be
equal. For example: Remember that the sign for
a subset, ⊆, incorporates
{0,1} ⊆ {0,1,2} and {0,1,2} ⊆ {0,1,2} an underscore in the same
✚ A ⊂ B means ‘A is a proper subset of B’. way as the ‘greater than
or equals’ symbol, ≥. This
This means that everything in A is also in B, but they are not equal. There
indicates that the subset
is at least one other value in B.
can be equal to its superset.
{0,1} ⊂ {0,1,2}
The symbol for a proper
✚ A = B means ‘A is identical to B’. subset, ⊂, does not
{0,1,2} = {0,1,2} incorporate an underscore
and cannot be the same as
Now test yourself its superset.

18 State whether each of these sets is finite, infinite or countably infinite

a A = { x | x ∈  }
b B = { x | x ∈  ∧ x > 0 ∧ x < 10}
c C = { x | x ∈  ∧ x > 0 ∧ x < 10}
19 What symbol is used to represent a subset?
20 What symbol is used to represent a proper subset?
21 A = {r, s, t}, B = {q}, C = {r, t, s}
a State whether C is a subset of A and explain why
b State whether C is a proper subset of A and explain why
c Write down the Cartesian product of A x B

Answers available online

Set operations
For each example, A = {1, 2, 3} and B = {3, 4, 5}.
✚ Membership of a set indicates that a value is in that set.
1 is a member of A.
✚ Union (represented as ∪) means joining both sets together and including
all elements, see Figure 4.7.
The ∪ symbol is similar to the logical ∨ symbol, indicating OR.
Figure 4.7
A ∪ B = {1, 2, 3, 4, 5}
✚ Intersection (represented as ∩) means joining both sets together and
including only those elements common to both, see Figure 4.8.
The ∩ symbol is similar to the logical ∧ symbol, indicating AND.
A ∩ B = {3}
✚ Difference (represented as , Δ or \) means joining both sets together and
including only those elements that are not common to both, see Figure 4.9. Figure 4.8
A \ B = {1, 2, 4, 5}

Now test yourself

22 A = {r, s, t} and B = {p, q, r, s}

List the values in the following sets.
a A ∩ B
b A ∪ B
c A \ B Figure 4.9

Answers available online 87

My Revision Notes AQA A-level Computer Science Third Edition

325487_04_MRN_AQA_CS_071-102.indd Page 88 6/29/21 9:59 PM f-0116 /103/HO02206/work/indd

Regular expressions
Regular expressions are a way of describing a set. They are also used to
Regular expression A
describe particular types of languages. They can be extremely useful to
sequence of characters
4 Theory of computation

identify items that meet certain criteria; for example, for validation checks.
used to describe a pattern.
Symbols used in regular expressions include: Used in searching and
✚ | meaning ‘or’; for example, a | b validation.
The set of valid inputs would be {a, b}
✚ + meaning ‘1 or more’; for example, ab+
The set of valid inputs would include {ab, abb, abbb, abbbb, …}
✚ * meaning ‘0 or more’; for example, ab*
The set of valid inputs would include {a, ab, abb, abbb, …}
✚ ? meaning ‘0 or 1’; for example, ab?
The set of valid inputs would be {a, ab}

Regular expressions can be represented using finite state machines, and it is

common for questions to appear in which a finite state machine should be
described as a regular expression.
a

b c
S0 S1 S2

a, b
c
a, b, c

S3

a, b, c
Figure 4.10 A finite state machine

In this machine the acceptable inputs include 0 or more ‘a’ values followed by
a single ‘b’ and a single ‘c’. Any other inputs would be sent to the trap state.
Therefore, the acceptable inputs would be written as: a*bc
b
a
S0 S1 S2

c
c
b, c a, b

S3

Figure 4.11 A finite state machine

In this finite state machine the accepting state can be reached by inputting ‘a’
followed by 0 or more repetitions of ‘bc’.
This would be written as a(bc)*

Regular languages
A formal language is made up of words that contain letters from the alphabet
and where those words fit a set of rules.
A regular language is a formal language that can be represented using a
regular expression. If the language cannot be represented using a regular
expression then it is not a regular language.
88
A regular language can also be represented using a finite state machine.

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_04_MRN_AQA_CS_071-102.indd Page 89 6/29/21 9:59 PM f-0116 /103/HO02206/work/indd

Now test yourself

23 Explain the meaning of each symbol when used in a regular expression:

a +

4 Theory of computation
b *
c ?
d | Revision activity
24 Identify another way to describe a regular expression. ✚ For each regular
25 List the values which would be accepted by each regular expression: expression you can find,
a ab(c|d) create the equivalent
finite state machine.
b ab?cd
✚ For each finite state
c (ab)|(cd+)
machine you can find
d ab*a
with an accepting state,
26 What type of language can be described using a regular expression? create the equivalent
regular expression.
Answers available online

Context-free languages
Backus–Naur form (BNF) and syntax
diagrams
Backus–Naur form
Backus–Naur form (BNF) is a method of describing syntax rules.
Backus–Naur form (BNF)
In BNF, each type of value is described as containing values from a set. A notation used to describe
the syntax of a language.
Each individual production rule is replaced with (::=) certain values. The
symbol ::= can be read ‘is defined as’. Syntax The set of rules for
a given language.
Values in <angle brackets> must be further broken down and defined
until all values are made up of terminal values (that is, values that cannot be Production rule The set
broken down any further). The symbol | means ‘or’. of acceptable inputs for a
given symbol.
In the example below, spaces are represented with an underscore.
Terminal A single value
For example, the syntax rules for entering a name could be defined as follows: that cannot be broken down
<fullname> ::= <title> _ <name> into smaller parts.

<title> ::= MR | MISS | MRS | MS

<name> ::= <word> | <name> _ <word>
<word> ::= <letter> | <word> _ <letter>
<letter> ::= A | B | C | D | E | F | G | H | I | J | K
| L | M | N | O | P | Q | R | S | T | U | V
| W | X | Y | Z
Line 1 – the set fullname is defined as a value from the set title followed by
a value from the set name.
Line 2 – the set title is defined as a value from the set of terminal values MR
or MISS or MRS or MS.
Line 3 - the set name is defined as either a value from the set word or as a
value from the set name followed by a value from the set word.
Line 4 – the set word is defined as either a value from the set letter or as a
value from the set word followed by a value from the set letter.
Line 5 – the set letter is defined as a value from the set of terminal values A
or B or C ... through to Z. 89

My Revision Notes AQA A-level Computer Science Third Edition

325487_04_MRN_AQA_CS_071-102.indd Page 90 6/29/21 9:59 PM f-0116 /103/HO02206/work/indd

Recursion is used in lines 3 and 4. To see how this works consider the set
Exam tip
word contains ‘DAVE’.
✚ Line 4 says if a word is not defined as a single letter then it is defined as a When asked in an exam,
word followed by a letter. BNF definitions that do
4 Theory of computation

✚ This would replace the word ‘DAVE’ with the word ‘DAV’ followed by the not use recursion can be
letter E. described using a regular
✚ The new word ‘DAV’ is replaced with the word ‘DA’ followed by the letter expression (such as
‘V’. fullname, title and
✚ The new word ‘DA’ is replaced with the word ‘D’ followed by the letter ‘A’. letter in the previous
✚ Finally, the word ‘D’ is replaced with the a value from the set letter, example). BNF definitions
which contains the terminal value ‘D’. that do use recursion
cannot be described using
The set might contain other values which need to be defined. For example, a regular expression (such
a fullname is made up of further sets title and name. The set might be as name and word in the
defined as containing one of a series of terminal values which do not need to previous example). This is
be defined. a simplification of the real
world but provides the
For example, acceptable titles include Mr, Miss, Mrs and Ms.
expected answer for A-level
The set might use recursion to allow for values that can be of variable length. exam requirements.
For example, a name is made up of a word, or a name followed by a word.
‘Dave’ is a word, and so can be accepted as a name.
‘Dave Andrew’ is a name followed by a word, and so can be accepted as a
name.
‘Dave Andrew Smith’ is a name followed by a word, and so can be accepted as
a name.

Syntax diagrams
Syntax diagrams are an alternative way to represent BNF.
Syntax diagram A
Definitions that contain terminals tend to be quite straightforward, diagram used to describe
demonstrating each possible value. For instance, a digit can be described as: BNF graphically.
Digit ::= 0 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9
Digit
0
1
2
3
4
5
6
7
8
9

Figure 4.12 A syntax diagram for a digit

Definitions that contain recursion are drawn using a loop, so that an integer
can be described as:
Integer ::= <digit> | <digit><integer>
Integer
digit

integer

Figure 4.13 A syntax diagram for an integer

90

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_04_MRN_AQA_CS_071-102.indd Page 91 6/29/21 9:59 PM f-0116 /103/HO02206/work/indd

Now test yourself

The rules for a name are given using the following syntax rules:
<name> ::= <word> | <word> _ <name>

4 Theory of computation
<word> ::= <letter> | <letter> _ <word>
<letter> ::= A|B|C|D|E|F|G|H|…
27 Give one example of a terminal value in this syntax.
28 Explain why the rule for a name cannot be described using a regular expression.
29 Represent the production rule for a word using a syntax diagram.

Answers available online

Revision activity
✚ Identify patterns and rules in constructions such as postcodes and addresses, and
try to represent these using BNF.
✚ For each production rule, create the equivalent syntax diagram.

Classification of algorithms
Comparing algorithms
It is important to be able to compare the efficiency of different algorithms in
Efficiency Being able
order to identify which is the best solution for each particular problem
to complete a task with
As the size of the data to be processed grows, the efficiency of the algorithm the least use of time or
becomes much more important. Sorting a list of 10 values with an inefficient memory.
algorithm may not cause a problem, but when sorting a list of 100 000 values it Time-wise complexity A
is much more important to choose wisely. measure of how much time,
The efficiency of an algorithm can be described by looking at two main aspects: or how many steps, will be
✚ Time-wise complexity: Algorithms that have a lot of steps or comparisons required to complete an
will take longer to compute for large data sets. algorithm.
✚ Space-wise complexity: Algorithms that store a lot of temporary values Space-wise complexity
while the processes take place will take up a lot of memory. A measure of how much
memory will be required to
An ideal algorithm will be efficient in terms of both time to execute and space
complete an algorithm.
needed for memory.
However, depending on the particular situation, sometimes it might be more
important to:
✚ choose an algorithm which is time-efficient even if it uses a lot of memory
to store values temporarily during execution
✚ choose an algorithm which is space-efficient even if it might take longer to
carry out the instructions.

Note
Recursive algorithms are often considered to be elegant solutions as they can be
described using few lines of code. They are typically inefficient however, particularly
space-wise, as the local variables for each call need to be stored while any further
calls are running.

Now test yourself

30 What is the key consideration in time-wise complexity?

31 What is the key consideration in space-wise complexity?
32 Why might a time-efficient algorithm not always the best choice?

Answers available online

91

My Revision Notes AQA A-level Computer Science Third Edition

325487_04_MRN_AQA_CS_071-102.indd Page 92 6/29/21 9:59 PM f-0116 /103/HO02206/work/indd

Maths for understanding Big-O notation

In order to understand Big-O notation it is important to understand some
underlying mathematical concepts.
4 Theory of computation

Functions
A function is a method of mapping one set of values to another.
For example, the function y = 2x maps each value of x to a new value, y.

x y Making links
−3 −6 The issues surrounding
−2 −4 function terminology,
including domains, co-
−1 −2
domains and sets, is
0 y = 2x 0 typically only assessed
1 2 in questions relating to
functional programming.
2 4 Functional programming
3 6 is covered in detail in
Chapter 12.
Figure 4.14 A mapping function

The input values are referred to as the domain, and the output values are
referred to as the co-domain.
If the domain (input values) is made of integers (natural numbers) and the
co-domain is also made of integers then this can be described using the
format → .
Consider each of the following graphs.
✚ The graph of y = 3 represents a constant function.
The value of y is constant and does not go up as x changes
✚ The graph of y = 2x represents a linear function.
Linear Rising in a straight
The value of y goes up in a straight line – so if x doubles, so does y. line; for example, the graph
✚ The graph of y = 2x2 represents a polynomial function. of y = 2x.
The value of x is raised by a power. Other examples include graphs with x3, Polynomial An expression
x4 and so on. involving powers; for
✚ The graph of y = 2x represents an exponential function. example, y = 2x2.
A value is raised by the power of x. Whenever x is used as a power the Exponential An expression
value of y will go up increasingly quickly. where the variable is used
✚ The graph of y = log2 x represents a logarithmic function. as an exponent, or power;
for example, y = 2 x.
The value of y goes up very slowly. Using y = log2 x is the opposite of using 2x.
✚ The graph of y = x! represents a factorial function. Logarithmic The opposite
of exponential. If y = 2 x,
The value of y goes up extraordinarily quickly. then x = log 2y.
1000
900
800
700
600
500
400
300
200
100
0
0 10 20 30 40 50 60 70 80 90 100

Key
y = log x y=x y = n log n y = x2 y = 2x y = x!
92
Figure 4.15 A graph representing different types of function

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_04_MRN_AQA_CS_071-102.indd Page 93 6/29/21 9:59 PM f-0116 /103/HO02206/work/indd

To better understand the meaning of a logarithm, consider this table showing

powers of 2; that is, the first few values of y = 2x:

20 21 22 23 24 25 26 27 28

4 Theory of computation
1 2 4 8 16 32 64 128 256

Applying a logarithm to the same base is the inverse of raising to a power; in

other words:
log2(2x) = x
This can be see if we consider the following values of y = log2 x:

log 2 1 log 2 2 log 2 4 log 2 8 log 2 16 log 2 32 log 2 64 log 2 128 log 2 256 Factorial The product of all
0 1 2 3 4 5 6 7 8 positive integers less than
or equal to a given integer.
Permutation One of the
Factorials and permutations different ways that a set can
It is also important to be familiar with the concept of a factorial. be arranged.
n! (pronounced ‘n factorial’) multiplies all of the positive integers up to and
including n. For example:
Now test yourself
5! = 5 × 4 × 3 × 2 × 1
6! = 6 × 5 × 4 × 3 × 2 × 1 33 Give an example of a
linear function.
The graph of n! grows more rapidly than an exponential function.
34 What type of function is
A factorial is used to describe the maximum number of permutations of a set the graph of y = 2x3?
of values. 35 Which graph rises most
steeply – an exponential
For example, if I want to visit cities A, B and C in any order then I could use
function or a logarithmic
any of the following permutations:
function?
{A,B,C}, {A,C,B}, {B,A,C}, {B,C,A}, {C,A,B}, {C,B,A} 36 What type of function is
If there are three values in the set, then there are 3! (= 6) possible y = 1?
permutations.
Answers available
If there are four values in the set, then there are 4! (= 24) possible permutations. online

Order of complexity
Time-wise efficiency is described using Big-O notation.
This is used to put the time efficiency of different algorithms into some sort of
order, and uses the notation O(n), where n is the number of steps and different
terms are used inside the brackets for each order of complexity.

Constant time
Finding the first item in a list, or jumping straight to a value at a given index,
always takes exactly one step.
The time-wise complexity for an algorithm that always takes the same
number of steps, no matter how big the list, is described using the time-wise
complexity O(1) for one step, O(2) for two steps, O(3) for three steps, and so on –
though O(1) is much more likely.

Logarithmic time
A binary search of n values will usually take very little time because the
possible list of remaining values is halved each time. This halving at each step
means that the maximum possible number of steps taken to find the right
item goes up on a logarithmic scale.
We can, therefore, state that the time-wise complexity of the binary search is 93
O(n log n).

My Revision Notes AQA A-level Computer Science Third Edition

325487_04_MRN_AQA_CS_071-102.indd Page 94 6/29/21 9:59 PM f-0116 /103/HO02206/work/indd

Linear time
In the worst case, a linear search of n items will take n steps. If the item is the
last one in the list then every item much be checked before it is found.
4 Theory of computation

We can therefore state that the time-wise complexity of the linear search is O(n).
Notes
The term loglinear is not
Loglinear (n log n) time
often used – simply writing
The merge sort is a very efficient algorithm which, much like the binary search, n log n is more common.
involves splitting lists in half repeatedly. Sorting is always a more complex task
than search, and so the time-wise complexity is not quite as small. Polynomial time refers to
time-wise complexity where
The time-wise complexity of the merge sort is O(n log n). n is raised to a power; for
example, O(n2), O(n3), O(n4),
Polynomial time and so on.

To sort a list of n values using the bubble sort, each pass would require (n – 1)
comparisons.
To fully sort the list it will be necessary to complete (n – 1) passes. The total
steps would therefore be (n – 1) × (n – 1).
However, for a large list, n ≈ n – 1 (that is, n is almost equal to (n – 1))
So, (n – 1)(n – 1) ≈ n2
We can, therefore, state that the time-wise complexity of the bubble sort is O(n2).
Other algorithms that can be expressed as O(n2) time-wise complexity include
the insertion sort and selection sort.

Exponential time
The time-wise complexity for an algorithm that uses exponential time will
be written as O(kn), where k is a constant and the value is raised to the power
of n.
An example could be trying every possible value of a passcode. If a passcode
uses four digits then there are 104 possible values to try. If the passcode uses
six digits then there are 106 possible values.

Factorial time
Factorial time is even worse than exponential time. The travelling salesman
problem is an example of a problem that has a time-wise complexity of O(n!).
To make it easier to remember these time complexities, and examples of each,
consider this graph:
Exam tip
Intractable Tractable Big-O notation
1000
O(n2) – bubble sort
Remember that searching is
900 O(n!)
always simpler than sorting,
800 O(kn)
and for each topic there
ime

700 O(n log n) – merge sort

is a good algorithm and a
ial t

600 bad algorithm. The good

nom

500 search and the good sort

poly

400 techniques can both be

300 described using logarithms
200 because they use divide
O(n) – linear search
100 and conquer techniques.
O(log n) – binary search
0 The bad search and the
0 10 20 30 40 50 60 70 80 90 100
bad sort both involve going
Figure 4.16 A graph showing the time complexities for different algorithms through every item in the
list, and so scale up with n
(rather than log n).

94

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_04_MRN_AQA_CS_071-102.indd Page 95 6/29/21 9:59 PM f-0116 /103/HO02206/work/indd

Now test yourself

37 What is the time-wise complexity for a merge sort?

38 Other than the bubble sort, give an example of an algorithm which has the time

4 Theory of computation
complexity O(n2).
39 Explain why the linear search has a time-wise complexity of O(n).
40 State the time-wise complexity for the binary search.
41 Explain why the bubble sort has a time complexity of O(n2).

Answers available online

Limits of computation
Although many problems can be solved with enough time and enough memory,
it is important to consider the physical constraints of computing hardware.
A time-efficient algorithm that uses a large amount of memory may not be
able to function effectively once all available RAM is used.
Similarly, while CPUs run at enormously high frequencies, time-complex
algorithms may still take longer than is reasonable to carry out their
processing.
As such, even if a problem can be computed in theory, hardware limits may
mean that it is not possible to compute it in a practical setting.

Classification of algorithmic problems

Problems that can be solved in polynomial or less time are considered to be
Tractable A problem which
tractable. This means that they can be solved in a reasonable amount of time.
can be completed in a
Problems that cannot be solved in polynomial or less time (for example, reasonable amount of time
exponential or factorial time) are considered to be intractable. This means (polynomial time or less).
that they cannot be solved in a reasonable amount of time. See Figure 4.15. Intractable A problem
Exam tip which can be solved,
but not in a reasonable
Exam questions often ask you to explain what is meant by the term intractable. It is amount of time (worse than
vital to point out that intractable problems can be solved, but not in a reasonable polynomial).
amount of time. Both parts of the statement are required, and it is important to
Heuristic An approach to
emphasise that intractable problems can be computed.
solving intractable problems
in a reasonable amount
In order to solve intractable problems, heuristic methods are often used. of time by accepting an
imperfect solution.
These include:
✚ reducing the complexity of the problem
✚ relaxing some of the constraints on the problem
✚ using an algorithm that uses a best-guess or estimate based on experience Exam tip
✚ accepting a close-to-optimal solution.
Make sure you memorise
Note these examples of heuristic
algorithms as this topic
Though not covered on the specification, Prim’s and Kruskal’s algorithms are interesting comes up often.
examples of heuristic solutions and could form part of an interesting NEA project.

Revision activity
Now test yourself
Investigate problems such
42 What is meant by a tractable problem? as the travelling salesman
43 What two features define an intractable problem? problem, the bin-packing
problem and the knapsack
44 Suggest three strategies for tackling an intractable problem.
problem to see examples of
Answers available online intractable problems.
95

My Revision Notes AQA A-level Computer Science Third Edition

325487_04_MRN_AQA_CS_071-102.indd Page 96 6/29/21 9:59 PM f-0116 /103/HO02206/work/indd

Computable and non-computable problems

Many problems are computable, even if they can’t always be completed
Computable A problem
efficiently.
that can be solved with a
4 Theory of computation

Some problems are non-computable. That means that they simply cannot be computer.
solved using a computer.

Halting problem
The halting problem is: given a computer program and an input, is it possible
Non-computable A
to determine whether that program would loop forever or would eventually
problem that cannot be
stop, without running it?
solved by a computer.
A mathematical proof shows that an algorithm to solve the halting problem
Halting problem A
cannot exist. specific example of a non-
Note computable problem that
proves that some problems
The proof for the halting problem is not required for A-level computer science, but it are non-computable.
involves demonstrating that a successful solution would create a paradox, which is
therefore impossible.

The significance of the halting problem is that it demonstrates that there are
some problems that cannot be solved using a computer.

Now test yourself

45 What is a non-computable problem?

46 Explain what is meant by the halting problem.
47 Explain why the halting problem is significant.

Answers available online

A model of computation
Turing machine
A Turing machine is a theoretical computing device with a single, fixed
Turing machine A model
program that is made up of:
of computation with a single
✚ a finite set of states
program which manipulates
✚ a finite alphabet of symbols symbols on a strip of tape.
✚ an infinite length of tape, with marked off squares
✚ a read-write head that can travel across the tape, one square at a time. Starting state The state a
Turing machine is in when it
Because a Turing machine uses a finite set of states, it can be described using starts its program.
a finite state machine (FSM), either using a state transition diagram or a state
transition table. Halting state A state with
no outgoing transitions, and
An FSM should have one state as the starting state. so the Turing machine will
Any states with no outgoing transitions are called halting states. stop.

Empty states can be recorded using a number of symbols, but it is typical for
AQA to use the # symbol.
Each transition is labelled with an input, output and an arrow indicating the
movement of the read-write head.
An example of Turing machine is described below using a state transition
diagram.

96

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_04_MRN_AQA_CS_071-102.indd Page 97 6/29/21 9:59 PM f-0116 /103/HO02206/work/indd

0|0 0|1
1|0
1|1 S1
S0 S1

4 Theory of computation
#|# #|#

SH

Figure 4.17 A state transition diagram representing the operation of a Turing

machine

The diagram in Figure 4.16 could be shown using this state transition table:

Starting state Input Next state Output Movement

S0 0 S0 0 ←
S0 1 S1 1 ←
S0 # SH # →
S1 0 S1 1 ←
S1 1 S1 0 ←
S1 # SH # →

An alternative format that could be used is a transition function. It shows the

starting state and input, followed by the next state, output and movement.
δ (delta) means ‘change’. The first transition function reads: ‘The change if the
state is 0 and the input is 0 is that the state should be set to 0, the output set to
0 and the read-write head should move left’.

δ (S0,0) = (S0, 0, ←) δ (S1,0) = (S1, 1, ←)

δ (S0,1) = (S1, 1, ←) δ (S1,1) = (S1, 0, ←)
δ (S0, #) = (SH, #, ←) δ (S1, #) = (SH, #, →)

The tape for the Turing machine is shown below, with an asterisk showing
the current position or the read-write head.

… # 0 1 1 0* … S0

When the first value is read, the Turing machine is in State 0 and the input
is 0.
Reading the transition function for this situation, the state should remain as
State 0, the output 0 should be written to the tape and the read-write head
should move to the left.
This leads to the result below:

… # 0 1 1* 0 … S0

When the next value is read, the Turing machine is in State 0 – and the input
is 1.
Reading the transition function for this situation, the state should be set to
State 1, the output 1 should be written to the tape and read-write head should
move to the left.

97

My Revision Notes AQA A-level Computer Science Third Edition

325487_04_MRN_AQA_CS_071-102.indd Page 98 6/29/21 9:59 PM f-0116 /103/HO02206/work/indd

This leads to the result below:

… # 0 1* 1 0 … S1
4 Theory of computation

Following the algorithm described by the state transition diagram or by

transition functions produces the following completed trace table:

… # 0 1 1 0* … S0
… # 0 1 1* 0 … S0
… # 0 1* 1 0 … S1
… # 0* 0 1 0 … S1
… #* 1 0 1 0 … S1
… # 1* 0 1 0 … SH

The purpose of this Turing machine is to calculate the twos complement of a Revision activity
binary number.
✚ Try creating your
Now test yourself own Turing machine
to carry out bit-wise
48 What four things are needed for a Turing machine? operations such as twos
49 Give two ways that a Turing machine can be represented complement, left shifts
50 What is the name for a state with no outgoing transitions? and right shifts.
51 A state transition table contains a starting state and input, plus what three other ✚ Find examples of simple
pieces of data? Turing machines online
and hand-trace them
Answers available online using your own inputs.

Universal Turing machines

It is possible to describe an algorithm for any computable problem using a
Turing machine. However, each Turing machine can only solve one problem
because the program is fixed. Universal Turing
machine (UTM) A Turing
A universal Turing machine (UTM) is a Turing machine which can read in the
machine that takes the
description of another Turing machine and its tape in order to simulate that
description of another
Turing machine. Turing machine and its tape
The UTM will then carry out the operations on the tape exactly as the other as inputs. It can simulate
Turing machine does. In other words, it can simulate any other Turing machine. any conceivable Turing
machine.
Because the UTM stores both the description of the Turing machine and its
tape on the UTM’s tape, the UTM acts as an interpreter.
As any single computable problem can be computed by a Turing machine,
therefore every possible computable problem can, in theory, be computed on a
Note
UTM.
A UTM was the theoretical conception of a modern computer, proposed Exam questions will only
be asked about Turing
by computer scientist and mathematician Alan Turing in the 1930s. As a
machines that have one
hypothetical machine it can be considered more powerful than any real
tape that is infinite in one
computer due to the infinite tape, which effectively provides the UTM with an
direction.
infinite amount of memory.

Now test yourself

52 What two inputs does a universal Turing machine take?

53 Why is a universal Turing machine considered more powerful than any real-world
computer?
54 What role does a universal Turing machine carry out when it reads in a Turing
machine’s description?

98 Answers available online

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_04_MRN_AQA_CS_071-102.indd Page 99 6/29/21 9:59 PM f-0116 /103/HO02206/work/indd

Summary
✚ Logic problems can be solved using logical deductions ✚ A regular expression is a notation used to describe a set
✚ An algorithm is a series of steps to complete a task, ✚ Regular expressions are used in searching and validation

4 Theory of computation
that always terminates ✚ The symbols *, +, ?, and | are used in regular
✚ Algorithms written in pseudo-code solve problems expressions and have specific meanings
using the standard programming constructs: sequence, ✚ Regular expressions can also be represented using a
assignment, selection, iteration finite state machine, and vice versa
✚ Algorithms can be hand-traced by completing a trace table ✚ It is possible to be asked to represent a finite state
✚ In recursive algorithms, trace tables should include the machine as a regular expression, and vice versa
call number ✚ A language is classed as a regular language if it can be
✚ It is important to be able to convert the description of represented as a regular expression
an algorithm, whether in English or pseudo-code, into ✚ The syntax rules of a language can be checked using
program code in your chosen language Backus–Naur form or syntax diagrams
✚ Where they are provided, always make sure to use the ✚ Rules in BNF that do not use recursion can be
exact variable names and prompts as they appear in represented using a regular expression
the exam paper ✚ If BNF uses recursion, then it can represent languages
✚ Abstraction is a technique for simplifying a problem that cannot be represented using regular expressions
by removing or hiding complexity and includes: ✚ The efficiency of algorithms can be described time-
representational abstraction, generalisation, information wise and space-wise
hiding, procedural abstraction, functional abstraction, ✚ Functions map one set of values (the domain) to
data abstraction and problem abstraction/reduction another set of values (co-domain)
✚ Decomposition is the technique of breaking a problem ✚ The number of permutations of a set of n objects is n!
into increasingly small sub-problems (where, for example, 4! = 4 × 3 × 2 × 1)
✚ Composition is the technique of combining individual ✚ Big-O notation is used to describe the time-wise
subroutines to create compound subroutines that solve efficiency of algorithms
larger problems ✚ Searching algorithms include the binary search, O(log n)
✚ Automation is the technique of designing, implementing, (logarithmic time), and the linear search, O(n) (linear time)
and executing solutions to complex problems ✚ Sorting algorithms include the merge sort, O(n log n),
✚ A finite state machine is a model of a system that can and the bubble sort, O(n2) (polynomial time)
be described using a fixed number of states ✚ Other algorithms use constant time, O(1), exponential
✚ Each time an input is provided to a finite state machine time, O(kn) and factorial time, O(n!)
there will be a transition to a new state ✚ Tractable problems can be solved in a reasonable
✚ A Mealy machine is a finite state machine where each (polynomial or less) time
input provides a specific output as well as a transition ✚ Intractable problems can be solved, but not in a
between states reasonable time (because of hardware limits, for example)
✚ Finite state machines can be described using state ✚ Heuristic methods such as reducing complexity or
transition diagrams and state transition tables accepting a close-enough solution are used to solve
✚ A set is an unordered list of values, with no duplicates intractable problems
✚ Sets can be described using set comprehension ✚ Some problems, such as the Halting problem, cannot
symbols; for example: be solved algorithmically
A = { x | x ∈ ∪ ∧ x ≥ 1 } ✚ The halting problem states that a program that can
check whether or not another program will loop
✚ Sets can also be described using compact
forever, given a specific input, cannot exist
representation; for example:
✚ The halting problem means that some problems cannot
{ anbn | n > 0 ∧ n < 4 } be solved by a computer
✚ The cardinality of a set refers to the number of ✚ A Turing machine is a theoretical computing device that
elements within it can carry out one fixed program
✚ Some sets are finite, and some are infinite ✚ A Turing machine has a finite set of states, a finite
✚ The elements within countably infinite sets can be alphabet, an infinite strip of tape and a read-write head
counted off by the natural numbers ✚ Any algorithm that can be computed can be described
✚ The Cartesian product of two sets contains the set of using a Turing machine, and vice-versa
all ordered pairs from those sets ✚ A Turing machine can be described using a finite state
✚ All values in a subset also belong in the superset machine, state transition diagrams, state transition
✚ All values in a proper subset belong in the superset, but tables and transition functions
this doesn’t contain all of the values from the superset ✚ A universal Turing machine can read in the description
✚ The union of two sets includes all values from both for any Turing machine and its tape in order to simulate it
sets combined ✚ A universal Turing machine is a theoretical machine,
✚ The intersection of two sets only contains those values more powerful than any real computer because it has
in both sets infinite memory / an infinite tape, and provides the
✚ The difference of two sets contains only the values that theoretical basis for modern computers
are in one set, but not both 99

My Revision Notes AQA A-level Computer Science Third Edition

325487_04_MRN_AQA_CS_071-102.indd Page 100 6/29/21 9:59 PM f-0116 /103/HO02206/work/indd

Exam practice

1 A 7-bit code, 010 1011 is processed by the finite state machine, as shown.
The code is processed from the most significant bit, so the first value processed will be 0.
4 Theory of computation

1|1

SA SB

1|0
0|0 0|1

a) State the output when the 7-bit code 010 1011 is processed. [1]
b) The last value that is output from the finite state machine is added to the 7-bit code before it is transmitted.
State the purpose of the final bit. [2]
c) The finite state machine could also be represented using a state transition table. Copy and complete the state
transition table below. [3]

Starting state Input Output Next state

SA 0 0 SA
SA 1
SB
SB

d) A Turing machine to achieve the same outcome is shown, along with its tape, below. The current position is
labelled with an asterisk (*).

SC

#|0 #|1

1|1

SA SB

1|0
0|0 0|1

… 1* 1 0 0 1 0 1 # # … SA

Copy and complete this trace table by hand-tracing the Turing machine. [4]

… 1* 1 0 0 1 0 1 # # … SA
… …
… …
… …
… …
… …
… …
… …
… …

e) As well as a finite set of states and a tape, identify two other features required by a Turing machine. [2]
➜
100

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_04_MRN_AQA_CS_071-102.indd Page 101 6/29/21 9:59 PM f-0116 /103/HO02206/work/indd

2 The subroutine Fold takes an array of integers as an argument.

The subroutine Copy takes two arguments – an array of integers and the index of the first element to be copied. This
subroutine will return a copy of the array, starting from the position indicated by the index argument. For example:
Calling Copy([9,8,7,6,5],1) would return [8,7,6,5].

4 Theory of computation
FUNCTION Fold (List)
IF Len(List) > 1 THEN
Head = List[0]
Tail = Copy(List,1)
return Head * Fold(Tail)
ELSE
return List[0]
ENDIF
ENDFUNCTION
a) Copy an complete this trace table by hand-tracing the subroutine call Fold([3,5,8,2]). [4]

Call Number List Head Tail Return

1 [3,5,8,2]
2
3
4
3
2
1

b) Explain why recursive algorithms are often considered to be space-wise inefficient. [2]
c) State the time-wise complexity for each of the following algorithms. [4]

Algorithm Time-wise complexity

Linear search
Merge sort
Binary sort
Insertion sort

d) Explain why the time-wise complexity for a bubble sort is O(n2). [2]
e) State what is meant by an intractable problem. [2]
f) Suggest two possible approaches to solving an intractable problem. [2]
g) Explain the significance of the halting problem. [1]
3 Abstraction and automation are essential techniques in computer programming.
a) Explain what is meant by abstraction. [2]
b) Copy and complete this table by entering the appropriate letter in each case. [6]
A Abstraction by generalisation
B Data abstraction
C Information hiding
D Composition
E Functional abstraction
F Problem abstraction ➜

101

My Revision Notes AQA A-level Computer Science Third Edition

325487_04_MRN_AQA_CS_071-102.indd Page 102 6/29/21 9:59 PM f-0116 /103/HO02206/work/indd

Description Abstraction
Combining individual functions to create compound functions
Removing unnecessary details until the problem can be
4 Theory of computation

recognised as one that has already been solved

Grouping features of the problem until it can be identified as a
particular type of problem
Hiding the details of how a task should be performed inside a
subroutine that returns a value
Creating structures such as stacks and queues using arrays and
pointers
Hiding the attributes that are stored within an object

c) List the four steps involved in automation. [4]

4 A language is partly described using the finite state machine shown.
a

c
S0 S1 S2
b

a, b
c a, b, c

S4

a) State which state is the accepting state. [1]

b) Write the regular expression described by this finite state machine. [1]
c) Explain why the language described by the finite state machine is a regular language. [1]
d) Below is an attempt at describing the language using the following BNF production rules.
State which BNF production rule cannot be represented as a regular expression and explain why not. [2]
<code> ::= <digit><digit><word> | <digit><word>
<word> ::= <letter> | <word><letter>
<letter> ::= a | b | c
<digit> ::= 1 | 2 | 3
e) Copy and complete this table by placing a Y or N next to each input that would be accepted by the BNF rules
above (regardless of whether the input would be accepted by the finite state machine). [3]

Input Accepted?
12ab
ab12
3ab1
2abc
321cba
21caabc

102

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_05_MRN_AQA_CS_103-138.indd Page 103 6/29/21 9:58 PM f-0116 /103/HO02206/work/indd

5 Fundamentals of data
representation

Number systems
Numbers can be grouped into different sets.
Many standard sets of numbers can be indicated using specialised set
notations such as , , , and .

Natural numbers
The set of natural numbers, , is the set of positive integers. A handy way to
Natural numbers Positive
remember this is that, when humans first started counting, these were the
integers, including 0.
numbers that were obvious. 0 people, one person, two people, three people,
Numbers used to count
and so on.
things.
={0,1,2,3,4,…}
The set of natural numbers is infinitely large, so we use the notation ‘…’ to
show that the set carries on forever.

Integer numbers
The set of integer numbers, , is the set of all integers, both positive and
Integer numbers Whole
negative.
numbers, including both
 = { … , −3 , −2 , −1 , 0 , 1 , 2 , 3 , … } positive and negative values.

Rational numbers
The set of rational numbers, , is the set of all numbers that can be
Rational numbers
represented as fractions. This includes integers, since all integers can be
Any number that can be
written as a fraction of 1 (for example, 3 = 3/1).
represented as a fraction,
 includes { … , −2 , −4/3 , 0 , 1/100 , … } including an integer.

Irrational numbers
The set of irrational numbers is the set of all real numbers which are not
Irrational numbers Any
rational numbers (in other words, numbers that cannot be represented as
number that cannot be
fractions). It does not use a specialised set notation, though it can be
represented as a fraction,
considered as the set of all real numbers, minus all rational numbers ( \ ).
including π and square roots
The set of irrational numbers includes { … , π , √2 , e , … } of non-square numbers.

Real numbers
The set of real numbers, , contains all rational and irrational numbers. It
Real numbers The
does not include imaginary numbers, such as i or √−2.
collection of all rational and
It is important to recognise the relationships between the sets. For example, irrational numbers.
the set of natural numbers is a subset of the set of integer numbers; the set of
integer numbers is a subset of the set of rational numbers; the set of rational
numbers and the set of irrational numbers are both subsets of the set of
real numbers.
103

My Revision Notes AQA A-level Computer Science Third Edition

325487_05_MRN_AQA_CS_103-138.indd Page 104 6/29/21 9:58 PM f-0116 /103/HO02206/work/indd

Making links
Sets are discussed in the Component 1 topic on regular expressions (Chapter 4), with
additional operators that can be used to define bespoke sets.
5 Fundamentals of data representation

Understanding sets is key to understanding the domain and co-domain of functions in

functional programming (Chapter 12).

Ordinal numbers
Ordinal numbers are used to describe the order in which numbers appear. For
example, in the ordered set {apple, banana, clementine}, ordinal numbers can
be used to indicate that ‘clementine’ is the third item in the set.

Counting and measurement

Natural numbers are used for counting. This is because the numbers are
Ordinal numbers
discrete and relate to physically countable objects.
Numbers used to count
Real numbers are used for measurement. This is because the numbers the order that something
are continuous and can be expressed in more flexible terms (for example, appears; for example, 1st,
calculating a hypotenuse using a square root, or a measurement of a circle 2nd, 3rd, and so on.
using π) Discrete Can only take
certain, countable values,
Now test yourself
such as shoe sizes or
1 State the specialised set notation and describe the numbers contained in each set. ordinal numbers.
a Integers Continuous Can take any
b Rational numbers value, such as the length of
c Real numbers an object.
d Natural numbers
2 Explain why an integer can also be described as a rational number.
3 Place at least one cross (×) in each row to show which sets each value belongs to.
A value can belong to more than one set.

Number    Irrational 
7
4.3
−2
√5
0

4 Which set of numbers is used as ordinal numbers?

Answers available online

Number bases
Numbers are traditionally represented in powers of 10, because humans
Number base The number
typically have 10 fingers and thumbs on which to count. This is referred to as
of digits available in that
base 10, denary, or decimal.
number system.
Decimal numbers can be written with the number 10 as a subscript; for Decimal Numbers with
example, 4310. base 10.
In computer systems, which are based on on/off circuits, numbers are Denary An alternative less
represented in powers of 2. This is referred to as base 2, or binary. ambiguous term for base 10
Binary numbers can be written with the number 2 as a subscript; for example, numbers.
104
0010 10112. Binary Numbers with base 2.

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_05_MRN_AQA_CS_103-138.indd Page 105 6/29/21 9:58 PM f-0116 /103/HO02206/work/indd

A different number system is used as a shorthand for binary that uses one
Hexadecimal Numbers
digit to represent each group of four binary numbers. Each set of four binary
with base 16.
numbers can have one of 16 values, from 0 to 15. This number system is
referred to as base 16, or hexadecimal.

5 Fundamentals of data representation

Exam tip
Hexadecimal numbers use the digits 0 to 9 just as in denary, followed by the
letters A–F. Hexadecimal numbers can be written with the number 16 as a Be careful not to refer to
subscript; for example, 2B16. numbers with a fractional
part (such as 12.3) as ‘decimal
When displaying binary numbers it is easy for humans to make typing errors, numbers’. This term should
or to misread the numbers due to the limited number of characters and the only be used to create a
length of the numbers. Binary numbers are also difficult to remember. So, distinction between different
hexadecimal is used as a shorthand for binary numbers. number bases. To avoid
Each block of four binary numbers can be represented using a single confusion, it is acceptable
hexadecimal digit, meaning that there is no wasted space. to use the term denary to
refer to numbers represented
For instance, the hexadecimal number 3399CC16 is much easier to recall and in base 10, and fractional
recognise than the binary equivalent 0011 0011 1001 1001 1100 11002. numbers to refer to numbers
with a fractional part.
Now test yourself

5 What name is used for the number system based on two digits? Note
6 What name is used for the number system based on ten digits?
It is important to remember
7 Why is it sometimes preferable to represent numbers from a computer system that computer systems
using hexadecimal? always store and process
data in a binary format.
Answers available online
These representations may
be used for input or output,
Converting between binary and decimal but computers are only
To fully understand how to convert between number bases you must be capable of storing values as
familiar with the concept of place value. We can write a decimal number such 0s and 1s.
as 503 more formally like this:

Place value 100 (=102) 10 (=101) 1 (=10 0)

Decimal number 5 0 3

To obtain the decimal number we multiply each decimal digit by its place
value: (5 × 100) + (0 × 10) + (3 × 1) = 503.
To convert a binary number to its decimal equivalent, we follow exactly the
same process. However the place values in binary are different because they
are powers of 2 rather than powers of 10. Least significant bit
(LSB) The right-most bit in a
Starting at the right-hand side, the place value of the least significant bit (LSB),
binary number. The bit with
has a value of 1. Each subsequent place value, moving left, is worth double the
the smallest place value.
previous value.

Place value 128 (=27) 64 (=26) 32 (=25) 16 (=24) 8 (=23) 4 (=22) 2 (=21) 1 (=20)
Binary number 0 1 1 0 1 0 1 0

The decimal value of the binary number is found by multiplying each binary
digit by its place value and adding them all together. In this example the
binary number 01101010 is expressed in decimal as:
(128 × 0) + (64 × 1) + (32 × 1) + (16 × 0) + (8 × 1) + (4 × 0) + (2 × 1) + (1 × 0)
= 64 + 32 + 8 + 2
= 10610
Most significant bit
To convert a decimal number to its binary equivalent there are two main
(MSB) The left-most bit in a
methods. The more intuitive method is to remove the largest possible power
binary number. The bit with
of 2, working from the most significant bit (MSB), and removing values one at the largest place value.
a time. 105

My Revision Notes AQA A-level Computer Science Third Edition

325487_05_MRN_AQA_CS_103-138.indd Page 106 6/29/21 9:58 PM f-0116 /103/HO02206/work/indd

Worked example
Convert 12310 to binary.
12310 = 64 + 59
5 Fundamentals of data representation

Place value 128 64 32 16 8 4 2 1

Binary number 0 1

5910 = 32 + 27

Place value 128 64 32 16 8 4 2 1

Binary number 0 1 1

This is repeated until the binary value is completed.

12310 = 64 + 32 + 16 + 8 + 2 + 1

Place value 128 64 32 16 8 4 2 1

Binary number 0 1 1 1 1 0 1 1

The next method is more easily written as an algorithm that can be

programmed.
In this method the decimal number is halved, and the remainder is written
in, starting with the LSB.

Worked example
Convert 12310 to binary.
12310 ÷ 2 = 61 remainder 1.

Place value 128 64 32 16 8 4 2 1

Binary number 1

6110 ÷ 2 = 30 remainder 1.

Place value 128 64 32 16 8 4 2 1

Binary number 1 1

3010 ÷ 2 = 15 remainder 0.

Place value 128 64 32 16 8 4 2 1

Binary number 0 1 1

This is repeated until the remaining decimal value reaches 0.

1510 ÷ 2 = 7 remainder 1 Exam tip

710 ÷ 2 = 3 remainder 1 Be very careful when

converting to binary to
310 ÷ 2 = 1 remainder 1 check how many binary
110 ÷ 2 = 0 remainder 1 digits should be used. If the
question states that the
010 ÷ 2 = 0 remainder 0
answer should be in 8-bit
binary, then the decimal
Place value 128 64 32 16 8 4 2 1
number 3 should be written
Binary number 0 1 1 1 1 0 1 1 as 0000 0011 and not as 11.

106

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_05_MRN_AQA_CS_103-138.indd Page 107 6/29/21 9:58 PM f-0116 /103/HO02206/work/indd

Now test yourself

8 Convert each of the following decimal values to 8-bit binary.

a 17

5 Fundamentals of data representation

b 84
c 133
9 Convert each of the following binary values to decimal.
a 0110
b 1001 1101
c 0111 1011

Answers available online

Converting to and from hexadecimal

When converting between binary and hexadecimal, each block of 4 binary
digits is converted to a single hexadecimal digit.
For binary values under 10, the conversion works exactly as it would with
decimal numbers.
10012 = 916
For binary values of 10 or more, the letters A-F are substituted:

Binary Decimal Hexadecimal Binary Decimal Hexadecimal

0000 0 0 1000 8 8
0001 1 1 1001 9 9
0010 2 2 1010 10 A
0011 3 3 1011 11 B
0100 4 4 1100 12 C
0101 5 5 1101 13 D
0110 6 6 1110 14 E
0111 7 7 1111 15 F

Because each binary number is split into blocks of 4 binary digits, even large
numbers are quick and simple to convert:
1001 11012 = 9D16
The reverse process involves expanding each hexadecimal digit back into its
binary equivalent.
F316 = 1111 00112
To convert from decimal to hexadecimal it is possible to convert to binary first:
17310 = 1010 11012 = AD16
Alternatively, the hexadecimal value can be found directly by dividing the
decimal number by 16 and representing the division and remainder in
hexadecimal.

Worked example
Convert 17310 to hexadecimal.
17310 ÷ 16
= 10 remainder 13
= AD16
As 1010 = A16 and 1310 = D16

When converting via binary, the reverse method can be used:

107
3A16 = 0011 10102 = 5810

My Revision Notes AQA A-level Computer Science Third Edition

325487_05_MRN_AQA_CS_103-138.indd Page 108 6/29/21 9:58 PM f-0116 /103/HO02206/work/indd

When converting directly from hexadecimal to decimal we use the concept of

Exam tip
place value.
It is important to be able to
Worked example convert between number
5 Fundamentals of data representation

Convert 3A16 to decimal. bases for Component 2,

and it is also important to
The number 3A16 is written as: understand the algorithms
used, as Component 1
Place value 16 (=161) 1 (=160) can sometimes include
Hexadecimal number 3 A programming tasks, trace
tables and finite state
3A16 = (16 × 3) + (1 × 10) = 5810 machines based around
number base conversion.

Now test yourself

10 Convert each of the following binary values to hexadecimal.

a 1011 0101 b 1101 1011 c 0111 1110
11 Convert each of the following hexadecimal values to binary.
a F3 b EA c C7
12 Convert each of the following decimal values to hexadecimal.
a 161 b 67 c 213
13 Convert each of the following hexadecimal values to decimal.
a 20 b 83 c 79

Answers available online

Units of information
Bits and bytes
A single binary digit is referred to as a bit (a shortening of the words
Binary digit).
Each bit can take the value of a 0 or a 1.
In order to improve the ease of reading binary data, bits are usually written in
blocks of four digits, referred to as a nibble.
For most purposes, eight binary digits are grouped together as a byte.
The number of unique combinations of values within a binary value is always
a power of 2. Exam tip
If there are 2 bits then there are 22 possible combinations. 22 = 4. If you are asked to
complete a list of all unique
00, 01, 10, 11
combinations (for example,
If there are 3 bits then there are 23 possible combinations. 23 = 8. for a trace table) then write
out the binary numbers in
000, 001, 010, 011, 100, 101, 110, 111
numerical order. That is, in
If there are 8 bits then there are 28 combinations. 28 = 256. order of the decimal values
0, 1, 2, 3, and so on.
In general, if there are n bits then there are 2n combinations.

Units
In modern computing a large amount of binary data is stored, transferred or
used, and so larger units are required. Decimal prefix A
shorthand used for
As is the case with all metric measurements (grams, metres, and so on), a multiples of 1000 (103)
decimal prefix is used for multiples of 1000 (103). bytes; for example, kilobyte,
1000 bytes = 103 bytes = 1 kilobyte, or 1 kB megabyte, gigabyte.
108

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_05_MRN_AQA_CS_103-138.indd Page 109 6/29/21 9:58 PM f-0116 /103/HO02206/work/indd

1000 kB = 106 bytes = 1 megabyte, or 1 MB

Note
1000 MB = 109 bytes = 1 gigabyte, or 1 GB
Since the number of bits
1000 GB = 1012 bytes = 1 terabyte, or 1 TB is larger than the number

5 Fundamentals of data representation

of bytes, it is common for
The convention is to use an uppercase B for bytes and a lowercase b for bits.
internet service providers to
Since 1 byte = 8 bits, a conversion between bytes and bits can be carried out advertise connection speeds
by multiplying or dividing by 8. in bits. 200 Mbps (megabits
per second) sounds more
100 kB = (100 × 8) kb = 800 kb
impressive than 25 MBps
1600 Mb = (1600 ÷ 8) MB = 200 MB (megabytes per second).

As the 1000 multiplier is based on powers of ten, an alternative approach for

the binary number base is to use a binary prefix which is based on powers of Binary prefix A shorthand
2 and uses multiples of 1024 (210). used for multiples of 1024
(210) bytes; for example
1024 bytes = 210 bytes = 1 kibibtye, or 1 KiB kibibyte, mebibyte, gibibyte.
1024 KiB = 220 bytes = 1 mebibyte, or 1 MiB
1024 MiB = 230 bytes = 1 gibibyte, or 1 GiB Exam tip

1024 GiB = 240 bytes = 1 tebibyte, or 1 TiB Remember that a letter ‘i’
in the name always refers
Now test yourself to the binary prefix (for
example, KiB, MiB, GiB).
14 State the number of bits in a:
Always read the question
a byte carefully to check whether
b nibble. you are being asked about
15 How many unique combinations can be made using five binary digits? the metric prefix or the
16 Without using a calculator, find the number of bytes in: binary prefix.
a 1 megabyte
b 2 kibibytes
c 120 TB
17 Which is bigger, 300 MB or 300 MiB?

Answers available online

Binary number system

Unsigned binary
In unsigned binary, all values are positive.
Unsigned binary A
For a 7-bit binary number the ranges are as follows. system that is only capable
of representing positive
Minimum value Maximum value numbers (0 or larger).
Binary 000 0000 111 1111
Decimal 0 127

The minimum value for any unsigned binary number is 0.

The maximum number for an unsigned binary number is 2n – 1, where n is the
number of bits.
A quick way to remember this is to understand that adding 1 to the maximum
would require an extra bit – for instance, adding 1 to a 7-bit number adding
would give 1000 0000, or 128, which requires 8 bits.

Exam tip
Remember that each additional binary digit always doubles the possible range of
values – for example, a 7 bit number has 128 possible values and an 8 bit number has
256 possible values. This fact is critical and questions relying on this knowledge can
crop up anywhere in Component 2, or Component 1.
109

My Revision Notes AQA A-level Computer Science Third Edition

325487_05_MRN_AQA_CS_103-138.indd Page 110 6/29/21 9:58 PM f-0116 /103/HO02206/work/indd

Unsigned binary arithmetic

Addition
5 Fundamentals of data representation

To add two unsigned binary numbers, write the numbers out one above the
other and add in columns, starting with the least significant bit.
To do this we need to remember sum and carry rules and note the following
when adding binary digits:
0+0=0
0+1=1
1+0=1
1 + 1 = 0 carry 1
1 + 1 + 1 = 1 carry 1
Binary digits can only be 0 or 1, so remember to carry if the answer would be 2 Exam tip
or more in decimal. Always write the numbers in binary.
Marks are given for the
0 0 1 1 0 0 1 0 + method as well as the
answer. Always show
1 0 1 1 0 1 0 1 your working for binary
1 1 1 0 0 1 1 1 ← sum arithmetic questions – never
just write the answer.
1 1 ← carry

Multiplication
To multiply two binary numbers the method involves:
✚ splitting the small number into a sum of powers of 2
✚ multiplying the larger number by each of these powers of 2
✚ adding together the results.

Worked example
Calculate (11 × 5)10 in binary.
Split the smaller number into powers of 2; in this case, 5 = 22 + 21 = 4 + 1
11 × 5 = 11 × (1 + 4)
= (11 × 1) + (11 × 4)
Multiplying by powers of 2 can be done by performing left shifts. For each power of 2,
perform the appropriate left shifts and then add the resulting values.
1011 × 1 = 1011 = 1110
1011 × 2 = 1 0110 = 2210
1011 × 4 = 10 1100 = 4410
Therefore, 1110 × 510 = 10112 + 10 11002
Write this out by putting the smaller number across the top and the larger number
across the bottom.

1 0 1 1 +
1 0 1 1 0 0
1 1 0 1 1 1
Note
1
It is not necessary to perform
Finally, convert the result back to decimal: binary division. Binary
subtraction is explained in
11 01112 = 5510
the next section, on two’s
110 We can check that this is correct by noting in decimal that 11 × 5 is indeed 55. complement binary.

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_05_MRN_AQA_CS_103-138.indd Page 111 6/29/21 9:58 PM f-0116 /103/HO02206/work/indd

Now test yourself Exam tip

18 Add the following binary numbers. You must show your

a 0011 + 0101 working in binary in order

5 Fundamentals of data representation

b 0010 1011 + 0110 0101 to demonstrate that you
can carry out binary
c 0101 1101 + 0011 0111
arithmetic accurately, but
19 Multiply the following binary numbers. once you have completed
a 0110 × 10 your answer carry out the
b 0001 0110 × 11 arithmetic on the decimal
c 0011 0001 × 101 values and convert the
result to binary in order to
Answers available online check your final answer.

Revision activity
✚ Use a random number generator to choose numbers and practise converting them to
binary and adding them. Convert the result back to decimal to check your accuracy.
✚ Use a random number generator to choose one large number (up to 256) and one
small number (up to 10), then practise converting them to binary and multiplying
them. Convert the result back to decimal to check your accuracy.

Signed binary using two’s complement

In order to represent negative binary numbers, a method known as two’s
Two’s complement A
complement is used. This method involves re-thinking the value of the most
representation of binary
significant bit.
numbers that includes
A normal 4-bit binary number is written using the following place values: both positive and negative
values.
8 4 2 1
1 1 1 0

In this case the number 11102 = 1410

In a two’s complement binary number, the left-most bit is negative but
otherwise the same rules for converting to decimal apply:

−8 4 2 1
1 1 1 0

In two’s complement, 11102 = −210

As the value 0 is considered to be a positive number, the possible range of
values for an 8-bit binary number is between −12810 (1000 00002) and +12710
(0111 11112). There are 128 possible positive values (including zero) and 128
possible negative values.
Notice that in two’s complement, all negative numbers begin with 1 and all
positive numbers plus zero begin with 0.

Exam tip
If you are asked to find the decimal representation of a two’s complement binary
number, or if you want to check your working, remembering to count the most
significant bit as a negative number is the quickest and simplest method.

Note
Remember that −1 is larger than −128. This can seem counter-intuitive but consider
that having −£1 (owing £1) is better than having −£100 (owing £100).
111

My Revision Notes AQA A-level Computer Science Third Edition

325487_05_MRN_AQA_CS_103-138.indd Page 112 6/29/21 9:58 PM f-0116 /103/HO02206/work/indd

There are two methods for converting between a positive binary number and
its negative equivalent in two’s complement.

Worked example
5 Fundamentals of data representation

Method 1: flip the bits and add one

One common method for taking a positive number and making it negative is to flip (or
invert) the bits and add one.

Original number: 0011 0100

Flip the bits: 1100 1011
Add one: 1100 1100

To check:

−128 64 32 16 8 4 2 1
0 0 1 1 0 1 0 0

0011 01002 = 32 + 16 + 4 = 5210

−128 64 32 16 8 4 2 1
1 1 0 0 1 1 0 0

110011002 = −128 + 64 + 8 + 4 = −5210

To change a negative number into a positive number, reverse the process. Subtract
one and then flip the bits.

Original number: 1001 1101

Subtract one: 1001 1100
Flip the bits: 0110 0011

1001 11012 = −9910

0110 00112 = 9910

Worked example
Method 2: flip from the right
Subtracting binary numbers can sometimes be complicated, and it is possible to carry
out the operations in the wrong order. Some people prefer a different method.
Regardless of the conversion (positive to negative, or negative to positive), copy the
binary sequence, starting with the right-most bit.
Copy the pattern up to and including the first 1 digit, and then flip all of the remaining
bits.

Original number: 0011 0100 Making links

Copy up to and including the first 1: ____ _100
This method of conversion
Flip the remaining bits: 1100 1100 is commonly described
using a finite state machine
We can see that this gives the same result as the first example in method 1.
or Turing machine using the
This method is symmetrical, in that it works the same whether the original number is techniques described in
positive or negative, and is also simpler to program. Chapter 4.

Exam tip Exam tip

Read the question carefully. The question will always indicate which representation is Both methods are perfectly
being used and the answers will be quite different if the question uses unsigned, valid, so use whichever
rather than two’s complement, representation. one you are most
112 comfortable with.

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_05_MRN_AQA_CS_103-138.indd Page 113 6/29/21 9:58 PM f-0116 /103/HO02206/work/indd

Now test yourself Revision activity

20 How can you tell instantly whether a two’s complement number is positive or Use a random number
negative? generator to choose

5 Fundamentals of data representation

21 Convert each of the following positive binary numbers into its negative equivalent numbers and practise
in two’s complement. converting them to
and from their negative
a 0110 1010
equivalent. Use online tools
b 0010 1101
to check your accuracy.
c 0000 0000
22 Convert each of the following negative binary two’s complement numbers into its
positive equivalent.
a 1000 0001
b 1111 1111
c 1001 0101
d 1101 1000

Answers available online

Subtraction
Using two’s complement binary it is possible to subtract two numbers.
For instance, 73 − 14 can be solved by adding +73 to −14.

Worked example
7310 = 0100 10012
1410 = 0000 11102
To convert 1410 to −1410 in binary, we flip the bits and add 1:
−1410 = 1111 00102
Now we perform the calculation: 73 +(−14):

0 1 0 0 1 0 0 1 +
1 1 1 1 0 0 1 0
0 0 1 1 1 0 1 1
1 1

Any remaining carry bits are discarded as an overflow.

The final result can be converted to a decimal number for checking. In this case, 0011
10112 = 5910, which is the correct answer.

To calculate the possible range of a two’s complement binary value containing

n binary digits, remember that the first digit is used to indicate the sign.
An 8-bit number two’s complement number can have the same maximum
value as a 7-bit unsigned binary number.
The possible range of negative values always seems to be one more than
the range of positive values. This is because 0 is considered to be a positive
number.

Number of bits Possible range

4 −8 to +7
6 −32 to +31
8 −128 to +127

In general, for an n bit number, the possible range is −2n − 1 to (2n − 1) − 1 113

My Revision Notes AQA A-level Computer Science Third Edition

325487_05_MRN_AQA_CS_103-138.indd Page 114 6/29/21 9:58 PM f-0116 /103/HO02206/work/indd

Now test yourself Revision activity

23 What is the possible range of values in a 7-bit two’s complement number? Create your own practice
24 Use binary arithmetic to add a positive and a negative number. questions by picking two
5 Fundamentals of data representation

decimal numbers. Convert

a 0011 1011 + 1101 1010
each one to binary and
b 0001 0110 + 1001 0101
perform the arithmetic,
c 0000 0001 + 1111 1111 and then convert the
25 Use two’s complement to solve these subtraction problems. answer back to its decimal
a 0010 0110 – 0001 1101 equivalent to check it.
b 0110 1100 – 0000 1001
c 0001 1011 – 0110 1100

Answers available online

Numbers with a fractional part

There are two main methods for dealing with number that have a fractional
Binary point Used to
part. In both cases the numbers have a binary point, which performs exactly
separate the whole number
the same role as a decimal point for decimal numbers.
and fractional parts of a
binary number.
Fixed point binary Fixed point binary A
In fixed point binary the position of the binary point is fixed. method for representing
Any value to the left of the binary point increases in powers of two, exactly as fractional numbers where
happens with the integer representation. the number of binary digits
before and after the decimal
0100.00002 = 8.010 point is fixed.
n
The digits to the left of the binary point are assigned the value 12 where n = 1,
2, 3, and so on.
These can be written as fractional values:

1 1 1 1
−8 4 2 1 . 2 4 8 16

0 1 1 0 . 1 0 1 0

Alternatively, they can be written using decimal fractions:

−8 4 2 1 . 0.5 0.25 0.125 0.0625

0 1 1 0 . 1 0 1 0

We can see that:

0110.10102 = 4 + 2 + 0.5 + 0.125 = 6.62510
Fixed point binary has the advantage of simplicity, though it does introduce
significant rounding errors as values such as 10.1 are very difficult to
represent accurately.
To find the binary representation of a decimal fraction, remove each
fractional power of two.

Worked example
Converting the decimal fraction 7.5625 into its fixed point binary representation.
Calculate the whole number part, in this case 7:

−8 4 2 1 .
0 1 1 1 .

114 Note that 7.5625 – 7 = 0.5625, which is the value left to find

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_05_MRN_AQA_CS_103-138.indd Page 115 6/29/21 9:58 PM f-0116 /103/HO02206/work/indd

1
The first value after the binary point is worth 2
= 0.5

−8 4 2 1 . 0.5

5 Fundamentals of data representation

0 1 1 1 . 1

0.5625 – 0.5 = 0.0625 left to find

1
Second value after the binary point is worth 4
= 0.25
0.25 can’t be taken from 0.0625

−8 4 2 1 . 0.5 0.25
0 1 1 1 . 1 0

1
Third value after the binary point is worth 8
= 0.125
0.125 can’t be taken from 0.0625

−8 4 2 1 . 0.5 0.25 0.125

0 1 1 1 . 1 0 0

1 Exam tip
Fourth value after the binary point is worth 16 = 0.0.625
0.0625 – 0.0625 = 0 Fixed point binary
questions may use two’s
−8 4 2 1 . 0.5 0.25 0.125 0.0625 complement or unsigned
0 1 1 1 . 1 0 0 1 binary representations.
Always check the question
Therefore, 7.562510 = 0111.10012 carefully.

Now test yourself

26 Convert the following two’s complement, fixed point numbers to decimal.

a 011.10100
b 10110.110
c 011011.01
27 Convert the following decimal numbers into their two’s complement, fixed point
binary equivalent.
a 12.5
b −3.75
c 9.875

Answers available online

Floating point binary

Particularly large or small fractional numbers represented using fixed point
Floating point binary A
binary can be very wasteful of memory, as they will need lots of 0s. Floating
method for representing
point binary provides a more efficient, but more complex, representation of
fractional numbers where
fractional numbers.
the position of the decimal
The number is represented with a mantissa and an exponent, in a similar way point is moved according to
to standard form in mathematics. For example, in decimal. 4 130 = 4.13 × 103 the value of the exponent.
where 4.13 is the mantissa and 3 is the exponent.
Mantissa The part of a
The mantissa provides the value of the number, while the exponent says how floating point binary number
far to move the decimal point which provides the value of
that number.
In floating point binary the mantissa is a binary fraction stored in two’s
complement format, where the MSB is −1 and remaining bits are fractions Exponent The part of a
after a binary point. For example, for an 8 bit mantissa: floating point binary number
which states how far to
−1 . 1 1 1 1 1 1 1 move the decimal point.
2 4 8 16 32 64 128
115
0 . 1 1 0 0 1 0 0

My Revision Notes AQA A-level Computer Science Third Edition

325487_05_MRN_AQA_CS_103-138.indd Page 116 6/29/21 9:58 PM f-0116 /103/HO02206/work/indd

The exponent is also stored in two’s complement format. For example, for a 4
Note
bit exponent:
Exam questions using the
−8 4 2 1 floating point representation
5 Fundamentals of data representation

0 0 1 1 will usually use two’s

complement for both the
mantissa and the exponent.
Worked example
Calculate the decimal value of the floating point binary number 01100100
0011.
To calculate the decimal value of the following floating point binary number:

0 1 1 0 0 1 0 0 0 0 1 1
Mantissa Exponent

The first step is to calculate the value of the exponent. In this case, +3.

−8 4 2 1
0 0 1 1

The second step is to move the binary point by that number of positions …

−1 . 1 1 1 1 1 1 1
2 4 8 16 32 64 128

0 . 1 1 0 0 1 0 0

… becomes …

−8 4 2 1 . 1 1 1 1
2 4 8 16

0 1 1 0 . 0 1 0 0

The third step is to convert this value in the same way as with fixed point binary:
1
4+2+ 4
= 6.25

The same method is used for negative numbers. Remember that both the
mantissa and exponent are in two’s complement, which means if the first bit
is 1 then the number is negative.

Worked example
Calculate the decimal value of the floating point binary number 10110000
1111.

1 0 1 1 0 0 0 0 1 1 1 1
Mantissa Exponent

The first step is to calculate the value of the exponent. In this case, −1.

−8 4 2 1
1 1 1 1

The second step is to move the binary point by that number of positions. However, for
a negative exponent, rather than moving the binary point to the left instead shift the
bits to the right, which achieves the same effect. This is because we do not want the
left-most bit to move, as it is used to indicate the sign …

−1 . 1 1 1 1 1 1 1
2 4 8 16 32 64 128

1 . 0 1 1 0 0 0 0
116
… becomes …

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_05_MRN_AQA_CS_103-138.indd Page 117 6/29/21 9:58 PM f-0116 /103/HO02206/work/indd

−1 . 1 1 1 1 1 1 1
2 4 8 16 32 64 128

1 . 0 0 1 1 0 0 0

5 Fundamentals of data representation

The third step is to convert the value in the same way as with fixed point binary:
1 1
−1 + 8
+ 16 = −1 + 0.125 + 0.0625 = −0.8125

It is also important to know how to convert a decimal number to its floating

point equivalent.
✚ First, start by writing the number in a fixed point format. Use the
minimum number of digits needed, remembering to use two’s complement
format. This means the left-most bit will indicate the sign.
✚ Second, move the binary point so that it comes straight after the first value
in order to find the mantissa
✚ Thirdly, count the number of positions that the binary point has been
moved by in order to find the exponent. Use however many digits are
needed, remembering to include the right-most bit to indicate the sign.
✚ Finally, write the full binary number down using the number of digits
specified in the question, adding leading or trailing digits as necessary

Making links
At the second step it is important that the two digits either side of the binary point
should be different. For example:

−1 . 1 1 1 1 1 1 1
2 4 8 16 32 64 128

0 . 1

Or …

−1 . 1 1 1 1 1 1 1
2 4 8 16 32 64 128

1 . 0

This is referred to as normalised floating point binary and is discussed in more detail
laer in this chapter.

Worked example
Write the decimal number 11.2510 as a floating point number with an 8 bit
mantissa and a 4 bit component.
1 Write the number in fixed point two’s complement.
11.2510 = 01011.012
2 Move the binary point so that it comes after the first value.
Mantissa = 0.101101 Exam tip
3 Count the number of positions that the binary point has been moved by and express
Floating binary is a topic
this value in two’s complement. In this case the binary point moved four positions:
that often trips students
Exponent = 410 = 01002 up. Remember that both
4 Finally, write the full binary number down using the number of digits specified in mantissa and exponent
the question. We were asked for an 8 bit mantissa, so we need to add another bit, will be written using two’s
of value 0, to the right of the mantissa. complement binary and
ensure you get lots of
0 1 0 1 1 0 1 0 0 1 0 0 practical practice converting
Mantissa Exponent between decimal and
binary numbers, both large
(positive exponents) and
small (negative exponents).
117

My Revision Notes AQA A-level Computer Science Third Edition

325487_05_MRN_AQA_CS_103-138.indd Page 118 6/29/21 9:58 PM f-0116 /103/HO02206/work/indd

When using floating point binary it is possible to represent very large and
very small numbers. Exam questions will usually specify the number of bits
to use for the mantissa and the exponent, with a mantissa of 8 bits and an
exponent of 4 bits being typical. Be prepared to work with exponents of up to
5 Fundamentals of data representation

6 binary digits.

Now test yourself Revision activity

28 Which part of a floating point number is used to describe how far to move the ✚ Practise converting
decimal point? large integers and small
29 Convert each of the following two’s complement, floating point binary numbers fractions, both positive
into their decimal equivalent. and negative, into
floating point form and
Mantissa Exponent use online tools to check
your accuracy.
a 0.11010 001001 ✚ Generate random
b 1.0110000 0010 binary numbers to
c 0.11000 11110 use as mantissa and
exponent. Guess the
30 Convert each of the following decimal numbers into their two’s complement, scale of the number first
floating point binary equivalent. Use a 6-bit mantissa and a 4-bit exponent in each (large or small, positive
case. or negative) and then
convert the number
a 12.5
into decimal. Use online
b −7.25
tools to check your
c 0.125 accuracy.
d −0.9375

Answers available online

Rounding errors
Due to the nature of fractional numbers which are written as fractions of
powers of 2, it is not possible to store some values with complete accuracy.
Representing 0.510 as a binary fraction is trivial, using 0.12.
Representing 0.110 accurately as a binary fraction is impossible, even using
an extremely large number of binary digits. It can only be represented
approximately as the binary value is rounded up or down.

Absolute and relative errors

The absolute error is the difference between the intended value and the
Absolute error The
actual value that is stored.
difference between the
The absolute error is always written as a positive number. intended value and the
actual value.
For example, the value 9.810 is represented using fixed point binary as
1001.11012 Relative error The
percentage difference
Converting this number back to decimal gives a value of 9.812510 between the intended value
This gives an absolute error of 0.0125. and the actual value.

The relative error is expressed as a percentage of the original value.

The relative error can be calculated by dividing the absolute error by the
original value and multiplying by 100. In our example above, the relative error
is calculated as:
(0.0125 ÷ 9.8) × 100 = 0.13%
Smaller values will typically lead to a larger relative error.

118

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_05_MRN_AQA_CS_103-138.indd Page 119 6/29/21 9:58 PM f-0116 /103/HO02206/work/indd

Worked example
Calculate the relative error when storing the decimal value 0.1310 in binary.
0.1310 is represented using 4 bits in binary as 0.0012.

5 Fundamentals of data representation

Converting this binary value back to decimal gives 0.12510.
The absolute error is 0.13 − 0.125 = 0.005.
The relative error is 0.005 ÷ 0.13 × 100 = 3.85%

Notice that this relative error is much larger than the previous example, even
though the absolute error was smaller.

Now test yourself

31 Will a large number be likely to have a larger or smaller relative error?

32 Calculate the absolute and relative errors for each of these values.
a 12.26 represented in binary as 1100.01
b 0.98 represented in binary as 0001.00
c 7.7 represented in binary as 0111.11

Answers available online

Range and precision

There are several considerations when making decisions over which format to
use for fractional numbers.
✚ Range: The difference between the largest and smallest values.
✚ Precision: The reduction of rounding errors.
✚ Speed of calculation: The time a processor will take to convert the
representation in order to carry out calculations or display the value.

Advantages Disadvantages
Fixed point Simpler to convert and allows for quicker A more limited range, particularly if large
binary processing. magnitude numbers and small magnitude numbers
need to be stored using the same system.
All values can be stored with the same level of
absolute precision. Limited precision.
Values are processed in a very similar manner to
binary integers, meaning that hardware can be
re-purposed without needing to be re-designed.
Floating A wider range of values are possible by using a The precision of two different values can be very
point binary large exponent. different due to the different exponents.
More precise as a small exponent can be used. Processing takes longer.

Normalisation of floating point form

Floating point numbers should be normalised. This means that the first and
Normalised The standard
second values in the mantissa should always be opposite.
way of writing floating point
This means that no space is wasted with leading zeroes in a positive number, binary, in which the first two
or leading ones in a negative number. bits are opposite to each
other.
In two’s complement binary the number 0000 0100 has the same value as the
number 0100 – the leading zeroes are not providing any useful information
and therefore we want to get rid of them. Similarly, the negative number
1111 1110 has the same value as 10 - the leading ones are not providing any
useful information and therefore we want to get rid of them. We can do this
by choosing an exponent that ensures the first and second values in the
mantissa are opposite.
119

My Revision Notes AQA A-level Computer Science Third Edition

325487_05_MRN_AQA_CS_103-138.indd Page 120 6/29/21 9:58 PM f-0116 /103/HO02206/work/indd

It is important to be able to recognise whether a floating point binary number

is normalised by checking that the first two values are opposite. For example:

−1 . 1 1 1 1 1 1 1
2 4 8 16 32 64 128
5 Fundamentals of data representation

0 . 1

Or …

−1 . 1 1 1 1 1 1 1
2 4 8 16 32 64 128

1 . 0

It is also important when converting numbers into floating point binary form
to ensure that they are normalised.

Now test yourself Exam tip

33 Give two reasons why someone might choose to use fixed point rather than Questions about whether
floating point binary. a floating point number
34 Give two reasons why someone might choose to use floating point rather than is normalised are very
fixed point binary. common. Remember that
the first two digits should
35 What is the quickest way to check if a floating point number is normalised?
always be opposite to
36 For each floating point number, state whether it is normalised. each other if the number is
normalised.
Mantissa Exponent
a 00101010 0110
b 10100 10110
c 11010110 11
d 0110 110101

Answers available online

Underflow and overflow

When adding or subtracting numbers it is possible that the result will not be
stored correctly.
An overflow can occur when adding or multiplying values.
Overflow Where the result
For example, adding the unsigned binary values 1101 and 0110: of a calculation is too large
to be stored in the available
1 1 0 1 + digits.
0 1 1 0
1 0 0 1 1
1 1

Assuming the answer should be stored using 4 bits, as with the two original
values, the extra 1 from the calculation cannot fit in the available space.
Depending on the configuration of the system:
✚ this may cause the leading 1 to be dropped, resulting in an incorrect result
(in this case suggesting that 13 + 6 = 3)
✚ the large result may spill over into the next block of memory.
On some occasions overflow errors are not a problem, for example when
adding a positive and negative number using two’s complement.
For example, calculating 7 – 4 using two’s complement binary:

0 1 1 1 +
1 1 0 0
1 0 0 1 1
120
1 1

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_05_MRN_AQA_CS_103-138.indd Page 121 6/29/21 9:58 PM f-0116 /103/HO02206/work/indd

The overflow digit can be safely dropped, as the solution is correct. This
is because the first digit is used to represent the sign, and so the overflow
simply allows for the leading bit to be flipped.
An underflow can occur when a subtraction takes place and the value will not

5 Fundamentals of data representation

Underflow Where the
fit in the space available.
result of a calculation is too
For example, adding the two’s complement values 1101 and 1000: small to be stored in the
available digits.
1 1 0 1 +
1 0 0 0 Now test yourself
1 0 1 0 1 37 Explain what is meant
1 by an underflow error.
38 In a computer game,
Dropping the leading 1, this gives the erroneous result −3 – 8 = 5. This happens a player’s score goes
because the correct result is −1110, which cannot be stored as a 4-bit binary up by 1 each time they
number. collect a coin. A player
has 255 coins and when
Note they collect the next
In the original version of the computer game Civilisation the character of Gandhi was coin their score suddenly
a peaceful character and was assigned an aggression score of 1. At a particular point jumps from 255 to 0.
in the game each leader’s aggression score would reduce by 2 points. As the values Explain the most likely
were stored as unsigned binary integers, this caused an underflow error and Gandhi cause of this error.
immediately became the most aggressive character in the game. In subsequent
Answers available
versions of the game the bug was fixed, however the character of Gandhi has always
online
been given a large aggression score as a reference to this famous mistake.

Information coding systems

Character form of a decimal digit
The previous topics have all been focused on how to represent a number
value in binary so that it can be processed as a number.
Sometimes a number value can be stored as a character value. Think of it as
the letter ‘3’ rather than the number 3.
This is useful in situations where the value is intended to be processed as text
rather than as a number. For example:
✚ telephone numbers have a leading zero and are not generally used in any
arithmetic
✚ postcodes include numbers and letters.
It is important to make sure you know whether you are dealing with the pure
binary representation (for example, unsigned integer, two’s complement, fixed
point or floating point) or the character code representation (for example, the
letter ‘9’),

ASCII and Unicode

In order to store or transmit text, each character is assigned a numeric value.
Character set The set
This is referred to as a character set. of all characters that
A character set lists of all the characters which can be understood by a given can be understood by a
computer system and their character codes. computer system, and their
associated character codes.

121

My Revision Notes AQA A-level Computer Science Third Edition

325487_05_MRN_AQA_CS_103-138.indd Page 122 6/29/21 9:58 PM f-0116 /103/HO02206/work/indd

There are two main character sets in use – ASCII and Unicode.
ASCII A 7-bit character set
that can represent up to 128
ASCII unique characters.
ASCII stands for the American Standard Code for Information Interchange.
5 Fundamentals of data representation

Unicode A variable size

ASCII is generally agreed to use 7 bits per character, though sometimes an 8th character set in which each
bit is used as a parity bit. character code can be 8, 16
Using 7 bits means that there are 128 possible unique values. or 32 bits in length.

The first 33 codes in the character set are reserved for special characters.
The other 95 values are used for letters and symbols, including:
✚ upper case letters ✚ punctuation
✚ lower case letters ✚ mathematical operators.
✚ digits 0-9
This is a section of the ASCII character set:

Character Decimal Code Binary Code

X 88 0101 1000
Y 89 0101 1001
Z 90 0101 1010
[ 91 0101 1011
\ 92 0101 1100
] 93 0101 1101

The character codes for upper case letters in ASCII are separated by 3210 from
lower case letters. This means that only one bit (the sixth bit) needs to be
changed to change cases:

Character Decimal Code Binary Code

A 65 0100 0001
a 97 0110 0001
B 66 0100 0010
b 98 0110 0010
C 67 0100 0011
c 99 0110 0011

Also note that the last 5 binary digits can be used to identify the letter:
Exam tip
A = 0100 0001 = 64 + 1
You are not expected
B = 0100 0010 = 64 + 2 to remember individual
character codes, but it is
Z = 0101 1010 = 64 + 26
useful to know that the last
When a key on a keyboard is pressed, the binary code is transmitted to the five digits can be used to
computer. identify the letter.

Text files are saved by storing the binary code for each character in the file.
Some codes refer to special characters which are not normally visible when a Note
file is displayed on a screen, such as STX (start of text), BS (backspace), LF (line
There are various specific
feed, or new line), and so on.
versions of Unicode
Unicode including UTF-8, UTF-16 and
UTF-32. It is not necessary
ASCII does not leave enough possible values to represent letters with accents,
to be aware of the
characters from other alphabets or emojis.
differences and for the sake
Unicode is an expanded character set which uses a varied character length of this qualification all forms
of between 8 and 32 bits per character. This allows for all 1 million+ valid of Unicode are referred to
characters to be uniquely represented. under that single term.

122 In situations where these are not needed, ASCII can be preferable as it uses
less storage space.

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_05_MRN_AQA_CS_103-138.indd Page 123 6/29/21 9:58 PM f-0116 /103/HO02206/work/indd

It is important to note that the first 128 Unicode characters and codes are the
same as ASCII, so the two systems are compatible. Note
Now test yourself The Unicode Consortium

5 Fundamentals of data representation

is responsible for defining
39 Describe two differences between ASCII and Unicode.
the code for each character
40 Explain why mobile phone text messages, also known as SMS messages, are and each year they decide
typically transmitted using ASCII rather than Unicode. which new emojis to add,
41 If the character A is represented using the character code 0100 0001, write out the and what their character
character codes needed to spell the word ‘BAD’. codes should be.

Answers available online

Error checking and correction

When data is transmitted or stored, there is a possibility that it contains an error.
It is very easy for a bit to misread, or for interference to cause an error in a bit
pattern.
There are many ways to try and identify errors, some of which include ways
to automatically correct them.

Parity bits
A simple solution is to count the number of 1s in a sequence of binary values
Parity bit A single binary
and add a parity bit.
digit added to some data
In odd parity, there should be an odd number of 1s. in order to help with error
checking.
The number of 1s is counted in the data and if there is already an odd number
then a 0 is added as the parity bit. Odd parity A method of
using parity in which each
001 1010 – Original data
block of data should have
0001 1010 – Data to be transmitted an odd number of 1s.
If there is an even number of 1s then an extra 1 is added as the parity bit, so
that the total number of 1s is now odd.
001 1011 – Original data
1001 1011 – Data to be transmitted
When the message is received, the number of 1s is counted again. Exam tip

If the number of 1s is still odd then it is assumed that the data is correct. The parity bit can be added
to the start or end of the
If the number of 1s is even, then there has been an error, and the data is
message. Always read the
re-requested. question carefully to check
1001 1011 – Message appears to be correct whether the question refers
to even parity or odd parity
1001 0011 – Message contains an error
and to check which is the
The parity bit is then discarded and the original value can be processed. parity bit.
With even parity, the process is identical except that the parity bit is chosen
Even parity A method of
so that the number of 1s is even.
using parity in which each
The receiver needs to know whether if the data has been sent using even or block of data should have
odd parity. an even number of 1s.
Overhead Additional data
Advantages Disadvantages
added to the original values.
Relatively small overhead Cannot help correct the error.
in terms of adding extra data.
If two bits are incorrect, then the error will not be identified.

Note
A technique known as 2D parity can help identify multiple errors and can also correct some
errors. This is an interesting technique but is not covered in the AQA A-level specification. 123

My Revision Notes AQA A-level Computer Science Third Edition

325487_05_MRN_AQA_CS_103-138.indd Page 124 6/29/21 9:58 PM f-0116 /103/HO02206/work/indd

Majority voting
Majority voting involves sending each bit repeatedly.
Majority voting
Each bit must be sent an odd number of times and, within that transmission, Transmitting each bit an odd
5 Fundamentals of data representation

whichever digit is seen most commonly is considered to be the correct one. number of times in order
to identify and correct any
For example the bit pattern 1010 is being sent using majority voting, and each
transmission errors.
bit is sent 3 times.
The transmission should be 111 000 111 000.
Due to a timing error, the message received reads 110 000 011 000.
In each block of bits, the majority is considered to be correct and so the data
stored is 1010.
An odd number of bits must be used in order to avoid a tie when deciding
which is the most common value. Repetitions of 3, 5 and 7 bits are all
acceptable.

Advantages Disadvantages
Makes it more likely that an error can be fixed. The extra overhead in data
transmission is very high.
Less chance that the data will have to be re-sent.

Checksums
A checksum is a value that can be found by applying an arithmetic algorithm
Checksum A value derived
to the original data.
by following an algorithm,
A very simple algorithm might be to add up all of the decimal digits in a large used to check for errors.
number and store the total.
For an original value of 374 268, the digits added together would total 30. Making links
When the transmission is received the same process is repeated and the total Hashing algorithms can be
is compared to the checksum. used to generate checksums.
While checksums appear
If the value 375 268 has been received then the sum of the individual digits in the specification for
will be 31. Since this does not match then an error has been identified and the Component 2, hashing
data should be re-requested. algorithms are explored as
A more complex algorithm might be to add the square of the digits in the even part of the data structures
positions to the cube of the digits in the odd positions. topic in Component 1 (see
Hash tables in Chapter 2).
83 + 62 + 23+ 42 + 73 + 32 = 512 + 36 + 8 + 16 + 343 + 9 = 924
Checksums are effective at identifying errors because two or more errors are Note
very unlikely to cancel each other out.
One common checksum
Simple checksums are not able to locate or correct the exact error. algorithm is the MD5
algorithm, and large
Advantages Disadvantages downloads will sometimes
The transmission overhead is smaller than The transmission overhead is larger than have an MD5 checksum file
using majority voting. using parity bits. that can be downloaded
Effective at identifying errors because two Simple checksums are not able to locate for checking once the
or more errors are very unlikely to cancel or correct the exact error. download is complete.
each other out.

Check digit A single digit

Check digits derived by following an
A check digit works in almost exactly the same way as a checksum, except algorithm, used to check for
the value must be a single digit. This is generally achieved by using a modulo errors.
operation, often modulo 10, which provides one of 10 possible outcomes. Modulo Finding the
One of the more famous uses of a check digit occurs in the ISBN-13 code remainder when one
which is used to uniquely identify books. number is divided by
124 another.

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_05_MRN_AQA_CS_103-138.indd Page 125 6/29/21 9:58 PM f-0116 /103/HO02206/work/indd

Each book is given a unique 12-digit decimal number.

Starting with the right-most digit, the first value is multiplied by 3, the next
by 1, the next by 3 and so on.

5 Fundamentals of data representation

The results are added and divided by 10, noting the remainder.
If the remainder is not 0 then a value is added, in the same manner as a parity
bit, so that the result of dividing by 10 is 0.

ISBN-13 9 7 8 1 4 7 1 8 6 5 8 2
Weight 1 3 1 3 1 3 1 3 1 3 1 3
Partial sum 9 21 8 3 4 21 1 24 6 15 8 6

Total = 126
Total % 10 = 6
Value to add = 4
The final 13-digit code is therefore 978-1-4718-6582-4 (the spacing is added to
aid readability).
When the code is entered, the same calculation is performed on the full code.

ISBN-13 9 7 8 1 4 7 1 8 6 5 8 2 4
Weight 1 3 1 3 1 3 1 3 1 3 1 3 1
Note
Partial sum 9 21 8 3 4 21 1 24 6 15 8 6 4
In some cases, such as
Total = 130 ISBN-10, there are 11
possible values for the
Total % 10 = 0
check digit. Using modulo
As the modulo result is 0, it can be assumed that the values have been 11 the final value could be a
recorded or transmitted correctly. digit (0-9) or an X.

Advantages Disadvantages
Lower overhead than a checksum, With only 10 or 11 possible values for the
meaning that data can be transmitted check digit, it is possible that an erroneous
more quickly. transmission will be identified as correct.

Now test yourself

42 Explain how a parity bit can be used to check a binary value for an error.
43 Describe one situation where an incorrect binary value could pass a parity check.
44 Explain the difference between odd and even parity. Revision activity
45 What rule is used to decide how many times to transmit each bit using majority voting? Carry out your own check
46 Describe the difference between a checksum and a check digit. digit calculations, using
the ISBN numbers on the
47 Of the four methods described in this Chapter, only majority voting is capable of
back of any textbooks and
correcting errors. Explain why this isn’t the most commonly used method.
revision guides you have
Answers available online with you.

Representing images, sound

and other data
Bit patterns, images, sound and other data
All data in a computer system must be represented using binary numbers.
In addition to numbers and text, techniques have been devised to represent 125
images and sounds using binary numbers as well.

My Revision Notes AQA A-level Computer Science Third Edition

325487_05_MRN_AQA_CS_103-138.indd Page 126 6/29/21 9:58 PM f-0116 /103/HO02206/work/indd

Analogue and digital

Data in the real world is typically analogue. This includes visible light, sound,
Analogue A continuous
and physical movement.
signal, usually represented
5 Fundamentals of data representation

Analogue signals are continuous as there are an infinite number of values as a curved wave.
between each measurement. They are usually represented as a wave with
smooth curves.
Amplitude

Time

Figure 5.1 An analogue signal

Data in computer systems is usually digital. This is because computer systems

Digital A discrete signal,
generally store and transmit data using systems with two states – on and off,
often represented using a
represented in binary.
stepped wave.
Digital signals have a finite set of values and use discrete measurements with
no intermediate values. They are usually represented as a stepping, or square,
signal.
ON
OFF

Figure 5.2 A digital signal

Analogue/digital conversion
In order for computers to read in analogue data, an analogue to digital
Data Values or information.
converter (ADC) must be used.
Analogue to digital
An ADC will use an analogue sensor to read in an analogue signal and
converter (ADC) Converts
produce a digital output.
an analogue signal into a
Microphones and digital cameras are examples of devices that use an ADC. digital signal.

Each digital value will be a close approximation to the analogue signal at any Signal The electric or
particular moment. electromagnetic impulses
that are used to transmit
A digital to analogue converter (DAC) is used to convert a digital signal into an data.
analogue signal.
Digital to analogue
This is used, for example, to convert a digital audio file into an analogue converter (DAC) Converts
signal which can be used to recreate the sound through a speaker. a digital signal to an
analogue signal.
Exam tip
Questions about ADCs and DACs are usually written in the context of either digital
images or digital sound. Make sure you are able to apply your understanding of ADCs
and DACs to the topics on the following pages.

Now test yourself

48 What is the main difference between an analogue signal and a digital signal?
49 Explain why it is not possible to store an analogue signal in a computer system.
50 What device is used to convert a digital signal so that it can be transmitted over an
analogue medium?
51 Other than a microphone, name one device which contains an ADC.

Answers available online

126

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_05_MRN_AQA_CS_103-138.indd Page 127 6/29/21 9:58 PM f-0116 /103/HO02206/work/indd

Bitmapped graphics
A bitmap (or raster) graphic is an image made up of individual blocks of colour
Bitmap An image format
called pixels.
that uses a grid of pixels.

5 Fundamentals of data representation

Each pixel is a single colour, and the colour of each pixel is recorded using a
Pixel The smallest
binary code.
addressable element of an
In black and white images each pixel can be represented using a single bit, image.
such as 1 for white and 0 for black. Colour depth The number
Data stored in memory of bits used to store the
colour of each pixel.
0 1 0 0 1 0 0 1

1 1 1 0 1 1 0 0

Pixels displayed in screen

Figure 5.3 Each pixel in this black and white image is represented by one bit of data

Colour depth
Increasing the number of bits per pixel increases the possible number of
colours that each pixel can represent. This is known as the colour depth.

Bits per pixel (colour depth) Possible colours

2 22 = 4
4 24 = 16
8 28 = 256
16 216 = 65 536

Most high-quality images use a colour depth of 24 bits per pixel – 8 bits each
for the amount of red, green and blue (RGB).
Increasing the colour depth allows for a more accurate representation of the
image, however it also increases the size of the image file.

Size and resolution

The size of an image is a measure of its pixel dimensions.
Size The pixel dimensions
size in pixels = width in pixels × height in pixels measured as the width x
height of an image in pixels.
Increasing the number of pixels allows the image to be more accurate,
however it also increases the size of the image file. Resolution The number
of pixels per unit of length,
The resolution of the image is the number of pixels per unit of length such as
also known as pixel density,
cm or inches. This is also known as the pixel density of an image. often measured in PPI or
Print and display resolutions are often measured using pixels per inch (PPI) or DPI.
dots per inch (DPI) (where dots and pixels mean the same thing). An image
with a width of 1024 pixels and a display resolution of 72 PPI would be 1024 ÷
Note
72 = 14.2 inches wide.
Digital screens, including
The overall file size for an image can be found by multiplying the number of
mobile phones, are usually
pixels by the number of bits per pixel.
described according to both
their pixel dimensions and
their display resolutions.

127

My Revision Notes AQA A-level Computer Science Third Edition

325487_05_MRN_AQA_CS_103-138.indd Page 128 6/29/21 9:58 PM f-0116 /103/HO02206/work/indd

file size = size in pixels × colour depth

This can also be expressed as:
file size = width (px) × height (px) × bits per pixel
5 Fundamentals of data representation

(Note that ‘pixel’ is often abbreviated to ‘px’.)

For example, for an image which is 2000 px wide, 2000 px tall and has a colour
depth of 24 bits per pixel:
size = 2000 px × 2000 px = 4 000 000 px Exam tip
file size = 4 000 000 px × 24 bits per px = 96 000 000 bits Remember that 1 byte =
8 bits. Always read the
There are 8 bits in a byte, so to convert this to bytes:
question carefully and make
96 000 000 bits ÷ 8 = 12 000 000 bytes sure that your answer is
presented in the right units.
= 12 MB
The major advantage of bitmap images is that they are an effective way to
store a photo-realistic image due to the possible complexity of the different
shades within each pixel.

Metadata
Metadata means data about the data.
Metadata Data about data.
While an image file contains the binary data that makes up the colour of each Additional data stored in a
individual pixel, additional data is also needed. file.

This includes the width, height and colour depth of an image which is
necessary to convert the raw binary data back into a usable image.
Other metadata might include the time and date the image was created or
saved, the camera settings used (for a photograph), the name of the software
used to edit it and GPS data.

Now test yourself

52 What is meant by a pixel?

53 Explain two effects of increasing the colour depth of an image.
54 Explain the impact of increasing the colour depth of an image by 1 bit per pixel.
55 What term refers to the number of pixels in an image?
56 Suggest four pieces of metadata that could be added to a digital photograph.
57 Calculate the colour depth for an image with up to 16 colours.
58 Calculate the file size in bytes for an image that is 600 pixels wide, 400 pixels tall
and with a colour depth of 16 bits per pixel.

Answers available online

Vector graphics
Vector graphics use geometry to describe the objects that make up an image.
Vector graphic An image
Each object is a shape (also known as a vector primitive), such as a line, made of lines and shapes.
polygon, circle, curve or text.
Vector primitive Simple
Properties are used to describe each object in the image so that it can be objects which can be
drawn as needed. These properties include co-ordinates, line colour, line combined to create a vector
thickness and fill colour. Vector graphic files store the details of these graphic.
properties. Polygon A shape made of
A vector graphic is made up of various vector primitives and the image can be straight lines, for example,
drawn to any scale. rectangle, hexagon, and so on.

128

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_05_MRN_AQA_CS_103-138.indd Page 129 6/29/21 9:58 PM f-0116 /103/HO02206/work/indd

5 Fundamentals of data representation

Start coordinate
Figure 5.4 Scaled vector graphics

Vector graphics versus bitmapped graphics

The file size of bitmap images is large as the colour for each pixel must be
stored.
With a vector graphic the data that must be stored for each object is much
smaller in size.
When a bitmap image is scaled up the pixels are enlarged, making them more
noticeable, and the image becomes blocky. The resolution decreases as there
are fewer pixels per inch.
With a vector graphic, the scale can be adjusted with no loss of quality as the
image is redrawn each time.
Vector graphics are:
✚ not able to store a photo-realistic image as the vector primitives require
each image to be made of relatively simple shapes and colours
✚ ideal for use in logos, fonts, and clipart as they can be scaled with no loss
Clipart A cartoon style
in quality and can be edited or split into separate objects.
image.
Bitmap Images Vector Images
Large file size. Small file size.
Loss in quality when scaled. No loss in quality when scaled.
Suitable for photo-realistic images. Unsuitable for photo-realistic images.
Less commonly used for logos, fonts, and Ideal for logos, fonts, and clipart.
clipart.

Now test yourself

59 Give three properties of vector objects.

60 Explain why scaling a vector graphic does not reduce the quality of the final image.
61 State one advantage and one disadvantage for storing an image as a bitmap
graphic.
62 Suggest two cases where a vector graphic would be preferable to a bitmap.
63 Suggest one reason why digital photographs are generally stored as a bitmap
rather than a vector.

Answers available online

Digital representation of sound

When analogue sounds are recorded digitally, the amplitude of the sound
Amplitude A measure of
wave is measured at fixed intervals of time.
how large a vibration or
The value of each sampling point is stored as a binary number. oscillation is. 129

My Revision Notes AQA A-level Computer Science Third Edition

325487_05_MRN_AQA_CS_103-138.indd Page 130 6/29/21 9:58 PM f-0116 /103/HO02206/work/indd

15
14
13
12
5 Fundamentals of data representation

11
10
9
Amplitude

8
7
6
5
4
3
2
1
0
0 1 2 3 4 5 6 7 8 9
Time
Figure 5.5 Sampling an analogue wave

The fixed interval of time is known as the sample rate and is measured in Hz,
Sample rate The rate at
where 1 Hz is one sample per second.
which samples are taken,
A sample rate of 48 kHz means that there will be 48 000 samples per second. typically measured in kHz.

Using a higher sample rate means that more samples will be captured: Sampling resolution The
✚ increasing the accuracy of the recording number of bits per sample.
✚ while also increasing the file size.
The sampling resolution refers to the number of bits per sample to measure
the amplitude. (This concept is similar to the colour depth of an image).
In Figure 5.5 each measurement of amplitude can take the integer values
0-15. This represents a 4 bit sample resolution (as 24 = 16). For a 3 bit sample
resolution the horizontal axis of Figure 5.5 would only have eight values
(as 23 = 8).
Using a larger sampling resolution means that the amplitude measurement of
each sample will be captured more accurately, again increasing the accuracy
of the recording and the file size.
The file size required for an audio recording can be found by multiplying the
Note
length of the recording (in seconds) by the sample rate and the sampling
resolution. If audio is recorded in
stereo then there will be
file size = length of recording (seconds) × sample rate (Hz) × sampling
two recordings – a left
resolution (bits per sample)
channel and a right channel,
For example, for a 2 minute recording at 48 kHz and 16 bits per sample, the file doubling the overall file
size is: size requirements; in this
example, the stereo file size
file size = 120 seconds × 48 000 Hz × 16 bps would be 23.04 MB
= 92 160 000 bits
= 11 520 000 bytes
= 11.52 MB

Nyquist’s theorem
If the sample rate of a sound wave is too low then it is possible that the
recording will miss some peaks or troughs.
In order to ensure that the recorded sound does not miss this data, Nyquist’s Nyquist’s theorem The
theorem states that the sample rate should be at least double the maximum rule that the sample rate
hearing range. should be at least double
the maximum frequency
Since humans can hear sounds up to around 20 kHz, a minimum sample rate
130 that can be heard.
of 40 kHz should be used.

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_05_MRN_AQA_CS_103-138.indd Page 131 6/29/21 9:58 PM f-0116 /103/HO02206/work/indd

Sounds recorded at a lower sample rate may still be recognisable, but will not
be as accurate.
Telephone calls typically use a lower sample rate, and while this is suitable for
speech, music played over a telephone line sounds tinny and lacks definition.

5 Fundamentals of data representation

Musical Instrument Digital Interface (MIDI)
Musical Instrument Digital Interface (MIDI), is a protocol and file format that
Musical Instrument
is used to capture certain information about musical notes. This information
Digital Interface (MIDI)
includes:
A type of file that uses
✚ Pitch (how high or low the note is) the instructions for how
✚ Note length to create each note rather
✚ Note volume than capturing the analogue
✚ Instrument sound wave.
MIDI files can be created directly using a computer program or captured as
someone plays a MIDI-enabled instrument such as a keyboard.
Rather than capturing the analogue sound wave generated at the time, a
MIDI file records information such as which notes should be played, at what
volume and for what length of time.
This means you cannot listen directly to a MIDI file – you need some kind of
computer hardware and software to interpret it and generate sounds.
A MIDI file can be modified, and each individual note can be edited to change
any of its properties. For instance, if a musician played the wrong note this
can easily be corrected in the MIDI file.
Once captured, a MIDI file can then be used to generate sounds from any
MIDI instrument. For instance, a MIDI file captured from a piano performance
could be used to generate the same performance but on a completely different
instrument such as a synthesiser.
MIDI produces much smaller files than audio files and are much more easily
edited and corrected. Most modern music recordings make at least some use
of MIDI.

Now test yourself

64 What value is being measured when sound is recorded?

65 Identify two effects of decreasing the sample rate of a sound recording?
66 What are the effects of increasing the sampling resolution?
67 Which rule predicts that sound quality will noticeably degrade when the sample
rate drops below 40 KHz?
68 Calculate the file size in MB for a 5 minute recording at 40 KHz and a sampling
resolution of 8 bits per sample.
69 Give two advantages for using MIDI files.

Answers available online

Data compression
Data compression means reducing the size taken up by data. This is useful as
Compression Reducing
it means that less storage space is used on storage devices, and less time is
the size of a file.
taken to transmit data.
Sound files and image files are often compressed as these files typically use Note
large amounts of data.
Video files are made up of
Other files, including text files, data files and program files can also be image data and audio data.
compressed. Video files are not explicitly
covered in the specification,
but it can be assumed that
131
similar techniques apply.

My Revision Notes AQA A-level Computer Science Third Edition

325487_05_MRN_AQA_CS_103-138.indd Page 132 6/29/21 9:58 PM f-0116 /103/HO02206/work/indd

Lossy compression
In lossy compression some data is removed from the file. Once removed this
Lossy compression
data cannot be replaced.
Reducing the size of a file
5 Fundamentals of data representation

In image files this can involve reducing the pixel dimensions or reducing the by permanently removing
colour depth. some data.

Reducing the resolution means that fewer pixels are needed. While this
reduces the image quality, it is often possible to reduce the number of pixels
quite significantly without making the image unusable, especially if the
image is to be viewed on screen.

Figure 5.6 A high resolution image on the left and a low resolution image on the right

Reducing the colour depth means that each pixel is stored using fewer bits.
The cost of this is that fewer colours are available which can lower the overall
accuracy of the image.
The advantage of using lossy compression is that the file size can be
significantly reduced, often with minimal noticeable effects. The major
downside is that the lost data cannot be recovered.
Text files, program code and raw data cannot be compressed with lossy
compression as the data would lose its meaning - for example removing every
third letter from a piece of computer code would stop the code from working.

Lossless compression
Lossless compression, as the name suggests, does not involve losing or
Lossless compression
removing any data.
Reducing the size of a file
One technique to achieve this is run length encoding (RLE). without removing any data.

A run is a sequence of identical values, and the data can be shortened by Run length encoding
recording the data in pairs of values indicating the run length and then the (RLE) A lossless
value. compression technique of
recording the length and
The text data AAAAAAAAABBBBBBBBAAAAAA could be shortened to 9A 8B 6A. value of each run in the data.
Another method of lossless compression is to use a dictionary method. Run A series of identical
Each unique value in the data is recorded as a key-value pair. values in a file or dataset.

For example the phrase: Dictionary A data structure

made of key:value pairs.
“how much wood could a woodchuck chuck if a woodchuck could chuck wood”
could be rewritten using the following dictionary.

Key Value
1 how
2 much
3 wood
4 could
5 a
6 woodchuck
7 chuck
132 8 if

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_05_MRN_AQA_CS_103-138.indd Page 133 6/29/21 9:58 PM f-0116 /103/HO02206/work/indd

The phrase could then be stored using just the keys: “1 2 3 4 5 6 7 8 5 6 4 7 3”.

Making links
Dictionaries are a type of data structure commonly used in programming, and

5 Fundamentals of data representation

are included in the specification for Component 1. You can find out more about
dictionaries and how they are used in programming in Dictionaries in Chapter 2.

The major advantage of lossless compression is that no data is lost and, when
Note
decompressed, the original file is recreated exactly as it was. Any file can
potentially be compressed using lossless compression. Other compression
techniques exist, and
The major disadvantage of lossless compression is that file size reductions are
data compression is a
usually much more limited than with lossy compression. Depending on the
highly important area of
number of runs, or the number of unique values in the file, it is even possible
research with far-reaching
that the compressed version will not be any smaller than the original. implications. Only those
Now test yourself techniques described
here are included in the
70 Explain the purpose of compressing data. specification.
71 Identify two methods of reducing the file size of an image using lossy
compression. Revision activity
72 Why can program code not be compressed using lossy compression? Find examples of simple
73 Other than the answer above, give one disadvantage for using lossless compression. 8-bit graphics using an
74 A digital image is made of repeating patterns. Explain how a dictionary could be image search and practise
used to compress the image file. representing them using run
length encoding
Answers available online

Encryption
Encryption is a method of scrambling data so that it cannot be understood.
Encryption A method of
Encryption is carried out by starting with a plaintext message and applying a hiding the meaning of a
cipher and a key in order to generate a ciphertext message. message.

Once received, the ciphertext message is then decrypted back into the Plaintext The original,
original plaintext message. unencrypted message.
Cipher An algorithm for
Caesar cipher encrypting data.
One of the most basic ciphers is the Caesar cipher, an example of a Key A value used to encrypt
substitution cipher. or decrypt data.
In this cipher each letter is shifted by a fixed amount, or a key. For example, Ciphertext The encrypted
with a key of 2, each plaintext letter is moved two places further up the form of the message.
alphabet.
Decryption A method of
converting an encrypted
Plaintext letter A B C D E … X Y Z
message back to its original
Ciphertext letter C D E F G … Z A B form.
Caesar cipher A
Making links substitution cipher in
which each letter is shifted
Writing a computer program to carry out a Caesar cipher is good practice, and the
according to the key.
kind of task that can appear in Component 1, Section B. In order to wrap the alphabet
back to the start, the modulo operator can be used. Make sure you are comfortable Substitution cipher
with the programming techniques in Chapter 1 and try writing your own programmed A cipher in which each
version of the Caesar cipher. plaintext letter is replaced
with a ciphertext letter.
The plaintext message DAZE would be encrypted as FCBG.
Once the message is received, along with the key, the process is reversed, and
the ciphertext message FCA would be decrypted as DAY. 133

My Revision Notes AQA A-level Computer Science Third Edition

325487_05_MRN_AQA_CS_103-138.indd Page 134 6/29/21 9:58 PM f-0116 /103/HO02206/work/indd

The Caesar cipher is easily cracked as there are a maximum of 25 possible

Brute force Trying every
keys (a key of 26 would mean that the plaintext and ciphertext messages
possible combination.
would be identical). Trying every possible key is known as a brute force attack.
In order to speed up the cracking of a message encrypted using the Caesar
5 Fundamentals of data representation

cipher, frequency analysis can be used. This means that the most commonly
used letters (typically E and T in the English language) are likely to occur most
often.
Testing the keys where the most frequent letters align with E or T can further
reduce the time taken to crack this encryption method, though with only 25
possible keys it is trivially quick to break this code with a computer.

Exam tip
Although not listed directly in the specification, you may be asked a question based
on a substitution cipher of a similar complexity to the Caesar cipher (for example, with
the letters of the alphabet written backwards, or using space as a 27th character).
Practise with both of these examples to ensure that you are comfortable with the
principles of substitution ciphers.

Vernam cipher
Vernam cipher A
The Vernam cipher is also known as a one-time pad.
cipher which uses a
The cipher starts with choosing a random sequence of characters or binary randomly generated key
digits as the key. This key is the one-time pad as it must only be used once. and is mathematically
unbreakable.
The next task is to write down the binary codes for each character in the
plaintext message and in the key. One-time pad The key
used in the Vernam cipher.
Plaintext message n o w
Key g p w
Plaintext (binary) 0110 1110 0110 1111 0111 0111
Key (binary) 0110 0111 0111 0000 0111 0111

The individual bits in the plaintext and key characters are passed through an
XOR gate in order to generate the cipher text.

Making links XOR A logical operation

which checks whether
In an XOR (or exclusive-OR) gate, the output is 1 if either input is a 1, but not both. If
both inputs are equal, or
both inputs are 0 or both inputs are 1, then the output is 0. Logic gates are explored in
different.
more detail in Chapter 6.

Plaintext (binary) 0110 1110 0110 1111 0111 0111

Key (binary) 0110 0111 0111 0000 0111 0111
Ciphertext (binary) 0000 1001 0001 0000 0000 0000

The decryption process is identical. The ciphertext is passed through an XOR

gate with the key, and the plaintext is produced.

Ciphertext (binary) 0000 1001 0001 0000 0000 0000

Key (binary) 0110 0111 0111 0000 0111 0111
Plaintext (binary) 0110 1110 0110 1111 0111 0111
Plaintext n o w

The rules regarding the key state that it should:

✚ be truly random
✚ be at least as long as the ciphertext
✚ not be repeated
134 ✚ not be re-used.

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_05_MRN_AQA_CS_103-138.indd Page 135 6/29/21 9:59 PM f-0116 /103/HO02206/work/indd

If these rules are followed then it is mathematically impossible to reproduce

the plaintext message without the key, because each character uses a
different, unrelated key to its neighbours and every possible decrypted value
is equally likely.

5 Fundamentals of data representation

The biggest flaw in the Vernam cipher is that both the sender and receiver
must know which key is being used. This means that either the key must be
transmitted in a separate message (which could be intercepted) or that both
parties must have a copy of the keys in advance, in which case it may be
possible to physically retrieve the key.
The Vernam cipher is the only cipher in existence to have been
mathematically proven to be completely unbreakable, even given infinite time
and infinite ciphertext.
All other ciphers are based on computational security – that is, using
algorithms which are so complex that they would take far longer than is
useful to break. Theoretically all ciphers other than the Vernam cipher are
breakable given enough time and data. Symmetric The same key
is used in encryption and
Making links decryption.
Both ciphers described in this chapter are symmetric, which means they use the Asymmetric Different keys
same key for encryption and decryption. The importance of asymmetric encryption are used for encryption and
and the use of public/private key encryption is discussed in Chapter 9 (internet decryption.
security). Note that both topics are often assessed in the same question.

Now test yourself

Revision activity
75 What is meant by the terms plaintext and ciphertext?
✚ Practise encrypting and
76 Decrypt the message ‘LTTI QZHP’ which has been encrypted using the Caesar decrypting messages
cipher and the key 5. using the Caesar cipher,
77 Describe two methods that could have been used to decrypt the message if you using online tools to
did not know the key. check your accuracy.
78 State three rules for choosing a key to be used in the Vernam cipher. ✚ Practise encrypting and
79 Name one cipher which is mathematically impossible to break. decrypting messages
using the Vernam cipher,
80 Explain why that cipher is not routinely used.
using online tools to
Answers available online check your accuracy.

Summary

Number systems ✚ Binary values are usually written with a space after
✚ Numbers are classified in sets, including  (natural),  each block of 4, called a nibble
(integers),  (rational),  (real) and irrational numbers ✚ A block of n binary bits can represent 2n possible
✚ Ordinal numbers are used to identify the position of a values
value in a set ✚ Multiples of 103 (1000) bytes are referred to as 1 kB,
✚ Natural numbers ( - positive integers, including 0) are 1 MB, 1 GB, 1 TB, and so on
used for counting ✚ Multiples of 210 (1024) bytes are referred to as 1 KiB,
1 MiB, 1 GiB, 1 TiB, and so on
✚ Real numbers ( ) are used for measurement
Number bases Binary number system
✚ Numbers can be represented using any base ✚ Unsigned binary is used to represent natural numbers,
with a maximum range between 0 and 2 n − 1 (where n
✚ You are expected to be familiar with numbers in base 2
is the number of bits)
(binary), base 10 (decimal) and base 16 (hexadecimal)
✚ It is important to be confident adding and multiplying
✚ You are expected to be able to convert between
binary numbers manually
numbers in different bases
✚ Two’s complement is used to represent signed
✚ Hexadecimal is used as a shorthand for binary because
(negative) binary numbers
each hexadecimal digit maps exactly to one binary nibble
✚ In two’s complement the MSB a negative number (for
Bits and bytes example, in a 4 bit number: −8, +4, +2, +1)
✚ A single binary digit is called a bit ✚ Subtracting binary numbers can be done by adding the
✚ Bits are usually grouped into blocks of 8, called a byte negative version of the second number
➜ 135

My Revision Notes AQA A-level Computer Science Third Edition

325487_05_MRN_AQA_CS_103-138.indd Page 136 6/29/21 9:59 PM f-0116 /103/HO02206/work/indd

✚ The range of a two’s complement number is from ✚ A check digit is similar to a checksum, but is only
−2 n − 1 to 2 n − 1 − 1, because 0 is included as a positive represented by a single digit, often making use of the
number modulo operator
✚ Fractional numbers can be represented using fixed
5 Fundamentals of data representation

Representing images, sound and other data

point or floating point binary
✚ In fixed point binary, the decimal place position is ✚ Analogue measurements are continuous rather than
fixed and values after this are weighted as 1 , 1 , 1 , and discrete and are often represented with a curved wave
2 4 8 ✚ Digital measurements are discrete and are often
so on
✚ In floating point binary the value is written using a represented with a stepped wave
mantissa and an exponent ✚ Data can be converted between analogue and digital
✚ The binary point in the mantissa is moved by the using an analogue to digital converter (ADC) or digital
number of places indicated in the exponent to analogue converter (DAC)
✚ Both exponent and mantissa are usually written using ✚ The most common use for a DAC is to convert a digital
two’s complement binary in exam questions audio signal to an analogue signal
✚ You are expected to be able to convert fractions ✚ Images can be represented using bitmap images or
between decimal and binary. vector images
✚ Binary fractions cannot represent all decimal fractions ✚ The size of an image typically refers to its width ×
with complete accuracy as they use inverse powers of height, in pixels (px)
2 ( 1 , 1 , and so on) ✚ Resolution refers to pixel density, measured in dots per
2 4 inch (DPI)
✚ The absolute error is the difference between the
✚ The colour depth of an image refers to how many bits
original value in decimal to be stored and the binary
are used to store the colour of each pixel
value that is actually stored
✚ For a bitmap image file: file size = width (px) × height
✚ The absolute error is always written as a positive value
(px) × colour depth
✚ The relative error is found by dividing the absolute error
✚ Metadata is additional data stored in a file, and for an
by the original value, multiplied by 100 and written as a
image file this can include the pixel dimensions, colour
percentage (%)
depth, data edited and GPS data
✚ Fixed point binary is quicker to process, but less
✚ Bitmap graphics are suitable for photo-realistic images
precise and more wasteful of storage
due to the large range of colours and their ability to
✚ Floating point binary takes longer to process, but is
represent images that are not made of simple shapes
more precise and more efficient in terms of storage
✚ Vector graphics are made by combining one or more
✚ In fully normalised floating point only one digit is placed
vector primitive objects, which describe the properties
before the binary place and this will be the opposite of
of each object, including its co-ordinates, line colour
the first digit after the binary place, in order to avoid
and fill colour
wasting space with leading digits
✚ Vector graphics are suitable for logos and clipart due to
✚ Overflow occurs when the result of adding binary
their ability to scale without quality loss and their small
values is too large
file size
✚ Underflow occurs when the result of subtracting binary
✚ Analogue sound is captured by measuring the
values is too small
amplitude of the sound wave at fixed intervals
Information coding systems ✚ The sample rate is measured in kHz and increasing the
✚ Numbers can be represented as a pure binary value, or sample rate increases the accuracy of the recording
as a character code ✚ Nyquist’s Theorem states that the minimum sample
✚ ASCII and Unicode are coding systems in which each rate should be at least double the maximum frequency
character has a unique binary code you wish to sample – for sound waves this maximum
✚ ASCII uses 7 bits per character, for a maximum of 128 frequency is hearing range (up to 20 kHz in humans)
characters ✚ The sample resolution is the number of bits per sample
✚ Unicode uses either 8, 16 or 32 bits per character, ✚ For a sound file: file size = length of recording
which is needed to represent other alphabets and (seconds) × sample rate (Hz) × sampling resolution (bits
symbols including emojis per sample)
✚ Error checking can be carried out using parity bits, ✚ MIDI stands for Musical Instrument Digital Interface and
majority voting, checksums, or check digits is used to represent information about each individual
✚ Parity bits use odd or even parity and append either a 0 note or sound in a piece of audio
or a 1 to a binary message to ensure that is has an odd ✚ Each MIDI note can be individually edited in order to
or even number of 1s change features such as its pitch or length without
✚ In majority voting each bit is transmitted a fixed, odd affecting the rest of the sound
number of times, and if all bits received are not the ✚ File compression is used to reduce the size of files in
same, the majority value is assumed to be correct order to improve storage capacity and/or transmission
✚ A checksum uses an algorithm to generate a value times
based on the data in the original message which can ✚ Image and sound files are most frequently compressed,
be recalculated once the message has been received though text files and program files can also be
to check that the message is identical compressed
136 ➜

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_05_MRN_AQA_CS_103-138.indd Page 137 6/29/21 9:59 PM f-0116 /103/HO02206/work/indd

✚ Lossy compression refers to compression techniques ✚ The encryption algorithm is known as a cipher and the
in which some data is removed (for example, pixels value used to encrypt and decrypt the data is known as
and/or colour depth in an image, samples and/or the key
sample resolution in a sound file ✚ The Caesar cipher is a substitution cipher in which each

5 Fundamentals of data representation

✚ Any data removed from a file during lossy compression letter is shifted along the alphabet by a fixed amount
cannot be recovered, but significant savings in file size dictated by the key
can be made for a relatively low loss of quality ✚ The Caesar cipher is trivially easy to break as there are
✚ Text and program files cannot be compressed with only 25 possible keys to try
lossy compression as they would become meaningless ✚ The Vernam cipher is the only mathematically secure
✚ Lossless compression refers to compression cipher
techniques in which no data is removed and the ✚ Each character to be encrypted is passed through an
decompressed file is identical to the original XOR gate along with a randomly selected key
✚ Run length encoding (RLE) is a lossless compression ✚ Because the key is completely random and each
technique in which each run of identical values is character is encrypted with a different key, the
recorded as a pair of run:length values ciphertext is distributed uniformly and each possible
✚ Dictionary-based methods use a table of key:value decryption value is equally likely
pairs in order to represent the data using its keys rather ✚ The Vernam cipher is only effective if the key is
than the full values in each case completely random and only ever used once
✚ Encryption is a method of scrambling data so that it ✚ The major problem with the Vernam cipher is that both
cannot be understood by anyone but the intended sender and recipient must know the exact key
recipient ✚ All other ciphers are based on computational
✚ The original message is referred to as plaintext and the complexity and are therefore potentially breakable
encrypted message is referred to as ciphertext given enough time and ciphertext

Exam practice

1 Four numbers are listed below. i) The number 7.8 is represented as 0111.1100.
a) For each number, tick one or more boxes to show Calculate the absolute and relative error. [3]
which sets it belongs to. Some numbers belong to j) Calculate the decimal value of the floating-point
more than one set. [4] binary number with a mantissa of 0.110 1000 and an
exponent of 0110. You must show your working. [2]
Natural Integer Rational Irrational Real k) Write the normalised floating point representation
17.4 of −426 using a 10 digit mantissa and a five digit
exponent. [3]
√2
3 The letter H is represented in ASCII using the binary
7 code 100 1000.
−12 a) State the binary codes for the letters I and K. [2]
b) Explain why someone writing in using a different
b) State which value from the table above is an ordinal alphabet might not be able to use ASCII encoding
number and explain the purpose of an ordinal and suggest an alternative system. [2]
number. [2] c) The letter H is transmitted using odd parity, with the
2. a) Represent the number 113 as an 8-bit unsigned parity bit appended before the most significant bit.
integer. [1] State the binary code which is transmitted. [2]
b) Represent the number -47 as an 8-bit two’s d) The binary code 1011 0101 is received, still using
complement integer. [1] odd parity. State whether this transmission would
be accepted. [1]
c) Use binary addition to find the value of 113 – 47. [2]
e) Identify two flaws with using a parity check. [2]
d) Using only binary values, calculate the value of
23 × 9. [3] f) Another transmission is sent using majority voting
in which each bit is transmitted three times. The
e) Name the problem that would occur if finding the
data received is 111 000 110 011 101. State the
value of 27 × 11 using 8-bit unsigned binary. [1]
original data. [1]
f) Represent the number 113 as a hexadecimal
g) Explain why it would not have been suitable to
number. [2]
transmit each bit four times. [1]
g) Explain why it may sometimes be preferable to
h) Some text has been encrypted using the Vernam
represent a value using hexadecimal rather than
cipher. Explain what is meant by encryption. [1]
binary. [2]
i) Describe why the message was not encrypted
h) Calculate the decimal value of the two’s
using the Caesar cipher. [1]
complement fixed-point binary number
1101.0110. [2]
➜ 137

My Revision Notes AQA A-level Computer Science Third Edition

325487_05_MRN_AQA_CS_103-138.indd Page 138 6/29/21 9:59 PM f-0116 /103/HO02206/work/indd

j) The received message is as follows, and the key is c) Identify two effects of increasing the colour
MOFK. The ASCII code for the letter H is 100 1000. depth of an image to 9 bits. [2]
Decrypt the message and show the plaintext d) Describe two advantages for saving the file
message. You must show your working. [4] as a vector graphic, rather than a bitmap. [2]
5 Fundamentals of data representation

000 0101 e) Identify two methods of lossy compression

000 1010 that could be applied to a bitmapped image. [2]
000 1010 f) Explain why run length encoding is unlikely
001 1011 to have a significant impact on the file size of a
bitmap image. [1]
4 A digital photograph is 4000 pixels wide, 3000 pixels tall
and has a colour depth of 8 bits. g) Explain why lossy compression could not be
applied to a program file containing source
a) Calculate the size of the file in MB. [2]
code. [2]
b) State the maximum number of colours in
the image. [1]

138

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_06_MRN_AQA_CS_139-157.indd Page 139 6/29/21 10:26 PM f-0116 /103/HO02206/work/indd

6 Fundamentals of computer
systems

Hardware and software

Relationship between hardware and software
The term hardware relates to the physical components of a computer system.
Hardware The physical
This includes the:
components of a computer
✚ internal components (including RAM and the CPU)
system.
✚ peripheral components (including secondary storage devices, printers,
monitors, keyboards and mice). Peripheral External
hardware devices, on the
The term software relates to the program that run on computer systems. This periphery of the computer
includes the: system.
✚ operating system
✚ utility programs Software The programs
that run on a computer
✚ application programs.
system.
Software requires hardware to carry out the instructions, and hardware
requires software in order to be given instructions to carry out.

Now test yourself

1 Identify three pieces of hardware commonly found as part of a laptop computer.

2 Identify three pieces of software commonly found installed on a laptop computer.

Answers available online

Classification of software
There are two broad classes of software: system software and application
software.
System software is intended to allow the computer system to run. This
System software
includes:
Software intended to allow
✚ operating systems ✚ libraries the computer system to
✚ utility programs ✚ translators. run.
Systems software is there to support the running of the computer system, Application software
rather than to achieve a specific outcome for the user. Software intended to allow
Application software is intended to allow the computer to serve a useful the end-user to achieve a
purpose. This includes: task.
✚ word processing software
✚ spreadsheet software
✚ web browsers.
General-purpose application software can be used to achieve multiple tasks,
for example word processing software can be used to write a letter, make
notes or create a display.
Special purpose application software is written with a more specific purpose in
mind and cannot easily be used for a different purpose. Examples include audio
editing software, flight simulator training software and computer games.

139

My Revision Notes AQA A-level Computer Science Third Edition

325487_06_MRN_AQA_CS_139-157.indd Page 140 6/29/21 10:26 PM f-0116 /103/HO02206/work/indd

System software
Operating systems:
✚ provide an interface between the user and the hardware
6 Fundamentals of computer systems

✚ run application programs

✚ manage resources and hardware.

Utility programs are individual programs which help configure, maintain or

optimise the computer system. Examples of utility programs include:
✚ virus scanning ✚ disk cleanup
✚ file manager ✚ backup software
✚ system monitor ✚ disk defragmenter.

Library programs are collections of resources or code that can be used by

other programs. They include pre-written subroutines which can be called by
other programs on the system.
Translators are used to translate high-level source code into low-level
machine code. Examples include:
✚ assemblers ✚ compilers ✚ interpreters.

Now test yourself

3 Name the four main types of system software.

4 State whether a word processor is an example of application software or system
software.
5 Name three different examples of utility programs.
6 Describe the purpose of a library.

Answers available online

Role of an operating system (OS)

An operating system (OS) serves three main purposes, all of which are
related. An OS:
✚ hides the complexities of the hardware
✚ provides an interface for the user
✚ manages and allocates system resources.

User interface
The user interface can be graphical (GUI), command line (CLI), menu driven or
voice controlled.

A GUI is commonly used on desktop and mobile devices, and is typically

User interface The tools
made up of windows, icons, menus and pointers. This is often shortened to
that are provided for a user
the acronym WIMP.
to interact with a computer
A CLI is often used for remote computers or servers. The lack of graphical system.
cues makes a CLI less user-friendly, but those confident using text-based Graphical user interface
commands can make changes quickly and efficiently. (GUI) An image-based
interface that uses
A CLI is also ideal for remote connections as no graphical data needs to be
windows, icons, menus and
transmitted.
pointers.
A menu-driven interface is often used on devices with few inputs, such as a
Command line interface
(non-touchscreen) watch or digital set top box.
(CLI) A text-only interface
Voice controlled interfaces can be used in situations where using a manual which is less user-friendly
input method are not practical. This includes phone-based systems and but often more powerful for
knowledgeable users.
digital home assistants.

140

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_06_MRN_AQA_CS_139-157.indd Page 141 6/29/21 10:26 PM f-0116 /103/HO02206/work/indd

Hardware resources
Running the computer hardware includes significant complexities, which
are hidden from the user because the operating system handles them in the
background.

6 Fundamentals of computer systems

Tasks include:
✚ Memory management: Programs are loaded into RAM when run; the most
frequently used instructions are loaded into the cache for faster access and
the least frequently used instructions are moved into virtual memory.
✚ CPU allocation: Individual instructions are passed to the CPU with
differing levels of priority. Multi-core processors require complex solutions
in order to make the most efficient use of the CPU, with minimal delays
while dependent tasks are still running.
✚ Peripheral management: Input/output devices need to be managed, and
each data stream needs to be processed in competition with the programs
already running.

Making links
In order to fully appreciate the hardware complexities which are hidden by the
operating system it is important to understand system architecture, including the
role of the CPU and RAM. This topic is discussed in Chapter 7 (Internal hardware
components).

Now test yourself

7 State the main purpose of an operating system.

8 Identify three aspects of hardware management which are run by the operating
system.
9 Describe two features of a graphical user interface.
10 Name two types of user interface other than a GUI.

Answers available online

Classification of programming
languages
Computer programming has evolved over the lifetime of computers and there
are several different classifications for programming languages.

Low-level languages
There are two types of low-level language.
Low-level language A
Machine code refers to the binary code which is directly acted upon by the programming language
processor. which describes exactly
how to interact with the
Each type of processor has its own machine code instruction set with
computer’s hardware.
individual commands for tasks such as fetching data from memory, saving
data to memory, adding values, etc. Machine code Each
instruction is represented
The only language that a computer understands is machine code. as a binary code.
Assembly language is a text-based equivalent to machine code. Remembering Assembly language Each
and correctly entering the exact binary codes is difficult, so assembly instruction is represented
language allows each binary instruction to instead be represented by a short as a text-based command.
code, which is closer to the English language. This makes it easier to program.

141

My Revision Notes AQA A-level Computer Science Third Edition

325487_06_MRN_AQA_CS_139-157.indd Page 142 6/29/21 10:26 PM f-0116 /103/HO02206/work/indd

For instance, the assembly language instructions:

ADD R3, R1, R2
would add the value in register 2 to the value in register 1 and store the result
6 Fundamentals of computer systems

in register 3.
Each assembly instruction maps to one machine code instruction, and each
type of processor has its own assembly language.

Making links
AQA has developed its own assembly language instruction set and it is normal to
include a question in which students are expected to read or write programs using
assembly language. The specific details are discussed in Chapter 7 (The processor
instruction set).

Now test yourself

11 Name the two different examples of low-level programming languages.

12 Explain why one of those languages is more commonly used by programmers.

Answers available online

High-level language
High-level languages use structured statements (for example, IF statements,
High-level language A
WHILE loops, and so on) in order to simplify the task of describing an
programming language
algorithm.
which uses keywords and
Keywords and constructs are written in an English-like form which makes constructs written in an
them more recognisable. English-like form.

High-level languages also include features such as local variables, parameters,

named constants and indentation – features which are not found in assembly
language code.

One line of code from a high-level language might require multiple machine
code instructions.

High-level languages include those you will have studied for Component 1,
such as:
✚ C# ✚ Python
✚ Delphi/Pascal ✚ Visual Basic.net
✚ Java
Imperative high-level languages are languages in which the commands are
carried out in a programmer-defined order. The term imperative refers to Imperative A language in
which commands are used
types of programming language where the programmer is describing how
to say how the computer
the computer should achieve the desired result. All of the examples above are
should complete the task.
imperative high-level languages.
Declarative A language
The alternative to imperative high-level languages is declarative high-level in which the programmer
languages, in which the programmer writes a program that describes what codes what they want to
the program should achieve, but not how. happen, but not how.

For example, the command for searching through a table of data does not
contain an explicit loop. Exam tip
Remember that imperative
Examples of declarative programming languages include:
refers to the specifics of
✚ SQL ✚ LISP how to solve the problem,
✚ Haskell ✚ Prolog whereas declarative refers
142 to what the program should
achieve.

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_06_MRN_AQA_CS_139-157.indd Page 143 6/29/21 10:26 PM f-0116 /103/HO02206/work/indd

Low-level versus high-level

Language Advantages Disadvantages Note

6 Fundamentals of computer systems

Low-level Faster / more efficient execution as Will only run on that platform – the
While all CPUs of the same
the program can be optimised for a programmer needs to know which
type use the same basic
particular processor. type of CPU and which operating
instruction set, the choice
system the code is being written for.
Requires less memory. of operating system also
More difficult for the programmer has an impact. This means
Provides direct control of the
to write code. that a program compiled
hardware.
for an Intel CPU running
High-level Is independent of CPU and will run Slower / less efficient execution. Windows will not run on the
on a wide range of platforms. same CPU running Linux.
Requires more memory.
Easier for the programmer to
Doesn’t provide direct access to
write code, and therefore create
the hardware.
complex programs.

Now test yourself

13 Describe one reason why high-level languages are more commonly used than
low-level languages.
14 Explain the meaning of imperative in relation to high-level languages.
15 Suggest one situation where a programmer may prefer to use a low-level
programming language.

Answers available online

Types of program translator

Programs written in any language other than machine code must be Source code Code written
translated. In order for a program to run it must be translated from the by a programmer.
original source code into object code.
Object code Low-level
There are three main types of translator. code, translated from the
source code.
Note
Translator Systems
Machine code refers to the code directly acted on by the CPU. Object code is a portion software for converting one
of machine code which is not yet fully linked to a complete program. The difference is form of code into object
subtle and it is beyond the scope of A-level. Either term is likely to be accepted. code.

Assemblers, compilers and interpreters

Assembler
An assembler is used to translate assembly language instructions into object
Assembler Translates
code instructions.
assembly language code
Because assembly language and machine code map one-to-one, each into object code.
assembly language instruction is translated directly into its binary equivalent.
Compiler Translates high-
level code into object code
Compiler as a batch.
A compiler is used to translate high-level code into object code. This is a more
complex process than with an assembler as one line of high-level code may
involve several machine code operations.
A compiler will process all of the code in a program in one batch, and will
produce an executable binary file (often with a .exe file extension, though not
always).
143

My Revision Notes AQA A-level Computer Science Third Edition

325487_06_MRN_AQA_CS_139-157.indd Page 144 6/29/21 10:26 PM f-0116 /103/HO02206/work/indd

Once compiled, the executable file can be run multiple times without the
need to translate the source code again.

Advantages Disadvantages Uses

6 Fundamentals of computer systems

The executable file is much The executable file will Producing a program that
quicker to run than the be compiled specifically will be run many times
source code. for one set of machine without frequent changes
code instructions and the to the source code.
Once compiled, the
executable file will only run
executable file can be run Producing a program where
on that platform.
without the need to translate the programmer wishes
the source code again. The whole program must to keep the details of the
be fully compiled before algorithms secret.
The programmer can share
it can be run and this can
the executable file with
slow down the process of
others, but the program
debugging.
cannot be easily edited or
algorithms in the source If there is an error in the
code copied as only the program then the program
machine code is present in will not compile, meaning
the file. that a partially working
program cannot be tested.
If the compiler detects any
errors then it will try to
inform the user of all errors
that have been found.

Interpreter
An interpreter achieves the same basic goal as a compiler – translating
Interpreter Translates and
high-level code into machine code. The main difference is that an interpreter
then runs high-level code
translates and then executes one line of code at a time.
one section at a time.
An interpreter doesn’t produce an executable file, but must be re-translated
each time it is run.

Advantages Disadvantages Uses

The program can be run The program must be Producing a program
on any platform that has translated each time it that may need to run
an interpreter for that is run, meaning that the on a variety of different
language, making it easier program will run more platforms.
to run the program on slowly than compiled code.
Debugging a large program
different systems.
The source code must be that would take a long time
Debugging can be quicker made available to anyone to compile.
as the whole program running the program, as
doesn’t need to be the user must translate the
translated in order to test source code at runtime.
one part.
The program will stop at
If there is an error in the first error it encounters,
the program then the and so multiple errors will
interpreter will not detect not be identified at once.
this until it reaches that
part of the program, and
so a partially completed
program will still run, up to
a point.

144

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_06_MRN_AQA_CS_139-157.indd Page 145 6/29/21 10:26 PM f-0116 /103/HO02206/work/indd

Bytecode
In order to help compiled programs run on a wider range of devices, some
Bytecode An intermediate
programming languages are compiled to bytecode, rather than object code.
code between high-level

6 Fundamentals of computer systems

The source code is translated using a compiler, to an intermediate language, and object code which can
called bytecode. When the program is executed the bytecode is executed in run in a virtual machine.
a virtual machine. This means that the physical computer system creates a
Virtual machine Software
virtual computer system that has a common instruction set, no matter what
that emulates or simulates a
platform the physical computer system is running.
computer system.
When the bytecode is executed it is translated by the virtual machine, into
object code that can be understood by the physical computer system.
This means that any platform that can support the virtual machine can also
run the program.
This allows the compiled program to be run on a much wider range of devices
without needing to be recompiled and also allows for optimisation by the
virtual machine for that platform.

Now test yourself

16 What is the common purpose of all translators?

17 Which translator would be used with code written in assembly language?
18 Describe the main difference between a compiler and an interpreter.
19 Identify one advantage and one disadvantage for using an interpreter over a compiler.
20 Explain the purpose of bytecode.

Answers available online

Logic gates
Logic gates are physical devices which take one or more inputs and produce
an output according to certain logical rules. Logic gate Device which
takes one or more binary
There are six basic logic gates which students are expected to recognise and outputs and produces a
be able to draw. single binary output.
The function of each logic gate can be described using a truth table, which Truth table A table showing
shows the possible inputs and the equivalent outputs. the possible inputs and their
corresponding outputs.
Each logic gate can also be represented using a Boolean expression.
Boolean expression A
mathematical notation for
logic gates and circuits.

Types of logic gate

Symbol Truth table NOT gate
A Q The NOT gate is also
referred to as an inverter, as
A Q 0 1 it inverts (or flips) the input
value.
Inverter or NOT Gate 1 0
Note that this is the only
Boolean expression Q = A Q = NOT A gate which only uses one
input.
Figure 6.1 A NOT gate and its truth table

145

My Revision Notes AQA A-level Computer Science Third Edition

325487_06_MRN_AQA_CS_139-157.indd Page 146 6/29/21 10:26 PM f-0116 /103/HO02206/work/indd

Symbol Truth table

A B Q
AND gate
The AND gate produces an
0 0 0
6 Fundamentals of computer systems

A output only if input A AND

Q input B are on.
B 0 1 0

2-input AND Gate Note that this gate looks like

1 0 0
a letter D, which is the last
letter in AND.
1 1 1
Boolean expression Q = A.B Q = A AND B

Figure 6.2 An AND gate and its truth table

Symbol Truth table

OR gate
A B Q
The OR gate produces an
output if either input A OR
0 0 0
A input B are on OR both.
Q
B 0 1 1 Note that each line on the
2-input OR Gate drawing for this gate is
1 0 1
curved, as are the letters
in OR.
1 1 1
Boolean expression Q = A+B Q = A OR B

Figure 6.3 An OR gate and its truth table

Symbol Truth table

XOR gate
A B Q
The XOR (or exclusive OR)
gate produces an output if
0 0 0
A either input A OR input B are
Q on but not both.
B 0 1 1

2-input Ex-OR Gate Note that this gate looks

1 0 1
very similar to an OR gate,
with an extra curved bar on
1 1 0
the input.
Boolean expression Q = A + B Q = A XOR B

Figure 6.4 An XOR gate and its truth table

Symbol Truth table

A B Q
NAND gate
The NAND gate is equivalent
0 0 1
A to an AND gate followed by
Q a NOT gate.
B 0 1 1
Note the round ‘nose’ on
2-input NAND Gate
1 0 1 the front to indicate that the
result should be inverted.
1 1 0
Boolean expression Q = A.B Q = NOT (A AND B)

Figure 6.5 A NAND gate and its truth table

146

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_06_MRN_AQA_CS_139-157.indd Page 147 6/29/21 10:26 PM f-0116 /103/HO02206/work/indd

Symbol Truth table

A B Q
NOR gate
The NOR gate is equivalent
0 0 1

6 Fundamentals of computer systems

A to an OR gate followed by a
Q NOT gate.
B 0 1 0

2-input NOR Gate Again, note the round ‘nose’

1 0 0
on the front.
1 1 0
Boolean expression Q = A+B Q = NOT (A OR B)

Figure 6.6 A NOR gate and its truth table

Now test yourself

21 State the name of the gate which produces an output of ‘on’ (or 1) when one input
is on, but not both.
22 State the difference in the symbols for an AND gate and a NAND gate.
23 State another name for an inverter.

Answers available online

Logic circuits
Logic circuits which contain two or more gates can be devised to solve
Logic circuit A solution to
practical problems.
a problem that uses one or
For example, the windscreen wipers (O) on a car may activate if the engine (E) more logic gates.
is switched on and either the wiper switch is activated (W) or the rain sensor
is activated (S).
The logic circuit for this scenario is shown below, along with the relevant
truth table.

E
O

W
S

Figure 6.7 Logic circuit diagram for a car’s windscreen wiper

E W S O
0 0 0 0
0 0 1 0
0 1 0 0
0 1 1 0
1 0 0 0
1 0 1 1
1 1 0 1
1 1 1 1

The circuit can also be represented using its Boolean expression: O = E.(W+S)
You may be asked a question which involves translating between any two of:
✚ a written description of a problem
✚ a logic circuit diagram
✚ a logical expression.
147

My Revision Notes AQA A-level Computer Science Third Edition

325487_06_MRN_AQA_CS_139-157.indd Page 148 6/29/21 10:26 PM f-0116 /103/HO02206/work/indd

Worked example
–
Draw a logic circuit to represent the Boolean equation A.B + A.C.
–
The circuit can be created by combining the output of A.B with the output of A.C, using an OR gate.
6 Fundamentals of computer systems

First, draw the part of the

B circuit that represents. A.B.

Second, draw the part of the

B
circuit that represents. A.C.

Finally, connect the two using

B Q
an OR gate.

Figure 6.8

148

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_06_MRN_AQA_CS_139-157.indd Page 149 6/29/21 10:26 PM f-0116 /103/HO02206/work/indd

When completing truth tables for complex circuits it can be helpful to include
intermediate points.

Worked example

6 Fundamentals of computer systems

Complete the truth table for the following logic circuit and write the
Boolean expression.
A
D
B
Q

C E

Figure 6.9 A more complex logic circuit

To complete the truth table we would consider the intermediate points labelled D and
E on the diagram, and given by the expressions:
D = A.B
E = B.C

Inputs Intermediate points Output

A B C D E Q
0 0 0 0 0
0 0 1 0 1
0 1 0 0 0
0 1 1 0 0
1 0 0 1 0
1 0 1 1 1
1 1 0 0 0
1 1 1 0 0

We would then complete the table by noting that:

Q = D⊕E

Inputs Intermediate points Output

A B C D E Q
0 0 0 0 0 0
0 0 1 0 1 1
0 1 0 0 0 0
0 1 1 0 0 0
1 0 0 1 0 1
1 0 1 1 1 0
1 1 0 0 0 0
1 1 1 0 0 0

Using the expression Q = D⊕E and substituting for D and E we can construct the full
Boolean expression:
– –
Q = A.B ⊕ B.C

149

My Revision Notes AQA A-level Computer Science Third Edition

325487_06_MRN_AQA_CS_139-157.indd Page 150 6/29/21 10:26 PM f-0116 /103/HO02206/work/indd

Now test yourself

– –
24 a Draw a logic circuit to represent the Boolean equation Q = A.B + A.B.
b Complete a truth table for the Boolean equation.
6 Fundamentals of computer systems

25 Complete a truth table for the logic circuit in Figure 6.10.

C
B

Figure 6.10

Answers available online

Adders
Two of the most important logic circuits are the half adder and the full adder.
Half adder A logic circuit to
The half adder is used to add two binary numbers so that: add two binary digits.
0 + 0 = 0 carry 0 Full adder A logic circuit to
add three binary digits.
0 + 1 = 1 carry 0
1 + 0 = 1 carry 0
1 + 1 = 0 carry 1
When written out as part of a larger binary addition it is common to use the
following layout.
0 1 1
1 + 0 + 1 +
1 1 1 0
The outputs are labelled as S (sum) and C (carry).

A
Exam tip
S
B You may be asked to
construct the circuit for a
half adder. If so, start with
the truth table and the
C two gates required can be
worked out from there.
Figure 6.11 A half adder logic circuit

The Sum is calculated using an XOR gate and the Carry is calculated using an
AND gate.

INPUT OUTPUT
A B C S
0 0 0 0
0 1 0 1
1 0 0 1
1 1 1 0

150

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_06_MRN_AQA_CS_139-157.indd Page 151 6/29/21 10:26 PM f-0116 /103/HO02206/work/indd

When adding two digit (or larger) numbers, it is necessary to add 3 inputs –
the two numbers being added and the carry from the previous digit. These are
labelled A, B and Cin (Carry in).
When adding three binary numbers there is an additional possible case:

6 Fundamentals of computer systems

1 + 1 + 1 = 1 carry 1
To do this we need a circuit called a full adder. This is made from two half
adders joined with an OR gate to generate the Sum and Carryout.
A
B S
Cin

Cout

Figure 6.12 A full adder logic circuit

The truth table gives some clues as to the overall purpose of the logic circuit.

INPUT OUTPUT
A B Cin Cout S
0 0 0 0 0
0 0 1 0 1
0 1 0 0 1 Exam tip
0 1 1 1 0
You will not be asked to
1 0 0 0 1
construct a full adder, but
1 0 1 1 0 you may be shown one and
1 1 0 1 0 asked to explain its function,
1 1 1 1 1 or to complete a truth table.

Now test yourself

26 State the two gates used in a half adder.

27 Describe the difference in purpose between a half adder and a full adder.
28 Explain why a full adder would be needed to add 4-bit binary numbers.

Answers available online

D-type flip-flop
The edge triggered D-type flip-flop is a logic circuit which can be used to store
Edge triggered D-type
a state. This concept is the basis of computer memory. flip-flop A logic circuit used
✚ The circuit takes two inputs, a data input, and a clock input signal which to store the state of an
changes from 0 to 1 and back again at regular intervals. input.
✚ When the clock input signal changes from 0 to 1 (called the rising edge),
the output changes to match the current data input signal. Rising edge The point
✚ In between the rising edges, at any other point in the clock signal, any at which the clock signal
changes from 0 to 1.
changes in the input data has no effect on the output – the old output will
still appear, regardless of the current data input. Volatile memory Memory
✚ This can be used to check if a value has changed since the last timing which can only store a value
signal (by comparing the current output of the signal to its input), or to when supplied with power.
store data to be used later.
It should be noted that this circuit still requires power and is therefore a type
of volatile memory. 151

My Revision Notes AQA A-level Computer Science Third Edition

325487_06_MRN_AQA_CS_139-157.indd Page 152 6/29/21 10:26 PM f-0116 /103/HO02206/work/indd

Exam tip
You are not expected to recall the inner workings of a D-type flip-flop. However, you
should remember that:
6 Fundamentals of computer systems

✚ it is used to store the state of an input

✚ it takes two inputs – data and clock.
And be prepared to answer a question in context (for example, leaving a car headlight
on for a short time even after the ignition is switched off – the headlight should go off
based on both inputs, the state of the ignition and the clock).

Now test yourself

29 State the two inputs required for a D-type flip-flop.

30 Explain the purpose of a D-type flip-flop.

Answers available online

Boolean algebra
Using Boolean algebra
Each logic gate and each logic circuit can be written as a Boolean expression.
More complex logic circuits can often be replaced with a simpler design, and
so Boolean expressions are checked to see if they can be simplified.
There are three main steps involved in simplifying a Boolean expression.

Boolean identities
A Boolean identity refers to an expression where there is only one input, and
Boolean identity A
therefore the circuit can be simplified.
relation that is always true.
For example the Boolean expression A.1 would produce the following truth
table.

A 1 A.1
0 1 0
1 1 1

As the second input is always on (represented as ‘1’), the output is always the
same as the value of A.
–
Another Boolean expression that can be simplified is A.A .
– –
A A A.A
0 1 0
1 0 0

As A and NOT A are opposite, the output from the AND gate must always be 0.
There are eight main Boolean identities to remember:
A+0=A A .A = A
A .0 = 0 A+A=A
–
A+1=1 A. A = 0
–
A .1 = A A+A=1

152

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_06_MRN_AQA_CS_139-157.indd Page 153 6/29/21 10:26 PM f-0116 /103/HO02206/work/indd

Exam tip
In the Boolean expressions above, whilst A is a single input it could represent a more
complex expression.

6 Fundamentals of computer systems

Take the expression Q = A + 0 which can be represented by the logic diagram:

O
Q
A

Figure 6.13 Q = A + 0
Now consider the following diagram:

O
B Q
C A

Figure 6.14
Now we can see that A is in fact the output of B.C and, therefore, the expression
above can be written as:
Q = B.C + 0, where A = B.C
and, since A + 0 = A,
B.C + 0 = B.C
In fact, it doesn’t matter what expression is contained within dotted box – if the box
has one output then you can think of it as: ‘box + 0 = box’. And the same goes for any
of the Boolean identities above.

Now test yourself

31 Simplify each of these Boolean expressions using a Boolean identity.

a B+1
b D+D
–
c C.C
d (A.B).1

Answers available online

Rearranging Boolean expressions

The rules for rearranging Boolean expressions work in exactly the way you
would expect based on traditional algebra taught in maths.
Boolean expressions are commutative.
This means that:
✚ A.B can be rewritten as B.A (just as 3 × 5 = 5 × 3).
✚ A + B can be written as B + A (just as 3 + 5 = 5 + 3)
Boolean expressions are associative.
This means that brackets can be rearranged as long as the operations are
equal. For example:
A + (B+ C) = (A + B) + C, and
(A.B).C = A. (B.C)
BIDMAS applies.
–
This means that in the Boolean expression A + A.(A + B) the AND operator
–
takes precedence over the OR operator between A and A.
153

My Revision Notes AQA A-level Computer Science Third Edition

325487_06_MRN_AQA_CS_139-157.indd Page 154 6/29/21 10:26 PM f-0116 /103/HO02206/work/indd

The first two terms cannot be removed as a Boolean expression because the
Note
A– must be multiplied first.
AND operations should
be considered to have the
Expanding and factorising brackets
6 Fundamentals of computer systems

same level of precedence

As with traditional algebra, brackets can be expanded, and factorising can as a multiply operation.
reduce common factors.
OR operations should be
Expanding brackets involves multiplying each term in the brackets by the considered to have the
value outside: same level of precedence
A.(B + C) = A.B + A.C as an addition operation.
Factorising is the reverse process, identifying common factors and removing
them:
Expanding brackets
A.B + B.C = B.(A + C)
The process of removing
Note that a single input can be factorised by itself: brackets by multiplying.
A + A.B = A.(1 + B)
Factorising The process of
This then leads to the use of Boolean identities to further simplify: removing a common factor.
A.(1 + B) = A.1 = A Common factor A term
Expanding two sets of brackets together can be completed remembering the which appears in 2 or more
acronym FOIL: parts of an expression.
–
(A + B). (A + B) =
First (the first term in each bracket): A.A +
–
Outer (the two outer most terms): A.B +
Inner (the two inner most terms): B.A +
–
Last (the last term in each bracket): B.B
– –
= A.A + A.B + B.A + B.B
–
We can then apply Boolean identities (since B.B = 0):
–
= A + A.B + B.A + 0
And again (since X + 0 = X):
–
= A + A.B + B.A
We can now factorise as A is a common factor:
–
= A. (1 + B + B)
And again apply Boolean identities:
= A.1
=A

Now test yourself

32 Simplify each of the following Boolean expressions

a A.B + B
–
b C.(D + D )
– –
c (A + B). (A + B )
d C.D.(C + D)

Answers available online

De Morgan’s laws
Given a Boolean expression with a bar over two or more inputs, the bar can be
split:
––
A + B = A.B
The reverse is also true, and the bar can be joined:
– – De Morgan’s laws Break
A + B = A.B
De Morgan’s laws can be simplified to a simple phrase: ‘Break the line, change the line, change the sign
154 (and vice versa).
the sign’.

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_06_MRN_AQA_CS_139-157.indd Page 155 6/29/21 10:26 PM f-0116 /103/HO02206/work/indd

Using this technique it is common to find you have a double bar. Applying the
Exam tip
first De Morgan law:
– – – – Any question set on
A + B = A. B
the exam paper must
Alternatively applying the second De Morgan law gives:

6 Fundamentals of computer systems

be solvable, or have
– – a reasonably simple
A + B = A. B
solution. If a question
Since passing a value through two NOT gates returns the same result, a looks particularly large or
double bar cancels itself out in both cases; that is: daunting then remember
– – that you would not be set
A + B = A.B
an impossible task.
A good strategy for simplifying Boolean expressions is to look for
opportunities to apply, in this order:
✚ Boolean identities
✚ De Morgan’s laws
✚ Expanding or factorising Now test yourself
After each step, start the list of strategies again, and if you hit a dead end then
33 Simplify the following
step backwards and look for an alternative route.
Boolean expressions.
Exam tip a A.B + B
b C.D + D + E
Most longer exam questions award one mark for each skill shown, and examiners are – – –
instructed to stop marking a question once they see the first mistake. Therefore, write c F + G + F.G
the full expression out each time, making just one change at a time. If making three
Answers available
changes at once, you are more likely to make a mistake and also to lose marks for
online
skills that you have used correctly.

Summary
✚ The term hardware describes the physical components of ✚ Assembly language code is written in a form closer to
a computer system English
✚ The term software refers to the programs that run on ✚ Each machine code and assembly language is specific
that system to one platform
✚ Software can be classified as systems software or ✚ Each assembly language instruction is equivalent to
application software one machine code instruction
✚ Systems software is concerned with allowing the ✚ High-level languages use structured statements and
computer system to run and includes the operating English-like keywords and constructs
system, utility programs, libraries, and translators ✚ High-level languages support the use of local variables,
✚ Application software is not associated with the parameters, named constants and indentation
operation of a computer but instead allows it to serve a ✚ Imperative high-level languages describe how a
useful purpose by performing tasks problem should be solved, step by step
✚ The operating system is a piece of systems software ✚ Declarative high-level languages describe what should
which hides the complexity of the hardware and happen, but not how
provides a user interface ✚ High-level languages make it easier for programmers to
✚ The operating system also performs resource write code and can be run on different platforms
allocation, specifically memory management, ✚ Low-level languages use less memory and can be
controlling CPU access and peripheral management optimised for the platform on which it runs
✚ Utilities are programs which deal with one aspect of ✚ An assembler is used to translate assembly language
running the computer system code into object code
✚ Examples of utility programs include disk ✚ A compiler or an interpreter can be used to translate
defragmenters, back-up software, virus scanners and high-level code into object code
disk cleanup ✚ A compiler translates all of the code at once and
✚ Libraries are collections of subroutines which can be produces an executable file which can be run without
accessed by other programs further translation
✚ Translators are programs that convert source code into ✚ An interpreter translates and then executes one line of
object code – assemblers, compilers and interpreters code at a time, and does not produce an executable file
✚ The low-level languages are machine code and ✚ A compiler will attempt to identify all errors when
assembly language compiling and ensures that the source code can be
✚ Machine code is a set of instructions for a particular kept secret, as only the executable file is distributed
CPU made up of individual binary codes ✚ A compiled program will only run on the system it has
been compiled for
155

My Revision Notes AQA A-level Computer Science Third Edition

325487_06_MRN_AQA_CS_139-157.indd Page 156 6/29/21 10:26 PM f-0116 /103/HO02206/work/indd

✚ An interpreter will re-translate the program each time ✚ A D-type flip-flop stores the state of the input and can
it is run, so that the same source code can be run on therefore be used as a unit of memory
more platforms without needing to provide a different ✚ A D-type flip-flop has two inputs – a data signal and a
executable file clock signal
6 Fundamentals of computer systems

✚ An interpreter can allow for faster debugging as a ✚ Boolean expressions use the ‘+’ symbol to mean
partially complete program will still run OR and the ‘.’ operator (meaning multiply) to mean
✚ When sharing an interpreted program, the source code AND
must be shared, and the program must be re-translated ✚ BIDMAS rules apply, and so AND operations should be
each time it can run applied before OR operations
✚ Bytecode is used in some languages as an intermediate ✚ Boolean identities can be used to simplify Boolean
step as this bytecode can then be executed in a virtual expressions, for example, A.1 = A
machine, allowing compiled programs to run on a wide ✚ Boolean expressions can be factorised, and brackets
range of platforms expanded, just as in normal algebra
✚ Logic gates take one or more binary inputs and ✚ A bar over any part of a Boolean expression refers to a
produce a single binary output NOT operator
✚ The six main logic gates are the NOT, AND, OR, XOR, ✚ De Morgan’s Law means that if you break a line, you
NAND and NOR gates change the sign
✚ Each logic gate has its own truth table, which describes ✚ The reverse is true, and two bars can be joined, with
the range of possible inputs and the corresponding the operator where the break was being changed
outputs ✚ A rule of thumb when simplifying Boolean expressions
✚ Logic gates can be combined to create a logic circuit, is to look for opportunities to use De Morgan’s Law,
which can be used to solve a larger range of problems followed by looking for Boolean identities and finally
✚ A half adder is used to add two binary numbers looking for opportunities to use brackets (factorising or
✚ A full adder is used to add three binary numbers expansion)

Exam practice

1 A computer has been newly set up, with an operating 3 a) State the name of this logic gate. [1]
system, disk defragmenter, hard disk drive and a
A
compiler Q
B
a) State which item from the list above is an item of
hardware. [1]
b) Copy and complete the truth table for this logic
b) State which category of system software a disk gate. [2]
defragmenter belongs to. [1]
c) Describe two different types of resource A B Q
management carried out by the operating 0
system. [2]
0
d) Identify one other feature of the operating
system. [1] 1
2 Dani has written a computer program using an 1
imperative high-level language.
a) Explain the meaning of the term imperative high- c) Draw the circuit diagram for the logical expression.
level language. [2] [4]
b) Suggest two reasons why Dani may have chosen Q = A.B + B.C
to use a high-level language rather than a low-level d) Copy and complete the truth table for the logic
language. [2] circuit described in part c). [3]
c) State the name of one low-level language. [1]
A B C Q
d) State the purpose of a compiler. [2]
e) Describe two advantages to Dani for choosing to
use a compiler. [4]
f) Describe one disadvantage to the end-user
of Dani’s program if she had chosen to use an
interpreter. [2]
g) Dani’s program is compiled to bytecode instead
of object code. Suggest one reason why she has
chosen this method. [1]
h) Explain the process of executing the bytecode. [2]

➜
156

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_06_MRN_AQA_CS_139-157.indd Page 157 6/29/21 10:26 PM f-0116 /103/HO02206/work/indd

e) Use the truth table to draw a simplified logic circuit. 4 Simplify each of these Boolean expressions.
–
[3] a) 1 ⊕ B [1]
f) State the name of this logic circuit. [1] b) A.B + B [3]
– –
c) (A + B). (A + B) [4]

6 Fundamentals of computer systems

A
B S d) A + B.B.C [5]
Cin

Cout

157

My Revision Notes AQA A-level Computer Science Third Edition

325487_07_MRN_AQA_CS_158-181.indd Page 158 6/29/21 9:57 PM f-0116 /103/HO02206/work/indd

7 Computer organisation and

architecture

Internal hardware components

The internal components of a computer system vary slightly between
Internal components The
computing platforms, however the basic requirements are the same.
components essential to a
In almost every case the components are connected to a single printed circuit computer system.
board (PCB) referred to as the motherboard, or logic board.
Motherboard The circuit
These components include: board to which all other
✚ processor components are connected.
✚ main memory
✚ I/O controllers
✚ buses
✚ control bus
✚ address bus
✚ data bus

The components
Processor
The processor, or central processing unit (CPU), is used to process
instructions.
Every other aspect of computer hardware is designed around passing
instructions and data to the CPU in order to be processed, and then returned.
In some systems it is possible to remove and replace the processor in order
to upgrade the system, while in other systems the processor is permanently
fixed and cannot be replaced.
The individual components of a processor are discussed later in this chapter.
CPU Electronic device used
to process instructions.
Main memory
Main memory Memory
Main memory stores data and instructions that are directly accessed by the
that can be directly
CPU. There are two main forms of main memory: RAM and ROM.
accessed by the CPU.
✚ RAM (random access memory), is volatile storage used to hold the
instructions and data for programs that are currently running. RAM Short term storage for
✚ ROM (read only memory) non-volatile storage used to hold the startup currently running programs
instructions for the computer and currently used data.

RAM can be easily upgraded in most desktop and laptop computers as ROM Long term storage for
each memory module is built onto a small circuit board with standardised startup instructions.
connectors. Volatile Data is lost when
electrical power is removed.
Non-volatile Data is
Note retained even when
electrical power is removed.
When a computer system runs out of main memory a portion of the secondary
storage can be allocated as virtual memory. The least frequently accessed contents Virtual memory A portion
of main memory are placed in virtual memory, which has significantly slower data of secondary storage used
access times due to the physical limitations (for example, spinning up a magnetic to store the least frequently-
hard drive). used instructions and data.
158

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_07_MRN_AQA_CS_158-181.indd Page 159 6/29/21 9:57 PM f-0116 /103/HO02206/work/indd

Addressable memory
Inside main memory, each block of memory is given a unique address so that
the correct data can be retrieved and passed to the CPU. Without using blocks,
the whole of the memory would need to be retrieved each time. Applications

7 Computer organisation and architecture

The address is written as a sequence of binary numbers (often represented
using hexadecimal).
Instructions and data from different programs are typically grouped together 87FF
in order to improve efficiency. Library programs
83FF

I/O controllers Utility programs

8200
I/O devices (input/output devices) are required to enable computer systems to
Operating
be used effectively. system
Instructions and data must be passed to the memory and processor and 8000
the results returned from the system in order for the computer to serve a Figure 7.1 A basic
practical purpose. memory map

External devices include:

✚ storage devices (such as HDDs, SSDs, memory cards, optical disc drives) I/O devices Input and
✚ input devices (such as keyboards, mice, webcams, microphones, sensors) output devices such as
✚ output devices (such as monitors, speakers, motors). keyboard, mice, printers
I/O controllers provide a hardware interface between the internal computer and monitors.
components and the I/O devices. I/O controller The physical
interface between an I/O
Note device and the internal
components.
In order for an external device to communicate with the computer system it needs a
hardware I/O controller and a software device driver.
Note
Each type of I/O controller requires specific hardware for that device; for I/O ports are general-
example: purpose controllers that
✚ a disk controller (for connecting a HDD or SSD) allow a wider array of
✚ a keyboard controller (for connecting a keyboard) devices to be connected,
✚ a video display controller (for connecting a monitor). for example, USB sockets
which allow the use of mice,
These controllers are typically built into the motherboard and use a physical
printers, and even remote-
socket for connecting the devices.
controlled toy rocket
launchers.
Now test yourself

1 Identify three pieces of internal hardware.

2 Describe the purpose of main memory.
3 Give one reason why memory should be addressable.
4 Identify three devices which connect through an I/O controller.

Answers available online

Buses
Three buses are used to allow communication between the processor, RAM
Bus A communication
and the I/O controllers. system for transferring data.

Control bus Control bus A bus for

The control bus is used to send and receive control signals to each device. transmitting control and
These signals include commands from the processor and status signals (for status signals.
example, indicating read or write mode). Bi-directional Data travels
The control bus is bi-directional. This means that the messages can be sent in both directions.
both to and from the processor.

159

My Revision Notes AQA A-level Computer Science Third Edition

325487_07_MRN_AQA_CS_158-181.indd Page 160 6/29/21 9:57 PM f-0116 /103/HO02206/work/indd

Address bus Address bus A bus for

The address bus is used to indicate the physical address of the data which is transmitting memory
to be accessed (for example, the specific block of memory in RAM). addresses.
The address bus is uni-directional. This means that the processor sends out
7 Computer organisation and architecture

Uni-directional Data only

the addresses but does not receive them. travels in one direction.
Data bus A bus for
Data bus
transmitting data.
The data bus is used to carry the actual data that is required to be
transmitted.
The data bus is bi-directional. However in many situations data only travels
one way along the data bus (for example, when data is passed to the video
display controller, or received from the input controller). Any status messages
are passed over the control bus, which is bi-directional.
The three buses work in unison. If some data is required to be fetched from
memory then:
✚ the control bus carries signals indicating the status of memory (for
example, ready to read, ready to write, busy or in error) and indicates that,
for example, a read operation is to take place
✚ the address bus carries the address of the memory which is to be read (in
this example), sent by the processor
✚ the data bus carries the data stored in the required memory location to the
processor.
It can be helpful to consider a graphical representation of a computer system
(Figure 7.2).

Monitor Keyboard

Video Keyboard
Processor RAM
controller controller

Address bus

Data bus

Control bus

Figure 7.2 A model of the internal components

Note the directionality of each bus.

✚ The control bus is bi-directional at all points.
✚ The address bus is uni-directional away from the processor.
✚ The data bus is sometimes bi-directional, but data only comes in from the
keyboard controller and out to the video controller.

Now test yourself

5 Name the three buses used in computer architecture.

6 What is transported using the control bus?
7 Which bus is always directed away from the processor?
8 Which bus is often shown on diagrams as being bi-directional at the processor and
RAM, but uni-directional at different I/O controllers?
9 On a diagram, how can you identify which bus is the control bus?

Answers available online

160

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_07_MRN_AQA_CS_158-181.indd Page 161 6/29/21 9:57 PM f-0116 /103/HO02206/work/indd

von Neumann vs Harvard architecture

Computer architecture refers to the underlying design principles.
In a von Neumann architecture the computer system uses main memory to

7 Computer organisation and architecture

store both instructions and data.
In a Harvard architecture the computer system uses separate memory for
instructions and for data.

von Neumann architecture A computer system design with one shared memory
for instructions and data.
Harvard architecture A computer system design with separate memory for
instructions and data.

Figure 7.3 von Neumann and Harvard architectures

von Neumann architecture Harvard architecture

Advantages It is flexible because each block Can read both instructions
of memory can be used for and data simultaneously,
instructions or data as needed. meaning that it can carry out
This allows for situations where the instructions more quickly
there are more instructions without having to wait for two
than data, or more data than sequential read operations to be
instructions. carried out.
Disadvantages Cannot read instructions and The ratio of data and
data simultaneously, meaning instructions stored in memory
processing is typically slower. is fixed, making this system less
effective for general-purpose
computing
Uses Used in almost all general- Used extensively in digital
purpose computers due to the signal processing (DSP)
increased flexibility. hardware. This includes
recording and playing back
audio, processing images in a
digital camera and working with
sensors.

Now test yourself Digital signal processing

Manipulating digitised data
10 Which architecture uses two separate areas of memory? relating to analogue signals.
11 For what purpose is Harvard architecture extensively used?
12 Why is von Neumann architecture preferable for general-purpose computer
systems?

Answers available online

161

My Revision Notes AQA A-level Computer Science Third Edition

325487_07_MRN_AQA_CS_158-181.indd Page 162 6/29/21 9:57 PM f-0116 /103/HO02206/work/indd

The stored program concept

Whereas early computers required that the program was input as it was run
Machine code Binary code
(for example, on punch cards), the stored program concept allows programs to
7 Computer organisation and architecture

instructions which can be

be stored in memory before being run. read and understood by a
✚ In the stored program concept, machine code instructions are stored in processor.
memory.
✚ Each instruction is then fetched, decoded, and executed, in order, by a Arithmetic Mathematical
processor. operations such as
addition, subtraction and
✚ The processor performs operations on the data – such as arithmetic or
multiplication.
logical operations.
Logical Operations using
computation logic such as
Now test yourself
AND, OR and NOT.
13 How is it possible to execute a computer program without the stored program concept?
14 What three steps need to be carried out in order to execute a stored program?

Answers available online

Structure and role of the

processor and its components
The processor and its components
The processor is made up several components, as outlined below.
Arithmetic logic unit
Arithmetic logic unit (ALU) Carries out arithmetic
The arithmetic logic unit (ALU) performs arithmetic and logical operations (mathematical) and logical
such as addition, subtraction and logical AND operations. operations.

Control unit Control unit (CU) Decodes

instructions and sends/
The control unit (CU) supervises the fetch-execute cycle, sends and receives
receives control signals.
control signals and decodes each instruction once it has been fetched from
memory. Fetch-execute cycle
The process by which
Clock instructions are fetched
from memory, decoded and
The clock generates timing signals which are used to synchronise the tasks
executed.
within the fetch-execute cycle.
Clock Generates timing
Clock speed is measured in Hz (where 1 Hz = 1 state change per second) and
signals.
most modern processors operate in the range 1 to 4 GHZ (up to 4 billion state
changes per second). General-purpose
registers Registers that
General-purpose registers can be used to store any
General-purpose registers are used to temporarily store data in the processor. data.
Registers Small areas
The result of a calculation may be placed in a register if it is known that it will
of memory inside the
be needed again imminently.
processor.
The exact number of general-purpose registers varies, however it is typically
Dedicated registers
in the region of 16-32 for modern desktop processors.
Registers which are
dedicated to holding data
Dedicated registers for a specific, functional
Some registers (called dedicated registers) are dedicated to a specific purpose. purpose.
The following registers are all used directly as part of the fetch-execute cycle.
✚ Program counter (PC) – stores the address of the next instruction
✚ Current instruction register (CIR) – stores the instruction currently being
162 executed

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_07_MRN_AQA_CS_158-181.indd Page 163 6/29/21 9:57 PM f-0116 /103/HO02206/work/indd

✚ Memory address register (MAR) – stores the address of the item in memory
Exam tip
currently being addressed. This could refer to an instruction or to data
✚ Memory buffer register (MBR) – stores the data that is currently being Be familiar with the
fetched from memory acronyms for each

7 Computer organisation and architecture

✚ Status register (SR) – stores the current state of the processor, such as dedicated register, but make
whether an overflow has been detected sure you can state each
one’s full name
Now test yourself

15 Name the five main components in a processor – both the full names and the
abbreviations.
16 What two tasks are carried out by the ALU?
17 Which part of the processor is used to decode instructions?
18 What data can be stored in a general-purpose register?

Answers available online

The fetch-execute cycle

The fetch-execute cycle (sometimes referred to as the fetch-decode-execute
cycle, or FDE cycle) describes the steps involved in executing a stored program.
At the start of each cycle the PC (program counter) already stores the address
of the next instruction.

Fetch
In this part of the cycle the next instruction is fetched from memory.
1 The contents of the PC are copied to the MAR
2 The contents of the PC are incremented.
3 The address bus transfers the contents of the MAR to main memory.
4 The data at that memory location is transferred to the processor via the
data bus.
5 This fetched data is stored in the memory buffer register (MBR).
6 The data (which is an instruction) is transferred to the CIR.

Decode
In this part of the cycle the contents of the CIR are decoded by the processor’s
control unit.

Execute
The final part varies significantly depending on the exact instruction.
If the instruction involves fetching data from memory then:
✚ the address of that data is stored in the MAR
✚ the address bus transfers the contents of the MAR to main memory
✚ the data at that memory location is transferred to the processor via the
data bus
✚ the fetched data is stored in the memory buffer register (MBR).
If the instruction involves writing data to memory, then the process will be
almost identical, except that the data will be transferred from the MBR to
main memory via the data bus.
Sometimes the cycle is described in five steps:
1 MAR ← [PC]
2 a PC ← [PC]+1
b MBR ← [Memory]addressed
3 CIR ← [MBR]
4 Decode instruction
5 Execute instruction 163

My Revision Notes AQA A-level Computer Science Third Edition

325487_07_MRN_AQA_CS_158-181.indd Page 164 6/29/21 9:57 PM f-0116 /103/HO02206/work/indd

In this style of question, the square brackets read ‘the contents of’ and the
Exam tip
arrow, ←, reads ‘copied to’. For example, Step 1 reads ‘The contents of the
program counter are copied to the memory address register’. Questions referring to the
fetch-execute cycle are
Note that steps 2a and 2b can occur simultaneously, as both tasks use
7 Computer organisation and architecture

typically longer-answer
different registers and do not rely on the result of the other operation.
questions. Make sure you
are confident explaining the
Now test yourself fetch part of the cycle in
19 Identify the three main stages in the fetch-execute cycle. particular, including the use
of both buses and registers.
20 Describe the first part of the fetch stage.
21 What is the purpose of the program counter?
22 Which register communicates via the address bus?
23 Which register communicates via the data bus?

Answers available online

The processor instruction set

Each instruction is made up of an opcode and an operand.
Opcode An operation code,
The opcode describes: describing what operation is
✚ what operation is to be carried out (for example, add, load, and so on) to be carried out.
✚ the addressing mode.
Operand The value on
The operand describes: which the operation is to be
✚ the value(s) which are to be processed. carried out.
Each processor has its own processor instruction set. Instruction set The list of
This instruction set describes the binary machine code for each available machine code values for
operation. each available operation.

Note that in machine code the addressing mode is included as part of the
opcode. In assembly language the addressing mode is typically shown as part
of the operand in order to make the assembly language instruction easier to
follow, however the addressing mode bit is still stored as part of the opcode.

opcode operand(s)
basic machine operation addressing mode
1 0 0 1 1 1 1 1 1 1 0 0 0 1 1
Figure 7.4 An example machine code instruction

CMP r1, #10 ‘compare the value in register 1 with the value 10’

Operation code Operands Addressing mode Comment

or opcode
Figure 7.5 An example assembly language instruction

Addressing modes
The addressing mode describes whether the operand is an immediate value,
or the address of a value.
Addressing mode Either
Immediate addressing is when the operand of an instruction contains the immediate or direct.
number itself. For example:
Immediate addressing
ADD R1, R2, #3 The operand is the value,
This instruction will add the number 3 to the value in register 2 and store the which is to be processed.
answer in register 1. Direct addressing The
Direct addressing is when the operand of an instruction contains an address operand is the address of
(either a register number or an address in memory). For example: the value which is to be
164 processed.
ADD R1, R2, R3

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_07_MRN_AQA_CS_158-181.indd Page 165 6/29/21 9:57 PM f-0116 /103/HO02206/work/indd

This instruction will add the value in register 3 to the value in register 2 and
Exam tip
store the answer in register 1.
To help remember the
The exact format of the addressing mode in machine code is specific to that
vocabulary consider that
processor’s instruction set.

7 Computer organisation and architecture

when using immediate
addressing the operand can
Now test yourself be processed immediately,
without needing to be
24 What is the purpose of an opcode? fetched. When using direct
25 What is the purpose of an operand? addressing the MAR and
26 Which addressing mode is used for a memory address? address bus must direct the
data from memory.
27 Why is an addressing mode necessary?

Answers available online

Machine code and assembly language

operations
Each processor has its own instruction set, detailing the specific operations
that can be carried out by that processor. Note

AQA has its own instruction set which is used in Component 2 examinations. The AQA Assembly
Language Instruction Set
can be found under ‘Other
Exam tip assessment resources’
on the AQA website, aqa.
The instruction set is always included in the question paper with details of what each
org.uk. Search for ‘7516-
operation does and how it should be structured, so you do not need to memorise this
7517’ and then scroll down
content.
to ‘Past papers and mark
Questions may ask you to write your own assembly language code, or to trace an schemes’.
existing algorithm (or both).
This section often causes some confusion for students, so while you don’t need to
memorise the instruction set, you do need to have practical experience of using it.

Almost all operations include a command and a register address. This is

Exam tip
where the result of each operation will be stored. For example:
Be careful not to mix up
ADD R0, R1, R2
your technical vocabulary.
The operation is ADD R0, the two operands are R1 and R2. The instruction includes
both the operation and the
Thus, the values in register 1 and register 2 will be added and the result stored
operand(s). Referring to the
in register 0.
whole instruction as the
The operand will either use a # symbol to indicate immediate addressing operation, or vice versa, can
(such as #25) or Rn to indicate direct addressing, where n is the address of a cost you marks.
register (such as R4).
At the start of the following examples the contents of the registers and main
memory are as follows.

Registers Main memory

R0 R1 R2 R3 R4 R5 100 101 102 103
10 20 5 0 8 12 25

Load Fetch a value from

Load and store memory into a register in
The operation LDR will load, or fetch, data from memory and store it in one of the processor.
the registers. Store Copy a value from
The operation STR will store data from a register to an address in memory. a register in the processor
and save it in memory.
LDR R0, 100
165
STR R2, 103

My Revision Notes AQA A-level Computer Science Third Edition

325487_07_MRN_AQA_CS_158-181.indd Page 166 6/29/21 9:57 PM f-0116 /103/HO02206/work/indd

The first instruction will fetch the item from memory location 100 and copy it
into register 0.
The second instruction will copy the item from register 2 into memory
address 103.
7 Computer organisation and architecture

Registers Main memory

R0 R1 R2 R3 R4 R5 100 101 102 103
8 20 5 0 8 12 25 5

Arithmetic
With arithmetic, remember that the first register is the location for storing
the result. Exam tip

ADD R4, R0, R1 Putting the answer into the

wrong register is a very
SUB R5, R1, #2 common mistake. For each
MOV R3, R5 assembly instruction, fill in
the operations as you would
The first instruction adds the value from register 1 to the value in register 0 a maths problem.
and stores the answer in register 4.
When you see the
The second instruction subtracts the decimal value 2 from the value in instruction SUB R5, R1,
register 1 and stores the answer in register 5. #2 think (but do not write!)
The third instruction moves the value FROM register 5 and stores it in register 3. R5 = R1 - #2.

Registers Main memory

R0 R1 R2 R3 R4 R5 100 101 102 103
8 20 5 18 28 18 8 12 25 5

Now test yourself

28 For the instruction LDR R2, 102, which one of these statements is true?
A The value 102 will be stored in register 2
B The value stored in register 2 will be copied into memory
C The value stored at memory address 102 will be copied into register 2
29 Write the assembly language code for storing the value in register 4 into the
memory address 101.
30 What symbol is used to indicate immediate addressing?
31 Explain the effect of executing the instruction MOV R4, R0.
32 If R1 = 10, R2 = 20 and R3 = 30, state the values stored in each register
after executing the instruction SUB R2, R1, R3.
Answers available online

Compare and branch Compare The operation

The compare and branch operations allow programmers to create selection used to check the condition
and iteration structures in assembly language code. between two values.

Remember that you cannot branch on a condition unless you have performed Branch Jump to a labelled
a comparison first. part of the program.
A selection structure can be created as follows: Selection A method
CMP R3, R5 of choosing whether to
execute the following
BEQ same instruction(s).
B diff
Iteration A method of
The first operation will compare the values in register 3 and register 5. The repeating a section of code.
result will be either:
Condition A comparison to
✚ EQ (equal to)
see if two values are equal,
✚ NE (not equal to) or whether one is greater/
166 ✚ GT (greater than – if R3 > R5) less than the other.
✚ LT (less than – if R3 < R5)

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_07_MRN_AQA_CS_158-181.indd Page 167 6/29/21 9:57 PM f-0116 /103/HO02206/work/indd

The second operation will branch (or goto) the part of the program with the
Label A named point in the
label ‘same’ if the result was EQ.
program.
Alternative operations that would also be based on the result of the
comparison would be:

7 Computer organisation and architecture

✚ BNE <label>
✚ BGT <label>
✚ BLT <label>
The third operation will always branch to the part of the program
labelled ‘diff’.
In this case the third operation can only be reached if the branch condition
in line 2 is not met. This structure works in the same way as a selection
statement with an IF and an ELSE component.
An iteration structure can also be created. For example, the code below would
create an infinite loop.
start:
ADD R0, R1, R2
Now test yourself
STR R0, 100
33 Which operation must
B start
immediately appear
And the code below would create a while loop that repeats until the value before a conditional
of register 0 goes beyond 100. After this the section of code labelled ‘done’ is branch?
executed. 34 Which opcode would
start: be used to branch if
ADD R0, R1, R2 operand 1 was less than
operand 2?
CMP R0, #100
35 True or false? Once
BGT done a branch has been
B start completed, the
Note that labels for sections of code are written using their name followed by execution returns to
the point at which the
a colon.
branch was made.
All other lines in the program are slightly indented (though always to the
same extent). Answers available
online
Bitwise logic
There are four bitwise logic operations and each one involves applying that
logical operation to each bit one at a time. Bitwise An operation that
AND R0, R1, R2 works on one bit at a time.

The values of register 1 and register 2 would be processed using a bitwise AND.

Worked example Making links

For the operation AND R0, R1, R2, calculate the value of register 0 if For a more in-depth
register 1 has the value 2010 and register 2 has the value 510. discussion of logical
In binary, register 1 is 0001 0100 and register 2 is 0000 0101. operations, see Logic circuits
in Chapter 6.
The AND operation means perform the AND operation on equivalent bits from each
register.
Starting with the right most bit (or LSB), we can see that
✚ 0 AND 1 = 0 ✚ 1 AND 1 = 1 ✚ and so on.
✚ 0 AND 0 = 0 ✚ 0 AND 0 = 0
This gives the result:
0001 0100
0000 0101 AND
0000 0100
where 0000 0100 is 410 and, therefore, register 0 has the value 4. 167

My Revision Notes AQA A-level Computer Science Third Edition

325487_07_MRN_AQA_CS_158-181.indd Page 168 6/29/21 9:57 PM f-0116 /103/HO02206/work/indd

Registers Main memory Exam tip

R0 R1 R2 R3 R4 R5 100 101 102 103
One possible question is
4 20 5 18 28 18 8 12 25 5 to describe the purpose of
7 Computer organisation and architecture

performing a bitwise AND

Exam tip operation with an operand
of 0. This can be used to
To find out if a value is even or odd, perform a bitwise AND operation with an operand reset a register to 0 as A.0
of 110. If the result is 0 then the LSB of the other operand must be 0, meaning that it is is always 0.
an even number. For example:
1110
0001 AND
0000
The result shows that the value (14) is even.
A similar technique could be used to test the MSB in order to check if a two’s
complement number is positive or negative. In this case the operand should be
a number whose MSB is 1 and all other bits 0. If the result is 0, then the two’s
complement number is positive, otherwise it is negative.
For example, to show that a) the 4-bit two’s complement number –1 is negative, b) the
4-bit two’s complement number 4 is positive, using a bitwise AND operation:
1111
1000 AND
1000
Non-zero result means the two’s complement number is negative.
0100
1000 AND
0000
Zero result means the two’s complement number is positive.

The other logical operations all use three characters in the AQA assembly
language.
ORR R1, R1, #10
EOR R2, R3, #20
MVN R3, R3
The first instruction will perform a bitwise OR between the value in register 1
and the decimal number 10, storing the result in register 1.
The second instruction will perform a bitwise XOR (exclusive OR) between the
value in register 3 and the decimal number 20, storing the result in register 2.
The third instruction will perform a bitwise NOT (move NOT) on the value in
Exam tip
register 3 and store the answer in register 3.
To find out if two values are
0001 0100 0001 0010
equal, an XOR operation can
0000 1010 OR 0001 0100 XOR 0001 0010 NOT be used. A XOR 2 (A⊕2)
will return 010 if and only if
0001 1110 0000 0110 1110 1101
A = 2.
Registers Main memory
R0 R1 R2 R3 R4 R5 100 101 102 103
4 30 6 237 28 18 8 12 25 5

168

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_07_MRN_AQA_CS_158-181.indd Page 169 6/29/21 9:57 PM f-0116 /103/HO02206/work/indd

Now test yourself

36 Convert each value to unsigned binary and carry out the bitwise operation.
a 12 AND 61

7 Computer organisation and architecture

b 17 OR 23
c 47 XOR 121
d NOT 213
37 What opcode is used for each bitwise logical operation from question 1?
38 What might be the purpose of the instruction AND R1, R0, #1?
39 Write two lines of assembly language which could be used to convert a positive
number, stored in register 3, to its negative equivalent using twos complement.
The result should be stored in register 4.
Logical shift An operation
Answers available online involving moving each bit in
a given direction.

Logical shifts Logical shift left Moving

all of the bits left, doubling
A logical shift will move all bits to the left or right a given number of places,
the value each time.
filling any gaps with zeroes.
Overflow When data is lost
LSL R0, R0, #1
because the result is too
This operation will apply a logical shift left, meaning each of the bits in the large.
value of register 0 will be moved one place to the left. For example: Logical shift right Moving
0000 0100 LSL, #1 all of the bits to the right,
halving the value each time.
0000 1000
Underflow When data is
This has the effect of doubling the value. lost because the result is
✚ A left shift multiplies a value by 2n , where n is the number of shifts. too small.
✚ Performing a left shift of magnitude 4 will multiply by 24 = 16
✚ This can lead to a loss of accuracy as any overflow is discarded
Now test yourself
LSR R3, R2, #1
40 Describe the result of
This operation will apply a logical shift right, meaning each of the bits in the applying a logical shift
value of register 2 will be moved one place to the right. The result will be left by one place.
stored in register 3.
41 Describe the result of
0000 0110 LSR, #1 applying a logical shift
right by one place.
0000 0011
42 Calculate the result of
This has the effect of halving the value. applying each of the
✚ A right shift divides a value by 2n, where n is the number of shifts. following instructions.
✚ This can lead to a loss of accuracy as any underflow is discarded. a LSL R0, #5, #2
For example: b LSL R0, #3, #4
c LSR R0, #12,
LSR R4, R2, #2
#2
0000 0110 LSR, #2 d LSR, R0, #18,
0000 0001 #3
43 The following
The original number is 6 and the right shift of 2 means ‘divide by 4’. The result
instructions are
is calculated as 1 which is not accurate due to underflow.
executed. Explain why
Registers Main memory the value in R0 at the
end of the process may
R0 R1 R2 R3 R4 R5 100 101 102 103
not be the same as the
4 30 6 3 1 18 8 12 25 5 value in R0 at the start.
LSR R0, R0, #1
HALT LSL R0, R0, #1
The only operation with more than three letters, the HALT instruction will Answers available
end the program. online
169

My Revision Notes AQA A-level Computer Science Third Edition

325487_07_MRN_AQA_CS_158-181.indd Page 170 6/29/21 9:57 PM f-0116 /103/HO02206/work/indd

Trace tables Hand trace Manually step

A common question is to ask students to hand trace an assembly language through each instruction in
program. an algorithm.
7 Computer organisation and architecture

Trace table A table for

Worked example recording the values
stored in each register and
memory location.
Registers Main memory
R0 R1 R2 R3 R4 R5 100 101 102 103
2 10 12 8 3 12 25
Making links
LDR R4, 100
Hand tracing algorithms is
CMP R4, #1 an essential skill which is
assessed using high-level
BLT labelb
pseudo-code in Component
AND R0, R0, #0 1. General advice for
labela: hand tracing algorithms is
discussed in Trace tables in
ADD R0, R0, R3 Chapter 4.
SUB R4, R4, #1
CMP R4, #0
BGT labela
STR R0, 103
labelb:
HALT
✚ In the first line of the trace table we write down the initial values of the registers
and memory locations.
✚ We work through the code and a new line to the trace table every time a value
changes. The first change is to load the value of memory address 100 to R4.
✚ R4 is not less than 1, so we do not branch, nor add anything to the trace table.
✚ The next change is R0 AND 0. The result is 0, stored in R0.
✚ Moving on to labela, the next change is adding R3 to R0 and storing the result
in R0.
✚ 1 is subtracted from R4.
✚ These last two lines are repeated until R4 = 0, at which point the value in R0 is
stored in memory address 103.

Registers Main memory

R0 R1 R2 R3 R4 R5 100 101 102 103
2 10 12 8 3 12 25
3
0
8 2
16 1
24 0 24

This program multiplies the value in register 3 by the value in memory address 100 and
stores the result in memory address 103.
Lines 2 and 3 of the program will jump to the HALT operation if the value in register 4
is less than 1.

170

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_07_MRN_AQA_CS_158-181.indd Page 171 6/29/21 9:57 PM f-0116 /103/HO02206/work/indd

Now test yourself

44 Copy and complete the trace table for the following program.
MOV R1, R0

7 Computer organisation and architecture

ADD R1, R1, R2
SUB R2, R1, R3
EOR R4, R2, R3
LSL R5, R4, #3

Registers
R0 R1 R2 R3 R4 R5
10 15 12

Answers available online

Interrupts
The processor is continually processing instructions, even when a computer
seems to be idle.
It is important to be able to get the processor’s attention and to alter the
sequence of instructions which is about to be processed. This might happen
for a number of reasons, such as:
✚ timing interrupts; for example, at fixed intervals the screen must be
redrawn
✚ program error interrupts; for example, if a program has attempted to
divide by 0
✚ hardware error interrupts; for example, a printer reports a paper jam
✚ I/O interrupts; for example, the user has pressed a key.
An interrupt is sent to the processor in order to ensure that it can deal with
this event.
In order for the processor to recognise this, the FDE cycle includes a step to
check for an interrupt.
Interrupt A signal sent to
Fetch
the processor in order to
alert it to an event.
Check for interrupt Decode Interrupt service routine
(ISR) A program which will
examine an interrupt and
Execute
handle the event.
Figure 7.6 A fetch-execute cycle model with a check for any interrupts
Stack A data structure in
Because programs will require a large number of instructions to be processed, which the most recently
when an interrupt is detected then that program’s execution must be paused. added item is returned first.
✚ The current state of the processor and its registers is saved on a stack.
✚ The source of the interrupt is identified.
✚ The appropriate interrupt service routine is called.
✚ The state of the processor is restored from the stack.
An interrupt service routine (ISR) is a software routine which will take an
interrupt and examine it in order to determine the best course of action.
This will usually involve adding new instructions to the queue of instructions 171
to be processed.

My Revision Notes AQA A-level Computer Science Third Edition

325487_07_MRN_AQA_CS_158-181.indd Page 172 6/29/21 9:57 PM f-0116 /103/HO02206/work/indd

Different ISRs are written for different types of interrupt as the responses to
different interrupts will need to deal with that specific event.

Now test yourself

7 Computer organisation and architecture

45 Give an example of an event that might trigger an interrupt.

46 When in the fetch-execute cycle is there a check for an interrupt?
47 Explain the purpose of an interrupt service routine.
48 Describe what happens when an interrupt is detected.

Answers available online

Factors affecting processor performance

There are many factors which can affect the performance of a processor.

Cores
A single-core processor can carry out one instruction at a time.
A dual-core processor can carry out two instructions at a time.
A quad-core processor can carry out four instructions at a time.
Single-core A processor
… and so on. containing a single CPU.
However, it is not true that a quad-core processor will be four times faster Dual-core A processor
than a single-core because: containing two CPUs, both
✚ some tasks cannot be run in parallel working simultaneously.
✚ other tasks may cause a delay while waiting for the result of another
Quad-core A processor
instruction
containing four CPUs, both
✚ there is a processing overhead in splitting the task into separate threads.
working simultaneously.
That said, a quad-core processor will typically complete a set of tasks more
Cache A small unit of
quickly than an equivalent single-core processor.
volatile memory placed on
the same circuit board as a
Cache memory processor for fast access.
A cache is a small amount of memory located on the processor chip which Clock A component in the
acts as main memory for the most frequently accessed instructions and data. processor for generating
The cache can be accessed much more quickly than RAM and, therefore, timing signals.
increasing the size of the cache reduces the average time it takes to fetch
frequently-used instructions and data from memory.

Note
Most desktop and laptop processors will have three levels of cache: L1, L2 and L3.
L1 is the smallest and offers the quickest access.
L3 is the largest and offers the slowest access – though it is still faster than accessing
RAM.

Clock speed
The clock is the processor component that generates timing signals.
✚ Most modern processors have a clock speed of at least 1 GHz, which means
there are 1 billion state changes per second.
✚ Increasing the clock speed means that each instruction is processed more
quickly.
✚ One significant downside of increasing clock speed is that this generates
more heat which will ultimately shorten the lifespan of the processor and
make it more prone to errors.
172

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_07_MRN_AQA_CS_158-181.indd Page 173 6/29/21 9:57 PM f-0116 /103/HO02206/work/indd

Word length
A binary word is a piece of binary data that can be processed in one unit. Word The maximum
✚ Using a 4-bit word length for example, means that only 4 bits can be number of bits that can
processed at once. This would mean that adding two 16-bit numbers would

7 Computer organisation and architecture

be processed in one
be time consuming and would take multiple cycles. instruction.
✚ Using a larger word length means that larger values can be processed
Address bus width The
in fewer cycles, improving the overall speed at which a computer will
maximum number of bits
complete its processing.
that can be used to address
Most modern desktop computers use a 32- or 64-bit word length. a memory location.
Data bus width The
Address bus width maximum number of bits
The address bus is used to transport the address of an item in main memory. that can be transferred at
one time.
The range of available addresses is limited by the size of the address bus.
If the address bus is:
✚ 8 bits wide, then only 256 (= 28) addresses can be used
✚ 16 bits wide, then 65536 (= 216) addresses can be used.
Increasing the address bus means that more memory can be addressed. Exam tip
In the case of word length,
Data bus width address bus width and
data bus width, remember
The data bus is used to transport data from memory to the processor, and
that adding 1 bit will
vice versa.
double the range of
Increasing the width of the address bus means that more data can be possible values. This often
transferred in one fetch cycle, meaning that less time is lost waiting for data catches students out.
to arrive at the processor.

Now test yourself

49 Identify three aspects of a processor which can affect its performance.

50 Explain the difference between increasing the number of cores and the clock speed.
51 a What term is used to describe a small, fast block of memory built onto the
same circuit board as the processor?
b What data is stored in this block of memory?
52 What is the significance of the word length?
53 What happens to the number of memory addresses that can be accessed if the
address bus width is increased by 1 bit?

Answers available online

External hardware devices

Input and output devices
There are many input devices and output devices. Four are covered in detail
here. Barcode reader A device
for reading barcodes.
Barcode reader
A barcode reader is an automated input device. Barcode A black and white
image consisting of vertical
The reader is made of three main components. bars of different widths to
✚ Light emitter: shines red light at a pattern of black and white vertical bars. represent data.
✚ Light sensor: reads the pattern of reflected light.
✚ Decoder: converts this pattern into digital data.
Barcodes are made of a series of vertical bars, printed in a black and white
pattern. They are commonly seen on products such as books as the barcodes
themselves are small, cheap and easy to manufacture. 173

My Revision Notes AQA A-level Computer Science Third Edition

325487_07_MRN_AQA_CS_158-181.indd Page 174 6/29/21 9:57 PM f-0116 /103/HO02206/work/indd

The barcode scanner needs direct line of sight in order to function. Obscured,
Note
folded or damaged barcodes cannot be read.
Barcodes can be thought of
as a one-dimensional data
7 Computer organisation and architecture

pattern, much like an array

of numbers. QR codes are a
two-dimensional equivalent,
using small squares of
black and white to store
more data using the same
technological principles. QR
codes are not covered in
Figure 7.7 A barcode reader and barcode the specification, however it
may be useful to recognise
Digital camera the similarities.
A digital camera uses a lens to focus light through the camera’s body and onto
an array of sensors. Digital camera A device
✚ Each sensor represents one pixel in the image. for taking photographs
✚ A colour filter is used to separate the amount of red, green, and blue light stored as digital data.
being captured.
✚ Each sensor reacts to the incoming light and generates an electrical signal. Pixel The smallest
addressable unit of an
✚ The signals are processed by an ADC and turned into binary data.
image.
✚ Each pixel is generated based on the combination or red, green and
blue light. ADC Analogue to digital
converter.
The major advantage of a digital camera as an input device is that it captures
a large amount of data compared to a barcode or other simple optical reader.
If the camera is to be used as an automated input device (for example,
to identify items passed through a checkout till) this means that a lot of
processing is required in order to derive any meaning from the data (as
opposed to a barcode scanner).
Digital cameras can be used effectively to recognise objects and to identify
faces, though these can be obscured, and identification is not always accurate.

3 4

Figure 7.8 A digital camera: light is let in through the shutter (1) and focused by the
lens (2); it is directed through RGB filters (3) before being focused onto the sensor
array (4)

Radio frequency identification

A radio frequency identification (RFID) tag is used to transmit a small amount
RFID Radio frequency
of data using a radio frequency over a short distance.
identification.
✚ Many RFID tags are passive, meaning that they have no internal power
source and do not broadcast for most of the time. Instead an RFID scanner RFID tag A small device
broadcasts a radio signal which provides enough power for the RFID tag to capable of emitting data
respond. using radio frequencies.
✚ Active RFID tags use a battery to allow them to broadcast the signal and RFID scanner A device for
have a longer range. reading the data from an
✚ RFID scanners are very accurate as it is difficult to distort or obscure the RFID tag.
signal accidentally.
✚ RFID tags do not require line of sight.
✚ However, RFID signals can be shielded using materials which block the
174 radio frequency signal.

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_07_MRN_AQA_CS_158-181.indd Page 175 6/29/21 9:57 PM f-0116 /103/HO02206/work/indd

✚ RFID tags are fairly cheap, as well as being small and light. They are
Exam tip
slightly more expensive than printing a barcode.
✚ RFID tags are often used for advertising on public transport, for contactless Barcode readers, digital
payment and contactless passports. cameras and RFID

7 Computer organisation and architecture

scanners are all methods
of data input. As well as
Laser printer understanding their main
A laser printer is a device for creating a physical copy of a document or image principles of operation it
on paper. is common to be asked
✚ When a document is sent to a printer it is converted into a bitmap image. to compare the suitability
✚ Inside the printer a round printer drum is coated in a static electrical of each device for a given
charge. A laser is aimed at the drum using a mirror, and the laser beam is scenario.
modulated (switched off and on) to draw the image.
✚ Wherever the laser beam touches the drum the charge is removed, and Laser A device for creating
this is where the toner is intended to stick. an intense beam of light.
✚ The toner is a powdered form of ink and is given the same static electrical
Printer A device for
charge as the drum. This means the charged areas of the drum will repel
creating a hard copy of a
the toner. The drum picks up the toner where the laser has removed the
document, usually on paper.
static charge.
✚ In a colour printer there are four different drums for each coloured toner. Drum A round device used
The four toner colours are known as CMYK (cyan, magenta, yellow, and to attract the toner.
key, which is black). Toner A powdered form of
✚ The paper is rolled over the drum in order to transfer the toner to the ink with a static charge.
paper. Once this is done the toner is fused to the paper using heated
rollers. CMYK In a colour printer
the colours used are cyan,
Laser printers take a short amount of time to warm up, making them slightly magenta, yellow and key
slower to print one page. However, they are much faster, much cheaper to run (black).
and far less prone to failures than inkjet printers.

Laser beam charges

the drum at certain Note
points to create Mirror directs
the image on The primary colours in
the beam onto
the drum the drum
printing (CMYK) are different
from the primary colours
Toner is dispersed in monitors (red, green and
onto the drum blue (RGB)). This is because
from roller Corona wires monitors use additive colour
to create a charge the to add light to an otherwise
negative of drum black display. Printers use
Paper in the image Drum
subtractive colour to absorb
light from an otherwise
white piece of paper.
Corona wires Corona wires Fuser melts
attract toner discharge the toner
from the drum the paper onto the paper
onto the paper
Figure 7.9 The inner workings of a laser printer

Now test yourself

54 What are the three main components in a barcode scanner?

55 What happens to light when it enters a digital camera?
56 What technology does an RFID scanner use to read data?
57 Give three advantages to using RFID as an automated input device rather than a
barcode scanner.
58 Name three different components in a laser printer.

Answers available online

175

My Revision Notes AQA A-level Computer Science Third Edition

325487_07_MRN_AQA_CS_158-181.indd Page 176 6/29/21 9:57 PM f-0116 /103/HO02206/work/indd

Secondary storage devices

Secondary storage refers to non-volatile storage of programs and data. It is
Secondary storage
necessary because all volatile data (that is, RAM) will be lost when the system
Persistent, non-volatile
7 Computer organisation and architecture

is powered down. Programs and data stored on a secondary storage device

storage.
can be retrieved at a later date.
Non-volatile Retains data
Exam tip when powered down.

Questions on storage devices generally fall into one of two categories: Volatile Loses data when
powered down.
1 Explain the workings of one particular type of storage device in detail.
2 Compare the advantages and disadvantages of different storage devices for a
given scenario. Exam tip
For questions of type 1, focus on learning and using the technical vocabulary and you Many students incorrectly
can construct your answer around these key terms. assume that secondary
storage only refers to
Mechanical hard disk drive backups. In order to avoid
this trap, whenever you
A mechanical hard disk drive (HDD) uses a metal disk, called a platter, which
see the term ‘secondary
is coated in a thin film of magnetic material.
storage’, just think ‘storage’.
✚ The film is made up of concentric rings, or tracks, each of which is split up
into sectors.
✚ Each sector is made up of thousands of magnetic charges, to indicate 0s or Hard disk drive (HDD) A
1s that represent data. storage device which saves
✚ The platter spins at high speed and a read/write head is moved over the data using magnetic film.
platter, which can both detect and change the magnetic charges in that Platter A metal disk used
sector. to store the data in a HDD.
Most mechanical hard drives use multiple platters, each with its own read/ Track A concentric ring on
write head. a platter.
Head
Sector A small section of
assembly Disks Cylinders
a track.
Read/write head The
device used to read and
write magnetic data to each
to sector.

Figure 7.10 HDD platters

track

sector

Figure 7.11 HDD tracks and sectors

176

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_07_MRN_AQA_CS_158-181.indd Page 177 6/29/21 9:57 PM f-0116 /103/HO02206/work/indd

Solid state disk drive Solid state disk drive

A solid state disk drive, often referred to as a solid state drive (SSD), has no (SSD) A storage device
moving parts. with no moving parts that
✚ An SSD is made up of NAND flash memory, and a controller. uses flash memory and a

7 Computer organisation and architecture

✚ NAND flash memory is non-volatile and is based on floating gate controller.
transistors. NAND flash memory
✚ The storage is split up into blocks, and each block is split up into pages. Memory which can be
✚ Pages cannot be over-written – they must be erased first. However the electrically erased and
whole block must be erased. reprogrammed.
✚ The controller is used to manage these complexities.
Controller A component of
SSDs have lower latency and faster access speeds than HDDs, due to the an SSD which manages the
lack of moving parts. However they have a limited number of read and write reading and writing.
actions before the flash memory degrades.
Floating gate transistor
As well as SSDs there are several types of solid state storage device, including Used as a memory cell in
SD cards and USB sticks. flash memory.

Solid state drives are gradually replacing HDDs in desktop and laptop Block A subdivision of the
computer systems. storage on an SSD.

Figure 7.12 shows the principle of NAND flash memory. When the control gate Page A subdivision of a
is turned on, electrons flow from the source to the drain, and some of those block.
electrons are attracted into the floating gate. When the control gate is turned Latency The time taken for
off, the electron flow stops and electrons in the floating gate are trapped the first signal to reach its
there. The presence, or not, of electrons in the floating gate corresponds to a destination.
‘1’ or ‘0’ state.
NAND memory Note
Insulating oxide
layers The spelling ‘disk’ is
Control Gate generally used in computing
terms, though the terms
Floating Gate CD and DVD are standards
defined using the spelling
‘disc’. You will not be
Source Drain
n+ n+
penalised for your choice of
electrons
spelling.

P = Substrate Optical disk A storage

medium which uses light to
Figure 7.12 The principle of NAND flash memory read the data.

Optical disk CD Compact disc, typically

used for audio and small
Optical disks include the CD (compact disc), DVD (digital versatile disc) and computer programs.
BluRay. Optical refers to light, and the data is read using a laser beam which is
DVD Digital versatile disc,
a form of light.
typically used for medium
✚ Data is stored on one spiral track that covers the disk, and this track uses
quality video and medium
pits (indentations) and lands (bumps).
sized computer programs.
✚ The laser beam is shone at the disk, focused at a point on the track
✚ The laser light is reflected back to a sensor while the disk spins at a BluRay A high-capacity
constant linear velocity. optical disc typically used
✚ Where a pit or land continues, a certain amount of light is reflected back to for high definition video and
the sensor, representing a 0. large computer programs.
✚ Where there is a transition between a pit and a land, the light is scattered Track A spiral line that
which means a different amount of light is reflected back to the sense – covers the CD, containing
this represents a 1. pits and lands.

Note Pits Small indentations in

the track.
Because the distance around the outside of the disk is much further than the inside
of the disk, the disk will spin at different speeds so that the portion of the track being Lands Small bumps in the
track. 177
read by the laser moves at a constant velocity relative to the laser.

My Revision Notes AQA A-level Computer Science Third Edition

325487_07_MRN_AQA_CS_158-181.indd Page 178 6/29/21 9:57 PM f-0116 /103/HO02206/work/indd

single spiral track runs from Now test yourself

the centre to outer part of disk
59 How is a platter used in
pits and lands
a mechanical hard disk
7 Computer organisation and architecture

drive?
60 What two components
make up a solid state
disk drive?
61 How is data stored on
an optical disk?

Answers available
Figure 7.13 The workings of an optical disk online

Comparing capacity and speed

It is important to be able to make a practical judgement about the suitability
for different storage devices in different circumstances.
The technology (and particularly the cost and the capacity) are constantly
changing; however the basic principles remain consistent.
The main areas for comparison are capacity and access speed, however cost
and robustness are also factors to consider.

Hard disk drives Solid state disk Optical disks Capacity The amount of
drives data that can be stored.
Capacity Typically have a very Have a range of Small storage
Access speed The time
large storage capacity capacities, typically capacity (700 MB for
taken to read or write data.
(frequently measured measured in GB and CDs up to 50 GB for
in TB). increasingly available BluRay). Cost This can mean the
in the TB range. cost per device, or cost per
Access time Access time is quite Access time is Access times are GB.
slow. extremely fast due extremely poor Robustness How likely
to the lack of moving and is by far the the device is to break or be
parts. slowest of the damaged.
three technologies
discussed here.
Cost The overall cost of The cost is higher The cost to produce
storage is very low. than for an equivalent one disk is very low,
capacity HDD. though the overall
cost per GB is quite
high.
Robustness Fairly robust, although Very robust Robust – they can
they can be damaged and difficult to be dropped without
if dropped and can accidentally damage. too much damage.
be affected by very However, there are However, they are
strong magnetic a limited number easily scratched and
fields. of read and write exposure to UV light
Exam tip
actions before the can degrade the When asked to compare
memory degrades. tracks. solutions it is vital to refer
Applications An effective solution An effective solution Suitable for back to the scenario given
for archived storage as a storage device in transporting data in the question, as the
of large quantities of a laptop or desktop in small batches, effectiveness of each
data. due to their fast such as computer solution will depend heavily
response time. programs, films and on the requirements in that
music. specific case.

178

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_07_MRN_AQA_CS_158-181.indd Page 179 6/29/21 9:57 PM f-0116 /103/HO02206/work/indd

Now test yourself

62 Which storage device has the fastest access time?

63 Which storage device typically has the largest capacity?

7 Computer organisation and architecture

64 Which storage device would be most suitable for use in a mobile phone? Justify
your answer
65 Which storage device would be most suitable for sharing copies of a computer
program? Justify your answer

Answers available online

Summary
✚ The internal hardware of a computer system includes ✚ Each instruction is made up of an opcode (including the
the processor, main memory, I/O controllers, and buses addressing mode) and an operand
✚ Main memory is addressable so that each section of ✚ When using immediate addressing the operand should
memory can be retrieved when needed be treated as the value (also known as the datum) to be
✚ I/O controllers are hardware interfaces between the operated on
internal components and external devices ✚ When using direct addressing the operand should be
✚ The three buses are the control bus, address bus and treated as the address of a value in memory or a register
data bus ✚ In assembly code programming each opcode is
✚ The control bus is used to transmit control and status represented as a short piece of text
signals ✚ AQA examinations use a consistent assembly language
✚ The address bus is used to transmit the address where instruction set which is provided in the exam paper – it
data is to be read or written, from the processor to is important to be familiar with using this assembly
memory language instruction set to write simple programs and
✚ The data bus is used to transmit data to and from to hand trace existing programs
different components ✚ Basic AQA assembly language operations include: load,
✚ von Neumann architecture uses one area of memory add, subtract, store, branching, compare and halt
for both instructions and data, and is used in general- ✚ Bitwise logical operations such as AND, OR (ORR), XOR
purpose computers (EOR) and NOT (MVN) are carried out one bit at a time
✚ Harvard architecture uses two distinct areas of ✚ Logical shifts involve shifting the binary values to the
memory, one for instructions only and one for data left (LSL) or right (LSR) a given number of times
only, and is used in embedded systems such as digital ✚ A logical shift left has the effect of doubling the value
signal processing systems (DSPs) each time
✚ Programs can be saved in memory so that they can ✚ A logical shift right has the effect of halving the value
be executed repeatedly without having to input the each time, though at the risk of loss of accuracy due to
program each time underflow
✚ Machine code instructions are fetched from main ✚ Interrupts are signals which need to get the processor’s
memory and executed by the processor attention as it carries out the F-E cycle
✚ The processor is made of five key components – the ✚ There are four main types of interrupt – timing,
arithmetic logic unit (ALU), control unit, clock, general- program error, hardware error and I/O interrupt
purpose registers and dedicated registers ✚ When an interrupt is triggered the state of the processor
✚ The ALU carries out all arithmetic and logical operations is stored on a stack, the source of the interrupt is
✚ The control unit decodes instructions and sends identified, the appropriate interrupt service routine (ISR)
control signals to other devices is called and the state of the processor is restored
✚ The clock generates a timing signal so that each ✚ An ISR is a program designed to decide on the best
process can be synchronised course of action to follow for a given interrupt
✚ General-purpose registers are used to store data that is ✚ Increasing the number of cores in a processor means
being worked on by the processor that more instructions can be carried out simultaneously
✚ Each dedicated register is used for a specific purpose, ✚ Increasing the clock speed of a processor means that
mostly related to the fetch-execute cycle each instruction takes less time to complete
✚ The dedicated registers are the program counter ✚ A cache is a small, very fast area of memory used for
(PC), memory address register (MAR), memory buffer storing the most frequently accessed instructions and data
register (MBR), current instruction register (CIR) and the ✚ The word length dictates the number of bits which can
status register (SR) be processed in a processor core in any one cycle
✚ The fetch-execute cycle is often referred to as the ✚ The address bus width limits the range of available
fetch-decode-execute cycle as there are three phases memory addresses
✚ It is important to be familiar with the steps involved in ✚ The data bus width limits the amount of data which can
the fetch phase of the F-E cycle be fetched from memory in one cycle
✚ Each processor has its own instruction set, describing ✚ A barcode reader emits a red light which is aimed at a
179
the machine code for each operation black and white image made of bars

My Revision Notes AQA A-level Computer Science Third Edition

325487_07_MRN_AQA_CS_158-181.indd Page 180 6/29/21 9:57 PM f-0116 /103/HO02206/work/indd

✚ The sensor in the scanner reads the reflected light and and off, to remove the electric charge on the drum and
the decoder converts this into a digital code attract toner to areas that should be black
✚ Barcodes are cheap, small and light, but require line of ✚ The drum rolls over the paper, transferring the toner
sight and can easily be obscured or damaged which is then fused to the paper
7 Computer organisation and architecture

✚ A digital camera uses a lens to focus the incoming ✚ A mechanical hard disk drive (HDD) uses platters split
light onto a sensor chip with an array of sensors which into tracks and sectors
produces an electrical signal ✚ The platter of a HDD is coated in a magnetic film which
✚ Each sensor represents one pixel can be read and written to using a read/write head
✚ A colour filter is used to separate the amount of red, ✚ A solid state disk drive contains NAND flash memory
green, and blue light being captured. and a controller
✚ Each sensor electrical signal is processed by an ADC ✚ The SSD data is stored using floating gate transistors
✚ Digital cameras collect a large amount of data, which and is split into blocks and pages
requires a lot of processing in an automated system ✚ An optical disk uses a single spiral track filled with pits
✚ RFID (radio frequency identification) is able to transmit and lands
data over a short distance using radio waves ✚ A laser beam is used to read the data on an optical
✚ Active RFID tags use a battery to allow them to disk, with a transition between a pit and a land
broadcast the signal and have a longer range. representing a 1
✚ Passive RFID tags take power from the radio frequency ✚ HDDs typically have a large capacity and a medium
energy taken from the signal sent by the RFID scanner access time
✚ RFID tags are cheap, small, and light, and do not require ✚ SSDs typically have a medium capacity and a fast
line of sight access time
✚ A laser printer applies electrical static charges to a ✚ Optical disks typically have a small capacity and a very
drum and toner and uses a laser beam, switching on slow access time

Exam practice

1 A general-purpose computer is being designed.

a) Name three internal components that must be included in the design. [3]
b) State whether the computer should use von Neumann or Harvard architecture. [1]
c) Justify your answer to part b). [2]
2 A computer system has a 2 GHz, dual core processor.
a) Explain why a 1.5 GHz, quad core processor might be able to complete a task more quickly. [2]
b) A computer has an address bus of 7 bits. Describe the impact of
increasing the size of the address bus by 1 bit. [2]
c) Identify two other ways to improve the performance of a processor and explain the impact of each. [4]
d) Describe the steps involved with fetching and executing an instruction from memory using the full names
for any registers involved. [6]
3 a) Look at this assembly language program. Follow the program to copy and complete the trace table. [4]
label1:
AND R4, R1, #1
CMP R4, #0
Registers Main memory
BNE label2
R0 R1 R2 R3 R4 R5 100 101 102 103
LSR R1, R1, #1 12 1
LSL R2, R2, #1
B label1
label2:
STR R1, 101
STR R2, 102
HALT
b) Explain the purpose of the instruction AND R4, R1, #1. [1]
c) Explain the effect of applying a logical shift left by 4. [2]
d) Explain the potential problem with applying a logical shift right. [2]
➜
180

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_07_MRN_AQA_CS_158-181.indd Page 181 6/29/21 9:57 PM f-0116 /103/HO02206/work/indd

4 Write an assembly language program to carry out the following high level algorithm.
The mathematical operator // represents an integer division.
Assume that register R1 holds the value of A, R2 holds the value of B, R3 holds the value of

7 Computer organisation and architecture

C and registers R4-R10 are available if needed. [8]
B ← A // 4
B ← B * 4
IF B = A THEN
C ← 1
ELSE
C ← 0
5 A new footwear shop is considering how to enter customer purchases into its computer system.
The options available are using a barcode scanner, digital camera or RFID scanner. Discuss the advantages and
disadvantages of each and identify a suitable recommendation. [8]
6 A student is buying a new laptop for college and is unsure whether to purchase a laptop with a mechanical
HDD or a solid state disk drive.
a) Identify two components that make up a mechanical HDD. [2]
b) Identify two components that make up a solid state drive. [2]
c) Describe the advantages and disadvantages for each option and identify a suitable recommendation. [6]

181

My Revision Notes AQA A-level Computer Science Third Edition

325487_08_MRN_AQA_CS_182-187.indd Page 182 6/29/21 9:56 PM f-0116 /103/HO02206/work/indd

8 Consequences of using computers

Moral, ethical, legal and cultural

issues and opportunities
The increasingly widespread use of technology has brought about many changes,
most of which feature a combination of positive and negative consequences.
It is important to be able to discuss both previous examples (case studies) and
potential future developments (hypotheticals). Hypothetical
Consideration of a ‘what
Questions on this topic may be asked in relation to some technical details
if…’ scenario.
covered in other chapters (for example, networks in Chapter 9, encryption in
Chapter 5, automated input devices in Chapter 7, or robust program design in
Chapter 1). However, no new technical knowledge is required specifically for
this topic.

Definitions
✚ Moral and ethical issues relate to what is right or wrong, and how
consequences might affect individuals and groups of people. For example, an
automated telephone answering system might mean that a company makes
more profit by reducing the need for human call handlers, however it might
mean that the user experience for the customer is less satisfying and will
likely have a negative consequence for the staff who are no longer needed.
✚ Legal issues relate to the law. This might refer to those making laws,
enforcing laws and those who may break laws.
✚ Cultural issues relate to how the culture of a society changes. For example,
the widespread availability of computing equipment and fast internet
access has allowed many more people to work from home. This has both
positive and negative consequences such as reduced travel costs, but also
a lack of social interaction that previously happened in the workplace.

Communication and information flow

Governments, companies, organisations, and even individuals are
increasingly able to capture vast quantities of data about people. This
includes:
✚ location data
✚ browsing data
✚ shopping habits
✚ contact details
✚ political opinions.
This data may be collected, stored, analysed and even sold without our
knowledge or understanding. With so much data available:
✚ behaviour can be monitored – for instance, enabling your mobile phone
location and Bluetooth allows third parties to collect an incredible level of
detail about where you are every second of the day
✚ multiple sources of data can be combined and analysed to build up very
detailed pictures about someone – for instance where they shop, how much
they spend, who they spend time with, how much they earn, how fit they are
✚ This data may be published and distributed outside the user’s control – for
instance, when material such as photographs are uploaded to social media
sites, their terms and conditions normally allow the site to reuse as they
182 wish – even though you might think it is ‘your’ photograph.

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_08_MRN_AQA_CS_182-187.indd Page 183 6/29/21 9:56 PM f-0116 /103/HO02206/work/indd

Users’ personal data and personal preferences can be used in a variety of

Personal data Data that
ways, with or without their consent, and the consequences of how this data is
relates to an identifiable
collected and how it is used has a significant impact on individuals, society, person.
and on the culture of that society.

8 Consequences of using computers

Moral Relating to the
Software engineers and computer scientists must make decisions about the principles of right and
algorithms they design, the systems they build and the software they write wrong.
in order to benefit, rather than harm, individuals and societies. They must
ensure that appropriate moral and cultural values are embedded within their Cultural Relating to
the ideas, customs and
software. This can mean many things, but examples could include:
behaviour of a society.
✚ voice recognition systems that struggle to interpret regional accents
✚ facial recognition systems that display racial bias.
In both of these examples, appropriate moral, ethical and cultural values have
not been embedded into the software.

Negative consequences may also Example

occur because of …
… a deliberate act. A programmer includes a secret login so that they can access data without
detection once the system is running.
… poor anticipation of consequences. A Google street view car captures and publishes images of someone getting
changed in front of a window.
… poor design decisions. A programmer chooses not to encrypt data before it is transmitted over a
network.
… a mistake. An error in the way the login process is handled means that a user can trick the
system into giving them administrator access.

The people who develop systems must think carefully about what data is to
be collected, how it can be collected, how it should be stored, what processing
will occur and how the results of the processing should be used.
Another point to consider is that software is scalable. This means that: Scalable Used to describe
✚ once written and compiled, the internet allows the same software to be a product, service or
easily distributed to almost anywhere in the world business that can cope with
✚ one piece of software could be installed on literally millions or even increased demand.
billions of computers.
This represents the tremendous impact that software developers and
computer scientists can have, and why they must consider all of the ethical,
moral, legal and cultural impacts of their code.

Now test yourself

1 Identify four items of personal data that are typically stored in online systems.
2 When creating software, suggest three ways that inappropriate consequences
could be introduced into the program.

Answers available online

Legislation
It is not necessary to have a detailed knowledge of computing-related
legislation, though it may be useful to have a basic knowledge of some key
examples.

Computer Misuse Act

This legislation is used to prosecute individuals who attempt to gain
unauthorised access to a computer system. This includes attempting to gain
access, copying data, editing data and deleting data. The emphasis is on
discouraging people from committing malicious acts on another system and
183
making hacking illegal.

My Revision Notes AQA A-level Computer Science Third Edition

325487_08_MRN_AQA_CS_182-187.indd Page 184 6/29/21 9:56 PM f-0116 /103/HO02206/work/indd

Data Protection Act Note

As organisations collect, store and process personal information, steps should
The Data Protection Act
be taken to ensure that the data cannot be accessed by unauthorised parties,
was updated in 2018 to
and that the data cannot be tampered with. This may well involve password
8 Consequences of using computers

implement the General Data

protection, data encryption, anti-malware and other protective measures. Protection Regulation (GDPR).
The Data Protection Act puts the emphasis on organisations who hold data A detailed understanding
to make sure that the data is protected. It is important to note that small, of legislation is not required
community-based organisations who keep a mailing list of email addresses however and either term
are subject to the Data Protection Act just as much as multi-national would be accepted.
corporations.

Making links
Encryption is an important aspect of ensuring that data cannot be understood.
Techniques for encryption are explored in Chapter 5.
Network security concerns, including malware, are discussed in Chapter 9.

Copyright, Designs and Patents Act

Often shortened to simply ‘copyright’, this legislation sets out what can and
cannot be legally reproduced or copied. Any original piece of work (such as
an image, script, video, music or software code) is automatically protected
by copyright law which means it cannot be copied, used, or re-distributed
without permission, with a few small exceptions.

Regulation of Investigatory Power Act (RIPA)

RIPA aims to describe what powers the police and other authorities have
to demand access to personal data stored by online services, such as
social networking companies, internet service providers and mobile phone
manufacturers.
The aim is to strike a balance between protecting individual freedom for
users, with the need to allow investigations that are in the interests of
national security, convicting criminals and helping to locate missing persons.

Other relevant legislation

Depending on the context of the question, other legal issues may well occur.
For example, in a partially automated factory there may be health and safety
considerations, driverless cars may be subject to road traffic legislation, and
consumer law may be relevant if the consequences affect customers.

Challenges for legislators

It takes a long time for new legislation to be written and for existing
legislation to be updated.
As technology continues to change at an increasingly rapid pace it can be very
difficult for legislators to keep up with these changes and to make sure that
the legislation is effective, unambiguous and relevant.

Now test yourself

3 What piece of legislation makes it illegal to distribute films, games and albums
without permission?
4 Other than computing-specific legislation, suggest two other legal issues relating
to self-driving cars.
5 Dave keeps a list of email addresses and phone numbers for the players in a small,
local football team. What legislation does he need to consider?
184
Answers available online

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_08_MRN_AQA_CS_182-187.indd Page 185 6/29/21 9:56 PM f-0116 /103/HO02206/work/indd

Case studies
In order to explore the topic of the consequences of computing, it can be
useful to look at past examples and to discuss both the technological details Making links

8 Consequences of using computers

and the moral decisions behind a particular event. Fully autonomous cars will
use machine learning to
Self-driving cars acquire the skills needed
to drive. It requires vast
Some cars currently include a semi-driverless mode, in which sensors are
amounts of data to learn
used to keep the car from drifting out of its lane and the car can steer itself as
from – there is more on Big
well as controlling throttle and brake application in order to avoid potential
Data in Chapter 11.
hazards. Drivers are instructed that they are still responsible for the vehicle
and should remain in control at all times.
Collisions which occur when drivers are using semi-driverless mode tend to
be high profile, and in many such cases the drivers have been distracted and
not in full control of the car at the time of the collision.
✚ Is it safe to force cars to stop if the driver lets go of the steering wheel?
✚ Could cameras be used to track drivers’ eye movements in order to ensure
they are paying attention to the road?
Many sensors are used on automated vehicles in order to identify potential
hazards and obstructions, however it is very difficult to create programs that
can accurately identify objects under all conditions and contexts.
Despite the higher profile new coverage of collisions involving automated
vehicles, the overall rate of accidents per mile is typically far lower than that
for manually driven cars.

Google street view

Google street view allows users to see locations before they visit. Vehicles
with roof-mounted cameras drive along roads, capturing images in all
directions. These images are then made available on the internet so that the
public can view them.
This is helpful for people planning journeys and also to help people look at the
surrounding area before deciding to visit or move to that area. These images
can also be used in news stories to help provide viewers and readers with a
geographical context.
There are several potential ethical issues relating to the use of Google
street view.
✚ Many people are concerned about an invasion of privacy and are unhappy
at the idea of Google storing and sharing images of them without
permission.
✚ Personal details such as faces and number plates are blurred using image
recognition techniques, though it cannot be guaranteed that all personal
data has been obscured.
✚ There are concerns that there are potential security risks with showing
exact routes and features. This includes areas of high risk such as
government buildings, but also private homes and businesses.

Malicious programmers
There have been many cases where a programmer has written code that will
collect personal data that is not necessarily needed or has created a method
by which they can collect data that they should not have access to.
Relevant legislation such as the Computer Misuse Act and the Data Protection
Act is appropriate for discussion here, and it is reasonable to ask how the
situation could have been prevented.
The sharing and auditing of code within an organisation is important in order
to be able to recognise situations where a programmer has, either maliciously 185

My Revision Notes AQA A-level Computer Science Third Edition

325487_08_MRN_AQA_CS_182-187.indd Page 186 6/29/21 9:56 PM f-0116 /103/HO02206/work/indd

or accidentally, introduced inappropriate code. Security vulnerabilities

can then be fixed and, if a programmer has made an inappropriate ethical
decision, this can be resolved.
8 Consequences of using computers

Exam tip
It is common for questions of this type to not have a ‘right’ and ‘wrong’ answer. Credit
is awarded for identifying, discussing and evaluating potential issues.

Marks are awarded for identifying relevant knowledge; for example, suggesting
appropriate input devices for collecting data or identifying the methods by which
wireless data transmissions can be intercepted.
To reach the top mark bands it is important to follow a line of reasoning, using your
knowledge to write in connected sentences in a way that makes sense and relates
to the context of the question. Explaining how each point links to the scenario and
adding as much technical detail and vocabulary as you can makes it more likely that
you will score well on this type of question.
Always make sure you back up any arguments or suggestions you make with
facts, logical arguments and technical details, as unsubstantiated statements don’t
demonstrate your understanding.

Now test yourself

6 What technological solutions could be used to encourage the drivers of semi-

driverless cars to maintain focus and control of the vehicle?
7 How can a company reduce the likelihood that a programmer will add
inappropriate or malicious code into a product?
8 What could a company such as Google do to help reduce complaints of invading
privacy when capturing and sharing street view images?

Answers available online

Hypotheticals
Another strategy for preparing for these types of questions is to ask ‘what if…?’.

Web-based social networking and cloud storage

systems
A system that allows users to upload and share information and resources is a
common context for questions on the consequences of computing.
A decision must be made regarding what data is to be collected and stored,
including geographical information, user preferences, and how much access
the administrators of the service should have to the content uploaded by
individual users.
In terms of legislation, it is the responsibility of the administrators to ensure
that data is protected. This means that data should be stored securely,
communication should be encrypted and password policies should be
enforced.
It is also important to consider copyright implications, as the creators of the
service could be held liable if the users are distributing material which cannot
legally be shared.
Users have the right to be forgotten, that is the right to have all personal data
removed from an organisation. It is a legal, but also an ethical consideration
as to whether any posts, files or other items uploaded are removed, or simply
anonymised.
186

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_08_MRN_AQA_CS_182-187.indd Page 187 6/29/21 9:56 PM f-0116 /103/HO02206/work/indd

The Internet of Things (IoT) Internet of Things

The Internet of Things refers to the use of internet enabled, ‘smart’ devices. (IoT) Physical devices
This includes: able to share data or to
✚ plugs ✚ doorbells be controlled using a

8 Consequences of using computers

✚ lights ✚ thermostats connection to the internet.
✚ televisions ✚ washing machines
✚ fridges ✚ security cameras Making links
✚ coffee machines ✚ digital assistants.
The Internet of Things and
Each device is designed to be accessed by the user over the internet so that
the increasing use of smart
they can access information and issue instructions to the device while away
devices means that a vast
from home.
amount of data is being
Login systems must be robust so that unauthorised users cannot access captured at high speed and
the device. A baby monitor, security camera or doorbell camera could give a in many different forms.
potential intruder detailed information about the layout and contents of a home This is referred to as Big
as well as information about typical routines and when the home is empty. Data, and the approaches
for storing and processing
Systems that control objects means that a malicious user, or an error in the code,
Big Data are discussed in
could cause devices to malfunction and potentially cause damage. For example: Chapter 11.
✚ a smart device for managing the environment in a tropical fish tank could
lead to temperatures too high or too low for the fish to survive
✚ altering a smart thermostat could lead to an empty house suffering frozen Revision activity
pipes or an increase in heating bills. Find or create a list of as
many smart devices as you
Now test yourself can. For each one, identify
9 Identify two potential legal issues with creating a system in which users can potential ways in which a
upload and share content. malicious user could take
advantage of the device if
10 Suggest two security risks associated with the use of smart, IoT devices.
they had access to it.
Answers available online

Summary
✚ Computer systems make it easier and more common ✚ Other legislation is often relevant depending on the
for organisations to collect personal data circumstances, including health & safety, road safety
✚ How that personal data is used has significant and consumer rights legislation
consequences that affect individuals, groups, and ✚ Legislation often lags behind changes in technology
larger societies as it takes time to recognise the need for it and then
✚ Moral and ethical consequences are those that affect create tightly defined legislation to deal with the
individuals and groups of people. These consequences consequences of that technology
can be positive or negative, and are usually elements ✚ Case studies of current and recent ethical
of both consequences of computing issues are an effective
✚ Cultural consequences affect the way that a society way of preparing for questions on this topic
works, thinks or behaves ✚ Hypothetical, what if, questions are a useful approach
✚ Computer scientists and software engineers have a to preparing for this topic
responsibility to ensure that systems are developed ✚ Questions on ethics will often include a technical
with these consequences in mind element that relies on knowledge of other topics such
✚ Design decisions should consider how data is to be as network security, data collection methods and
collected, stored, processed and used effective program design
✚ Consideration should be taken of the impact on ✚ To reach the top mark bands it is important to follow a
individuals and on society at large line of reasoning and to back up your points with facts,
✚ Computer-based legislation includes the Computer logical arguments and technical details
Misuse Act, Data Protection Act, Copyright, Regulation of
Investigatory Powers Act, and Designs and Patents Act

Exam practice

1 A company wishes to produce an automated garage Discuss a range of technologies that could be used
door opener. The door should automatically open when to allow the automated garage door opener to
the homeowner returns. function and consider the moral, ethical, legal and
cultural consequences of using this device. [12] 187

My Revision Notes AQA A-level Computer Science Third Edition

325487_09_MRN_AQA_CS_188-213.indd Page 188 6/29/21 9:52 PM f-0116 /103/HO02206/work/indd

9 Fundamentals of communication
and networking

Communication
Communication methods
In serial data transmission each bit is sent one after another. This can be done
Serial data transmission
with a single wire, or a single track on a circuit board.
Data is sent down one wire,
In parallel data transmission multiple bits are sent simultaneously. To achieve one bit after another.
this there must be multiple wires or multiple tracks on which to send the
Parallel data transmission
data. Data is sent on several wires,
In theory it is quicker to send data over a parallel connection as more data simultaneously.
can be transferred at once. However, there are several reasons why most Crosstalk Interference
communication is sent serially: caused when two or
✚ Crosstalk occurs when the signal in one wire causes electrical interference more worse are in close
with the signals on the neighbouring wires. proximity.
✚ Data skew can occur, meaning that one or more bits can be read incorrectly
Data skew When data that
by the receiver if the transmission is not correctly synchronised.
was sent at the same time
✚ The hardware for serial communication is simpler (and, therefore, cheaper)
arrives at a slightly different
to produce.
time to each other.
In synchronous data transmission both sender and receiver must be in sync.
Synchronous data
This means that a common clock must be used and the timing signal must be
transmission Where data
sent in addition to the data. is sent along with a timing
In asynchronous data transmission there is no common clock. signal.

The sender will send a start bit at the beginning of a transmission to indicate Asynchronous data
that the data is about to be sent. A stop bit will be sent at the end of that transmission Where data is
transmission. A start bit will be 0 and a stop bit will be 1. sent without a timing signal.

The process then repeats, and the receiver is able to use the start bit to Clock Used to provide a
synchronise its clock to that of the sender. This synchronisation is carried out timing signal.
each time a start bit is received. Start bit A bit sent at the
start of a message in order
Additionally, a parity bit might be added before the stop bit in order to allow
to provide timing data.
error checking to occur.
Stop bit A bit sent to mark
Finally, the transmission is often shown in reverse, indicating which bits will
the end of a message.
be received first by the receiver. If so, then the MSB would be on the right
hand side. Parity bit A bit used in
error detection.
A typical transmission of an ASCII character using odd parity might work as
follows: ASCII A 7-bit code used for
representing characters as
Character to transmit: M binary numbers.
ASCII value: 7710 or 100 11012
Parity bit : 1 Exam tip
Data transmitted: Parity bits are sometimes
included in questions on
Stop bit Parity bit Data Start bit start and stop bits.
1 1 1 0 1 1 0 0 1 0

188 Notice that the data is represented in reverse order.

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_09_MRN_AQA_CS_188-213.indd Page 189 6/29/21 9:52 PM f-0116 /103/HO02206/work/indd

Making links
The use of ASCII to represent characters using binary numbers is an important aspect
of data transmission as non-numeric data must often be communicated and is

9 Fundamentals of communication and networking

discussed at length in Chapter 5.
Due to the potential for data to be damaged during transmission, a number of
methods exist to try to detect and correct these errors. These methods, including
parity, are discussed in Chapter 5.

Now test yourself

1 Describe the main difference between serial and parallel data transmission.
2 In synchronous data transmission, what is transmitted other than the data itself?
3 In asynchronous data transmission, what is transmitted other than the data itself?

Answers available online

Communication basics
When data is transmitted a signal is sent which changes at fixed time
intervals to indicate a new value. 11
Each signal can represent more than one binary digit by using more than one
10
possible value. For example, an electrical current can be sent with a different
voltage.
01
The baud rate is the number of state changes per second and is measured in
Hz. For example, at a baud rate of 100 kHz there are 100 000 changes to the 00
signal per second.
0 1 2 3 4
The bit rate is the maximum number of bits that can be transferred per
Time (1/1000s of a second)
second. This can be calculated by multiplying the baud rate by the maximum
number of bits per signal. Figure 9.1 Sending multiple
bits per value at a baud rate
The bandwidth describes the range of possible signals, and a larger of 1 kHz and a bit rate of 2000
bandwidth means that a larger number of distinct values can potentially be bits per second
sent in one signal.
The bit rate is directly proportional to the bandwidth; in other words, if the Baud rate The number of
bandwidth doubles, then there are twice as many possible unique symbols state changes per second.
that could be sent. This means that the bit rate doubles as well. Bit rate The number of bits
The latency describes the time it takes for the first signal to reach its that can be transmitted per
destination. This is not linked to the baud rate, bit rate or bandwidth of a second.
connection. For instance, a high bandwidth connection can have a high or low Bandwidth The range of
latency depending on a number of other factors such as physical distance. possible signals that can be
sent in one signal.
A protocol is a set of rules or standards, and a number of transmission
protocols exist. These are essential as different systems need to be able to Latency The time taken for
communicate with each other. the first signal to reach its
destination.
Now test yourself Protocol A set of rules or
4 What term refers to the range of possible values in one received signal? standards.

5 What term refers to the time it takes for a signal to arrive at its destination?
6 What term refers to the number of signals that can be sent or received per second?
7 How is the total bit rate calculated?
8 What is the definition of a protocol?

Answers available online

189

My Revision Notes AQA A-level Computer Science Third Edition

325487_09_MRN_AQA_CS_188-213.indd Page 190 6/29/21 9:52 PM f-0116 /103/HO02206/work/indd

Networking
Network topology
9 Fundamentals of communication and networking

The topology of a network describes the physical or logical layout of that

Topology The physical
network. or logical arrangement of
connections in a network.
Bus topology
In a bus topology a main cable, or bus, is used to transmit all data across the Bus topology A network
network. arranged with a main cable,
or bus, connecting all
Each device connects to the bus, meaning that no central switch is necessary. devices.
Server Workstation

Terminator Terminator

Printer
Figure 9.2 A bus network

Advantages of a bus network:

✚ Simple cabling requirements.
✚ Easier to add new devices as the new cabling doesn’t need to go all the way
back to a central point.
Disadvantages of a bus network:
✚ Affected by high traffic as all traffic must use the main bus.
✚ Weaker security as all data is transmitted to all devices.
✚ Weaker reliability due to the reliance on the main bus – if that fails, the
whole network fails.

Star topology
In a star topology, a central switch is used to connect each device on the Star topology A network
network together. All data transmissions are passed through a link to the arranged with a switch (or
switch which then forwards the data on to the intended recipient. hub) at the centre.
Many small networks, and almost all home networks, are based on this
Switch A device that
topology.
receives and forwards data
on a network.
Link A physical connect
between two devices.

190
Figure 9.3 A star network

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_09_MRN_AQA_CS_188-213.indd Page 191 6/29/21 9:52 PM f-0116 /103/HO02206/work/indd

Advantages of using a star network include:

✚ copes well with high traffic as only the links between the switch and the
relevant devices are used, meaning that data can be transferred between
the switch and different devices at the same time

9 Fundamentals of communication and networking

✚ improved security – the data is only sent to and from the intended device
✚ improved reliability – if one device or link fails then the rest of the network
is unaffected.
Disadvantages of using a star network include:
✚ cabling requirements can be complex as multiple cables are needed
✚ adding new devices means running an extra cable back to the central switch
✚ adding devices quite far away from the switch increase the cabling
complexities
✚ if the central switch fails then the whole network fails.
It is possible to use a hub instead of a switch at the centre of a star topology. A Logical bus network
hub will broadcast all signals it receives, creating a logical bus network topology. topology A star network
This means that whilst it has been arranged as a physical star network, it that uses a hub as the
actually operates like a bus network – for instance, all data will be transmitted to central device, which
all devices. This approach is not advised or common practice any more. causes it to operate like a
bus network.
Note
An increasingly common network topology is the mesh network. In a mesh network
each device can be connected to two or more other devices, allowing for more
possible routes across the network.
The internet is the most significant example of a mesh network, and many larger Wi-Fi
networks make use of a mesh topology.
The mesh is not covered in the specification and is unlikely to appear in an exam, however
it is an important topology to understand if you wish to study networking in more detail.

Logical versus physical topologies

The physical topology refers to the physical layout or arrangement of the
cabling and connections between devices on the network.
The logical topology refers to the way that the data flows around a network.
It is possible for a physical star topology to behave as if it were a bus topology
by transmitting all data to all devices.

Now test yourself

9 What device is always used in a star network? 13 Give two advantages and one disadvantage for using
10 What does the word ‘bus’ refer to in a bus network? a star network.
11 What is the definition of a physical topology?
12 How can a network have a different physical topology
and logical topology? Answers available online

Types of networking between hosts

Client–server
Client–server network
In a client–server network, a server controls access to a centralised resource; A network in which clients
for example: make requests to servers.
✚ files, in a file server
✚ emails, in an email server Client A device which
✚ web pages, in a web server. makes requests.
Server A device which
The client requests access to the resource and the server then processes the
controls centralised access
request, decides whether access should be granted and provides the service
or response.
to a resource. 191

My Revision Notes AQA A-level Computer Science Third Edition

325487_09_MRN_AQA_CS_188-213.indd Page 192 6/29/21 9:52 PM f-0116 /103/HO02206/work/indd

Client–server networks are used where the central resource must be

managed; for example:
✚ selling seated tickets to a show
✚ assigning seats on an aeroplane
9 Fundamentals of communication and networking

✚ managing a centralised backup

✚ managing centralised security (anti-virus, firewalls, software updates, and
so on)
✚ managing a centralised login system.
In client–server networks security is increased as there can be a centralised
login or permissions system. However, configuration of the network is more
complex and running a separate central server requires additional work.
For small, simple networks a server can be a fairly low powered machine such
as a standard workstation. For most situations, however, a server should be a
specialised computer with a lot of processing power and memory.
One example of a client–server network is a school network with a centralised
file server, print server, and email server.

Peer-to-peer
In a peer-to-peer network all devices have an equal status, and no resource is
centrally controlled.
Each resource can be stored on one or multiple devices on the network and
can be accessed by any other device on that network. Peer-to-peer network A
network in which all devices
This effectively shares the storage and processing load across the devices and
are peers.
is useful for situations where no central control is needed, or when a large
number of requests might cause delays if relying on a central server. Peer Of equal standing,
able to act as a client or a
Peer-to-peer networks are frequently used for accessing or sharing large files, server.
as segments of the file can be stored on and copied from any device on the
network with that segment.
Peer-to-peer networks don’t require a central server and are typically easier to
set-up and maintain. Management of security is more difficult, however, and
it is difficult to ensure data consistency across the network.
One example of a peer-to-peer network is a simple home WiFi network in
which files can be sent to or from any device on the network.

Advantages Disadvantages
Client–server Better security – software and security If the server goes down then all clients are affected.
updates managed centrally, logins and
Access times may be slow if the server gets too many
access to files/folders also controlled.
requests from different clients at the same time.
Data backups easier to manage as all
More expensive because of the costs involved with
held in one place.
maintaining the central server.
A hacker targeting the server can bring down the network.
Peer-to-peer Easy to set up. Less secure – all devices have to have software and security
updates run individually, so outdated security is more likely.
Less expensive to set up – doesn’t
require specialist hardware. Multiple versions of files on different devices.
Network unaffected if one device fails. Each device needs to be backed up individually.

Exam tip
Questions on client–server or peer-to-peer networks tend to be longer discussion
and comparison questions. Make sure that you consider the context of the question
and remember that a server is always used in systems where centralised control is
important.
192

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_09_MRN_AQA_CS_188-213.indd Page 193 6/29/21 9:52 PM f-0116 /103/HO02206/work/indd

Now test yourself

14 Describe the role of a server.

15 Describe the role of a client.

9 Fundamentals of communication and networking

16 What is meant by a peer-to-peer network?
17 Give three advantages and two disadvantages for using a client–server network.

Answers available online

Wireless networking
Wireless networking uses radio frequencies in order to transmit data over a
Wireless network A
relatively small area (such as a house or office), called a local area network
network that allows devices
(LAN).
to transmit data using radio
Wi-Fi is an international communication standard for wireless networking. frequencies.

No cabling is required in a wireless network, meaning that devices can be Wi-Fi A set of technology
moved easily without disrupting communication. standards that allows
devices to communicate
In order to connect to a wireless network, a wireless access point must be using radio frequencies.
used. This is a device which broadcasts and receives the wireless signal to
and from the devices on the network. Wireless access point
A hardware device for
A wireless access point can be, and frequently is, connected to a wired allowing other devices
network in order to allow both wired and wireless connections, as necessary. to connect to an existing
Each device connecting to the wireless network requires a wireless network wireless network.
adapter. This can be built into the device (for example, on a computer’s Wireless network
motherboard) or added later (for example, through an expansion card or USB adapter A hardware device
wireless adapter). for enabling devices to
communicate using Wi-Fi.
Each wireless network has a service set identifier (SSID) which appears as the
‘name’ of the network. Enabling the broadcast of the SSID makes it easier for SSID Service set identifier-
new users to join a network. an identifier, or name, for a
wireless network.

Wireless security
The nature of wireless networking means that any device within range of the Note
wireless signal will be able to intercept and read the data being transmitted,
Wi-Fi is not an acronym and
without needing to gain physical access to the network.
is not a directly shortened
There are several methods for dealing with this security risk: form of any longer words.
✚ Disabling SSID broadcast. Wi-Fi is not the only wireless
Broadcasting of the network’s SSID can be disabled. Disabling the communication method.
broadcast means that only users who are aware of the network and know Bluetooth, 5G, NFC and
its SSID will be able to connect. Zigbee all make use of
✚ Strong encryption radio frequencies to allow
wireless communication.
Data sent over a wireless network should be strongly encrypted. Several
encryption methods exist including WEP and WPA. Most wireless
networks use WPA2 encryption as this offers significantly stronger
encryption than other methods.
Encryption A method of
Only users with the encryption key are able to send and receive data over scrambling data so that it
the wireless network, and while other devices can read the data being cannot be understood.
transmitted, this cannot be understood without the key. MAC address A physical
✚ MAC address whitelist address, uniquely assigned
A MAC address (or media access control address) is a unique hardware to each piece of network
address assigned to each network interface card (which includes wireless hardware.
network adapters). Whitelist A list of things
By enforcing a whitelist, only those MAC addresses on the whitelist will be considered to be acceptable
able to access the network and any other devices will be refused access. or trustworthy.
193

My Revision Notes AQA A-level Computer Science Third Edition

325487_09_MRN_AQA_CS_188-213.indd Page 194 6/29/21 9:52 PM f-0116 /103/HO02206/work/indd

Now test yourself

18 What technology does a wireless network use for communication?

19 What two hardware devices are needed to create a wireless network?
9 Fundamentals of communication and networking

20 What is an SSID?
21 What two pieces of information are usually required for a device to access a
wireless network?
22 Identify three ways to ensure that wireless data transmissions are secure

Answers available online

Avoiding collisions
If two or more devices attempt to transmit data at the same time, there will Collision Where two items
be a collision and the data will be unreadable. of data are transmitted at
To avoid this there are two systems in use, often used together. the same time, causing both
to be lost.
CSMA/CA stands for carrier sense multiple access with collision avoidance.
CSMA/CA Carrier sense
The aim of CSMA/CA is to allow devices to recognise when a duplicate
multiple access/collision
broadcast is occurring and to wait a random period of time before avoidance. A method
broadcasting again. It works as follows: of collision avoidance
✚ Sending device checks for traffic. by checking for existing
✚ If another device is broadcasting, the sending device waits before repeating transmissions.
the process.
✚ If no other device is broadcasting, the data is transmitted. RTS/CTS Request to send /
clear to send. A method
RTS/CTS stands for request to send / clear to send. of collision avoidance by
Using RTS/CTS the device wishing to transmit sends an RTS request and waits requesting clearance to
for a CTS response from the receiver. If the CTS response is not received, the transmit.
sender wait a random amount of time before sending the RTS again.
The random amount of time is used to ensure that, if two devices attempt to
broadcast at the same time, it is unlikely that both will attempt to broadcast at
exactly the same time again.
Once a CTS response has been received, the data is transmitted.
Finally, the receiver should respond with an acknowledgement (ACK). If this is
not received then the data is resent.
The full process can be written as follows.
✚ Sending device checks for traffic.
✚ If another device is broadcasting, the sending device waits before repeating
the process.
✚ If no other device is broadcasting, an RTS signal is sent.
✚ The receiving device sends a CTS response if it is ready to receive the
transmission.
✚ If no CTS response is received, the sending device waits a random period
of time before repeating the process.
✚ If a CTS response is received, the data is transmitted.
✚ The receiving device sends an acknowledgement (ACK) once all data has
been received.
✚ If the sending device does not receive an ACK then it repeats the process.

Now test yourself

23 What is the main aim of using CSMA/CA and RTS/CTS?

24 How does CSMA/CA improve wireless transmissions?
25 When a wireless device sends a request to send, what does it aim to receive?
26 When a wireless device has finished sending a message using RTS/CTS, how does
it know if the transmission has been received?
194
Answers available online

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_09_MRN_AQA_CS_188-213.indd Page 195 6/29/21 9:52 PM f-0116 /103/HO02206/work/indd

The internet
The internet and how it works

9 Fundamentals of communication and networking

The internet is a global collection of networks, all able to communicate with
each other.
Note

Packet switching As described in Network

topology above, the mesh
With so many networks connected together, the internet uses a mesh topology is increasingly
topology in which each network is potentially connected to many other common, but is not covered
networks. in the AQA specification and
This means that data can take many different routes in order to reach its so you should not be asked
intended destination. any questions relating to
this topic.
Each time data is sent over the internet it is split into equal sized portions,
or packets. These packets contain a portion of the original data as well as a
header that provides the packet number and information relating to each Mesh topology A network
layer of the TCP/IP stack (see TCP/IP Layers below). topology in which each
device can be directly linked
Each packet can take different routes across the internet, meaning that
to several other devices.
bottlenecks can be avoided and if a packet is lost then only that packet needs
to be re-sent rather than the whole of the original data. Packet A small parcel of
data, including a header.
This is known as packet switching.
Header Extra data added
to a packet such as the
destination address or
packet number.
Bottleneck A congested
path within a network.
Packet switching
Source Destination The process of sending
computer computer
individual packets on
different routes.
Router A network
Packet 1
hardware device for
Packet 2
Packet 3 connecting different
networks together.
Figure 9.4 An example of packet switching showing the paths of three different
packets

While a switch is used inside a network to direct packets to their destinations,

a different device is needed to connect a network to other networks and the
internet.

Router
✚ A router is used to connect different networks together. When a router
receives a packet it will read the destination IP address (see IP address
section below).
✚ If the packet is intended for that router’s network then the packet will
passed into the network. If the packet is intended for another network then
it will be forwarded to another router.
✚ A router will try to pass the packet to its destination via the fastest route
possible. This is done either by using the fewest number of steps or using
the route that is least congested at that moment.
✚ In order to complete this task a routing table is created which is used to
store the routes to particular network destinations.

195

My Revision Notes AQA A-level Computer Science Third Edition

325487_09_MRN_AQA_CS_188-213.indd Page 196 6/29/21 9:52 PM f-0116 /103/HO02206/work/indd

Gateway Gateway A network

✚ A gateway also connects different networks together. The unique function hardware device for
of a gateway is to connect networks that use different protocols. connecting networks that
✚ While a router can check the destination address and then forward a use different protocols.
9 Fundamentals of communication and networking

packet, a gateway will strip most of the header data away and create new
headers so that the packet can be transmitted through the next network.

Now test yourself

27 What is the name for a small parcel of data?

28 What hardware device is used to connect networks using different protocols?
29 What hardware device is used to connect networks using the same protocol?
30 Give two advantages for using packet switching.

Answers available online

URLs and domain names

A uniform resource locator (URL) is a method for addressing a resource on
Making links
another network.
There are several key
The most commonly known example of a URL is a web address, though URLs
network protocols covered
are also used for other purposes including email and file transfer (FTP).
in this chapter. They are
A URL is made up of several component parts. discussed in more detail
✚ Protocol: HTTP or HTTPS for webpages, FTP for file transfer, and so on. later in Chapter 9.
✚ Domain name: The identifier for the network. A number of protocols use
✚ Path: The names of any folders and/or the file to be accessed. URLs to locate resources.
More information on
http://www.awebsite.co.uk/index.html specific protocols can be
found in Chapter 9.

protocol domain name path

Figure 9.5 An example URL
Uniform resource
locator (URL) A standard
Domain names are used to identify the organisation or individual on the
structure for addressing
internet, and use a string format with each component separated by full stops.
an online resource using
The elements of a domain name are processed from the right hand side, with alphanumeric strings.
the top level domain (TLD) typically indicating the nature or location of the
Protocol The standard
domain’s subject, such as .co.uk, .gov, .org or rule for that type of
The central part of the domain name indicates the organisation or individual communication.
that the resource is linked to. For example, www.aqa.org.uk is a website run Domain name The
by and about the AQA exam board. identifier for an individual
The left-most part of the domain name is the hostname, which usually or organisation’s online
refers to a specific computer or server within that domain. It is typical for presence.
webservers to use the www hostname (www.awebsite.co.uk), while email Path The location of the file
servers will often use the mail hostname (mail.awebsite.co.uk), etc. or folder on a server that is
being addressed.
A fully qualified domain name (FQDN) addresses an exact resource and can
only be interpreted in one way. In order to achieve this they must include a Top level domain (TLD) The
hostname. right-most term in a domain
✚ awebsite.com/index.html is not a fully qualified domain name name, such as .com or .uk
✚ www.awebsite.com/index.html is a fully qualified domain name Hostname The device
However, when packets are sent to a server over the internet the packet is within the network that is
addressed to an IP address rather than to a domain name. being addressed.
Fully qualified domain
name (FQDN) A domain
name that addresses an
exact resource, including a
196 hostname.

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_09_MRN_AQA_CS_188-213.indd Page 197 6/29/21 9:52 PM f-0116 /103/HO02206/work/indd

IP address IP address A numerical

An IP address is a numeric addressing method that identifies devices on the address for a device on a
internet. The current most version uses four 8-bit numbers, each written in network.
decimal between 0 and 255, separated by full stops. An IP address is typically

9 Fundamentals of communication and networking

Dotted-decimal A format
represented in dotted-decimal format; for example, 107.162.140.19. for writing an IP address
Examples: with decimal values
separated by dots.
123.456.789.123 – invalid, no number can be greater than 255.
Domain name system
217.83.4 – invalid, there must be four numbers. (DNS) A system for looking
4.0.0.1 – valid. up domain names to find
their corresponding IP
IP addresses are long, difficult to remember and easy to mistype. The IP address.
address for a particular network or server can also change from time to time,
Internet registry A body
and so humans use domain names which are much easier to remember and
responsible for allocating IP
are much more meaningful.
addresses.
In order to translate between a domain name and an IP address the domain
name system (DNS) is used.
DNS uses a lookup table to find the corresponding IP address for a given
domain name, just as a mobile phone contacts list is looked up when a phone
call is made.
A computer will send a request to a DNS server in order to find the correct
IP address, and will then send packets to the required server with that IP
address.

Connection established using IP

address received from DNS

Home Laptop
Request for
IP address BBC Server
IP address sent

DNS Server
Figure 9.6 Carrying out a DNS lookup before requesting data from a server

Internet registries are organisations responsible for the allocation of IP

addresses. There are a very small number of registries, each covering a
different geographical area, in order to ensure that address allocation is
strictly controlled.

Now test yourself

31 Name the three main parts of a URL.

32 State three rules describing the format of an IP address.
33 Explain why domain names are used instead of IP addresses.
34 What system is used to convert a domain name to an IP address?
35 Describe the process of accessing a web page, based on entering a URL
36 How is it possible to avoid two domain names having the same IP address?

Answers available online

197

My Revision Notes AQA A-level Computer Science Third Edition

325487_09_MRN_AQA_CS_188-213.indd Page 198 6/29/21 9:52 PM f-0116 /103/HO02206/work/indd

Internet security
The nature of the internet means that data is routed across several other networks Firewall A software or
before it reaches its destination. This means that packets can be intercepted and hardware service that
9 Fundamentals of communication and networking

that any device connected to the internet can also be subject to a potential attack. blocks or allows individual
packets from entering or
Firewalls leaving.
A firewall is used to check and potentially stop packets entering or leaving Stateful inspection
the network. It can be a piece of software on each individual computer or a Inspecting the data
hardware device that acts as a proxy server. contained in a packet.
✚ Stateful inspection refers to examining the contents of a packet in order to Proxy server A device that
decide whether the data itself is suspicious. The firewall keeps a record of sits between a private and a
all current connections in order to identify whether a packet is part of an public network.
ongoing communication.
✚ A proxy server is a physical device that sits between a private network and Packet filtering Stopping
packets based on their IP
a public network. All data to and from the private network goes via the
address or protocol.
proxy server. It can therefore stop packets that have left a computer before
they reach the public network and can prevent packets from the public
network entering the private network, before they get to the computer. Now test yourself
✚ Packet filtering refers to stopping packets based on their destination or
source IP address or their protocol. A firewall can be configured to allow 37 Is a firewall a piece of
web traffic, but to block FTP packets, or can refuse to send or receive hardware or a piece of
packets addressed to or from a suspicious IP address. This can help prevent software?
data being unknowingly sent from the device and can stop a malicious 38 One way in which a
program downloading other malicious files. firewall secures network
traffic is packet filtering.
Proxy servers give each device inside a private network a level of anonymity as
Identify two others.
the public network can only communicate with the proxy server. It also allows
for packet filtering to occur for the entire network, at an institutional level, 39 Explain how packet
regardless of the configuration of each individual device within a network. filtering is used to keep
a device secure.

Encryption Answers available

Encryption is a method of scrambling data so that, if intercepted, it cannot be online
understood.

Making links
Encryption is an important principle that is used extensively in network communications,
but also in non-networking situations as well. The basic principles of encryption, as
well as two examples of symmetric encryption ciphers, are discussed in Encryption in Encryption Scrambling
Chapter 5. data so that it cannot be
understood.
Symmetric encryption refers to encryption where the same key is used to Symmetric encryption The
encrypt and decrypt the data. Symmetric encryption can be secure, however same key is used to encrypt
the need for both the sender and the receiver to have access to the same key and to decrypt a message.
creates a significant security risk.
Key A value used to encrypt
In asymmetric encryption a different key is used to encrypt and to decrypt or decrypt an encrypted
the data. Knowing the key used to encrypt the data does not allow the data to message.
be decrypted. This works as follows:
Asymmetric encryption
✚ The intended recipient of some data generates a public key and a private
Different keys are used to
key that are mathematically related. The public key is made publicly
encrypt and to decrypt a
available, but the private key is not.
message.
✚ The sender uses the public key to encrypt the data which cannot be
decrypted without the private key. Therefore, only the intended recipient Public key A key that
can decrypt the data and they are never required to share the private key. is made public. Has a
matching private key.
A B
Data sent Private key A key that is
Data encrypted Data decrypted kept secret. Has a matching
using B’s public key using B’s private key public key.
198
Figure 9.7 Private/public key encryption

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_09_MRN_AQA_CS_188-213.indd Page 199 6/29/21 9:52 PM f-0116 /103/HO02206/work/indd

This method of encryption is used in almost all online transactions, including

login pages and financial transactions. Removing the need to transmit the
private key (as is the case in symmetric encryption) vastly reduces the risk of
the data being breached.

9 Fundamentals of communication and networking

Digital certificates
One potential flaw in the use of asymmetric encryption is that an impersonator
Digital certificate A digital
could share a public key, claiming it is the public key for an intended recipient.
document indicating that a
Digital certificates are provided by a certification authority and include a serial public key is valid.
number, an expiry data, the name of the organisation, their domain name, Certification authority A
and a copy of the public key. body capable of providing
If a website has a public key but no valid certificate this will trigger a warning digital certificates.
when trying to communicate with that server as the authenticity of the public
key cannot be verified.

Digital signatures
A digital signature uses a checksum to ensure that a message has not been Digital signature A
altered during transmission. method of checking that an
encrypted message has not
The original message is passed through a checksum algorithm to produce
been altered.
a digest. The digest is then encrypted with the sender’s private key which
means that it can be decrypted by anyone with the sender’s public key. Checksum A value,
calculated by an algorithm,
The digest is included at the end of the original message before it is encrypted based on the contents of
with the recipient’s public key and transmitted. the original data.
When the recipient receives the message it is decrypted with the recipient’s Digest A checksum value
private key, and the digest can then be additionally decrypted using the used in a digital signature.
sender’s public key.
The checksum algorithm can be applied to the decrypted message and
compared to the decrypted digest. If the two do not match it is assumed that
the message was altered in some way during transmission.
Sender Recipient
Private key Private key
Making links
Public key Public key Checksum algorithms are
Decrypt with used in a variety of stations
Checksum recipient’s to verify that data has not
Original algorithm private key Original
Digest Encrypted been damaged or altered,
message message message
not just in networking. The
Encrypted
Encrypt with sender’s
digest
topic of error checking,
private key including a more detailed
Separate message and digest
look at checksum
Append to original message Encrypted Decrypt with
Transmit algorithms, is discussed in
digest sender’s
Encrypted public key Chapter 5.
Digest
digest
Original Encrypt with
message Compare Exam tip
recipient’s Checksum
Encrypted public key Encrypted Original algorithm
Digest It is very easy to mix up the
digest message message
terms digital certificate and
Figure 9.8 Private/public key encryption including a digital signature digital signature. Remember
that you cannot award
Now test yourself yourself a certificate, that
must be done by someone
40 Who can issue a digital certificate? else (a certification
41 What is the purpose of using a digital signature? authority). You can sign a
42 In a digital signature, why is the digest encrypted using the sender’s private key? document yourself though,
and this method uses the
43 Identify at least six of the 10 steps involved in using a digital signature.
encrypted checksum/digest.
Answers available online 199

My Revision Notes AQA A-level Computer Science Third Edition

325487_09_MRN_AQA_CS_188-213.indd Page 200 6/29/21 9:52 PM f-0116 /103/HO02206/work/indd

Malware Malware Malicious software.

There are several types of malware that are used to try to cause damage or Virus Malicious code
intercept data. that attaches to another
9 Fundamentals of communication and networking

Some malware will attempt to delete or destroy data, others will capture data program and self-replicates
such as usernames and passwords which can be sent back to the malware’s when executed.
creator. Trojan A malicious program
that pretends to be a useful
Virus program, does not
✚ A virus is a small piece of malicious code which attaches itself to an self-replicate.
existing program. The virus remains inert until the program it is attached
Worm A malicious program
to runs, at which point the malicious code is executed.
that copies itself over a
✚ A virus will typically add a copy of itself to another program and will
network.
often attempt to distribute itself to other devices by automatically sending
emails with an infected attachment or infecting files in shared workspaces. Anti-virus Software
✚ The key feature of a virus is that the host program must be run in order to that scans programs for
execute the malicious code. malicious code.
Bug A mistake in a
Trojans computer program that
✚ A Trojan is a piece of malware that is disguised as a useful program. When causes unexpected results.
the program is run then the malicious code is executed.
Edge-case A problem that
✚ Unlike viruses, Trojans do not attempt to replicate themselves into other
only occurs in an extreme
programs. Trojans are typically spread through user interaction, for
setting.
example downloading a program or opening an email attachment.

Note
A Trojan gets its name from the story of the Trojan horse. A Greek carpenter built a
giant statue of a horse which was then filled with soldiers and left as a gift for the
Trojans who were at war with the Greeks. The Trojans brought the horse into their
fortified city and the soldiers were able to climb out at night and open the gates,
allowing the rest of the army in. This is also why the word Trojan is usually capitalised.

Worms Note
A worm functions in a manner similar to a virus in that it is self-replicating.
However, worms do not need to attach themselves to a file in the way that a The detailed nature of
virus does, and instead spreads over a network. specific techniques to avoid
security flaws is beyond the
Once a worm is active it exploits vulnerabilities in the systems software to scope of this topic, however
replicate and transmit its code to other devices on the network. it is a tremendously
important and a growth
Protective measures industry in the field of
There are many ways to protect a system from malware. computer science and
✚ Firewalls can be used to monitor packets being sent across a network. They software engineering.
can identify unusual or suspicious communication and prevent packets
from being sent or received.
✚ Anti-virus programs scan files and emails for malicious code. It is vital to
keep anti-virus software up to data as new malicious code is always being
created.
✚ Education for users can help prevent people from running suspicious files
or downloading content from suspicious emails.
✚ Code quality is key. Bugs, errors, and edge-cases in software programs can
all be exploited by malicious programmers.

Now test yourself

44 What is meant by malware? 47 Which type of malware can copy itself over a network
45 Name two types of malware that can copy without needing other users to run it?
themselves. 48 Identify four ways of protecting against malware.
200 46 Name one type of malware that does not copy itself.
Answers available online

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_09_MRN_AQA_CS_188-213.indd Page 201 6/29/21 9:52 PM f-0116 /103/HO02206/work/indd

The Transmission Control

Protocol/Internet Protocol

9 Fundamentals of communication and networking

TCP/IP
There are lots of processes involved in sending a data packet around the
Making links
internet. To simplify this, different processes are separated into layers and
each layer follows a set of rules, or protocols, to complete each process. Ports are discussed and
explained in Sockets and
TCP/IP is a fundamental protocol used in network communications and is
ports below.
split into four layers.

TCP/IP layers Transmission Control

Each data packet is processed at each layer in order to add the additional Protocol / Internet
data necessary so that the packet can be directed and checked until it arrives Protocol (TCP/IP) One of
safely at its destination. the main protocols used in
network communications.
At the application layer an individual application will format the data so that
it can be recognised by that application. This is done using a protocol, many Application layer Formats
of which are described in the next section. An example is HTTP – the protocol data according to the
used to transmit websites to web browsers. application to be used.

This formatted data will then be dealt with at the transport layer, which is Transport layer Adds
responsible for two main jobs: a port number, packet
✚ Checking that all packets have arrived and are in the correct order. To number and error detection
data.
achieve this the transport layer adds a packet number to the original data.
✚ Identifying which application layer software should deal with the packets. Port number Extra data
To achieve this the transport layer adds a port number to the packet. The added to a packet that
port number indicates which application layer protocol should be used identifies what application
when the packet is received. layer protocol should be
used to process the data.
The network layer deals with addressing and routing of the data packet,
carrying out three main jobs. Network layer Adds
✚ Adding the sender IP address and recipient IP address. sender and receiver IP
✚ Routing a packet to the next host. addresses.
✚ Adding error checking bits (when sending data) and checking for errors Link layer Deals with the
(when receiving data). physical medium used for
transferring the data.
Exam tip
TCP/IP stack The use of
Always try to be as specific as possible. If asked about the function of the network the TCP/IP layers to add
layer, the response ‘adding an IP address’ is not detailed enough. ‘Adding the sender’s header data which is then
IP address’ is a much more specific answer. processed in reverse order.

The link layer is responsible for the physical transmission of the packet. This
differs between data sent over a wired network and a wireless network, for
example.
Once a packet has been sent from one device to another it is processed in the
reverse order. For this reason it is often referred to as the TCP/IP stack.

Making links
A stack is a LIFO (last in, first out) data structure which is useful in a range of
situations. This, and other data structures, are explored in Component 1 and it is
expected that you will be familiar with using stacks in a programming context. For
more detail please refer to Stacks in Chapter 2.

201

My Revision Notes AQA A-level Computer Science Third Edition

325487_09_MRN_AQA_CS_188-213.indd Page 202 6/23/21 7:37 PM f-0119a /103/HO02206/work/indd

If a packet has been sent to a wireless access point then the receiving device
Note
will remove the data relating to the link layer in order to check the destination
address, before adding new link layer data required to send the packet over a There are several models for
wired part of the network to a router. the TCP/IP stack, including
9 Fundamentals of communication and networking

4-layer, 5-layer and even

The router might then send the packet over the internet to another router
7-layer versions. AQA
which will, again, strip off the link layer data, check the destination address explicitly use the 4-layer
and add new link layer data to send the packet to the next router. model described here.
After arriving at the destination, the link layer data will be stripped off, the IP
address verified at the network layer and the transport layer will then check
the packet for errors, check the port number and attempt to reorganise the
packets in the correct order before passing the data back to the application
Exam tip
layer for processing.
To help remember the four
  layers in the TCP/IP stack
 it can be useful to use a
mnemonic. For example,
  application layer  
All Teachers Need Llamas.
transport layer  If the image of your teacher
network layer  hanging around on their
day off with a pet llama isn’t
link layer  memorable enough you can
always create your own.
Figure 9.9 The layers in the TCP/IP stack

Now test yourself

49 Name the four layers of the TCP/IP stack, in order.

50 Why is the TCP/IP system sometimes referred to as the TCP/IP stack?
51 How can the transport layer identify the purpose of each packet?
52 Identify one task carried out on the network layer.

Answers available online

Sockets and ports

An IP address is sufficient to identify a device on a network. However, more
information is needed to help identify what should happen to a packet once it
reaches its destination.
A port number is added to each packet at the transport layer which
indicates which protocol, and therefore which type of application, should deal
with the data.
Some commonly used ports include:

Port Number(s) Purpose Protocol Exam tip

80 Unencrypted web pages HTTP
Make sure you can
443 Encrypted webpages HTTPS remember the relevant port
25 Receiving mail POP3 numbers associated with
110 Sending email SMTP different protocols.

21 File transfer FTP

22 Remote shell access SSH

When the port number and IP address are both known, this is commonly
written using a colon as a separator; for example:
107.162.140.19:80
202

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_09_MRN_AQA_CS_188-213.indd Page 203 6/29/21 9:52 PM f-0116 /103/HO02206/work/indd

The combination of IP address and port number is described as a socket. The

Socket The endpoint,
socket provides an end-point for the communication.
or final destination, of a
Individual sockets can be opened or closed by a firewall, either allowing or packet. Described using
blocking traffic addressed to that socket. both the IP address and

9 Fundamentals of communication and networking

port number.

MAC addressing Network interface card

Each network interface card (NIC) has a unique Media Access Control (NIC) The component in
any device that physically
(MAC) address. This is sometimes referred to as the physical address as it is
connects to the network.
permanently configured at the hardware level and does not change.
Media Access Control
The link layer uses MAC addressing to identify exactly which device it is
(MAC) address A
sending the data to within that network in order to ensure that the data is
hardware, or physical,
sent to the correct device.
address that does not
change.
Now test yourself

53 What port number is associated with email?

54 What service or protocol is associated with port 22?
55 What is the difference between a packet sent over port 80, and one sent over
port 443?
56 What two pieces of information are needed to make up a socket?
57 Why does the link layer of the TCP/IP stack use a MAC address rather than an
IP address?

Answers available online

Standard application layer protocols

There are several common application layer protocols, each one relating to a
different practical purpose.
It is not necessary to know all of the standards and rules within each protocol,
but it is important to be able to identify the purpose for each.

Protocol Full name Purpose

FTP File Transfer Protocol Transferring files
HTTP Hypertext Transfer Protocol Transferring (unencrypted) web pages
HTTPS Hypertext Transfer Protocol Secure Transferring encrypted web pages
POP3 Post Office Protocol version 3 Transferring email
SMTP Simple Mail Transfer Protocol Transferring email
SSH Secure Shell Remote management

Exam tip
FTP
It is helpful to have had
An FTP server manages access to a store of files and a client using FTP client practical experience using
software can request access to those files. Permissions can be granted on FTP software, although
a file by file and folder by folder basis, allowing the client to potentially questions will not be asked
download and/or upload files remotely. in reference to specific
FTP servers can also be configured to require a login, or to provide examples of FTP programs.
anonymous access.

HTTP/HTTPS
A web server functions in a similar way, storing the resources and layout
information for web pages. A web browser acts as a client, requesting
particular information which is generated and returned by the web server.
203

My Revision Notes AQA A-level Computer Science Third Edition

325487_09_MRN_AQA_CS_188-213.indd Page 204 6/29/21 9:52 PM f-0116 /103/HO02206/work/indd

Once the data has been received the web browser renders the data in the
appropriate format on the screen.
Data that does not need to be secure can be sent using HTTP, which is
unencrypted. Data that does need to be secure is sent over HTTPS, which is
9 Fundamentals of communication and networking

encrypted using private/public key encryption.

SMTP/POP3
An email server receives email messages and stores them securely until the
data is accessed. If an email client program is used, then the email messages
are downloaded to the client software. If a web browser is used then the
emails stay on the server but are transmitted as a web page.
New emails are composed on the client device and sent to the email server
which forwards the messages on to the intended recipient’s email server.
SMTP is the protocol used for sending emails from a device to an email server.
POP3 is a protocol used for receiving emails from the email server.

Note
There are several other
protocols in common usage,
for example FTPS is the
secure equivalent of FTP,
however only those listed
above are required for the
AQA specification.

Figure 9.10 SMTP and POP protocols

SSH
SSH is an encrypted method for remotely managing a computer and is often Now test yourself
used to provide remote access to a server. The client is provided with a
58 What protocol is used to
command line interface which can be used to interact with the server.
connect to a remotely
SSH can be used to access data using other application layer protocols. For manage a computer?
instance, an SSH client can connect on port 80 to carry out web requests using 59 What service is accessed
commands such as GET, or port 25 to carry out email requests. using POP3 and SMTP?
60 What is FTP used for?
Exam tip 61 How can SSH be used to
access data that relies
It is helpful to have had practical experience using SSH to connect to a remote on other protocols?
computer and it is possible that questions will check your knowledge that SSH can
be used to carry out commands using other application layer protocols. However, it Answers available
is not necessary to be familiar with commands specifically relating to remote server online
management.

204

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_09_MRN_AQA_CS_188-213.indd Page 205 6/29/21 9:52 PM f-0116 /103/HO02206/work/indd

IP address structure
An IP address is a numeric addressing method that uses four 8 bit numbers,
Network identifier The
each between 0 and 255, separated by full stops. An IP address is typically
first part of an IP address,

9 Fundamentals of communication and networking

represented in dotted-decimal format, such as 107.162.140.19.
indicating the network.
Each IP address is made up of a network identifier and a host identifier.
Host identifier The second
A typical IP address for a device on a home network is 192.168.0.3. This part of an IP address,
network uses a 16-bit network identifier and a 16-bit host identifier. All indicating the device on that
devices on this network will have the same network identifier; in other words, network.
they will all be in the format 192.168.*.*. A device on the network can then Subnet A subdivision of a
have any IP address between 192.168.0.0 and 192.168.255.255. network.
For our home network example the network identifier is 192.168.0.0 and the
host identifier is 0.0.0.3
Written in binary, the two parts of the IP address can sometimes be easier to spot.

Network identifier Host identifier

1 1 0 0 0 0 0 0 . 1 0 1 0 1 0 0 0 . 0 0 0 0 0 0 0 0 . 0 0 0 0 0 0 1 1
1 1 0 0 0 0 0 0 . 1 0 1 0 1 0 0 0 . 0 0 0 0 0 0 0 0 . 0 0 0 0 0 1 0 1

In order to separate devices on the network this can be further divided into
subnets, which restricts the range of IP addresses for that that particular type
of device.
A software company’s network is going to be split so that there is one subnet
for admin purposes and one subnet for development work.
✚ Devices on the admin network will have an IP address between 192.168.0.0
and 192.168.191.255.
✚ Devices on the development network will have an IP address between
192.168.192.0 and 192.168.255.255.
Therefore, if a device has an IP address of 192.168.192.23, then we know that it
is on the development network.
192.168.192.23:

Network identifier Host identifier

1 1 0 0 0 0 0 0 . 1 0 1 0 1 0 0 0 . 1 1 0 0 0 0 0 0 . 0 0 0 1 0 1 1 1

Subnet masking
Subnet masking is used to identify the network identifier in an IP address. Subnet mask A 32-bit
This is achieved by applying a logical AND between the IP address of any number used to isolate the
device on the network and the subnet mask. network identifier in an IP
address.
For a 192.168.x.x network this can be achieved using the subnet mask 255.255.0.0
on a device on the network, such as one with the address 192.168.0.3 Logical AND The operation
of applying an AND function
11000000.10101000.00000000.00000011 192.168.0.3 between the individual bits
11111111.11111111.00000000.00000000 AND 255.255.0.0 in two binary numbers.
AND
11000000.10101000.00000000.00000000 192.168.0.0 Exam tip

This has the effect of cancelling out, or zeroing, the host identifier. With this The most likely question
example it is fairly straightforward to see how this works using the decimal on this topic is to be asked
representation of the IP address, though it is easier to work directly in the to write out a subnet mask
binary format. given the number of bits
allocated to the network
In the example above the first 16 bits are used for the network identifier and identifier and to calculate the
the last 16 bits are used for the host identifier. This means that there could be maximum number of devices
216 (65,536) possible devices on the network. that can be connected to the 205
network at the same time.

My Revision Notes AQA A-level Computer Science Third Edition

325487_09_MRN_AQA_CS_188-213.indd Page 206 6/29/21 9:52 PM f-0116 /103/HO02206/work/indd

In the following example the first 26 bits are used for the network identifier
Making links
and the last 6 bits are used for the host identifier. This means that there could
be 26 (64) possible devices on the network. For a device on this network with For more on logical AND
the IP address 230.117.156.130 then: see Chapter 6.
9 Fundamentals of communication and networking

11010110.01110101.10011100.10000010 230.117.156.130
11111111.11111111.11111111.11000000 AND 230.117.156.192
AND
11010110.01110101.10011100.10000000 230.117.156.128

Now test yourself

62 Name the two parts of an IP address.

63 If a network uses a 28-bit network identifier, how many devices can exist on that
network?
64 State the subnet mask for an IP address with a 24-bit network identifier.

Answers available online

IP standards
The examples of IP address so far have all referred to IPv4 (version 4).
IPv4 Internet Protocol
IPv4 was the default standard for IP addresses for a large number of years. version 4; uses 32 bits to
However the range of possible IP addresses for a 32 bit number is 232 (4.3 represent each address.
billion), and this has proven to be insufficient as more and more devices are IPv6 Internet Protocol
being connected to the internet. version 6; uses 128 bits to
As a result, IPv6 was introduced. represent each address.

Note
The problem with the limited supply of IPv4 addresses was identified many years ago
and IPv6 was introduced back in 1995; however, the widespread implementation and
rollout has taken several decades and is still not complete.
Many older routers, switches and devices have hardware that is not capable or
configured for IPv6 and will stop working when IPv4 is no longer used. Therefore both
standards are used on all new devices for both backwards and forwards compatibility.

While IPv4 uses 32 bits for each address, IPv6 uses 128 bits and is typically
represented in hexadecimal. This provides over 3.4 × 1038 possible addresses.

Public and private IP addresses

Before the full rollout of IPv6, a working solution for the shortage of IPv4 Public IP address An
addresses is to separate IP addresses into public and private IP addresses. address that can be
Public IP addresses must be unique so that networks connected to the accessed from any device
internet can be addressed. on the internet.

Private IP addresses are only used within networks. These must be unique Private IP address An
within that network, but can be re-used in other networks. address that can only
be accessed within that
There are three main classifications for private IP addresses: network.
✚ The 10 range (10.0.0.0 – 10.255.255.255) with a subnet mask of 255.0.0.0
✚ The 172 range (172.16.0.0 – 172.31.255.255) with a subnet mask of 255.240.0.0
✚ The 192 range (192.168.0.0 – 192.168.255.255) with a subnet mask of
255.255.0.0

206

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_09_MRN_AQA_CS_188-213.indd Page 207 6/29/21 9:52 PM f-0116 /103/HO02206/work/indd

Note
You are not expected to memorise all of the ranges for private IP addresses, though
it is useful to recognise that 192.168.*.* addresses are private as this is the default in

9 Fundamentals of communication and networking

most home networks.

Routable An address that

Public addresses are classed as routable because a device on another network
can be accessed from any
can directly address a device with a public IP address on another network. device on the internet.
Private addresses are classed as non-routable because a device on another Non-routable An address
network cannot directly address a device with a private IP address on another that can only be accessed
network. within that network.

Dynamic Host Configuration Protocol (DHCP)

The assignment of private IP addresses within a network is handled by a
DHCP server. Dynamic host
configuration protocol
Dynamic host configuration protocol (DHCP) is a process which simplifies (DHCP) A system for issuing
the process of configuring a device to connect to a network and avoids the private IP addresses within
accidental duplication of an IP address within a network which can occur if a network.
configuring the addressing manually.
The configuration provided by the DHCP will include an IP address and subnet
mask.
✚ A new device connecting to a network will send a request to discover a
DHCP server. The DHCP server will then offer a configuration to the new
device which can accept or decline the configuration.
✚ Once the client has accepted the configuration, the DHCP server records
this so that it cannot offer the same configuration to another device.
✚ Each DHCP configuration will have an expiry time so that if that device is
disconnected from the network it can retain temporary ownership of that
configuration (and IP address) for the next time it connects as long as that
is within a reasonable period.
On a small network a DHCP process would typically be handled a single
device that functions as a switch, router and DHCP server all in one.

Now test yourself

65 What are the two active versions of the IP standard called?

66 Why is IPv6 necessary?
67 Is the IP address 192.168.10.4 public, or private?
68 Explain why private addresses are non-routable.
69 Describe the process for assigning an IP address connecting to a network for the
first time.

Answers available online

Network Address Translation (NAT)

When a device with a public address sends a request, for example to a
webserver, the response is sent to the sender’s IP address.
When a device with a private address sends a request then this address is Network address
non-routable, and the response cannot be returned directly to that IP address. translation (NAT)
In this situation a router uses network address translation (NAT) to change Substituting a private IP
the sender’s IP address to a public IP address. address for a public IP
✚ For example, a router has a public IP address of 230.117.156.130 and a address, or vice versa.
computer inside the network has a private IP address of 192.168.0.3.
207

My Revision Notes AQA A-level Computer Science Third Edition

325487_09_MRN_AQA_CS_188-213.indd Page 208 6/29/21 9:52 PM f-0116 /103/HO02206/work/indd

✚ The computer inside the network sends a request for a web page. The
router changes the sender’s IP address on the packets to 230.117.156.130 so
that the response will be returned to the router.
✚ When the response is received the router changes the destination IP
9 Fundamentals of communication and networking

address for each packet to 192.168.0.3 and the computer inside the network
receives the packets.

Note
The form of NAT described here works well if only one computer inside the network
needs to communicate, but if two or more computers are requesting web pages at the
same time then the router will not know which device to send the responses to.
In this case dynamic NAT can be used, with the router using a pool of public IP
addresses, issuing a different IP address for each device so that the router knows
which private IP address to send each response to.

Port forwarding
Port forwarding can be used in conjunction with network address translation
Port forwarding Using the
by inspecting port numbers to aid the process.
port number of a packet
When a device with a private IP address sends a request the router will use to identify which private IP
NAT to change the sender’s IP address and can also use the port number in address should be used.
order to help identify which device to send the response to.
If one device has sent a web request on port 80 and another has sent an email
request on port 20 then the router will know which device to send each response
to even though both responses are returned to the same public IP address.

Note
Port forwarding can also be used when multiple devices inside a network want to use
the same port number. A router can be configured so that all messages received on
a given port number, such as 4000, are forwarded to one private IP address, and all
messages received on another port number, such as 4010, are forwarded to another
private IP address. This is useful when hosting a server inside a private network.

Now test yourself

70 Which device is responsible for carrying out network address translation?

71 Why is network address translation required?
72 When a response packet arrives back at the network, how can the intended
private IP address be identified?

Answers available online

Client–server model
Client–server
In the client–server model the server controls access to a centralised resource.
✚ When a client wants to access a resource (such as a file or a web page) then Server A device that
controls access to a
the client sends a request to the server.
centralised resource.
✚ The server processes the request and returns an appropriate response (which
could be the requested resource, a login page, a refusal, or something else). Client A device that makes
requests of a server.
In the case of many websites, rather than simply storing a series of HTML
files, the data that makes up the content of the website is stored in a database.
When a request to view a webpage is sent, the server queries the database
208 in order to generate the exact content on the webpage based on the current

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_09_MRN_AQA_CS_188-213.indd Page 209 6/29/21 9:52 PM f-0116 /103/HO02206/work/indd

values in its database. This means that websites can be automatically updated
Websocket A
to reflect live data (such as view counts, likes, new posts, etc.).
communications protocol
The basic client–server model is relatively passive, with the server only using TCP to create a full-
responding when a request is received from a client. In some situations, duplex connection.

9 Fundamentals of communication and networking

however, a more interactive communication is required. Full-duplex Both devices
Using the Websocket protocol a full-duplex socket connection is opened can send transmissions at
between a web browser and a server, using a TCP connection. This means the same time.
that both client and server have a persistent connection and either device can
communicate at any time during the process.
Note
CRUD and REST
Ideally, the database server
Connecting to a database can be carried out using a variety of methods, and web server should be
including: separate in order to improve
✚ A shell connection (SSH) using SQL commands security; however, it is
✚ A programmed solution using REST possible for both to be run
REST stands for representational state transfer and uses HTTP request from the same server.
methods in order to communicate with the database. This allows program
code, such as JavaScript, to communicate with the database server using the REST Representational
HTTP protocol. state transfer. A software
style that allows web
In the manipulation of databases the acronym CRUD is used to refer to the
services to interact.
main database operations. Using REST, HTTP request methods are used in
their place, as follows. JavaScript A programming
language often used for
web applications.
Database SQL command HTTP request Meaning
operations CRUD A mnemonic for
Create INSERT POST Add data to a database the four main database
operations.
Retrieve SELECT GET Search, or query, a database
Update UPDATE PUT Update or change a record
Delete DELETE DELETE Remove a record Note
Programmers create their
own APIs, and APIs that
Making links
use REST are referred to as
Databases are a critical component to understanding modern data storage and the RESTful.
implementation of a huge number of systems. Chapter 10 explains the principles of
database design as well as the DDL and SQL syntax used to create and work with
database tables. Application
programming interface
(API) A tool that allows
The REST application programming interface (API) runs on the database other programs to make
server. In order to access the commands: calls or requests.
✚ HTML pages that are stored on a web server include JavaScript code. JavaScript object
✚ The HTML pages are accessed through the client’s web browser, which notation (JSON) A file
then executes the JavaScript code on the client side. format for human-readable
✚ The JavaScript code makes calls to the REST API on the web server which text used to transmit data
then processes the request using REST to communicate with the database objects.
server over HTTP and returns the required data from the database to the
web server, and then on to the web browser. Extensible markup
language (XML) A file
Data communication between the server and the client web application is format for human-readable
commonly carried out using one of two formats: JSON (JavaScript Object text used to transmit data
Notation) or XML (Extensible Markup Language). objects.

209

My Revision Notes AQA A-level Computer Science Third Edition

325487_09_MRN_AQA_CS_188-213.indd Page 210 6/29/21 9:52 PM f-0116 /103/HO02206/work/indd

XML (Extensible Markup Language) JSON (JavaScript Object Notation)

–<document> {
–<person id="1"> "ID": 1,
9 Fundamentals of communication and networking

<first-name>Dave</first-name> "Firstname": "Dave",

<last-name>Smith</last-name> "Lastname": "Smith",
<email>dave@smith.com</email> "Email": "dave@smith.com"
</person> },
–<person id="2"> {
<first-name>Nicola</first-name> "ID": 2,
<last-name>Tandy</last-name> "Firstname": “Nicola",
<email>nictandy@hotmail.com</email> "Lastname": "Tandy",
</person> "Email": “nictandy@hotmail.com"
–<person id="3"> },
<first-name>Hermione</first-name> {
<last-name>Belicchi</last-name> "ID": 3,
<email>minniebel@gmail.com</email> "Firstname": "Hermione",
</person> "Lastname": "Belicchi",
</document> "Email": “minniebel@gmail.com"
}

Figure 9.11 (left) An XML document for storing details of users; (right) A JSON
document for storing details of users

JSON is generally preferable to XML as it is:

✚ easier for a human to read
✚ more compact Parse Process a string of
✚ easier to create symbols or letters.
✚ easier and, therefore, quicker for computers to parse (process).

Exam tip
You are not expected to write exact syntax using either JSON or XML, though you
should be able to recognise which is which. Remember that XML looks a lot like HTML
with its use of tags.

Now test yourself

73 What protocol is used to create a full-duplex connection between the client and
the server?
74 State the REST calls for each of the following database commands.
a SELECT
b INSERT
c DELETE
d UPDATE
75 What two formats are typically used to send data between the server and the
web application?
76 Give two advantages for using JSON.

Answers available online

210

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_09_MRN_AQA_CS_188-213.indd Page 211 6/29/21 9:52 PM f-0116 /103/HO02206/work/indd

Thin- versus thick-client computing

Thin-client computing is a model where the client is a low-powered
Thin-client Almost all
computer with limited processing power and limited memory. The majority
processing is done on the

9 Fundamentals of communication and networking

of the processing is carried out by the server and so a high level of system server side.
performance for the client is not needed.
One recent example of a thin client is a Google Chromebook, in which the user Now test yourself
mostly uses a web browser to access software and files stored online. The
Chromebook needs very little computing power as the programs are all run on 77 A Google chrome book
the server and the browser simply needs to render the results to the screen. is an example of what
kind of client?
Many offices and companies make use of desktop thin-client systems as the
78 What kind of client
thin-clients are cheap to produce and require very limited software. They are
would be most suited to
very easy to manage are very easily replaced as the device itself does very
3D rendering for a digital
little other than access the server’s resources. The devices can often be very
animation?
small as well.
79 Give two advantages
Thin clients are generally used where processing requirements are low; for using a thin-client.
for example, for office or productivity software such as word processing, 80 Give two disadvantages
spreadsheets and email. for using a thin-client.
Thick-client computing is a model where the client is at least a reasonably
Answers available
powerful computer with appropriate processing power and memory. A thick
online
client carries out much more of the processing.
A desktop PC or laptop is an example of a thick client. The computer may be
used to access files stored on a server; however, the editing and processing of
Thick-client A significant
these files is carried out on the client itself. amount of processing is
Thick clients are more expensive and require more powerful components, done on the client side.
however they are also more flexible and can be used even if the connection to
the server is not available at that time.
Thick clients are generally used for more powerful tasks such as image
editing, audio editing, video rendering and animation.

Summary

Communication Networking
✚ Serial transmission involves sending one bit at a time ✚ A protocol is a set of rules or standards
down a single wire ✚ A topology describes the physical or logical structure of
✚ Parallel transmission involves sending bits a network
simultaneously down multiple wires ✚ In a star topology all devices are connected to a central
✚ Parallel transmission is hampered by crosstalk and data switch
skew ✚ In a bus topology all devices are connected to a central
✚ Synchronous data transmission requires an additional cable or bus
timing signal to be transmitted ✚ Star topologies require more cabling, but cope better
✚ In asynchronous data transmission, start, stop and with high levels of traffic and are more secure
parity bits are added to provide the receiver with timing ✚ A physical star topology can act as a logical
data and error detection bus network by using a hub which forwards all
✚ Baud rate refers to the number of state changes per transmissions to all devices
second in a transmission ✚ In client–server networking a server controls access to
✚ The bit rate is the maximum number of bits that can be a centralised resource and a client requests access to
transferred per second that resource
✚ The bit rate can be higher than the baud rate if each ✚ In peer-to-peer networking all devices have an equal
signal can represent more than one value status and can act as a client or server in any interaction
✚ The bandwidth describes the range of values that can ✚ Wi-Fi is a protocol to provide a wireless network over a
be sent in a single signal small area using radio frequencies
✚ The bit rate is directly proportional to the bandwidth ✚ A wireless network adapter allows wireless devices to
✚ Latency refers to the time taken for the first part of a connect to a network
transmission to reach its destination ✚ A wireless network access point provides a connection
point to the wired part of the network ➜ 211

My Revision Notes AQA A-level Computer Science Third Edition

325487_09_MRN_AQA_CS_188-213.indd Page 212 6/29/21 9:52 PM f-0116 /103/HO02206/work/indd

✚ Wireless networks require encryption as any device in ✚ A digital certificate is issued by a certification authority in
range can potentially pick up the signals order to ensure that the public key is accurate and genuine
✚ Each wireless network has an SSID, which can be ✚ A digital signature is a digest or checksum that is used
enabled for ease of access or disabled for security to check that an encrypted message has not been
9 Fundamentals of communication and networking

✚ A wireless network can use a MAC address whitelist to altered in any way
block or allow individual devices from accessing that ✚ The digest is encrypted using the sender’s private
network key so that the receiver knows that it is genuine once
✚ CSMA/CA is used by wireless devices to prevent them decrypted using the sender’s public key
from broadcasting at the same time as other devices ✚ Malware is a term meaning malicious software (or
on the network software that causes harm)
✚ RTS/CTS involves sending a request to send and waiting ✚ A virus is a form of malware that attaches malicious
for a clear to send response before transmitting code to an existing program and can self-replicate
when executed
The internet
✚ A Trojan is a form of malware that pretends to be a
✚ The internet is a global connection of millions of helpful program but causes damage when executed. It
networks in a mesh topology is not self-replicating
✚ A router is a device for connecting networks together ✚ A worm is a form of malware that replicates itself across
that use the same protocol a network without attaching itself to another program
✚ A gateway is a device for connecting networks together ✚ Malware can be protected against using firewalls, anti-
that use different protocols virus programs, improved user education and improved
✚ Data is split into packets and additional data is added to code quality
a header before the packet is transmitted
✚ Packet switching means that each packet can be TCP/IP
routed across a different route in order to avoid routes ✚ TCP/IP is a fundamental communication protocol, and
that are slow, busy or have failed is split into four layers
✚ A URL, or uniform resource locator, is a method of ✚ The application layer deals with the formatting of data
addressing an online resource and contains a protocol, specific to an application; for example, HTTP
a domain name and a path ✚ The transport layer adds a packet number and port
✚ An IP address is a numeric address that uniquely number to each packet so that the packets can be
addresses a device on a network reassembled and the completed data passed to the
✚ Domain names are used in web browsers in preference correct application layer protocol
to IP addresses as they are easier to remember, easier ✚ The network layer adds the IP address of both the sender
to type accurately and are more meaningful and the recipient as well as adding error checking bits
✚ A hostname appears before the domain name and ✚ The link layer is used to transmit the packet over a
specifies the host, or device, which is being addressed; physical medium; for example, wired (ethernet) cable,
for example, in mail.awebsite.com and www.awebsite. Wi-Fi, and so on.
com the hostnames are mail and www ✚ The four layers of the TCP/IP stack can be remembered
✚ A fully qualified domain name (FQDN) must include using the mnemonic All Teachers Need Llamas
both a hostname and a domain name ✚ A socket is an endpoint for communication based on
✚ DNS (domain name system) is used to look up domain an IP address and a port number
names to find the corresponding IP address ✚ A MAC address is a unique physical address linked to
✚ Internet registries assign IP addresses to networks in an individual network interface card installed in a device
order to avoid accidental duplication ✚ Well known ports include 80 and 443 for web traffic,
✚ A firewall can be a software or hardware service used 25 and 110 for email, 21 for file transfer and 22 for shell
to stop suspicious transmissions access
✚ Packet filtering involves stopping packets based on ✚ Application layer protocols include FTP, HTTP, HTTPS,
their IP address or port number POP3, SMTP and SSH
✚ A hardware firewall can act as a proxy server if it sits ✚ FTP is used for file transfer
between a private network and a public network by ✚ SSH is used for remote management using a TCP
providing a layer of anonymity connection to a remote port
✚ Stateful inspection involves examining the data content ✚ HTTPS and HTTPS are used to transfer HTML from a
of a packet in order to decide whether it is suspicious web server to a web browser which it renders into text,
✚ Symmetric key encryption means using a cipher in images and other media
which the same key is used to encrypt and decrypt a ✚ POP3 and SMTP are used for email
message ✚ An IP address is split into a network identifier and a
✚ Asymmetric key encryption means that a different key host identifier
must be used to either encrypt or decrypt a message ✚ A subnet mask is used to identify the network identifier
✚ Private/public key encryption uses two mathematically part of an IP address
related keys – one public and one private ✚ IPv4 is used throughout the AQA specification, but IPv6
✚ The sender encrypts the message using the receiver’s is a newer format with a much wider range of possible
public key so that only the receiver can decrypt the addresses
212 message using their private key ➜

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_09_MRN_AQA_CS_188-213.indd Page 213 6/29/21 9:52 PM f-0116 /103/HO02206/work/indd

✚ IP addresses in the 10, 172 and 192 range are only used ✚ The Websocket protocol allows a full-duplex
inside private networks and are non-routable because connection between a web browser and a sever over
they can be duplicated in other networks TCP
✚ Public IP addresses are routable because they are ✚ A web browser can connect to a database using a

9 Fundamentals of communication and networking

unique across the internet RESTful API
✚ DHCP automates the process of assigning private IP ✚ CRUD is an acronym for database operations Create,
addresses in a network Retrieve, Update, Delete
✚ Network address translation replaces a sender’s ✚ Each database operation has an equivalent REST
private IP address with a public IP address so that the function
response can be received over the internet ✚ Data is transferred using JSON or XML
✚ Port forwarding uses the port number in a socket with a ✚ JSON is generally preferred as it is easier to read, easier
public IP address so that the router in a private network to parse, easier to create and is more compact
knows which device to forward the response to ✚ A thin client is a device with little processing power and
✚ In a client–server model, a client (for example, a memory, used with a more powerful server as most
browser on a laptop) sends a request message to a processing is done on the server side
server (for example, a web server), which responds ✚ A thick client is a device with more processing power
with a message back to the client and memory, used in situations where most processing
is done on the client side

Exam practice

1 The number 82 is sent using asynchronous Describe the process by which the web browser
transmission at a rate of 30 megabits per second finds the IP address for the webpage as well as
a) Explain the difference between the baud rate how it retrieves and displays the content of that
and the bit rate of the data transmission. [2] webpage. [5]
b) State the maximum bit rate of the transmission c) The webpage is transmitted using a secure
if the bandwidth is doubled. [1] protocol.
c) Explain why increasing the bandwidth of the Justify the decision to use asymmetric rather than
transmission will not improve the latency. [2] symmetric encryption. [2]
d) Copy and complete the table below to indicate the d) Explain the principles of operation of asymmetric
data that will be transferred, using even parity. [3] encryption and the significance of using a digital
certificate. [5]
Stop Parity Start 4 A router connecting a private network to the internet
bit bit bit uses the public IP address 17.43.0.61 and the private
IP address 10.0.0.1.
The private network contains several workstations, a
2 A network is configured using a physical star topology
file server, a DHCP server and a printer.
a) Explain the operation of a star topology. [2]
a) Why does the router have two IP addresses? [2]
b) Explain how a physical star topology can function
b) Explain how, when a device connects to the
as a logical bus topology. [2]
network, it is possible to ensure that it is given a
c) The network is to be extended to allow Wi-Fi
unique IP address. [2]
connections. Identify two advantages for
c) When an email request is sent to a public email
allowing wireless connections. [2]
server, the workstation sending the request initially
d) Identify two pieces of hardware required to
has a non-routable address.
create a Wi-Fi network. [2]
Explain what this means and identify one concept
e) Describe two methods of ensuring that a
that can be used to overcome this problem. [3]
wireless network can be secured. [4]
d) A user on the network wishes to download a file
f) CSMA/CA with RTS/CTS are used to reduce the
from the file server and a socket is used.
number of collisions on the wireless network.
State what is meant by a socket and identify two
Explain the steps involved in CSMA/CA and
pieces of information that will be required to
RTS/CTS. [5]
connect to the file server. [3]
3 A private network is connected to the internet e) The network is to be subdivided into several
using a router subnets. Each subnet will use a 27-bit network
a) State the name of an alternative device that can be identifier and will include one switch and one
used to connect a private network to the internet, printer.
and the circumstances in which this device would Write the subnet mask that will be needed and
be chosen. [2] state the maximum number of workstations
b) A user enters the URL for a webpage. that can be connected to one subnet. [2]
213

My Revision Notes AQA A-level Computer Science Third Edition

325487_10_MRN_AQA_CS_214-223.indd Page 214 6/29/21 9:51 PM f-0116 /103/HO02206/work/indd

10 Fundamentals of databases

Conceptual models and entity

relationship modelling
Data models are made up of entities and relationships (the links between
them). Entity A type of real-world
✚ An entity is a type of object which we wish to describe. object about which data
✚ Each entity is made up of attributes, or characteristics, of that type of should be stored.
object
Relationship The way in
For example, we might be designing a data model with an entity called which two entities relate to
‘Doctors’. For the data model we are considering, the attributes of the each other.
entity ‘Doctors’ might include their name, ID number and area of medical
Attribute A single
specialism. A different data model might define different attributes – it
characteristic of an entity.
depends on the purpose of the model. Relationships between entities can be
described as: E-R diagram An entity-
✚ one-to-one relationship diagram.
✚ one-to-many
✚ many-to-many.
These relationships can be represented using an entity-relationship diagram
(E-R diagram). In an E-R diagram, the entities (or tables in a database) are
represented as boxes, and the relationships are represented by lines.
For example, patients are registered with a doctor’s surgery. But each doctor’s
surgery has many different patients. This is an example of a one-many
relationship:

Doctor surgery Patient

Figure 10.1 An E-R diagram showing a one-many relationship

The single line connected the left-hand box represents the ‘one’ and the
forked lines on the right represent the ‘many’. The surgery can have many
patients, but each patient only has one surgery.
Within the doctor’s surgery, it is however possible for a patient to see more
than one doctor. This is an example of a many-many relationship:

Doctor Patient

Figure 10.2 An E-R diagram showing a many-many relationship

Many-many relationships are not easily implemented in databases –

instead we wish to turn the many-many relationship into two one-many
relationships. To do this we add a third entity.
In this example, when a doctor sees a patient we can define another entity
called Appointment. This allows us to create the following E-R structure
which is much more straightforward to implement in a database:

Doctor Appointment Patient

Figure 10.3 An E-R diagram showing a three table database using two one-many
214 relationships

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_10_MRN_AQA_CS_214-223.indd Page 215 6/29/21 9:51 PM f-0116 /103/HO02206/work/indd

In the above diagram we can say that:

Exam tip
✚ Each doctor can have many appointments, but each appointment is with
one doctor It is a common mistake to
✚ Each patient can have many appointments, but each appointment is with draw the relationships the

10 Fundamentals of databases
one patient wrong way round. Say to
yourself ‘one doctor can
It is quite common to be asked to draw in the relationships between entities, have many appointments’
and this includes both the direct relationships and the indirect relationships. (or the equivalent for your
In this case we would draw in the many-many relationship between the question) and remember
doctor and the patient. that the ‘many’ symbol
should attach to the ‘many’
entity.
Doctor Appointment

Patient

Figure 10.4 An E-R diagram showing all of the relationships between three entities

Sometimes a four-entity structure is used. In these cases there are up to six

possible relationships: Exam tip
It is normal to be asked
Doctor Appointment to draw in three possible
relationships with 4-table or
5-table examples. In these
questions you should draw
in the three relationships
you are most confident
about and should not draw
Patient Procedure
any extras, as you will
be marked down for any
Figure 10.5 An E-R diagram showing the relationships between four different entities incorrect relationships.

E-R diagrams can also be written in a format that shows the attributes within
each table in brackets, for instance:
Doctor (DoctorID, FirstName, LastName, Specialism)
Patient (PatientID, FirstName, LastName, PhoneNumber,
DateOfBirth)
Appointment (ApptDate, ApptTime, DoctorID, PatientID,
Notes)
The key fields are traditionally shown using an underline. (For an explanation
of key fields, see the next section.)

Now test yourself

1 What two things should be included in an E-R diagram?

2 In an E-R diagram, what is represented by a straight line with no ‘prongs’ at either
end?
3 Draw an E-R diagram to show the structure of a database for a hotel, containing
three tables – room, customer and booking.

Answers available online

215

My Revision Notes AQA A-level Computer Science Third Edition

325487_10_MRN_AQA_CS_214-223.indd Page 216 6/29/21 9:51 PM f-0116 /103/HO02206/work/indd

Relational databases
When databases are created, each entity in a data model is represented in a
Table The representation of
database as a table.
10 Fundamentals of databases

an entity within a database,

Each attribute in a table is represented as a field. that is a structure for storing
Each time we add a new item to a table this is represented as a record. attributes.
A flat file database contains only one table. Attribute A characteristic
A relational database contains two or more tables. This allows the data to be of an object.
stored with minimal duplication of data. Field A single item of data
For example, rather than storing the following data: from a record.

OrderID FirstName LastName Email ItemOrdered Price Record The data about one
item from a table.
1 Dave Smith dave@smith.com Motherboard £89.99
2 Dave Smith dave@smith.com CPU £132.99 Flat file database A
database made up one
3 Dave Smith dave@smith.com RAM £79.99
table.
4 Dave Smith dave@smith.com GPU £219.99
Relational database A
This could be split into two tables: database made up of two or
more, related tables.
Table: Customers
CustomerID FirstName LastName Email
1 Dave Smith dave@smith.com

Table: Orders
OrderID CustomerID ItemOrdered Price
1 1 Motherboard £89.99
2 1 CPU £132.99
3 1 RAM £79.99
4 1 GPU £219.99

This reduces the need for data entry, for data storage and reduces data Primary key An attribute
duplication. It also means that any changes to a customer’s details only need used to uniquely identify a
to be made once, reducing the risk of inconsistencies in the data. record.
✚ Each entity is made up of attributes, or characteristics, of that type of
Composite primary key A
object. Attributes are represented in tables as fields.
primary key made up of two
✚ Each table is populated with records, in which each record describes one
or more attributes.
item in that table.
✚ Each table requires a primary key, which is an attribute with values that Foreign key An attribute
uniquely identify each record in that table. This is important to avoid in one table that links to
ambiguity. There cannot be any duplicate values in a primary key field. the primary key in a related
✚ A primary key can be based on a single attribute or can be composed of table.
several attributes combined. This is known as a composite primary key, or
a composite key for short. Exam tip
✚ A foreign key is an attribute in a table which matches the primary key in a
related table. This creates the link between the two tables. AQA sometimes refer
to database tables as
In the Customers table above, the CustomerID is a primary key.
‘relations’ and sometimes
In the Orders table, OrderID is a primary key and the CustomerID is a as ‘entities’. Be careful
foreign key. not to confuse the terms
‘relation’ and ‘relationship’.
Now test yourself

4 Why is it preferable for to use a relational database of a flat-file database?

5 Why is it necessary to have a primary key?
6 What term is used for a primary key made up of more than one attribute?
7 What term is used when a primary key is used as a field in another table in order to
create a relationship between those entities?
216
Answers available online

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_10_MRN_AQA_CS_214-223.indd Page 217 6/29/21 9:51 PM f-0116 /103/HO02206/work/indd

Database design and

normalisation techniques

10 Fundamentals of databases
The rules around making sure that a database design is efficient are referred Note
to as normalisation.
There are several
✚ Normalisation ensures that there is less redundancy and less
normalisation rules, and
inconsistency in the data.
the AQA specification
✚ A poorly designed database might result in data being stored twice because
considers these up to and
the same attributes exist in different tables, or because different records
including the third normal
include the same data (for example, storing the address for a customer
form. It is not necessary
every time they make a purchase). to know which rule refers
✚ A poorly designed database might result in erroneous data because to which specific level of
duplicate data has been incorrectly updated or deleted (for example, if normalisation.
the address for a customer is stored every time they make a purchase but
needs to be changed then it is possible to miss one or more records with
the old address). Normalisation Structuring
✚ A fully normalised database is atomic. Each table, or group or attributes, a database in a way
cannot be broken down any further and so there are no repeating groups designed to reduce data
of attributes. redundancy and improve
integrity.
For example:
✚a ‘name’ attribute could be broken down into ‘first name’ and ‘last Redundancy Not needed,
name’ for example data that is
✚ an address attribute could be broken down into ‘house number’, ‘street’,
unnecessarily stored twice.
‘town’, ‘county’ and ‘postcode’. Inconsistency The result is
✚ A fully normalised database has no partial dependencies. A partial not always the same.
dependency is when an attribute depends on part of a composite key but
Third normal form A
not the whole composite key (note, this point is only relevant to tables with
database in which all data is
composite keys). atomic, there are no partial
For example, a table for an appointment with a doctor uses a composite dependencies and no non-
key: key dependencies.
appointment(date, time, doctorName, patientID, notes,
doctorSpecialism) Exam tip
The doctorSpecialism attribute depends on the doctorName, but not on
If you are asked what it
the rest of the attributes in the composite primary key.
means for a database to be
The solution is to create a separate doctor entity: fully normalised remember
doctor(doctorID, doctorName, doctorSpecialism) that data must be atomic,
and that all of the attributes
appointment(date, time, doctorID, patientID, notes)
in the table depend on
✚ A fully normalised database has no non-key dependencies. The value of the whole primary key
each attribute must only depend on the key. If the value of an attribute can (including all parts of
be determined using any other attribute, then it is not independent. If this a composite key, if the
is the case, then the table normally needs splitting into two further linked primary key is composite)
tables. and no other attributes.
For example, an address entity is designed as follows:
address(addressID, houseNumber, street, town, county, Now test yourself
postcode) 8 What are two
Two neighbours have a different house number, but the same street, town, advantages of
county and postcode. The attributes for street, town and county can be normalising a database?
determined from the postcode as well as the primary key addressID, as 9 What does the term
everyone on that street has the same postcode. atomic mean?

To resolve this, it can then be split into two tables: 10 What are two other
requirements for a fully
address(addressID, houseNumber, postcode) normalised database?
postcode(postcode, street, town, county) Answers available
online 217

My Revision Notes AQA A-level Computer Science Third Edition

325487_10_MRN_AQA_CS_214-223.indd Page 218 6/29/21 9:51 PM f-0116 /103/HO02206/work/indd

Structured Query Language (SQL)

It is common to be asked to write SQL code during the Component 2 exam.

Database definition language (DDL)

10 Fundamentals of databases

SQL can be used to

✚ create tables Database definition
✚ manipulate tables. language (DDL) The
The code used to create tables is known as database definition language (DDL) language used to create
database tables. A subset of
and an example in SQL is as follows:
SQL commands can be used
CREATE TABLE Purchase ( as a database definition
language.
PurchaseID INT,
CustomerID INT,
StoreID VARCHAR(6),
Date DATETIME,
TotalCost CURRENCY,
LoyaltyCard BOOL,
PRIMARY KEY (PurchaseID),
FOREIGN KEY (CustomerID) REFERENCES Customer(CustomerID))

Figure 10.6 The DDL code to create a database table

In the example above it is key to note that:

Note
✚ each field is declared using its name, followed by its data type
✚ SQL commands (in other words, anything not referring to the name of a It is possible to declare a
table, name of an field or a value) are typically written in all caps, although field and indicate it is a
this is convention and not a requirement of the syntax primary key in one line; for
✚ the primary key is declared using the command PRIMARY KEY followed by example:
the name of the field in parentheses CustomerID INT
✚ the foreign key is declared in a similar way to a primary key, followed by PRIMARY KEY
a reference to the corresponding field in the related table; the keys do not
need to have the same name in different tables, though it is often easier to Both methods are
follow the structure of the database if they do. acceptable.

There are many possible data types that can be used in SQL. Any valid type
will be accepted but, for numeric types it is important to consider whether an
integer or non-integer type should be used.
Example data Appropriate datatypes Notes Now test yourself
fixed length text CHAR(10) Each entry must be exactly the
11 Give two SQL data
length specified.
types that could be used
variable length text VARCHAR(10) Each entry can be any length up to to store a fractional
the specified length. number.
whole number INT Should only be used for numbers 12 Suggest three SQL data
that can never be fractional. types suitable for a date
fractional number FLOAT, REAL, FLOAT and REAL are both used for and time.
CURRENCY fractional numbers – they just have 13 Give one way of
different levels of precision. declaring a customer ID
CURRENCY should only be used for as a primary key in SQL.
financial data.
Answers available
date or time DATE, TIME, DATETIME DATETIME stores both a date and a
online
time.
yes or no BOOLEAN, TINYINT A Boolean is ultimately stored as a 0
or 1 and in some database systems
will be automatically converted to a
TINYINT (an integer type that is one
218 byte in size).

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_10_MRN_AQA_CS_214-223.indd Page 219 6/29/21 9:51 PM f-0116 /103/HO02206/work/indd

Retrieve (SELECT)
To retrieve data from a query, we use three SQL commands:
SELECT <fields>

10 Fundamentals of databases
FROM <table>
WHERE <condition>
The first two are mandatory but WHERE is optional. These can also be written
in one line:
SELECT ProductName, Price FROM Products WHERE Price < 100
An asterisk (*) can be used as a wildcard in order to select all fields:
SELECT * FROM Products WHERE Price < 100
It is possible to use logical operations such as AND, OR and NOT to further
refine queries:
SELECT * FROM Products WHERE Price < 100 AND Type = "Canned"
The results from queries can be further improved by sorting the results using
the ORDER BY command:
SELECT * FROM Products WHERE Price < 100 ORDER BY Price ASC
SELECT * FROM Products WHERE Price < 100 ORDER BY Price DESC

Complex queries Note

Complex queries are used to retrieve data from two or more tables. In order
If you are tackling a
to complete this it is necessary to state the primary and foreign key fields, to
database project as part
avoid duplicate responses.
of your NEA programming
There are several possible formats for achieving this, however the most project, the use of
accessible format is used here. Valid answers using other formats (such as composite queries is
JOIN and UNION) will be accepted by AQA. good evidence towards a
higher level of technical
The example below is based on the following database structure: complexity. This refers to
Doctor (DoctorID, FirstName, LastName, Specialism) additional commands such
as ORDER BY, SUM, MAX,
Patient (PatientID, FirstName, LastName, PhoneNumber, DateOfBirth) MIN, and AVG applied to
Appointment (ApptDate, ApptTime, DoctorID, PatientID, Notes) the results of a query.

SELECT Patient.FirstName, Patient.LastName, PhoneNumber,

ApptTime
FROM Doctor, Patient, Appointment
WHERE Date = "2023-03-26”
AND Appointment.DoctorID = Doctor.DoctorID
AND Appointment.PatientID = Patient.PatientID
ORDER BY Time ASC
The query above would return the first name, last name and phone number of
each patient as well as the time of the appointment for all appointments on 26
March 2023. This might be needed if someone needed to send a text message
reminder to all patients for that day.
Points of note:
✚ There are two FirstName fields in the database, which is why the table
name has been included as a prefix. This can be added to every field but
is not necessary where only one field exists; for example, there is only one
PhoneNumber field so we do not need to state the table it is in.

219

My Revision Notes AQA A-level Computer Science Third Edition

325487_10_MRN_AQA_CS_214-223.indd Page 220 6/29/21 9:51 PM f-0116 /103/HO02206/work/indd

✚ It is important to add all tables used in the FROM field, but also important
Exam tip
to exclude any tables in the database not relevant to this query.
✚ In this case there are two one-many links between the tables, and so When writing complex
the foreign keys and their corresponding primary keys (DoctorID and queries, always work
10 Fundamentals of databases

PatientID) must be referenced in the query. through the items in the

same order. Identify the
fields, identify the tables
Now test yourself
they come from and identify
14 What are the three main clauses of a select query? the main condition(s).
Then add in the foreign
15 What symbol is used to select all fields in a table?
key constraints using AND
16 What clause is used to sort the results into order? clauses.
17 Write a query that would select all the fields from the Sales table where the Value
of each sale was more than £500.
18 What additional clauses need to be added for a complex query?

Answers available online

Create (INSERT)
To insert data into a database it is easiest to insert an entire record in one go.
In order to do this, the data must be presented in the same order as the fields
appear in the table:
INSERT INTO Appointment VALUES
("2023-03-27","13:50",113,26850,"Routine checkup")
It is possible to enter data by specifying the fields in a different order. It is also Exam tips
possible to exclude some fields altogether, if they are not required:
Don’t forget to include the
INSERT INTO Appointment(DoctorID, PatientID, ApptTime,
key word VALUES when
ApptDate)
inserting data, so that the
VALUES (127,39254,"09:30","2023-04-03") system knows you are
referring to data values and
It is generally easiest to use the first format in exam questions.
not field names.
Note that dates and times are always entered using speech marks. Always remember to
✚ Dates are formatted “YYYY-MM-DD”; for example, "2023-01-17". include quotation marks
✚ Times are formatted “HH:MM:SS” or “HH:MM”; for example, "09:17:02". around date and time values
✚ The datetime format uses both, with the date first and a space before the on database questions,
time; for example, "2023-01-17 09:17:02". but never around Boolean
Note that Boolean values can be entered as TRUE, FALSE, 1, or 0. values.

Boolean values must NEVER be written in quotation marks.

Now test yourself
Update (UPDATE)
19 What is the first
Sometimes it is necessary to change a value already in the database. This
word used in an SQL
might be a single instance, or for all instances.
command to add data to
The structure of an update query should always read: a table?
UPDATE <table> 20 What word must be
used just before the
SET <field> = <value> brackets, and is easily
WHERE <condition> forgotten?
21 What is significant about
For example, the doctor with DoctorID 127 is ill and a replacement doctor
the order in which the
has been brought in for the day, with DoctorID 180.
values must be written?
UPDATE Appointment 22 What format would be
SET DoctorID = 180 used to write the date
12 November 2022?
WHERE DoctorID = 127
AND ApptDate = "2023-04-02" Answers available
online
220 Whilst there are two conditions here, note that only one condition may be
needed.

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_10_MRN_AQA_CS_214-223.indd Page 221 6/29/21 9:51 PM f-0116 /103/HO02206/work/indd

Delete (DELETE) Now test yourself

A delete query is used to remove data from the database.
23 What are the three
This command can be used to remove a single record, a number of records, or main clauses in an

10 Fundamentals of databases
all records from a table. update query?
DELETE FROM Doctor 24 What are the two main
clauses in a delete
WHERE DoctorID = 128 query?
To delete all records, no WHERE clause is needed. 25 How can you delete all
data from a table?
Though presented in a different order here, the four main types of database
manipulation can be remembered using the acronym CRUD. This stands for: Answers available
✚ create ✚ update online
✚ retrieve ✚ delete.

Making links
Databases are often run on servers, with clients accessing databases remotely. In
order to achieve this over a web interface, a RESTful API must be used which maps
HTTP requests to the equivalent SQL command. More detail on RESTful APIs and web-
based database access are discussed in Chapter 9.

Database manipulation SQL implementation

Create INSERT INTO <table>
VALUES (…,…,…)
Retrieve SELECT <fields>
FROM <tables>
WHERE <condition>
AND <condition>
SORT BY <field> DESC
Update UPDATE <table>
SET <field> = <value>
WHERE <condition>
Delete DELETE FROM <table>
WHERE <condition>

Revision activity
Using the W3Schools website you can create your own database tables using SQL
syntax.
Go to www.w3schools.com, scroll down to Learn SQL, and then click the Try It Yourself
button.
1 Use the command DROP TABLE <tablename> to delete the existing tables.
2 Choose a past paper question and create the tables using SQL code.
3 Make up some suitable data and add this to the database using an INSERT query.
4 Create SELECT queries, including complex queries, according to the questions on
the paper.
5 Create UPDATE and DELETE queries to alter the data in the database, running
SELECT queries after each one to check the results.

221

My Revision Notes AQA A-level Computer Science Third Edition

325487_10_MRN_AQA_CS_214-223.indd Page 222 6/29/21 9:51 PM f-0116 /103/HO02206/work/indd

Client–server databases
In a client–server database system it is possible for two clients to attempt to
Concurrent At the same
access, or alter, the data in a database at the same time. This is referred to as
10 Fundamentals of databases

time.
concurrent access.
If two or more clients attempt to alter data concurrently then it is possible
that the integrity of the database will be compromised.
For example, a bank balance is currently £1000. Now test yourself
✚ A payment of £10 is made into the account and so the balance is checked,
£10 added to the figure and the new balance of £1010 is written to the 26 What does concurrent
account. The write process was slightly slow, however. mean?
✚ At the same time, another transaction took place in which the account 27 Why is concurrent
holder spent £800. This processing was slightly quicker and the balance access a potential
was reduced to £200; however, the first transaction had not yet completed problem?
and, when it did, the balance was overwritten as £1010. 28 How can record locks
There are several possible methods to preserve the integrity of the data. prevent problems with
✚ Record locks: When a record is initially accessed it is locked and cannot concurrent access?
be accessed again until the lock has been removed. In the case above the 29 Identify two other
record for that customer would be locked by the first transaction until it methods of preserving
was complete. Only then could the second transaction be processed. data integrity against
✚ Serialisation: The database system manages the transactions so that only concurrent access
one can be carried out at a time. As both transactions were received at the issues.
same time the second transaction would be held in a queue momentarily Answers available
until the first transaction had completed. online
✚ Timestamp ordering: Each time access to the database is made, the
timestamp is recorded. In the above example, the second transaction
would change the timestamp at which the balance was last changed.
When the first transaction attempt to complete it would see that the
timestamp had changed and so would start again.
✚ Commitment ordering: Transactions are arranged into an order that avoids
potential conflicts, using an algorithm within the database system.

Summary

Conceptual data modelling and E-R modelling ✚ In a database normalised to the third normal form,
✚ Entity-Relationship diagrams show the relationships data should be atomic, data should have no partial-
between entities as: one-to-one, one-to-many or key dependencies and should have no non-key
many-to-many dependencies
✚ Entities contain attributes which each describe a SQL
characteristic of that entity
✚ SQL commands are used to create and manipulate
✚ Underlining can be used to identify the key fields
database tables
Relational databases ✚ Tables are created using the command CREATE
✚ Relational databases are designed to represent data TABLE ( … )
using multiple related entities ✚ Fields are declared by stating their name, followed by
✚ Data is stored in records, where each record describes their data type
one object of that type ✚ Text fields are declared using the data types CHAR(n)
✚ A primary key is an attribute which must have a unique or VARCHAR(n), where n refers to the fixed (for
value in order to identify a record CHAR) or maximum (for VARCHAR) length of the text
✚ A composite primary key is a primary key made up ✚ Dates and times are declared using the data types
of two or more attributes which must have a unique DATE, TIME, or DATETIME
combination ✚ Primary keys are declared using the command
✚ A foreign key is used to reference a primary key from PRIMARY KEY (<fieldname>)
another table in order to create a relationship between ✚ Foreign keys are declared using the command
those entities FOREIGN KEY (<foreignkey>) REFERENCES
<table>(<primarykey>)
Database design and normalisation ✚ The four main database manipulation commands can
✚ Normalisation is the technique of ensuring that a be remembered using the acronym CRUD
222 database design reduces data redundancy and ✚ Creating data is done using an INSERT query:
improves data integrity INSERT INTO <table> VALUES (…,…,…) ➜
Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads
325487_10_MRN_AQA_CS_214-223.indd Page 223 6/29/21 9:51 PM f-0116 /103/HO02206/work/indd

✚ Retrieving data is done using a SELECT query: ✚ Data can be sorted once retrieved by adding an extra
SELECT <fields> FROM <table> WHERE clause SORT BY <fieldname> ASC or SORT BY
<condition> <fieldname> DESC
✚ Updating data is done using an UPDATE query: ✚ Dates and times are written using the format "YYYY-

10 Fundamentals of databases
UPDATE <table> SET <field> = <value> MM-DD HH:MM:SS"
WHERE <condition> ✚ Boolean values are written as either TRUE, FALSE, 1,
✚ Deleting data is done using a DELETE query: DELETE or 0
FROM <table> WHERE <condition>
Client server databases
✚ Complex queries can retrieve data from more
than one table by including additional conditions ✚ In a client–server database system it is possible for
linking the primary and foreign keys; for two or more clients to attempt to access a record
example, AND <table>.<foreignkey> = concurrently, threatening the integrity of the database
<table>.<primarykey> ✚ This can be avoided using record locks, serialisation,
timestamp ordering or commitment ordering.

Exam practice

1 A community cinema with a single screen wishes to c) Design the remaining relation, underlining the
use a database to store the details of film showings attribute(s) for the primary key. [2]
and ticket bookings. d) The database should be fully normalised. Define
✚ Each film is identified with a unique number and what it means for a database to be fully
the film’s title, certificate and running length (in normalised. [2]
minutes) are recorded. e) Write the SQL code needed to create the Film
✚ Each showing is identified with a combination of the table. [3]
film’s ID, date of showing, and time of showing. f) Write the SQL code needed to add a new customer
✚ Each customer is identified with a unique code
named Nicola Tandy. Their email address is
made up from the first three letters of their first nicolatandy@hotmail.com and they are not
name and the first five letters of their surname. The currently a member of the loyalty scheme. [2]
customer’s first name, last name and email address
g) A number of showings have had the wrong start
are stored, as well as whether or not they are a
time recorded incorrectly. Write the SQL code
member of the cinema’s loyalty scheme.
needed to change all films scheduled to start
✚ Each ticket is identified with a unique number and
before 9am so that they will now start at 9:30am. [3]
the seat number is stored.
h) The cinema has cancelled a Halloween special.
a) The entity-relationship diagram below is
Write the SQL code needed to delete all showings
incomplete.
for the 30 October 2023. [2]
Draw on three additional relationships. [3]
i) Write an SQL query to show a list of all the names
and email addresses for loyalty scheme members.
Film Showing [3]
j) Write an SQL query to show the title and start
times for all 18 certificate films on the 23 May
2023. [5]
k) The owners of the community cinema are
considering splitting the building into several
screens, each with a different number of seats.
Customer Ticket
Describe the changes that would need to be made
to the database in order to accommodate this,
b) The Film, Showing and Customer relations ensuring that tickets could only be booked for
have already been defined. seats that exist, and are available. You do not
Film(FilmID, Title, Certificate, Length) need to provide any SQL code. [4]
Showing(FilmID, Date, StartTime) l) Customers can book tickets in the cinema, and
Customer(CustomerID, FirstName, LastName, also online, allowing concurrent access to occur.
Email, LoyaltyMember) Describe one problem that could occur. [2]
State the type of primary key used in the m) Identify two possible methods for avoiding
Showing table. [1] issues caused by concurrent access. [2]

223

My Revision Notes AQA A-level Computer Science Third Edition

325487_11_MRN_AQA_CS_224-226.indd Page 224 6/29/21 10:13 PM f-0116 /103/HO02206/work/indd

11 Big Data

Big data
Defining Big Data
Modern computer systems can be required to collect and process a huge
amount of data that lacks structure. This is referred to as Big Data. Big Data Data that cannot
easily be stored, retrieved,
Simply having a lot of data doesn’t necessarily mean that this is Big Data. and processed because it
There are three characteristics that are used to define Big Data: volume, is too varied, too large or
velocity and variety. acquired too rapidly.
Volume The quantity of
Volume data to be stored, typically
This refers to an amount of data so large that it cannot fit into a single server. too much to fit on a single
server.
Such a large volume of data must be processed across multiple servers
simultaneously. This has an impact because relational databases do not work Velocity The rate at which
well across multiple devices. Therefore, even structured data becomes hard to data is generated or collected.
analyse and process if there is a lot of it. Variety Data in many
forms; that is, unstructured.
Velocity
This refers to the rate at which data is generated or collected – for example Exam tip
social media platforms receive hundreds of millions of digital images per day. If you are asked what
A high velocity of data means that one server would be unable to cope with constitutes Big Data,
processing the data, particularly if the processing relies on data currently remember the three Vs of
stored on another server. volume, velocity and variety,
but ensure that you explain
your answers. Simply
Variety recalling the word will
This refers to the unstructured format of data and the different forms of data not be sufficient to show
being received or captured, including images, video, audio, text and numeric understanding.
data.
The variety of data means that it will not fit into a row-and-column format Now test yourself
that is required for processing in a traditional database table.
1 What are the three Vs of
These three issues combined make it difficult to store data, retrieve data and Big Data?
process data on demand. 2 Why is it a problem if
data is captured by the
Machine learning techniques are often used to process Big Data, as they allow
system in a wide range
the system to identify patterns in the data which make it easier to extract
of forms?
meaningful information.
3 Suggest one source of
Typical sources of Big Data include: Big Data.
✚ systems with networked sensors
✚ video surveillance systems Answers available
✚ social media platforms. online

Processing Big Data

When there is so much data that it cannot fit on one server, the data storage
and data processing must be distributed across multiple devices. This is Distributed processing
known as distributed processing. A system that uses multiple
processors or servers to
process data.
224

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_11_MRN_AQA_CS_224-226.indd Page 225 6/29/21 10:13 PM f-0116 /103/HO02206/work/indd

Functional programming is an effective solution for distributed processing

Functional programming
as it is easier to write correct code that can be distributed to run across more
A programming paradigm
than one server. based around using
Functional programming supports three main features which help with functions to process data.

11 Big Data
distributed programming: immutable data structures, statelessness, and Immutable data
higher-order functions. structures Data structures
✚ Immutable data structures: Rather than changing data and re-assigning a that cannot be changed.
new value to a variable, data is not changed in functional programming. Statelessness A system in
Instead, the data is processed by a series of functions, and the result is which the processing does
returned, leaving the original data unaltered. not depend on the state of
✚ Statelessness: The functions in a functional program don’t change another part of the program.
depending on another factor elsewhere in the program. This means High-order function A
that the state of another part of the system will not impact the function that can take another
process being applied. This is important for distributed processing as function as a parameter.
it may not be possible to check the states of other parts of
the system. Now test yourself
✚ Higher-order functions: Higher-order functions can take another function 4 What is meant by
as an argument. This means that functions can be chained together to distributed processing?
produce a result rather than having to store the intermediate values in 5 What type of
additional variables. programming is typically
used in distributed
processing?
Making links
6 What are the three
Functional programming is a different paradigm, or style of programming, to main features of that
procedural and to object-oriented programming. Your ability to understand and to type of programming?
write functional programs is assessed in Component 2 and this topic is explored in
much more detail in Chapter 12. Answers available
online

Representing Big Data

Big Data can be represented by a fact-based model, in which each fact
captures a single piece of information. Graph schema A graphical
A fact-based model is commonly drawn using a graph schema. tool for representing fact-
A graph schema uses nodes and edges to describe facts, and the relationships based models.
between facts. Node An item in a graph.
✚ Oval nodes are used to represent objects.
Edge A connection
✚ Rectangle nodes are used to represent individual items of data.
between nodes in a graph.
✚ Solid edges are used to represent relationships between objects.
✚ Dashed edges are used to represent links between objects and their data.
✚ Edges can be directed.

Tutor Group: Tutor Group:

12AP 12GG

Student: Student:
Dave Smith Nicola Tandy
Teaches Teaches

Teacher:
Siblings
Andrea Pla

Student:
Freya Smith Teaching Room:
CP02

Tutor Group:
12AP

Figure 11.1 A graph schema with nodes (the shapes) and edges (the lines) showing 225
facts about staff and students at a college

My Revision Notes AQA A-level Computer Science Third Edition

325487_11_MRN_AQA_CS_224-226.indd Page 226 6/29/21 10:13 PM f-0116 /103/HO02206/work/indd

In this example:
Now test yourself
✚ the teacher and the students are objects and so they are shown in ovals,
✚ each object has some data that relates to that person, which is shown in 7 In a graph schema, what
rectangles. is represented by a
11 Big Data

✚ dashed edges are used to connect the data to the person (for example, rectangular node?
Freya is in tutor group 12AP) 8 What is represented by
✚ solid edges are used to connect the objects (in this case, people). an oval node?
✚ sometimes, no arrow is shown (for example, Freya and Dave are siblings) 9 What kind of edge
because this is not a one-way relationship. should connect an item
✚ sometimes, an arrow is shown because Andrea teaches Dave and Nicola of data to an object?
(they don’t teach Andrea)
10 Using the figure above,
✚ for all solid edges, a label is added to describe the relationship.
state three things you
know about Dave Smith.
Summary
Answers available
✚ Big Data refers to data that does not fit into standard data structures online
✚ Big Data can be described as data that has one or all of the following
characteristics: volume, velocity, and variety
✚ Volume: data can be classified as Big Data if the volume is so big that it will not fit
on one server
✚ Velocity: data can be classified as Big Data if it is being generated or captured at an
extremely high rate
✚ Variety: data can be classified as Big Data if it is unstructured, and hence cannot
be represented using the table structures within relational databases
✚ Machine learning techniques are often used to identify patterns and to extract
meaningful information from Big Data
✚ Distributed processing is often required in order to process the large quantities of
data across multiple servers
✚ Functional programming is often used as the programming paradigm of choice for
processing Big Data because it lends itself to distributed processing
✚ Functional programming supports immutable data structures, statelessness and
high-order functions
✚ Big Data can be represented using a fact-based model
✚ A graph schema is a graphical representation of a fact-based model
✚ Objects are represented using an oval node
✚ Items of data are represented using a rectangular node
✚ Connections between objects are represented using a solid edge, which can be
directional
✚ Connections between data and objects are represented using a dashed edge

Exam practice

1 Describe three characteristics that might result Copy and complete the graph schema, adding the
in a set of data being classified as Big Data. [3] following information. [4]
2 Explain why functional programming is a suitable ✚ Andy Frey is 32 years old.
method to use for processing Big Data. [2] ✚ Rahal Meyrick treated a cat called Shadow. The cat
is black.
3 This is a graph schema for a fact-based model:
✚ Both vets are employed by a vet surgery called
Soucek’s Pets, based in Crewe.
Age:
27
✚ The email address for the vet surgery is info@
soucekspets.co.uk.

Vet: Vet:
Andy Frey Rahal Meyrick
Treated

Patient: Patient:
Colin Shadow

Species: Breed: Species:

Dog Schnauzer Cat
226

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_12_MRN_AQA_CS_227-234.indd Page 227 6/29/21 10:12 PM f-0116 /103/HO02206/work/indd

12 Functional programming

Functional programming
paradigm
Function type
A function is a rule that takes a value from a set of inputs (A) and assigns it to
an outputs that is contained within set B: Function A rule that takes
one or more values as an
For example, a function f that takes a positive integer as an input, and
input and produces an
produces an output which is the double of each input, can be written as
output.
follows:
Set A collection of objects.
f: {0,1,2,3,4,…} → {0,2,4,6,8,…}
Domain A set from which
The domain is the set that contains all of the input values.
all input values are chosen.
The co-domain is the set that contains all of the output values.
Co-domain A set from
It is not necessary for all values in the co-domain to be included in the output. which all output values are
For instance, in this example the co-domain does not contain the numbers 1, chosen.
5 and 9. However, all of the values in the co-domain are natural numbers.
The domain and co-domain are always subsets of objects of a given data
type (for example, integers, rational numbers, characters, and so on). In the
function above, the domain is the set of natural numbers ().
The co-domain is also the set of natural numbers (), although not every
value from that set will be included.
A function f has a function type:
f: A → B Making links
A is the argument type and B is the result type. They refer to the data types When describing the
from which the domain and co-domain are taken. For instance, for the domain and co-domain it is
example above where the domain and co-domains are taken from the set of important to be familiar with
natural numbers (integers), we can say: the main sets of , ,  and
f:  →  , as well as the principles
of subsets. The sets are
or described in Chapter 5. The
f: integer → integer topic of subsets and their
mathematical descriptions
However, this notation only describes the data type of the domain and co-
are discussed in Chapter 4.
domain and does not indicate the purpose or effect of the function.

First-class object
A first-class object is an object that can:
✚ appear in expressions First-class object An
✚ be assigned to a variable object that can be assigned
✚ be assigned as an argument as an argument or returned
✚ be returned from a function. from a function.

In most programming languages a numeric value is a first-class object.

FUNCTION AddOne(StartNum)
EndNum = StartNum + 1
Return EndNum
ENDFUNCTION 227

My Revision Notes AQA A-level Computer Science Third Edition

325487_12_MRN_AQA_CS_227-234.indd Page 228 6/29/21 10:12 PM f-0116 /103/HO02206/work/indd

Here, the value of StartNum has been passed as an argument and used in an
expression. The value resulting from that expression has been assigned to a
variable and returned from the function.
In functional programming, a function is also a first-class object. This means
12 Functional programming

that functions can be passed as arguments returned from other functions.

This does not mean that the function is called, and the result is passed, but
that the function itself is passed in its entirety.
A higher-order function is a function that takes a function as an argument, or
Higher-order function
returns a function, or both.
A function that takes a
Now test yourself function as an argument, or
returns a function, or both.
1 A function takes a whole number as an input and returns half of its value.
a What is the domain of the function?
b What is the co-domain of the function?
2 A function takes a whole number as an input and returns the square of its value.
Copy and complete the description of the function:
f: {1,2,3,4,…} → { …}
3 What are the four features of a first-class object?
4 What are the three possible features of a higher-order function?

Answers available online

Function application
Function application is the process of applying a function to its arguments.
For example, a function to add two values takes two integer arguments as an
input and produces a single integer result as an output.
add(6,2) → 8
The domain of the function is described as  × , which is the Cartesian
product of  and . This means that the input contains two values, both of
which are integers.

Note
The Cartesian product of two sets contains ordered pairs of the values from both sets.
For example:
{0,1,2,…} × {0,1,2,…} = {(0,0),(0,1),(0,2),(1,0),(1,1),(1,2),(2,0),
(2,1),(2,2),…}
This means that the function call must be either add(0,0) or add(0,1) or
add(0,2) or … etc.
Since the domain only contains integers, the call add(1,0.5) would not be accepted.

The co-domain for the function is .

The type of the function is written as:
f:  ×  → 
It is important to note that this function only takes one argument, but that
argument is a pair of values, such as (6,2).

Partial function application

Partial function application is about providing some of a function’s arguments Partial function
and producing a function that takes the remainder of the arguments. application Providing
Imagine that, when the function add is given an argument of (6,2), a new some of a function’s
function is created: arguments and producing
228 a function that takes the
fx: add (6,2) becomes fy: add _six (2) reminder of the arguments.

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_12_MRN_AQA_CS_227-234.indd Page 229 6/29/21 10:12 PM f-0116 /103/HO02206/work/indd

The function add _six will add 6 to a single integer (since it takes an integer
input, and produces an integer output): fy:  → 
For example: fy: 2 → 8

12 Functional programming
Note that the new function is an anonymous function and does not need to
be written by the programmer; it is simply a method of describing how the
partial function application occurs.
The whole process can be described by saying that the function add takes one
integer input in order to create a function that takes the second integer input.
The result of carrying out the new function produces an integer output.
This is written as: fx:  → ( → )
which means that function fx takes an integer argument and produces
another function that takes an integer argument, which produces an integer
output.
This can also be written as: fx:  →  → 

Composition of functions
Functional composition involves combining two or more functions together.
Functional composition
Function fx takes an integer input and returns the double of that value: The process of combining
fx a = 2a two or more functions.

For example: fx: 2 → 4

Function fy takes an integer input and returns the square of that value:
fy b = b2
For example: fy: 3 → 9
Function fz takes an integer input and finds double the value of the square of
the input.
For example: fz: 4 → 32 (42 = 16; 2 × 16 = 32)
Function h can be described as the composition of fx and fy, and can be
written as: fx ° fy
This can be read aloud as ‘fx of fy’.
We can also state that: fx ° fy a = 2(a2) Note
It is important to apply the functions in the correct order. In the case of Functions can have a range
fx ° fy, the function fy is applied to the original input and the function of identifiers. In some
fx is applied to the result returned from function fy. scenarios functions will
In terms of domains and co-domains it can be said that: be given a single letter to
identify them,
fy: A → B (A is the domain of function fy and B is its co-domain)
fx: B → C (B is also the domain of function fx and C is its co-domain)

Now test yourself

5 A function fx takes a two positive integers as arguments and returns the product
of those arguments: fx [a,b] = a*b
a Write the type of the function using function application.
b Write the type of the function using partial function application.
6 fx a = 2*a
fy b = b2
fz c = c – 1
a What is the result of fx ° fy 2?
b What is the result of fz ° fx 10?
c What is the result of fy ° fx ° fz 3?
229
Answers available online

My Revision Notes AQA A-level Computer Science Third Edition

325487_12_MRN_AQA_CS_227-234.indd Page 230 6/29/21 10:12 PM f-0116 /103/HO02206/work/indd

Writing functional programs

For the AQA examination, a form of pseudo-code is used to represent Note
12 Functional programming

functional programs.
It is useful to have
A function that takes a single argument will typically be written using the done some practical
following format: programming in a functional
fx a = a + 2 programming language.
However, this is not
This means that when the function fx is called, whatever value is passed to necessary and questions
the function as an argument, that value + 2 will be returned. For example: will only be asked on
fx 3 → 5 Component 2, the written
exam.
A function that takes two arguments will typically be written using the
following format:
fy [b,c] = b + c
This means that when the function fy is called, the value returned will be the
sum of the values passed as arguments. For example:
fy [7,4] → 11
A function combination will typically be written using the following format:
fz [d,e] = fx (fy [d,e])
This means that the function fy will be applied to the values passed as
arguments. The function fx will then be applied to the result. For example:
fz[3,1] → 6 (since fy [3,1] → 4 and fx 4 → 6)
There are three higher-order functions that you are expected to be familiar
with. Each one takes two arguments: a function, and a list.

Map
Map is a higher-order function that applies a given function to each item in
Map A higher-order
a list of values. Once complete, a new list containing the results is returned.
function that applies a given
(Lists are described in more detail in the next section.)
function to each item in a
The map function will typically be written in the following format: list.

fu f = map fx f
From the previous section, remember that fx a = a + 2. This means that
the function fx will be applied to each item in the list f, creating a new list of
values which is returned. For example:
If list = [3,9,25,10]
fu list → [5,11,27,12]

List1 List2
1 3
2 Apply function 4
3 5
fx a = a + 2
4 6
5 7
6 8
… …

Figure 12.1 Using the map function to apply a function to a list, in this case the
function will output the square of the input
230

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_12_MRN_AQA_CS_227-234.indd Page 231 6/29/21 10:12 PM f-0116 /103/HO02206/work/indd

Filter
Filter is a higher-order function that uses a conditional operator to generate a
Filter A higher-order
new list.
function that produces a

12 Functional programming
The filter function will typically be written in the following format: new list containing all items
from the input list that meet
fv g = filter (<10) g
a given condition.
This means that the function fv will return a new list containing all of the
values in list g that meet the criteria in parentheses (in this case, values less
than 10).
If list = [3,9,25,10]
fv list → [3,9]

List1 List2
1 1
2 Apply function 3
3 5
Filter = odd
4 …
5
6
…

Figure 12.2 Using the filter function to create a new list according to a condition.
Filters can use inequalities (for example, <10) or filters such as ‘odd’ or ‘even’

Fold (or reduce)

Fold is a higher-order function that reduces a list to a single value by
Fold A higher-order
repeatedly applying a combining function. Some languages refer to this as
function that reduces a
‘fold’, some refer to it as ‘reduce’.
list to a single value by
The fold function will typically be written in the following format: repeatedly applying a
combining function.
fw h = fold (+) 0 h
This means that each item in list h will be repeatedly added to the starting
value of 0. The result will be a single value, which will be returned; for
example, fw list → 47.
The process involves adding each value from the list, starting with the value
of 0 as shown in the table below.

List Head of list Tail of list Result of applying

combining function
0
Note
[3,9,25,10] 3 [9,25,10] 3
[9,25,10] 9 [25,10] 12 Using subtraction and
division as part of a fold
[25,10] 25 [10] 37
function is more complex,
[10] 10 [] 47 as the order in which the
items are combined is more
If a fold function uses a product operator it is typical to see the following: important. It is not expected
fw h = fold (*) 1 h that a question using a
subtraction or division as a
In this case the items in the list are combined by multiplying each of them combining function would
together. The first values is multiplied by 1. If the function is defined fw h = be asked.
fold (*) 0 h, then the first value (and each subsequent value) in the list will
be multiplied by 0 then the result will always be 0.
231

My Revision Notes AQA A-level Computer Science Third Edition

325487_12_MRN_AQA_CS_227-234.indd Page 232 6/29/21 10:12 PM f-0116 /103/HO02206/work/indd

Now test yourself

7 Write a function, fx, that will multiply a number by 6.

8 Write a function, fy, that will take two numbers, and subtract the second from the first.
12 Functional programming

9 Write a function, fz, that will take two numbers, carry out function fy, and then
carry out function fx on the result.
10 What will be the result of using the function map fx [1,2,3,4]?
11 What will be the result of using the function filter (even) [1,2,3,4]?
12 What will be the result of using the function fold (*) 1 [1,2,3,4]?
13 Describe the purpose of the fold function.

Answers available online

Lists in functional programming

Lists in functional programming can be described as being made up of a head Head The first item from
and a tail. a list.
The list [3,9,25,10] can be described in the format 3:[9,25,10], where the Tail The remainder of a list,
head is the value 3 and the tail is a list of values [9,25,10]. without its head.
✚ The head is always a single value, whereas the tail is always a list.
✚ A list can be empty, and this is represented using empty square brackets,
[ ]
There are several standard list processing operations and it is expected that
students will have practical experience of using each one. Although there is
no specific language in which this should be carried out, the AQA pseudo-
code for functional programming is similar in syntax to the language Haskell,
one of the more popular functional programming languages.

Process Description Haskell code Result

Return the head of Identifies the first head [1,2,3,4,5] 1
a list. element in the list.
Return the tail of Identifies all of the tail [1,2,3,4,5] [2,3,4,5]
a list. other elements
apart from the
head.
Test for an empty Checks whether let MyList =
list. there are any [4,8,15,16,23,42]
elements in the list. null MyList False
let MyList = []
null MyList True
Return the length of Identifies how many length 5
a list. elements there are [1,2,3,4,5]
in a list.
Construct an empty Creates a list that let emptylist emptylist
list. has no elements = []
[]
in it.
Prepend an item to Adds an item to the let SetA = SetA Prepend Add something to
a list. beginning of a list. [1,2,3,4] the start of a list.
[1,2,3,4]
let SetB = [0] Append Add something to
SetB
++ SetA the end of a list.
[0,1,2,3,4]
Append an item to Adds an item to the let SetC = SetA SetC
a list. end of a list. ++ [0]
[1,2,3,4,0]

232

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_12_MRN_AQA_CS_227-234.indd Page 233 6/29/21 10:12 PM f-0116 /103/HO02206/work/indd

It is possible to combine list processing operations, and it is common in

exam questions to repeatedly combine head and tail functions. For example,
consider the following operation:
head(tail(list)) → 9

12 Functional programming
This is because Making links
tail(list) → [9,25,10] Recursion is a programming
technique in which a
and the head of this list is 9.
subroutine will call itself
Using head and tail it is also possible to describe the application of the fold repeatedly until it reaches
function in more detail. a non-recursing, base case.
Re-visit the table in the Fold section above to see how the head:tail structure is Recursion is explored in
Recursive techniques in
used to recursively call the same combining function, reaching the base case
Chapter 1.
in which the tail is an empty list.

Now test yourself

14 Given that list = [2,3,5,7,11,13], state the result of applying the function:
a head(list)
b tail(list)
c head(tail(tail(list))).
15 How is an empty list represented?
16 State the result of [1] ++ list.
17 Explain how head and tail can be used to recursively carry out a filter function.

Answers available online

Summary

Functional programming paradigm Writing functional programs

✚ The set that contains all of the values that can be used ✚ The map function is a higher-order function that applies
as the input is called the domain a given function a list of values; for example: map +2
✚ The set that contains all of the values that can be used [1,3,5] → [3,5,7]
as the output is called the co-domain ✚ The filter function is a higher-order function that filters
✚ Not all of a co-domain’s members need to be outputs, a list according to a condition; for example: filter
so if the outputs are all even numbers the co-domain (>20) [7,35,41] → [35,41]
can still be described as the set of integers ✚ The reduce or fold function that reduces a list to a
✚ A first-class object can appear in an expression, be single value by repeatedly applying a combining function;
assigned to a variable, be assigned as an argument or for example: fold (+) 0 [1,2,3,4] → 10
be returned as a function call
Lists in functional programming
✚ In functional programming a function is a first-class
object ✚ A list can be written in the format head:[tail],
✚ A higher-order function is a function that can take a where the head is the first item in the list and the tail is
function as an argument, return a function, or both the remainder of the list; for example: 1:[2,3,4].
✚ In function application a function is applied to its ✚ The tail can be an empty list, which is written []
arguments; for a function that takes one argument this ✚ The higher-order functions map, filter and fold all
is written in the format f: integer → rational use recursion, repeatedly carrying out their function on
(or similar, depending on the domain and co-domain) the head and calling itself on the tail, ending with the
✚ If a function takes two arguments, the Cartesian base case of an empty list
product is used to describe the domain; for example: ✚ Other commands for working with lists include:
f: integer x integer → integer ✚ testing for an empty list: null listName
✚ In partial function application a function takes some ✚ returning the length of a list: length listName
arguments and produces another function that takes ✚ creating an empty list: let list = []
the rest of the arguments; for example, add(6,4) → ✚ prepend an item to the start of a list: let list =
add _ six 4 → 10 [0] ++ list
✚ The function type of partial function application can ✚ append an item at the end of a list: let list =
be described as add: integer → (integer list ++ [5]
→ integer) or integer → integer →
integer
✚ Functional composition combines two functions so that
fx o fy a → fx (fy (a)) 233

My Revision Notes AQA A-level Computer Science Third Edition

325487_12_MRN_AQA_CS_227-234.indd Page 234 6/29/21 10:12 PM f-0116 /103/HO02206/work/indd

Exam practice

1 A list of values is defined as list = [8,1,4,7,3].

The function double is defined as double x = x * 2.
12 Functional programming

a) State the result of applying the functions head(tail(tail(list))). [1]

b) State the result of applying the function fold (+) 0 list. [1]
c) State the result of applying the function map double tail(list). [2]
2 Four functions are defined as follows.
fw a = a + 10
fx [b,c] = b / c
fy d = filter (>0) d
fz e = map fx e
Lists of values are defined as follows:
m = [17,-6,28,0,-30]
n = [-2,7,16,-3]
o = [[12,3],[4,1],[24,4]]
a) The co-domain of function fw is , state the function’s domain. [1]
b) The domain of function fx is  x . State the function’s co-domain. [1]
c) Copy and complete the table below by ticking each higher-order function. [2]

Function Higher-order function?

fw
fx
fy
fz

d) Calculate the results of each of the following function calls. [3]

Function call Result

fy m
fz o
map fw (fy n)

3 The function fold (+) 0 [5,2,6,8] is called.

a) State the purpose of the fold function. [2]
b) State the result of the function call. [1]
c) Explain how you arrived at your answer to part b, and the recursive
steps that you followed. [3]

234

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_13_MRN_AQA_CS_235-240.indd Page 235 6/29/21 10:12 PM f-0116 /103/HO02206/work/indd

13 Systematic approach to problem

solving
There are many models used to describe or define approaches to software
development. Whichever model is used, the stages of development are very
similar.

Analysis

Evaluation Design

Exam tip
It is not necessary to be
familiar with any specific
model of software
Testing Implementation
development. However,
make sure you are familiar
with the five key stages
Figure 13.1 Stages of system development described below.

Making links
The NEA programming project is broken down into the same five steps outlined below. It
is rare for questions on this topic to appear specifically on the Component 2 examination,
though the knowledge here will be very helpful to supporting you in the NEA.

Aspects of software development

Analysis
The first stage of solving any problem is the analysis. The aim of this stage is
to identify the details of that problem and to make sure that the problem is
well understood.
In projects where the analysis is not given enough time or attention there is a
risk that the solution will not fully solve the problem, or that the solution will Analysis The first stage
not be suitable for the end-user. of systems development,
in which the problem is
The first part of the analysis is focused on understanding the problem: identified, and solutions
✚ Defining the problem proposed.
This is usually a short summary, giving the big picture.
Stakeholders The people
✚ Identifying the stakeholders with an interest in the
These include the client, the end-users and anyone else directly affected problem or the solution.
by the problem.
✚ Understanding the current system
This can be done by asking users how they deal with the current situation,
researching the context of the problem and observing the current system
in progress.

235

My Revision Notes AQA A-level Computer Science Third Edition

325487_13_MRN_AQA_CS_235-240.indd Page 236 6/29/21 10:12 PM f-0116 /103/HO02206/work/indd

The next step is to outline the potential solution.

✚ Examine existing solutions
This involves looking at existing products or solutions that already exist.
For example, looking at several other stock-taking systems in order to
13 Systematic approach to problem solving

identify what features are common, which are most used, and which are
unnecessary or irrelevant to this scenario.
✚ Identify required features
Having looked at the current system, compared existing solutions and
discussing the needs with a variety of stakeholders, it is important to
narrow down a list of key features that the solution must include.
✚ Development platforms
It is important to consider whether the solution should take the form of
a desktop program, mobile program, web-based platform or some other
form. It is also important to consider what programming languages could
be used in the development and details such as whether a database system
or file management system is necessary.
✚ Modelling
A model for the data must be created in order to understand the problem
and identify a suitable solution. Describing both the problem, and the
proposed solution, using diagrams, charts and tables is an effective way
to generate and refine the data model and to understand how the solution
might fit together. It can also help to demonstrate ideas to a client, who Modelling Graphical
may not be a computing expert. descriptions such as
flowcharts, E-R diagrams,
This is an example of using abstraction to help model aspects of the class diagrams and
external world for use in a computer program hierarchy charts.
✚ User Interface
User interface The inputs
A user interface is an extremely important consideration for any computer and outputs of a program.
system. The solution could involve a command line interface, a graphical
Agile A method of project
user interface (GUI), a menu driven interface or even a voice-controlled
management in which a
interface. Sketches and ideas for prompts, dialog boxes and other elements
rough prototype is created
of the user interface should be produced as early as possible.
and repeatedly improved.
Client discussion is key to analysis, both in order to help to understand the
problem and to ensure that both parties agree on the proposed solution. You
should show your ideas, plans, sketches and models to your client to help
them understand your proposed solution and you may wish to take an agile
approach, producing ever-more complete prototypes throughout the life of the
project.

Making links
The agile approach makes use of rapidly built prototypes to plan and develop the
project as you go. This can be a risky strategy and this is discussed in more detail in
the design section.

The final stage is to create a list of detailed, formal objectives. Each objective
should be SMART – specific, measurable, achievable, relevant, and timely. Any
object that does not meet these criteria will be either difficult to measure,
difficult to achieve, or will add little value to the overall solution.
The client may have questions to help them understand the proposal, the
developer may have questions to help clarify certain parts of the problem, and
the client might well have additional ideas or concerns about the proposed
solution that are important to consider.

236

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_13_MRN_AQA_CS_235-240.indd Page 237 6/29/21 10:12 PM f-0116 /103/HO02206/work/indd

Now test yourself

1 Identify three stakeholders in a self-service checkout system for a supermarket.

2 Suggest three different examples of modelling for a computer system.

13 Systematic approach to problem solving

3 What does the acronym SMART stand for?

Answers available online

Design
The design of the solution is something easily overlooked or rushed, as people
Design The second stage
are often keen to get straight into the practical development of the solution.
of systems development,
As projects become larger and more complex, however, the time taken in the
in which the details of the
design phase becomes increasingly important to avoid problems later in the
solution are planned.
development.
Top level design An
The design should consist of several sections. outline design that shows
how the whole program will
Top level design fit together.
A top level design gives an overview of the solution. The aim is to look at the Flowchart A diagram using
big picture in terms of what the project aims to achieve. Details of specific symbols to represent parts
algorithms can be skipped at this stage in order to provide a clear design that of an algorithm and arrows
covers the whole solution in one place. to show which step to
follow next.
The nature of the top-level design will depend very much on the type of
project. For example: Entity-relationship
diagram A diagram
✚ a flowchart might be ideal for a heavily algorithm-based project
showing the relationship
✚ an entity-relationship diagram might be more suitable for a heavily
between different entities in
database centred project
a database.
✚ a class diagram can help to summarise an object-oriented project
✚ a hierarchy chart can help to show how a large task is split into Class diagram A diagram
subroutines. showing the structure of
and relationship between
A combination of two or three diagrams might be needed to give this top-level classes in an object-
overview. oriented program.
Hierarchy chart A diagram
Making links
showing which subroutines
Class diagrams are an effective design tool to help describe the structure of an object- will call which other
oriented program. Object oriented programming is explored in Chapter 1. subroutines.

Hierarchy charts are commonly used in procedural programming to describe how

subroutines are related to each other and how a larger problem can be tackled by Making links
being split into smaller sub-problems. Hierarchy charts are also discussed in Chapter 1.
Databases are an essential
component of many
Database design computer-based systems.
If the project requires the use of a database then it is important to design the Component 2 typically
structure of the database carefully. This should include: includes a large proportion
✚ E-R diagrams of questions on the topic of
✚ a data dictionary (showing the data type and keys for each table) databases, and this topic
✚ designs for all required queries; this includes INSERT, UPDATE, and DELETE is discussed in detail in
Chapter 10.
queries as well as SELECT queries.

Algorithm design
Algorithm A sequence of
Algorithms are a key component to any software project. It is therefore
steps to follow in order to
necessary to plan algorithms before they are programmed. It is common
achieve an outcome.
to create the broad, top-level design initially and then plan each individual
subroutine in more detail. Pseudo-code A form of
program design that uses
It is not usually necessary to design every single algorithm as some are code which is not specific to
relatively straightforward. However, more complex and significant algorithms one programming language.
should always be planned in advance. This can be done with pseudo-code, 237
flowcharts, or other forms of modelling as appropriate.

My Revision Notes AQA A-level Computer Science Third Edition

325487_13_MRN_AQA_CS_235-240.indd Page 238 6/29/21 10:12 PM f-0116 /103/HO02206/work/indd

Data structure design Making links

It may be necessary to use abstract data structures such as stacks, queues,
Complex data structures
dictionaries, hash tables or vectors. If this is the case then it would be
such as stacks, queues,
appropriate to describe how these could be implemented. For example, a
13 Systematic approach to problem solving

dictionaries and vectors

stack can be created using a one-dimensional array and a pointer to the are discussed in detail in
top of the stack. Chapter 4. Making good use
If data is to be stored in files it is also important to consider how these files of these data structures
will be organised; for example: can be an effective way of
✚ will there be one file per user, or will all data be stored in the same file? demonstrating your abilities.
✚ what format will the file take?
Abstract data structure
Modular structure A data structure which
In a modular structure, each part of the program achieves one task. These can be implemented using
modules are combined to create the overall solution by passing parameters a combination of other,
and returning the results of each individual part of the program. simpler data structures.

Using a modular structure means that parts of the program are loosely Modular Made up of
coupled. This means that if one part of the program needs to be replaced independent units or parts.
then the rest of the program should be unaffected. This can be achieved by
avoiding the use of global variables and creating subroutines that solve a
general problem (for example, a subroutine in a card game should shuffle
an array of cards rather than the whole deck. This way the whole deck can
be passed if needed, but a smaller selection of cards can also be shuffled if
needed).
Modular programs are easier to debug, easier to update, require less code (as
modules can be re-used) and modules can be re-used in future projects.

User interface
Though the user interface should have been considered in the analysis phase,
the design phase is likely to be much more detailed and so a comprehensive
design of the user interface should be included.
This involves designs for all screens, pages, prompts, responses, and error
messages that may be required as part of the solution.
The design phase is often an iterative process, meaning that it is unlikely that Iterative Repeating
the whole project would be designed in one go before being implemented. a process in order to
In practice it makes much more sense to design part of the project and then approach an end point.
implement that section. This means that practical problems can be overcome
quickly and future iterations of the designs can take this into account. Agile A method of project
management in which a
rough prototype is created
Agile approach and repeatedly improved.
You may wish to take an agile approach to solving the problem. This involves
starting with a rough prototype that is repeatedly improved. In each phase
a relatively small part of the solution is quickly planned, implemented and
tested.
Phases are very short and approaches can be quickly tried and discarded if
not suitable. This can be an effective solution, but comes with risks as it is
possible to spend several phases working towards an end-goal, only to realise
that there is a critical flaw that requires a complete re-write to solve.

Now test yourself

4 What would typically be left out of a top-level design?

5 Which aspects of a database should be included in the design?
6 Why is it preferable to use a modular design?
238 Answers available online

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_13_MRN_AQA_CS_235-240.indd Page 239 6/29/21 10:12 PM f-0116 /103/HO02206/work/indd

Implementation
Implementation refers to the creation of the algorithms and data structures
Implementation: The
required to solve the problem.
third stage of systems

13 Systematic approach to problem solving

It is important to write, debug and test code in small chunks as the program development, in which the
is developed, rather than waiting until the end of the program. The modular program code and data
design discussed in the previous section makes this much easier, as each structures are created.
subroutine can be tested individually as well as testing the program as a whole.
Debug: Identify and remove
It is typical to use an iterative process to code, test and refine the errors.
implementation, with a focus on the core elements of the program first,
adding in additional elements later on. There is a risk in focusing on
implementing each section to perfection that time will run out and the client Making links
will be left with an excellent start, but an incomplete overall solution.
The implementation phase
draws on concepts covered
Now test yourself in Chapter 4.
7 What two things should be created during the implementation stage?
8 When is it appropriate to test the implementation?

Answers available online

Testing
Testing should be carried out throughout the development of the solution. The
aim of testing is not only to focus on whether the system works with expected Testing The fourth stage
data, but to test that it copes appropriately with extreme and invalid data. of systems development, in
which the program is tested
For example, a database designed to track shoe sizes up to and including size to ensure it functions as it
13 should be tested with: should.
✚ normal data (such as 6, 8, 11),
✚ boundary data (such as 13) Boundary data Data on
the edge of what should be
✚ erroneous data (such as 14, 15, 106, -2, ‘orange’).
accepted.
This will ensure that the boundaries are set appropriately and that incorrect
Erroneous data Data that
or invalid inputs are dealt with cleanly.
is in error or invalid.
Defensive programming should be used to prevent the program from
Defensive programming
crashing. For example, exception handling should be used to deal with invalid
Programming intended
inputs such as a string being entered where a number is expected.
to deal with unforeseen
Making links circumstances.
Exception handling
Exception handling is an effective technique for dealing with invalid inputs. The syntax
A program technique
and structure of exception handling is discussed in Chapter 1.
that uses a try … catch
structure to deal with errors
Acceptance testing should also take place towards the end of the project. that would otherwise cause
At this point, the client and/or end-users should be provided with the the program to crash; for
opportunity to test the solution for themselves in order to check: example, if the user enters
✚ they are happy that the solution solves the original problem a string where a number is
✚ it functions as it should expected.
✚ it does not contain any errors or problems.
Acceptance testing
Acceptance testing should be at an appropriate time in the software Testing by the client and/
development schedule so that if the feedback suggests that improvements are or end-user to ensure that
necessary, there is time to correct them. the system meets the
specification.
Now test yourself

9 What three types of test data should be used?

10 What is the meaning of defensive programming?
11 What is the final stage of testing?

Answers available online 239

My Revision Notes AQA A-level Computer Science Third Edition

325487_13_MRN_AQA_CS_235-240.indd Page 240 6/29/21 10:12 PM f-0116 /103/HO02206/work/indd

Evaluation
The final stage of any project is to evaluate the success of the project. To do
this the original objectives should be reviewed, using the testing process as Evaluation The fifth and
13 Systematic approach to problem solving

evidence, and each item marked as complete or incomplete. final stage of systems
✚ The most fundamental aspect of an evaluation is to consider whether the development, in which the
original problem has been solved. It is possible to make a fully functioning, success of the project is
complex program that doesn’t fully meet the requirements initially set out. considered.
✚ User feedback should be considered. This includes feedback from the
client, from the end-users and from other stakeholders.
It is perfectly acceptable for feedback to suggest further improvements and
developments for the future, and this does not necessarily mean that the
project is not a success. There are always opportunities to further refine and
extend any project.
Now test yourself

12 Who should provide feedback used in an evaluation?

13 What is the fundamental question that an evaluation should aim to answer?
Answers available online

Summary
✚ Large software development projects can be split into ✚ Testing is focused on ensuring that the program works
five sections for valid data and copes well with boundary data and
✚ Analysis is focused on identifying and describing the erroneous data
problem, establishing system requirements from potential ✚ Testing also includes acceptance testing with intended
users, creating data models and agreeing a specification users of the system
✚ Design is focused on planning the data structures, ✚ Evaluation is focused on reflecting on the finished
algorithms, modular structure and user interface solution and making sure that it solves the original
required to solve the problem problem
✚ Implementation is focused on the creation of the ✚ An iterative approach may be used at each stage as
program code and data structures part of a prototyping/agile approach

Exam practice

1 Anne-Marie has been asked to develop a computer program for a company that
manufactures vending machines.
a) Copy and complete the table by entering the appropriate letter,
A-E, to match the activities with the stages of systems development. [3]
A Analysis D Testing
B Design E Evaluation
C Implementation
Tasks Stage
Creating a flowchart to describe an algorithm for
part of the program
Reviewing the overall success of the finished
program
Trying out other vending machines in order to see
what features they include
Entering invalid data to see how the program copes
Writing the program code

b) Identify two stakeholders for this project. [2]

c) List three things that should be included in the design for the project. [3]
d) Anne-Marie is using a modular design for the program.
Explain what is meant by a modular design and why this is useful. [3]
e) The vending machine uses a numeric keypad allows the user
to enter a 3-digit number between 100 and 199 to choose their item.
Identify three types of input that should be tested and give an
240
example value for each. [6]

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_Gloss_MRN_AQA_CS_241-256.indd Page 241 6/29/21 10:11 PM f-0116 /103/HO02206/work/indd

Glossary

Glossary
Term Definition Page
Absolute error The difference between the intended value and the actual value. 118
Abstract data A data structure which can be implemented using a combination of other, simpler 238
structure data structures.
Abstract data type A complex data structure in which the complexity of how the data is stored or 38
accessed is typically hidden from the programmer.
Abstract method A method that has no program code and is intended to be overridden. 29
Abstraction Making a problem simpler by removing or hiding features. 81
Acceptance testing Testing by the client and/ or end-user to ensure that the system meets the 239
specification.
Access specifier (or A keyword that sets the accessibility of an attribute or method. 25
access modifier)
Access speed The time taken to read or write data. 178
ADC Analogue to digital converter. 174
Address bus A bus for transmitting memory addresses. 160
Address bus width The maximum number of bits that can be used to address a memory location. 173
Addressing mode Either immediate or direct. 164
Adjacency list A method of representing a graph by listing, for each node, just the nodes that are 44
connected to it directly.
Adjacency matrix A method of representing a graph using a grid with values in each cell to show which 44
nodes are connected to each other.
Aggregation The idea that one class can be related to another using a ‘has a’ relationship. 26
Agile A method of project management in which a rough prototype is created and 236, 238
repeatedly improved.
Algorithm A sequence of instructions that are followed in order to solve a problem. 53, 71,
237
Amplitude A measure of how large a vibration or oscillation is. 129
Analogue A continuous signal, usually represented as a curved wave. 126
Analogue to digital Converts an analogue signal into a digital signal. 126
converter (ADC)
Analysis The first stage of systems development, in which the problem is identified, and 235
solutions proposed.
Anti-virus Software that scans programs for malicious code. 200
Append Add something to the end of a list. 232
Application layer Formats data according to the application to be used. 201
Application A tool that allows other programs to make calls or requests. 209
programming
interface (API)
Application Software intended to allow the end-user to achieve a task. 139
software
Arguments The actual values that are passed to the subroutine at runtime. 19
Arithmetic Mathematical operations such as addition, subtraction and multiplication. 162
Arithmetic logic Carries out arithmetic (mathematical) and logical operations. 162
unit (ALU)
Array A data structure for holding values in a table 34
ASCII A 7-bit code used for representing characters as binary numbers. 122, 188
Assembler Translates assembly language code into object code. 143
241

My Revision Notes AQA A-level Computer Science Third Edition

325487_Gloss_MRN_AQA_CS_241-256.indd Page 242 6/29/21 10:11 PM f-0116 /103/HO02206/work/indd

Assembly language Each instruction is represented as a text-based command. 141

Assignment Changing the value stored in a variable. 11, 72
Associated class (or A class or object that is used to make up another object. 26
Glossary

associated object)
Association A type of aggregation where the container object can be destroyed without 26
aggregation destroying its associated objects.
Asymmetric Different keys are used for encryption and decryption. 135
Asymmetric Different keys are used to encrypt and to decrypt a message. 198
encryption
Asynchronous data Where data is sent without a timing signal. 188
transmission
Attribute A single characteristic of an entity. 214, 216
Attributes The properties that an object of that type has, implemented using variables. 23
Automation Designing, implementing and executing a solution to solve a problem automatically. 83
Backus–Naur form A notation used to describe the syntax of a language. 89
(BNF)
Bandwidth The range of possible signals that can be sent in one signal. 189
Barcode A black and white image consisting of vertical bars of different widths to represent 173
data.
Barcode reader A device for reading barcodes. 173
Base case The case in which a recursive function terminates and does not call itself. 22
Base class (or A class whose attributes and methods are inherited by another class. 25
parent class)
Baud rate The number of state changes per second. 189
Bi-directional Data travels in both directions. 159
Big Data Data that cannot easily be stored, retrieved, and processed because it is too varied, 224
too large or acquired too rapidly.
Binary Numbers with base 2. 104
Binary file A file that uses binary values to represent each item of data. 37
Binary point Used to separate the whole number and fractional parts of a binary number. 114
Binary prefix A shorthand used for multiples of 1024 (210) bytes; for example kibibyte, mebibyte, 109
gibibyte.
Binary search A searching algorithm in which the middle item is checked and half of the list is 62
discarded.
Binary tree A tree in which each node can have no more than two child nodes, each placed to the 64
left or right of the preceding node so that the tree is always in order.
Binary tree search A searching algorithm in which a binary tree is traversed. 64
Bit rate The number of bits that can be transmitted per second. 189
Bitmap An image format that uses a grid of pixels. 127
Bitwise An operation that works on one bit at a time. 167
Block A subdivision of the storage on an SSD. 177
Blu-Ray A high-capacity optical disc typically used for high-definition video and large computer 177
programs.
Boolean expression A mathematical notation for logic gates and circuits. 145
Boolean identity A relation that is always true. 152
Bottleneck A congested path within a network. 195
Boundary data Data on the edge of what should be accepted. 239
Branch Jump to a labelled part of the program. 166
Breadth first search Searching a graph by checking every node one step from the starting point, then 53
(BFS) every node two steps away, etc.
242 Brute force Trying every possible combination. 134

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_Gloss_MRN_AQA_CS_241-256.indd Page 243 6/29/21 10:11 PM f-0116 /103/HO02206/work/indd

Bubble sort A sorting algorithm in which pairs of items are sorted, causing the largest item to 65
bubble up to the top of the list.
Bug A mistake in a computer program that causes unexpected results. 200

Glossary
Bus A communication system for transferring data. 159
Bus topology A network arranged with a main cable, or bus, connecting all devices. 190
Bytecode An intermediate code between high-level and object code which can run in a virtual 145
machine.
Cache A small unit of volatile memory placed on the same circuit board as a processor for 172
fast access.
Caesar cipher A substitution cipher in which each letter is shifted according to the key. 133
Call The process of running a subroutine by stating its identifier, followed by any required 18
arguments in parentheses.
Call stack A data structure that stores the stack frames for each active subroutine while the 20
program is running.
Capacity The amount of data that can be stored. 178
Cardinality The number of values in a finite set. 86
Cartesian product The set of ordered pairs of values from both sets. 86
CD Compact disc, typically used for audio and small computer programs. 177
Certification A body capable of providing digital certificates. 199
authority
Character set The set of all characters that can be understood by a computer system, and their 121
associated character codes.
Check digit A single digit derived by following an algorithm, used to check for errors. 124
Checksum A value, calculated by an algorithm, based on the contents of the original data. 124, 199
Cipher An algorithm for encrypting data. 133
Ciphertext The encrypted form of the message. 133
Circular queue A queue which wraps around in a circle. If implemented using an array, the last index 39
is followed by the first index.
Class The definition of the attributes and methods of a group of similar objects. 23
Class diagram A diagram showing the structure of and relationship between classes in an object- 237
oriented program.
Client A device that makes requests of a server. 191, 208
Client–server A network in which clients make requests to servers. 191
network
Clipart A cartoon style image. 129
Clock A component in the processor for generating timing signals. 162,
172, 188
CMYK In a colour printer the colours used are cyan, magenta, yellow and key (black). 175
Co-domain A set from which all output values are chosen. 227
Collision Where two items of data are transmitted at the same time, causing both to be lost. 47, 194
Colour depth The number of bits used to store the colour of each pixel. 127
Command line A text-only interface which is less user-friendly but often more powerful for 140
interface (CLI) knowledgeable users.
Common factor A term which appears in 2 or more parts of an expression. 154
Compare The operation used to check the condition between two values. 166
Compiler Translates high-level code into object code as a batch. 143
Composite primary A primary key made up of two or more attributes. 216
key
Composition Combining parts of a solution together to create a solution made of component parts. 83
Composition A type of aggregation where, if the container object is destroyed, its associated 26
aggregation objects are also destroyed. 243

My Revision Notes AQA A-level Computer Science Third Edition

325487_Gloss_MRN_AQA_CS_241-256.indd Page 244 6/29/21 10:11 PM f-0116 /103/HO02206/work/indd

Compression Reducing the size of a file. 131

Computable A problem that can be solved with a computer. 96
Concatenate Join two (or more) strings together; for example, ABC + DEF = ABCDEF. 16
Glossary

Concurrent At the same time. 222

Condition A comparison to see if two values are equal, or whether one is greater/less than the other. 166
Constructor A special method (with the same name as the class) that is called when the object is 24
instantiated.
Container class (or A class or object that contains, or is made up of, other objects. 26
container object)
Continuous Can take any value, such as the length of an object. 104
Control bus A bus for transmitting control and status signals. 159
Control unit (CU) Decodes instructions and sends/ receives control signals. 162
Controller A component of an SSD which manages the reading and writing. 177
Convex The process of multiplying two (or more) vectors by scalar values that are both 50
combination positive and add up to exactly 1.
Cost This can mean the cost per device, or cost per GB. 178
Countable set A set with the same number of values as a subset of the natural numbers N. 86
Countably infinite A set whose values will go on forever, but can still be counted (for example, N). 86
sets
CPU Electronic device used to process instructions. 158
Crosstalk Interference caused when two or more worse are in close proximity 188
CRUD A mnemonic for the four main database operations. 209
CSMA/CA Carrier sense multiple access/collision avoidance. A method of collision avoidance by 194
checking for existing transmissions.
Cultural Relating to the ideas, customs and behaviour of a society. 183
Data Values or information. 126
Data bus A bus for transmitting data. 160
Data bus width The maximum number of bits that can be transferred at one time. 173
Data skew When data that was sent at the same time arrives at a slightly different time to each 188
other
Data structure The concept of storing multiple values together with a single identifier 34
Data type A definition of what type of data can be stored in that variable. 15
Database definition The language used to create database tables. A subset of SQL commands can be 218
language (DDL) used as a database definition language.
De Morgan’s laws Break the line, change the sign (and vice versa). 154
Debug The process of identifying and removing errors from program code. 13, 239
Decimal Numbers with base 10. 104
Decimal prefix A shorthand used for multiples of 1000 (103) bytes; for example, kilobyte, megabyte, 108
gigabyte.
Declaration The creation of a variable in memory. 10
Declarative A language in which the programmer codes what they want to happen, but not how. 142
Decomposition A method of solving a larger problem by breaking it up into smaller and smaller 22
problems until each problem can’t be broken down any further.
Decomposition Breaking a problem into smaller subproblems. 83
Decryption A method of converting an encrypted message back to its original form. 133
Dedicated registers Registers which are dedicated to holding data for a specific, functional purpose. 162
Defensive Programming intended to deal with unforeseen circumstances. 239
programming
Definite iteration, Iterating a fixed number of times (also known as a count-controlled loop, implemented 12
or count-controlled as a FOR loop).
244 iteration

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_Gloss_MRN_AQA_CS_241-256.indd Page 245 6/29/21 10:11 PM f-0116 /103/HO02206/work/indd

Denary An alternative less ambiguous term for base 10 numbers. 104

Dense graph A graph with many edges. 44
Depth first search Searching a graph by heading in one direction until you reach a dead end, then 53

Glossary
(DFS) stepping back one node at a time.
Design The second stage of systems development, in which the details of the solution are 237
planned.
Dictionary A data structure based on key– value pairs. 48, 132
Digest A checksum value used in a digital signature. 199
Digital A discrete signal, often represented using a stepped wave. 126
Digital camera A device for taking photographs stored as digital data. 174
Digital certificate A digital document indicating that a public key is valid. 199
Digital signal Manipulating digitised data relating to analogue signals. 161
processing
Digital signature A method of checking that an encrypted message has not been altered. 199
Digital to analogue Converts a digital signal to an analogue signal. 126
converter (DAC)
Direct addressing The operand is the address of the value which is to be processed. 164
Directed graph A graph some edges can only be traversed in one direction, shown with an 43
arrowhead.
Disconnected graph A graph in which two or more nodes are not connected to the rest of the graph. 43
Discrete Can only take certain, countable values, such as shoe sizes or ordinal numbers. 104
Distributed A system that uses multiple processors or servers to process data. 224
processing
Domain A set from which all input values are chosen. 227
Domain name The identifier for an individual or organisation’s online presence. 196
Domain name A system for looking up domain names to find their corresponding IP address. 197
system (DNS)
Dot product A calculation used to help find the angle between two vectors. 50
Dotted-decimal A format for writing an IP address with decimal values separated by dots. 197
Drum A round device used to attract the toner. 175
Dual-core A processor containing two CPUs, both working simultaneously. 172
DVD Digital versatile disc, typically used for medium quality video and medium sized 177
computer programs.
Dynamic The size of, and the memory assigned to that data structure can change. 38
Dynamic host A system for issuing private IP addresses within a network. 207
configuration
protocol (DHCP)
E-R diagram An entity-relationship diagram. 214
Edge A line used to represent the relationship between two nodes. Also known as an arc. 43, 225
Edge triggered A logic circuit used to store the state of an input. 151
D-type flip-flop
Edge-case A problem that only occurs in an extreme setting. 200
Efficiency Being able to complete a task with the least use of time or memory. 91
Encapsulation The concept of grouping similar attributes, data, and methods together in one object. 24
Encryption A method of hiding the meaning of a message. 133,
193, 198
Entity A type of real-world object about which data should be stored. 214
Entity-relationship A diagram showing the relationship between different entities in a database. 237
diagram
Erroneous data Data that is in error or invalid. 239
245

My Revision Notes AQA A-level Computer Science Third Edition

325487_Gloss_MRN_AQA_CS_241-256.indd Page 246 6/29/21 10:11 PM f-0116 /103/HO02206/work/indd

Evaluation The fifth and final stage of systems development, in which the success of the project 239
is considered.
Even parity A method of using parity in which each block of data should have an even number 123
of 1s.
Glossary

Exception handling A program technique that uses a try … catch structure to deal with errors that would 239
otherwise cause the program to crash; for example, if the user enters a string where a
number is expected.
Expanding brackets The process of removing brackets by multiplying. 154
Exponent The part of a floating point binary number which states how far to move the decimal 115
point.
Exponential An expression where the variable is used as an exponent, or power; for example, 92
y = 2x.
Exponentiation The raising of one number to the power of another. 14
Extensible markup A file format for human-readable text used to transmit data objects. 209
language (XML)
Factorial The product of all positive integers less than or equal to a given integer. 93
Factorising The process of removing a common factor. 154
Fetch-execute cycle The process by which instructions are fetched from memory, decoded and executed. 162
Field A category of data within a record or group of records. 37, 216
File A persistent collection of data, saved for access and accessible once the program or 37
subroutine has finished running.
Filter A higher-order function that produces a new list containing all items from the input list 231
that meet a given condition.
Finite set A set with a fixed number of values. 86
Finite state A computational model which has a fixed number of potential states. 83
machine (FSM)
Firewall A software or hardware service that blocks or allows individual packets from entering 198
or leaving.
First In, First Out Those items placed into the queue first will be the first ones to be accessed. 39
(FIFO)
First-class object An object that can be assigned as an argument or returned from a function. 227
Fixed point binary A method for representing fractional numbers where the number of binary digits 114
before and after the decimal point is fixed.
Flat file database A database made up one table. 216
Floating gate Used as a memory cell in flash memory. 177
transistor
Floating point A method for representing fractional numbers where the position of the decimal point 115
binary is moved according to the value of the exponent.
Flowchart A diagram using symbols to represent parts of an algorithm and arrows to show which 237
step to follow next.
Fold A higher-order function that reduces a list to a single value by repeatedly applying a 231
combining function.
Foreign key An attribute in one table that links to the primary key in a related table. 216
Frequency analysis The process of examining how often something occurs. A useful tool for trying to 48
break some encryption methods.
Full adder A logic circuit to add three binary digits. 150
Full-duplex Both devices can send transmissions at the same time. 209
Fully qualified A domain name that addresses an exact resource, including a hostname. 196
domain name
(FQDN)
Function A subroutine that returns a value. 19
Function A rule that takes one or more values as an input and produces an output. 227
246

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_Gloss_MRN_AQA_CS_241-256.indd Page 247 6/29/21 10:11 PM f-0116 /103/HO02206/work/indd

Functional The process of combining two or more functions. 229

composition
Functional A programming paradigm based around using functions to process data. 225
programming

Glossary
Gateway A network hardware device for connecting networks that use different protocols. 196
General case A case in which a recursive function is called and must call itself. 22
General-purpose Registers that can be used to store any data. 162
registers
Getter A function used to return the value of an attribute to another object. 25
Global variables Variables that can be accessed from any subroutine. 20
Graph A data structure designed to represent the relationships between items. 43
Graph schema A graphical tool for representing fact-based models. 225
Graph traversal Inspecting the items stored in a graph data structure. 53
Graphical user An image-based interface that uses windows, icons, menus and pointers. 140
interface (GUI)
Half adder A logic circuit to add two binary digits. 150
Halting problem A specific example of a non-computable problem that proves that some problems are 96
non-computable.
Halting state A state with no outgoing transitions, and so the Turing machine will stop. 96
Hand-trace Also known as dry run. Without using a computer, completing a table to record the 53, 72,
values of each variable as the program is executed line-by-line. 170
Hard disk drive A storage device which saves data using magnetic film. 176
(HDD)
Hardware The physical components of a computer system. 139
Hash table A data structure for holding data that is indexed according to its key. 47
Hashing algorithm An algorithm for calculating a key for each value. 47
Head The first item from a list. 232
Header Extra data added to a packet such as the destination address or packet number. 195
Heuristic An approach to solving intractable problems in a reasonable amount of time by 95
accepting an imperfect solution.
Hexadecimal Numbers with base 16. 105
Hierarchy chart A diagram showing which subroutines will call which other subroutines. 237
Hierarchy charts A diagram that shows which subroutines call which other subroutines. More complex 22
versions will also show what data is passed and returned.
High-level language A programming language which uses keywords and constructs written in an English- 142
like form.
High-order function A function that can take another function as a parameter. 225
Higher-order A function that takes a function as an argument, or returns a function, or both. 228
function
Host identifier The second part of an IP address, indicating the device on that network. 205
Hostname The device within the network that is being addressed. 196
Hypothetical Consideration of a ‘what if…’ scenario. 182
I/O controller The physical interface between an I/O device and the internal components. 159
I/O devices Input and output devices such as keyboard, mice, printers and monitors. 159
Identifier A technical term for the name of a variable. 15
Immediate The operand is the value, which is to be processed. 164
addressing
Immutable data Data structures that cannot be changed. 225
structures
Imperative A language in which commands are used to say how the computer should complete 142
the task. 247

My Revision Notes AQA A-level Computer Science Third Edition

325487_Gloss_MRN_AQA_CS_241-256.indd Page 248 6/29/21 10:11 PM f-0116 /103/HO02206/work/indd

Implementation The third stage of systems development, in which the program code and data 239
structures are created.
In-order tree Inspecting or displaying the value of each node after checking the node on the left but 57
traversal before checking the node on the right.
Glossary

Inconsistency The result is not always the same. 217

Indefinite iteration, Iterating until a condition is met (also known as a condition-controlled loop, 12
or condition- implemented in AQA pseudo-code as a WHILE loop or DO-WHILE loop).
controlled iteration
Index A value indicating the position of a value within an array or list. 34
Infinite sets A set whose values will go on forever. 86
Infix notation A method of writing mathematical or logical expressions with operators placed 60
between operands; for example, 2 + 2.
Information hiding Controlling access to data stored within an object. 24
Inheritance The idea that one class can use the attributes and methods defined in another class. 25
Instantiation The process of creating an object based on its class definition. 24
Instruction set The list of machine code values for each available operation. 164
Integer division A division operation in which any fractional part is discarded, equivalent to rounding 63
down to the nearest whole number.
Integer numbers Whole numbers, including both positive and negative values. 103
Interface A collection of subroutines that can be implemented by unrelated classes. 28
Internal The components essential to a computer system. 158
components
Internet of Things Physical devices able to share data or to be controlled using a connection to the 187
(IoT) internet.
Internet registry A body responsible for allocating IP addresses. 197
Interpreter Translates and then runs high-level code one section at a time. 144
Interrupt A signal sent to the processor in order to alert it to an event. 171
Interrupt service A program which will examine an interrupt and handle the event. 171
routine (ISR)
Intractable A problem which can be solved, but not in a reasonable amount of time (worse than 95
polynomial).
IP address A numerical address for a device on a network. 197
IPv4 Internet Protocol version 4; uses 32 bits to represent each address. 206
IPv6 Internet Protocol version 6; uses 128 bits to represent each address. 206
Irrational numbers Any number that cannot be represented as a fraction, including π and square roots of 103
non-square numbers.
Iteration The repetition of a process or block of code. 12, 72,
166
Iterative Repeating a process in order to approach an end point. 238
JavaScript A programming language often used for web applications. 209
JavaScript object A file format for human-readable text used to transmit data objects. 209
notation (JSON)
Key A value used to encrypt or decrypt data. 133, 198
Label A named point in the program. 167
Lands Small bumps in the track. 177
Laser A device for creating an intense beam of light. 175
Last In, First Out Those items placed into the stack most recently will be the first ones to be accessed. 41
(LIFO)
Latency The time taken for the first signal to reach its destination. 177, 189
Least significant bit The right-most bit in a binary number. The bit with the smallest place value. 105
(LSB)
248

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_Gloss_MRN_AQA_CS_241-256.indd Page 249 6/29/21 10:11 PM f-0116 /103/HO02206/work/indd

Library A collection of modules that provide related functionality. 19

Linear Rising in a straight line; for example, the graph of y = 2x. 92
Linear queue A queue in which data is stored in a line. 39

Glossary
Linear search A searching algorithm in which each item is checked one-by-one. 62
Link A physical connect between two devices. 190
Link layer Deals with the physical medium used for transferring the data. 201
List A data structure similar to an array, commonly used in Python in place of an array. 34
Load Fetch a value from memory into a register in the processor. 165
Local variables Variables that are declared within a subroutine and can only be accessed by the 19
subroutine during the execution of that subroutine.
Logarithmic The opposite of exponential. If y = 2x, then x = log2y. 92
Logic circuit A solution to a problem that uses one or more logic gates. 147
Logic gate Device which takes one or more binary outputs and produces a single binary output. 145
Logic problem A puzzle which is intended to be solved using logical reasoning. 71
Logical Operations using computation logic such as AND, OR and NOT. 162
Logical AND The operation of applying an AND function between the individual bits in two binary 205
numbers.
Logical bus network A star network that uses a hub as the central device, which causes it to operate like a 191
topology bus network.
Logical shift An operation involving moving each bit in a given direction. 169
Logical shift left Moving all of the bits left, doubling the value each time. 169
Logical shift right Moving all of the bits to the right, halving the value each time. 169
Lossless Reducing the size of a file without the loss of any data. 48, 132
compression
Lossy compression Reducing the size of a file by permanently removing some data. 132
Low-level language A programming language which describes exactly how to interact with the computer’s 141
hardware.
MAC address A physical address, uniquely assigned to each piece of network hardware. 193
Machine code Binary code instructions which can be read and understood by a processor. 141, 162
Main memory Memory that can be directly accessed by the CPU. 1258
Majority voting Transmitting each bit an odd number of times in order to identify and correct any 124
transmission errors.
Malware Malicious software. 200
Mantissa The part of a floating point binary number which provides the value of that number. 115
Map A higher-order function that applies a given function to each item in a list. 230
Matrix A rectangular, two-dimensional collection of values. 35
Mealy machine A finite state machine in which each input has a corresponding output as well as a 85
transition between states.
Media Access A hardware, or physical, address that does not change. 203
Control (MAC)
address
Memory The location where instructions and data are stored in a computer. 15
Merge sort A sorting algorithm in which items are split into single items and then merged into 66
sorted pairs, fours, eights and so on.
Mesh topology A network topology in which each device can be directly linked to several other 195
devices.
Metadata Data about data. Additional data stored in a file. 128
Methods Processes or actions that an object of that type can do, implemented using 23
subroutines.
Modelling Graphical descriptions such as flowcharts, E-R diagrams, class diagrams and hierarchy 236
charts. 249

My Revision Notes AQA A-level Computer Science Third Edition

325487_Gloss_MRN_AQA_CS_241-256.indd Page 250 6/29/21 10:11 PM f-0116 /103/HO02206/work/indd

Modular Independent of other subroutines. 20, 238

Module A file that contains one or more subroutines that can be imported and used in another 19
program.
Glossary

Modulo Finding the remainder when one number is divided by another. 124
Modulus The remainder of the division of one number by another. 14
Moral Relating to the principles of right and wrong. 183
Most significant bit The left-most bit in a binary number. The bit with the largest place value. 105
(MSB)
Motherboard The circuit board to which all other components are connected. 158
Musical Instrument A type of file that uses the instructions for how to create each note rather than 131
Digital Interface capturing the analogue sound wave.
(MIDI)
Named constant A variable whose value cannot be changed while the program is running. 16
NAND flash Memory which can be electrically erased and reprogrammed. 177
memory
Natural numbers Positive integers, including 0. Numbers used to count things. 103
Nested Placing a programming structure inside another programming structure. 12, 35
Network address Substituting a private IP address for a public IP address, or vice versa. 207
translation (NAT)
Network identifier The first part of an IP address, indicating the network. 205
Network interface The component in any device that physically connects to the network. 203
card (NIC)
Network layer Adds sender and receiver IP addresses. 201
Node Used to represent an item of data, drawn as a circle. Also known as a vertex. 43
Node An item in a graph. 225
Non-computable A problem that cannot be solved by a computer. 96
Non-routable An address that can only be accessed within that network. 207
Non-volatile Data is retained even when electrical power is removed. 158, 176
Normalisation Structuring a database in a way designed to reduce data redundancy and improve 217
integrity.
Normalised The standard way of writing floating point binary, in which the first two bits are 119
opposite to each other.
Number base The number of digits available in that number system. 104
Nyquist’s theorem The rule that the sample rate should be at least double the maximum frequency that 130
can be heard.
Object A specific instance of a class. 24
Object code Low-level code, translated from the source code. 143
Object-oriented An approach to solving a problem using a model based on real-world objects that are 22
programming designed to interact in a realistic way.
Odd parity A method of using parity in which each block of data should have an odd number of 123
1s.
One-time pad The key used in the Vernam cipher. 134
Opcode An operation code, describing what operation is to be carried out. 164
Operand A value or object that is to be operated on. 60, 164
Operation A function to be applied; for example, +, –, AND, OR. 60
Operators Symbols used to indicate a function. 14
Optical disk A storage medium which uses light to read the data. 177
Ordinal numbers Numbers used to count the order that something appears; for example, 1st, 2nd, 3rd, 104
and so on.
Overflow Where the result of a calculation is too large to be stored in the available digits. 120, 169
250 Overhead Additional data added to the original values. 123

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_Gloss_MRN_AQA_CS_241-256.indd Page 251 6/29/21 10:11 PM f-0116 /103/HO02206/work/indd

Overriding A method in a subclass re-defining a method inherited from a base class. 27

Packet A small parcel of data, including a header. 195
Packet filtering Stopping packets based on their IP address or protocol. 198

Glossary
Packet switching The process of sending individual packets on different routes. 195
Page A subdivision of a block. 177
Paradigm A particular style or approach to designing a solution to a problem. 22
Parallel data Data is sent on several wires, simultaneously 188
transmission
Parameters The variables that a subroutine needs in order for the subroutine to run. 19
Parity bit A single binary digit added to some data in order to help with error checking. 123, 188
Parse Process a string of symbols or letters. 210
Partial function Providing some of a function’s arguments and producing a function that takes the 228
application reminder of the arguments.
Pass The transfer of a value, or the value of a variable, to a subroutine. 18
Pass Travelling through a list from start to finish exactly once. 65
Path The location of the file or folder on a server that is being addressed. 196
Peer Of equal standing, able to act as a client or a server. 192
Peer-to-peer A network in which all devices are peers. 192
network
Peripheral External hardware devices, on the periphery of the computer system. 139
Permutation One of the different ways that a set can be arranged. 93
Personal data Data that relates to an identifiable person. 183
Pits Small indentations in the track. 177
Pixel The smallest addressable element of an image. 127, 174
Plaintext The original, unencrypted message. 133
Platter A metal disk used to store the data in a HDD. 176
Pointer A value that stores an address. In the context of queues this is usually the index of the 39
front or rear item.
Polygon A shape made of straight lines, for example, rectangle, hexagon, and so on. 128
Polymorphism Literally ‘many forms’ – the ability for two methods with the same name to carry out 27
their actions using different code.
Polynomial An expression involving powers; for example, y = 2x 2. 92
Port forwarding Using the port number of a packet to identify which private IP address should be used. 208
Port number Extra data added to a packet that identifies what application layer protocol should be 201
used to process the data.
Post-order tree Inspecting or displaying the value of each node after checking for all child nodes. 57
traversal
Postfix/Reverse A method of writing mathematical or logical expressions in which the operators 60
Polish notation appear after their operands; for example, 2 2 +.
Pre-order tree Inspecting or displaying the value of each node before moving on to other nodes. 57
traversal
Prepend Add something to the start of a list. 232
Primary key An attribute used to uniquely identify a record. 216
Printer A device for creating a hard copy of a document, usually on paper. 175
Priority queue A queue which stores a priority for each value so that the items with the highest 39
priority can be accessed first.
Private An access specifier that protects that attribute or method from being accessed by any 25
other object.
Private IP address An address that can only be accessed within that network. 206
Private key A key that is kept secret. Has a matching public key. 198
251

My Revision Notes AQA A-level Computer Science Third Edition

325487_Gloss_MRN_AQA_CS_241-256.indd Page 252 6/29/21 10:11 PM f-0116 /103/HO02206/work/indd

Procedural-oriented An approach to solving a problem using subroutines to tackle smaller subproblems. 22

programming
Procedure A subroutine that does not return a value. 19
Glossary

Production rule The set of acceptable inputs for a given symbol. 89

Protected An access specifier that protects that attribute or method from being accessed by 25
other objects unless they are instances of that class or a subclass.
Protocol A set of rules or standards. 189, 196
Proxy server A device that sits between a private and a public network. 198
Pseudo-code A format for program code that is not specific to one programming language. Pseudo- 10
code is useful for describing an algorithm that could be coded in one of several
different languages.
Pseudo-code A form of program design that uses code which is not specific to one programming 237
language.
Public An access specifier that allows that attribute or method to be accessed by any other 25
object.
Public IP address An address that can be accessed from any device on the internet. 206
Public key A key that is made public. Has a matching private key. 198
Quad-core A processor containing four CPUs, both working simultaneously. 172
Queue A data structure in which items are stored in the order in which they should be 39
accessed.
RAM Short term storage for currently running programs and currently used data. 158
Rational numbers Any number that can be represented as a fraction, including an integer. 103
Read/write head The device used to read and write magnetic data to each to sector. 176
Real numbers The collection of all rational and irrational numbers. 103
Record A collection of related data. 37, 216
Recursion The process of a function repeatedly calling itself. 22
Recursive An algorithm that uses a sub-routine which calls itself. 66, 80
Redundancy Not needed, for example data that is unnecessarily stored twice. 217
Registers Small areas of memory inside the processor. 162
Regular expression A sequence of characters used to describe a pattern. Used in searching and 88
validation.
Rehashing The process of recalculating the keys for all existing data using a new hashing 47
algorithm.
Relational database A database made up of two or more, related tables. 216
Relationship The way in which two entities relate to each other. 214
Relative error The percentage difference between the intended value and the actual value. 118
Resolution The number of pixels per unit of length, also known as pixel density, often measured 127
in PPI or DPI.
REST Representational state transfer. A software style that allows web services to interact. 209
Return To pass a value or the contents of a variable back to the place in the program where 19
the function was called.
RFID Radio frequency identification. 174
RFID scanner A device for reading the data from an RFID tag. 174
RFID tag A small device capable of emitting data using radio frequencies. 174
Rising edge The point at which the clock signal changes from 0 to 1. 151
Robustness How likely the device is to break or be damaged. 178
ROM Long term storage for start-up instructions. 158
Rooted tree A tree with a rooted node, from which all edges leading away from that root. 45
Rounding Reducing the number of digits used to represent a value while maintaining its 14
approximate value.
252 Routable An address that can be accessed from any device on the internet. 207

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_Gloss_MRN_AQA_CS_241-256.indd Page 253 6/29/21 10:11 PM f-0116 /103/HO02206/work/indd

Router A network hardware device for connecting different networks together. 195
RTS/CTS Request to send/clear to send. A method of collision avoidance by requesting 194
clearance to transmit.

Glossary
Run A series of identical values in a file or dataset. 132
Run length A lossless compression technique of recording the length and value of each run in the 132
encoding (RLE) data.
Sample rate The rate at which samples are taken, typically measured in kHz. 130
Sampling resolution The number of bits per sample. 130
Scalable Used to describe a product, service or business that can cope with increased 183
demand.
Scalar A single number used to represent or adjust the scale of a vector. 48
Scalar-vector Multiplying each element of a vector by a number in order to increase the scale of a 49
multiplication vector.
Scaling Changing the scale of a vector by multiplying the vector by a scalar. 49
Scope The visibility of variables (either local or global). 19
Secondary storage Persistent, non-volatile storage. 176
Sector A small section of a track. 176
Selection A program structure that makes a decision about which block of code to execute, 11, 72,
typically implemented using an IF statement. 166
Self-documenting Program code that uses naming conventions and programming conventions that 13
code help other programmers to understand the code without needing to read additional
documentation.
Sequence Executing instructions in the order they are written. 72
Serial data Data is sent down one wire, one bit after another 188
transmission
Server A device which controls centralised access to a resource. 191, 208
Set A collection of objects. 227
Set comprehension A collection of rules to define which values are in a set. 86
Set theory A branch of mathematical logic in which sets are collections of objects. 86
Setter A procedure used to allow another object to set or update the value of an attribute. 25
Shortest path The route between two nodes on a graph that incurs the least cost. In an unweighted 67
graph the cost for each path can be assumed to be 1.
Signal The electric or electromagnetic impulses that are used to transmit data. 126
Single-core A processor containing a single CPU. 172
Size The pixel dimensions measured as the width x height of an image in pixels. 127
Socket The endpoint, or final destination, of a packet. Described using both the IP address 203
and port number.
Software The programs that run on a computer system. 139
Solid state disk A storage device with no moving parts that uses flash memory and a controller. 177
drive (SSD)
Source code Code written by a programmer. 143
Space-wise A measure of how much memory will be required to complete an algorithm. 91
complexity
Sparse graph A graph with few edges. 44
SSID Service set identifier – an identifier, or name, for a wireless network. 193
Stack A data structure in which items are added to the top and removed from the top, much 41, 171
like a stack of plates.
Stack frame The collection of data associated with a subroutine call. 20
Stakeholders The people with an interest in the problem or the solution. 235
Star topology A network arranged with a switch (or hub) at the centre. 190
Start bit A bit sent at the start of a message in order to provide timing data. 188 253

My Revision Notes AQA A-level Computer Science Third Edition

325487_Gloss_MRN_AQA_CS_241-256.indd Page 254 6/29/21 10:11 PM f-0116 /103/HO02206/work/indd

Starting state The state a Turing machine is in when it starts its program. 96
State transition A diagram showing the states, inputs and transitions in a FSM. 84
diagram
Glossary

State transition A table showing the states, inputs and transitions in a FSM. 84
table
Stateful inspection Inspecting the data contained in a packet. 198
Statelessness A system in which the processing does not depend on the state of another part of the 225
program.
Static The size of, and the memory assigned to that data structure is fixed and cannot be 38
changed.
Static method A method within a class that can be called without the need to create an object of 29
that class.
Stop bit A bit sent to mark the end of a message. 188
Store Copy a value from a register in the processor and save it in memory. 165
Subclass (or child A class that inherits the attributes and methods from another class. 25
class)
Subnet A subdivision of a network. 205
Subnet mask A 32-bit number used to isolate the network identifier in an IP address. 205
Subroutine A named block of code designed to carry out one specific task. 18
Subset A selection of the values found in a set. 87
Substitution cipher A cipher in which each plaintext letter is replaced with a ciphertext letter. 133
Substring A series of characters taken from a string. 17
SWITCH An alternative selection structure to an IF statement that is slightly quicker to execute 11
if used with exact values rather than a range.
Switch A device that receives and forwards data on a network. 190
Syllogism An argument that uses logical reasoning. 71
Symmetric The same key is used in encryption and decryption. 135
Symmetric The same key is used to encrypt and to decrypt a message. 198
encryption
Synchronous data Where data is sent along with a timing signal. 188
transmission
Syntax The strict rules and structures used within a specific programming language. 10, 89
Syntax diagram A diagram used to describe BNF graphically. 90
System software Software intended to allow the computer system to run. 139
Table The representation of an entity within a database, that is a structure for storing 216
attributes.
Tail The remainder of a list, without its head. 232
TCP/IP stack The use of the TCP/IP layers to add header data which is then processed in reverse 201
order.
Terminal A single value that cannot be broken down into smaller parts. 89
Testing The fourth stage of systems development, in which the program is tested to ensure it 239
functions as it should.
Text file A file that uses text encoding (such as ASCII or Unicode) to store data. 37
Thick-client A significant amount of processing is done on the client side. 211
Thin-client Almost all processing is done on the server side. 211
Third normal form A database in which all data is atomic, there are no partial dependencies and no non- 217
key dependencies.
Time-wise A measure of how much time, or how many steps, will be required to complete an 91
complexity algorithm.
Toner A powdered form of ink with a static charge. 175

254 Top level design An outline design that shows how the whole program will fit together. 237

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads

325487_Gloss_MRN_AQA_CS_241-256.indd Page 255 6/29/21 10:11 PM f-0116 /103/HO02206/work/indd

Top level domain The right-most term in a domain name, such as .com or .uk 196
(TLD)
Top-down approach A method of planning solutions that starts with the big picture and breaks it down into 22
smaller subproblems.

Glossary
Topology The physical or logical arrangement of connections in a network. 190
Trace table A table for recording the values of variables as a program runs. 72, 170
Track A concentric ring on a platter. 176, 177
Tractable A problem which can be completed in a reasonable amount of time (polynomial time 95
or less).
Translation Moving a vector by adding another vector to it. 49
Translator Systems software for converting one form of code into object code. 143
Transmission One of the main protocols used in network communications. 201
Control Protocol /
Internet Protocol
(TCP/IP)
Transport layer Adds a port number, packet number and error detection data. 201
Tree A graph which has no cycles (it is not possible to loop back around to a previously 45
traversed part of the tree).
Tree traversal Inspecting the items stored in a tree data structure. 57
Trojan A malicious program that pretends to be a useful program, does not self-replicate. 200
Truncation Removing any value after a certain number of decimal places. 14
Truth table A table showing the possible inputs and their corresponding outputs. 145
Turing machine A model of computation with a single program which manipulates symbols on a strip 96
of tape.
Two’s complement A representation of binary numbers that includes both positive and negative values. 111
Underflow Where the result of a calculation is too small to be stored in the available digits. 121, 169
Uni-directional Data only travels in one direction. 160
Unicode A variable size character set in which each character code can be 8, 16 or 32 bits in 122
length.
Uniform resource A standard structure for addressing an online resource using alphanumeric strings. 196
locator (URL)
Universal Turing A Turing machine that takes the description of another Turing machine and its tape as 98
machine (UTM) inputs. It can simulate any conceivable Turing machine.
Unsigned binary A system that is only capable of representing positive numbers (0 or larger). 109
User interface The tools that are provided for a user to interact with a computer system. 140, 236
Variety Data in many forms; that is, unstructured. 224
Vector A data structure used to represent a position (for example, a two-dimensional vector 48
represents a position on a 2D plane).
Vector addition Adding two vectors together in order to perform a translation. 49
Vector graphic An image made of lines and shapes. 128
Vector primitive Simple objects which can be combined to create a vector graphic. 128
Velocity The rate at which data is generated or collected. 224
Vernam cipher A cipher which uses a randomly generated key and is mathematically unbreakable. 134
Virtual machine Software that emulates or simulates a computer system. 145
Virtual memory A portion of secondary storage used to store the least frequently-used instructions 158
and data.
Virtual method A method that may be overridden (this is the default in many languages). 29
Virus Malicious code that attaches to another program and self-replicates when executed. 200
Volatile Data is lost when electrical power is removed. 158, 176
Volatile memory Memory which can only store a value when supplied with power. 151
Volume The quantity of data to be stored, typically too much to fit on a single server. 224 255

My Revision Notes AQA A-level Computer Science Third Edition

325487_Gloss_MRN_AQA_CS_241-256.indd Page 256 6/29/21 10:11 PM f-0116 /103/HO02206/work/indd

Websocket A communications protocol using TCP to create a fullduplex connection. 209

Weighted graph A graph in which each edge has a value or cost associated with it. 43
Whitelist A list of things considered to be acceptable or trustworthy. 193
Glossary

Wi-Fi A set of technology standards that allows devices to communicate using radio 193
frequencies.
Wireless access A hardware device for allowing other devices to connect to an existing wireless 193
point network.
Wireless network A network that allows devices to transmit data using radio frequencies. 193
Wireless network A hardware device for enabling devices to communicate using Wi-Fi. 193
adapter
Word The maximum number of bits that can be processed in one instruction. 173
Worm A malicious program that copies itself over a network. 200
XOR A logical operation which checks whether both inputs are equal, or different. 134

256

Check your understanding and progress at www.hoddereducation.co.uk/myrevisionnotesdownloads