Discrete Mathematical Structures

Download as pdf or txt
Download as pdf or txt
You are on page 1of 275

Authored by B. V.

Senthil Kumar and Hemen Dutta

Discrete Mathematical
Structures
Mathematics and Its Applications: Modelling,
Engineering, and Social Sciences
Series Editor: Hemen Dutta

Discrete Mathematical Structures


A Succinct Foundation
B. V. Senthil Kumar and Hemen Dutta

Concise Introduction to Logic and Set Theory


Iqbal H. Jebril and Hemen Dutta

Tensor Calculus and Applications


Simplified Tools and Techniques
Bhaben Kalita

For more information about this series, please visit:


www.crcpress.com/Mathematics-and-its-applications/book-series/MES
Authored by B. V. Senthil Kumar and Hemen Dutta

Discrete Mathematical
Structures
A Succinct Foundation
CRC Press
Taylor & Francis Group
6000 Broken Sound Parkway NW, Suite 300
Boca Raton, FL 33487-2742
c 2020 by Taylor & Francis Group, LLC

CRC Press is an imprint of Taylor & Francis Group, an Informa business

No claim to original U.S. Government works


Printed on acid-free paper
International Standard Book Number-13: 978-0-367-14869-0 (Hardback)

This book contains information obtained from authentic and highly regarded sources.
Reasonable efforts have been made to publish reliable data and information, but the
author and publisher cannot assume responsibility for the validity of all materials or the
consequences of their use. The authors and publishers have attempted to trace the copyright
holders of all material reproduced in this publication and apologize to copyright holders if
permission to publish in this form has not been obtained. If any copyright material has not
been acknowledged, please write and let us know so we may rectify in any future reprint.
Except as permitted under U.S. Copyright Law, no part of this book may be reprinted,
reproduced, transmitted, or utilized in any form by any electronic, mechanical, or other
means, now known or hereafter invented, including photocopying, microfilming, and
recording, or in any information storage or retrieval system, without written permission
from the publishers.

For permission to photocopy or use material electronically from this work, please access
www.copyright.com (http://www.copyright.com/) or contact the Copyright Clearance
Center, Inc. (CCC), 222 Rosewood Drive, Danvers, MA 01923, 978-750-8400. CCC is a
not-for-profit organization that provides licenses and registration for a variety of users. For
organizations that have been granted a photocopy license by the CCC, a separate system
of payment has been arranged.
Trademark Notice: Product or corporate names may be trademarks or registered
trademarks, and are used only for identification and explanation without intent to infringe.

Library of Congress Cataloging-in-Publication Data


Names: B. V. Senthil Kumar, author. | Dutta, Hemen, 1981- author.
Title: Discrete mathematical structures : a succinct foundation / by B.V.
Senthil Kumar and Hemen Dutta.
Description: Boca Raton, FL : CRC Press/Taylor & Francis Group, 2020. |
Series: Mathematics and its applications : modelling, engineering, and
social sciences
Identifiers: LCCN 2019009359 | ISBN 9780367148690 (hardback : alk. paper) |
ISBN 9780429053689 (ebook)
Subjects: LCSH: Discrete mathematics
Classification: LCC QA297.4 .K86 2020 | DDC 518/.25–dc23
LC record available at https://lccn.loc.gov/201900935

Visit the Taylor & Francis Web site at


http://www.taylorandfrancis.com

and the CRC Press Web site at


http://www.crcpress.com
Contents

Preface ix
Authors xi

1 Logics and Proofs 1


1.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Proposition . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.3 Compound Propositions . . . . . . . . . . . . . . . . . . . . . 1
1.4 Truth Table . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.5 Logical Operators . . . . . . . . . . . . . . . . . . . . . . . . 2
1.5.1 Negation . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.5.2 Conjunction . . . . . . . . . . . . . . . . . . . . . . . . 2
1.5.3 Disjunction . . . . . . . . . . . . . . . . . . . . . . . . 3
1.5.4 Molecular Statements . . . . . . . . . . . . . . . . . . 3
1.5.5 Conditional Statement [If . . . then] [ → ] . . . . . . . 3
1.5.6 Biconditional [If and only if or iff] [ ↔ or ] . . . . . 4
1.5.7 Solved Problems . . . . . . . . . . . . . . . . . . . . . 4
1.5.8 Tautology . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.5.9 Contradiction . . . . . . . . . . . . . . . . . . . . . . . 6
1.5.10 Contingency . . . . . . . . . . . . . . . . . . . . . . . 6
1.5.11 Equivalence Formulas . . . . . . . . . . . . . . . . . . 6
1.5.12 Equivalent Formulas . . . . . . . . . . . . . . . . . . . 7
1.5.13 Duality Law . . . . . . . . . . . . . . . . . . . . . . . . 7
1.5.14 Tautological Implication . . . . . . . . . . . . . . . . . 7
1.5.15 Some More Equivalence Formulas . . . . . . . . . . . . 7
1.5.16 Solved Problems . . . . . . . . . . . . . . . . . . . . . 8
1.6 Normal Forms . . . . . . . . . . . . . . . . . . . . . . . . . . 9
1.6.1 Principal Disjunctive Normal Form or Sum
of Products Canonical Form . . . . . . . . . . . . . . . 9
1.6.2 Principal Conjunctive Normal Form or Product
of Sum Canonical Form . . . . . . . . . . . . . . . . . 10
1.6.3 Solved Problems . . . . . . . . . . . . . . . . . . . . . 10
1.7 Inference Theory . . . . . . . . . . . . . . . . . . . . . . . . . 13
1.7.1 Rules of Inference . . . . . . . . . . . . . . . . . . . . 14
1.7.2 Solved Problems . . . . . . . . . . . . . . . . . . . . . 15
1.8 Indirect Method of Proof . . . . . . . . . . . . . . . . . . . . 19
1.8.1 Method of Contradiction . . . . . . . . . . . . . . . . 19
1.8.2 Solved Problems . . . . . . . . . . . . . . . . . . . . . 19

v
vi Contents

1.9 Method of Contrapositive . . . . . . . . . . . . . . . . . . . . 21


1.9.1 Solved Problems . . . . . . . . . . . . . . . . . . . . . 21
1.10 Various Methods of Proof . . . . . . . . . . . . . . . . . . . . 21
1.10.1 Trivial Proof . . . . . . . . . . . . . . . . . . . . . . . 21
1.10.2 Vacuous Proof . . . . . . . . . . . . . . . . . . . . . . 22
1.10.3 Direct Proof . . . . . . . . . . . . . . . . . . . . . . . 22
1.11 Predicate Calculus . . . . . . . . . . . . . . . . . . . . . . . . 22
1.11.1 Quantifiers . . . . . . . . . . . . . . . . . . . . . . . . 23
1.11.2 Universe of Discourse, Free and Bound Variables . . . 23
1.11.3 Solved Problems . . . . . . . . . . . . . . . . . . . . . 24
1.11.4 Inference Theory for Predicate Calculus . . . . . . . . 28
1.11.5 Solved Problems . . . . . . . . . . . . . . . . . . . . . 28
1.12 Additional Solved Problems . . . . . . . . . . . . . . . . . . 32

2 Combinatorics 39
2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
2.2 Mathematical Induction . . . . . . . . . . . . . . . . . . . . . 39
2.2.1 Principle of Mathematical Induction . . . . . . . . . . 39
2.2.2 Procedure to Prove that a Statement P (n) is True for
all Natural Numbers . . . . . . . . . . . . . . . . . . . 40
2.2.3 Solved Problems . . . . . . . . . . . . . . . . . . . . . 40
2.2.4 Problems for Practice . . . . . . . . . . . . . . . . . . 56
2.2.5 Strong Induction . . . . . . . . . . . . . . . . . . . . . 57
2.2.6 Well-Ordering Property . . . . . . . . . . . . . . . . . 57
2.3 Pigeonhole Principle . . . . . . . . . . . . . . . . . . . . . . . 58
2.3.1 Generalized Pigeonhole Principle . . . . . . . . . . . . 58
2.3.2 Solved Problems . . . . . . . . . . . . . . . . . . . . . 58
2.3.3 Another Form of Generalized Pigeonhole Principle . . 60
2.3.4 Solved Problems . . . . . . . . . . . . . . . . . . . . . 61
2.3.5 Problems for Practice . . . . . . . . . . . . . . . . . . 69
2.4 Permutation . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
2.4.1 Permutations with Repetitions . . . . . . . . . . . . . 71
2.4.2 Solved Problems . . . . . . . . . . . . . . . . . . . . . 72
2.4.3 Problems for Practice . . . . . . . . . . . . . . . . . . 79
2.5 Combination . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
2.5.1 Solved Problems . . . . . . . . . . . . . . . . . . . . . 81
2.5.2 Problems for Practice . . . . . . . . . . . . . . . . . . 85
2.5.3 Recurrence Relation . . . . . . . . . . . . . . . . . . . 87
2.5.4 Solved Problems . . . . . . . . . . . . . . . . . . . . . 87
2.5.5 Linear Recurrence Relation . . . . . . . . . . . . . . . 88
2.5.6 Homogenous Recurrence Relation . . . . . . . . . . . . 88
2.5.7 Recurrence Relations Obtained from Solutions . . . . 89
Contents vii

2.6 Solving Linear Homogenous Recurrence Relations . . . . . . 90


2.6.1 Characteristic Equation . . . . . . . . . . . . . . . . . 91
2.6.2 Algorithm for Solving k th -order Homogenous Linear
Recurrence Relations . . . . . . . . . . . . . . . . . . . 91
2.6.3 Solved Problems . . . . . . . . . . . . . . . . . . . . . 92
2.7 Solving Linear Non-homogenous Recurrence Relations . . . . 95
2.7.1 Solved Problems . . . . . . . . . . . . . . . . . . . . . 96
2.7.2 Problems for Practice . . . . . . . . . . . . . . . . . . 102
2.8 Generating Functions . . . . . . . . . . . . . . . . . . . . . . 103
2.8.1 Solved Problems . . . . . . . . . . . . . . . . . . . . . 103
2.8.2 Solution of Recurrence Relations Using
Generating Function . . . . . . . . . . . . . . . . . . . 106
2.8.3 Solved Problems . . . . . . . . . . . . . . . . . . . . . 106
2.8.4 Problems for Practice . . . . . . . . . . . . . . . . . . 116
2.9 Inclusion–Exclusion Principle . . . . . . . . . . . . . . . . . . 117
2.9.1 Solved Problems . . . . . . . . . . . . . . . . . . . . . 118
2.9.2 Problems for Practice . . . . . . . . . . . . . . . . . . 131

3 Graphs 135
3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
3.2 Graphs and Graph Models . . . . . . . . . . . . . . . . . . . 135
3.3 Graph Terminology and Special Types of Graphs . . . . . . 138
3.3.1 Solved Problems . . . . . . . . . . . . . . . . . . . . . 140
3.3.2 Graph Colouring . . . . . . . . . . . . . . . . . . . . . 145
3.3.3 Solved Problems . . . . . . . . . . . . . . . . . . . . . 145
3.4 Representing Graphs and Graph Isomorphism . . . . . . . . 149
3.4.1 Solved Problems . . . . . . . . . . . . . . . . . . . . . 151
3.4.2 Problems for Practice . . . . . . . . . . . . . . . . . . 155
3.5 Connectivity . . . . . . . . . . . . . . . . . . . . . . . . . . . 156
3.5.1 Connected and Disconnected Graphs . . . . . . . . . . 158
3.6 Eulerian and Hamiltonian Paths . . . . . . . . . . . . . . . . 161
3.6.1 Hamiltonian Path and Hamiltonian Circuits . . . . . . 163
3.6.2 Solved Problems . . . . . . . . . . . . . . . . . . . . . 164
3.6.3 Problems for Practice . . . . . . . . . . . . . . . . . . 169
3.6.4 Additional Problems for Practice . . . . . . . . . . . . 169

4 Algebraic Structures 173


4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . 173
4.2 Algebraic Systems . . . . . . . . . . . . . . . . . . . . . . . . 173
4.2.1 Semigroups and Monoids . . . . . . . . . . . . . . . . 174
4.2.2 Solved Problems . . . . . . . . . . . . . . . . . . . . . 175
4.2.3 Groups . . . . . . . . . . . . . . . . . . . . . . . . . . 183
4.2.4 Solved Problems . . . . . . . . . . . . . . . . . . . . . 186
4.2.5 Subgroups . . . . . . . . . . . . . . . . . . . . . . . . . 192
4.2.6 Cyclic Groups . . . . . . . . . . . . . . . . . . . . . . . 193
viii Contents

4.2.7 Homomorphisms . . . . . . . . . . . . . . . . . . . . . 195


4.2.8 Cosets and Normal Subgroups . . . . . . . . . . . . . 197
4.2.9 Solved Problems . . . . . . . . . . . . . . . . . . . . . 202
4.2.10 Permutation Functions . . . . . . . . . . . . . . . . . . 208
4.2.11 Solved Problems . . . . . . . . . . . . . . . . . . . . . 211
4.2.12 Problems for Practice . . . . . . . . . . . . . . . . . . 216
4.2.13 Rings and Fields . . . . . . . . . . . . . . . . . . . . . 217
4.2.14 Solved Problems . . . . . . . . . . . . . . . . . . . . . 220
4.2.15 Problems for Practice . . . . . . . . . . . . . . . . . . 222

5 Lattices and Boolean Algebra 223


5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . 223
5.2 Partial Ordering and Posets . . . . . . . . . . . . . . . . . . 223
5.2.1 Representation of a Poset by Hasse Diagram . . . . . 224
5.2.2 Solved Problems . . . . . . . . . . . . . . . . . . . . . 226
5.2.3 Problems for Practice . . . . . . . . . . . . . . . . . . 230
5.3 Lattices, Sublattices, Direct Product, Homomorphism
of Lattices . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231
5.3.1 Properties of Lattices . . . . . . . . . . . . . . . . . . 231
5.3.2 Theorems on Lattices . . . . . . . . . . . . . . . . . . 233
5.3.3 Solved Problems . . . . . . . . . . . . . . . . . . . . . 236
5.3.4 Problem for Practice . . . . . . . . . . . . . . . . . . . 240
5.4 Special Lattices . . . . . . . . . . . . . . . . . . . . . . . . . 240
5.4.1 Solved Problems . . . . . . . . . . . . . . . . . . . . . 242
5.4.2 Problems for Practice . . . . . . . . . . . . . . . . . . 247
5.5 Boolean Algebra . . . . . . . . . . . . . . . . . . . . . . . . . 248
5.5.1 Solved Problems . . . . . . . . . . . . . . . . . . . . . 252
5.5.2 Problems for Practice . . . . . . . . . . . . . . . . . . 255

Bibliography 257
Index 259
Preface

The aim of this book is to provide a concise introduction to some significant


topics covered in the subject Discrete Mathematics. Various themes of discrete
mathematics have extensive applications in several courses in disciplines like
Computer Science, Engineering, and Information Technology. This motivated
us to provide a resource in the form of a textbook for undergraduate
and postgraduate students adopting relevant courses in Computer Science,
Mathematics, Engineering, Information Technology, etc. While in some
courses this book can be adopted as a text book, there are several other
courses where this book will serve the need of a reference book. This book
also will serve as a Handbook for the teachers in class room teaching as it
contains many solved problems and problems for practice which are included
at the end of each subtopic. In order to meet the needs of the learners,
we have included essential topics in this book, such as Logics and Proofs,
Combinatorics, Graphs, Algebraic Structures, Lattices and Boolean Algebra.
In Chapter 1: Logics and Proofs, vital concepts from basic level of
logic and proofs are provided. In Chapter 2: Combinatorics, we have
dealt with the idea of mathematical induction with more number of solved
problems for better understanding of the reader. Also, various other topics
in combinatorics like pigeonhole principle, permutations and combinations,
and recurrence relation are discussed in meticulous approach. In Chapter 3:
Graphs, fundamental notion of graph theory and various types of graphs are
introduced with lot of illustrations. In Chapter 4: Algebraic Structures,
the elementary tools of discrete mathematics such as algebraic structure,
semigroup, monoid, abelian group, subgroup, cosets, Lagrange’s theorem,
normal subgroup, homomorphism of groups, rings and fields are provided with
sufficient theorems with proofs and examples. In Chapter 5: Lattices and
Boolean Algebra, the concepts of partially ordered sets, lattices, Boolean
algebra and their properties are focussed. These concepts are useful in many
different types of computational circuits.
We thank all the authors who have contributed extensively in the field
of Discrete Mathematics and our family members and friends who have
motivated us to bring our objective in the form of a textbook. We are indebted
to the Management of Nizwa College of Technology, Nizwa, Oman for their
constant support and encouragement during the preparation of this book. We
would like to thank the editors in charge for this book project and supporting

ix
x Preface

staff at CRC Press, Taylor & Francis Group, for their timely cooperation in
publishing this book. We also welcome productive suggestions and comments
to improve the quality of the book for next edition.

B. V. Senthil Kumar, Nizwa, Oman


Hemen Dutta, Guwahati, India
Authors

B. V. Senthil Kumar is serving in the Section of Mathematics, Department


of Information Technology, Nizwa College of Technology, Nizwa, Oman. His
areas of interest are solution and stability of functional, differential, and
difference equations; operations research, statistics; and discrete mathematics.
He obtained his Ph.D. Degree in 2015 and has 18 years of teaching and research
experience. He has published more than 40 research papers in National and
International reputed journals. He has authored the books titled Functional
Equations and Inequalities: Solution and Stability Results (published by
World Scientific Publishing Company) and Probability & Queueing Theory
(published by KKS Publishers, Chennai, India) to his credit. He has also
contributed some book chapters. He has delivered invited talks in various
institutions and also organised many academic and non-academic events. He
is a member of many mathematical societies.

Hemen Dutta is a faculty member at the Department of Mathematics,


Gauhati University, India. He did his Master of Science in Mathematics, Post
Graduate Diploma in Computer Application, and Ph.D. in Mathematics from
Gauhati University, India. He received his M.Phil in Mathematics from
Madurai Kamaraj University, India. He currently teaches subjects like
real analysis, functional analysis, algebra, mathematical logic, computer
applications, etc. His primary research interest includes areas of mathematical
analysis. He has to his credit several research papers, some book chapters,
and few books. He has delivered talks at different institutions and organised a
number of academic events. He is a member of several mathematical societies.

xi
1
Logics and Proofs

1.1 Introduction
In this chapter, we discuss propositional logic and various methods of proving
validity of propositions. The concept of logic has many applications in
computer science to develop computer programs, to verify the logic of program
and also in electronics to design circuits.

1.2 Proposition
A proposition (or statement) is a declarative sentence which is true or false,
but not both. Consider, for example,
(i) The year 2000 is a leap year.
(ii) 5 + 3 = 7.
(iii) x = 1 is a solution of x3 = 1.
(iv) Close the door.
In the above, (i)–(iii) are propositions, whereas (iv) is not a proposition.
Moreover, (i) and (iii) are true, while (ii) is false.

1.3 Compound Propositions


Many propositions are composite, that is, composed of subpropositions and
various connectives discussed in the next section. Such composite propositions
are called compound propositions. A proposition is said to be primitive if it
cannot be broken into smaller propositions, that is, if it is not composite.

Examples:
(i) Apples are red, and milk is white.
(ii) Jack is brilliant or is a hardworking student.

1
2 Discrete Mathematical Structures

Note:
The truth value of a compound proposition is obtained by the truth values
of its subpropositions together with the way in which they are connected to
form the compound propositions.

1.4 Truth Table


A truth table lists all possible combinations of truth values of the propositions
in the left most column and the truth values of the resulting propositions in
the right most column.

1.5 Logical Operators


1.5.1 Negation
If P is a statement, then negation of P written as ¬P or ∼ P is read as “Not
P ”. The truth table for the operator “negation” is shown below.
Negation
P ¬P
T F
F T

Example:
P : Apple is red.
¬P : Apple is not red.

1.5.2 Conjunction
The conjunction of two statements P and Q is the statement P ∧ Q which
is read as “P and Q”. The statement P ∧ Q has a truth value T whenever
both P and Q have the truth value T ; otherwise, it has a truth value F . The
conjunction is defined by the truth table below.

Conjunction
P Q P ∧Q
T T T
T F F
F T F
F F F
Logics and Proofs 3

Example:
P : John worked hard.
Q: John passed the examination.
P ∧ Q: John worked hard, and he passed the examination.

1.5.3 Disjunction
The disjunction of two statements P and Q is the statement P ∨ Q and has a
truth value F only when both P and Q have truth value F ; otherwise, it has
a truth value T . The disjunction is defined by the truth table shown below.
Disjunction
P Q P ∨Q
T T T
T F T
F T T
F F F

Example:
P : 2 + 4 = 6 (T ).
Q: 2 > 10 (F ).
P ∨ Q: 2 + 4 = 6 or 2 > 10 is true.

1.5.4 Molecular Statements


The statements that contain one or more atomic statements and some
connectives are called molecular statements.
Examples: ¬P , P ∧ ¬Q, ¬P ∨ ¬Q, etc.

1.5.5 Conditional Statement [If . . . then] [ → ]


If P and Q are any two statements, then the statement P → Q which is
read as “If P then Q” is called a conditional statement. Here, P is called
“antecedent”, and Q is called “consequent”. The truth table is shown below.
If . . . then
P Q P →Q
T T T
T F F
F T T
F F T

Note:
P → Q has a truth value F if P has the truth value T and Q has the truth
value F . In all the remaining cases, it has the truth value T .
4 Discrete Mathematical Structures

Example:
P : It is hot.
Q: 2 + 3 = 5.
P → Q: If it is hot, then 2 + 3 = 5.

1.5.6 Biconditional [If and only if or iff] [ ↔ or ]


If P and Q are any two statements, then the statement P ↔ or P Q which
is read as “P if and only if Q” is called biconditional statement. The statement
P ↔ Q has the truth value T whenever both P and Q have identical truth
values. The truth table is shown below.
If and only if
P Q P ↔Q
T T T
T F F
F T F
F F T

Example:
P : John is rich.
Q: John is happy.
P ↔ Q: John is rich if and only if he is happy.

1.5.7 Solved Problems


1. Give the contrapositive statement of the statement “If there is rain,
then I buy an umbrella”.
Solution.
Let P : “There is rain” and Q: “I buy an umbrella”.
Then the given statement is P → Q.
Its contrapositive is ¬Q → ¬P .
2. Construct the truth table for P → ¬Q.
Solution.
The truth table is shown below.

Truth Table for P → ¬Q


P Q ¬Q P → ¬Q
T T F F
T F T T
F T F T
F F T T
Logics and Proofs 5

3. Find the truth table for P → Q.


Solution.
The truth table is shown below.

Truth Table for P → Q


P Q P →Q
T T T
T F F
F T T
F F T

4. Construct the truth table for the compound proposition


(P → Q) ↔ (¬P → ¬Q).
Solution.
The truth table is shown below.

Truth Table for (P → Q) ↔ (¬P → ¬Q)


P Q P → Q ¬P ¬Q ¬P → ¬Q (P → Q) ↔ (¬P → ¬Q)
T T T F F T T
T F F F T T F
F T T T F F F
F F T T T T T

5. What are the contrapositive, the converse, and the inverse of the
following conditional statement?
“If you work hard, then you will be rewarded”.
Solution.
P : You work hard.
Q: You will be rewarded.
¬P : You will not work hard.
¬Q: You will not be rewarded.
Converse: Q → P : If you will be rewarded, then you work hard.
Contrapositive: ¬Q → ¬P : If you will not be rewarded, then you
will not work hard.
Inverse: ¬P → ¬Q: If you will not work hard, then you will not
be rewarded.
6. Construct a truth table for the compound proposition
(P → Q) → (Q → P ).
Solution.
The truth table is shown below.
6 Discrete Mathematical Structures

Truth Table for (P → Q) → (Q → P )


P Q P → Q Q → P (P → Q) → (Q → P )
T T T T T
T F F T T
F T T F F
F F T T T

1.5.8 Tautology
A statement formula which is true regardless of the truth values of the
statements which replace the variables in it is called a tautology or a
universally valid formula or a logical truth.
Example: P ∨ ¬P is a tautology.

1.5.9 Contradiction
A statement formula which is false regardless of the truth values of the
statements which replace variables in it is called a contradiction.
Example: P ∧ ¬P is a contradiction.

1.5.10 Contingency
A statement formula which is neither tautology nor contradiction is called
contingency.
Example: P → Q is a contingency.

Note: To determine whether a given formula is a tautology or a contradiction,


construct the truth table. But this process is very lengthy since the truth table
will have 2n rows for n statements.

1.5.11 Equivalence Formulas


Let A and B be two statement formulas, and let p1 , p2 , . . . , pn denote all the
variables occurring in both A and B. If the truth value of A is equal to the
truth value of B for every one of the 2n possible sets of truth values assigned
to p1 , p2 , . . . , pn , then A and B are said to be equivalent.
Assuming that the variables and assignment of truth values to the variables
appear in the same order in the truth tables of A and B, the final columns in
the truth tables for A and B are identical if A and B are equivalent.
Examples:
(i) ¬¬P is equivalent to P .
(ii) P ∨ ¬P is equivalent to Q ∨ ¬Q.
Logics and Proofs 7

Remark: We know that A B is true whenever A and B have identical


truth values. This means A is equivalent to B (⇔) if and only if A B is a
tautology.

1.5.12 Equivalent Formulas


(1) P ∨ ¬P , P ∧ ¬P [Idempotent laws]
(2) (P ∨ Q) ∨ R ⇔ P ∨ (Q ∨ R), (P ∧ Q) ∧ R ⇔ P ∧ (Q ∧ R)
[Associative laws]
(3) P ∨ Q ⇔ Q ∨ P , P ∧ Q ⇔ Q ∧ P [Commutative laws]
(4) P ∨ (Q ∧ R) ⇔ (P ∨ Q) ∧ (P ∨ R), P ∧ (Q ∨ R) ⇔ (P ∧ Q) ∨ (P ∧ R)
[Distributive laws]
(5) ¬(P ∨ Q) ⇔ ¬P ∧ ¬Q, ¬(P ∧ Q) ⇔ ¬P ∨ ¬Q
[De Morgan’s laws]

1.5.13 Duality Law


Two formulas A and A? are duals of each other if either one can be obtained
from the other by replacing ∧ by ∨ and ∨ by ∧. The connectives ∨ and ∧ are
also called duals of each other.
Examples:
Write the duals of (i)(P ∨ Q) ∧ R, (ii)(P ∧ Q) ∨ T.
The duals are (i)(P ∧ Q) ∨ R, (ii)(P ∨ Q) ∧ T.

1.5.14 Tautological Implication


A statement A is said to tautologically imply a statement B if A → B is a
tautology. We use the notation ⇒.
Remark: To prove P ⇒ Q, we assume P to be true and prove Q to be true.
Otherwise, assume Q to be false, and prove P to be false also. Construction
of truth table is another method for proving the implication.

1.5.15 Some More Equivalence Formulas


1. P ∧ ¬P ⇔ F , P ∨ ¬P ⇔ T [Complement laws]
2. P ∨ T ⇔ T , P ∧ F ⇔ F [Dominance laws]
3. P ∧ T ⇔ P , P ∨ F ⇔ P [Identity laws]
4. P ∨ (P ∧ Q) ⇔ P , P ∧ (P ∨ Q) ⇔ P [Absorption laws]
5. ¬(¬P ) ⇔ P [Double Negation law]
6. P → Q ⇔ ¬Q → ¬P [Contrapositive law]
7. P → Q ⇔ ¬P ∨ Q [Conditional as disjunction]
8. P Q ⇔ (P → Q) ∧ (Q → P ) [Biconditional as conjunction]
8 Discrete Mathematical Structures

1.5.16 Solved Problems


1. Using truth table, show that the proposition P ∨ ¬(P ∧ Q) is a
tautology.
Solution.
The truth table is shown below. Since all the entries in the last
column are T , the given proposition is a tautology.
Truth Table of P ∨ ¬(P ∧ Q)
P Q P ∧ Q ¬(P ∧ Q) P ∨ ¬(P ∧ Q)
T T T F T
T F F T T
F T F T T
F F F T T

2. Express A ↔ B in terms of the connectives {∧, ¬}.


Solution.
A ↔ ⇔ (A → B) ∧ (B → A)
⇔ (¬A ∨ B) ∧ (¬B ∨ A).

3. Show that (p → r)∧(q → r) and (p∨q) → r are logically equivalent.


Solution.
(p → r) ∧ (q → r)
⇔(¬p ∨ r) ∧ (¬q ∨ r) (conditional as disjunction)
⇔(¬p ∧ ¬q) ∨ r (Distributive law)
⇔¬(p ∨ q) ∨ r (De Morgan’s law)
⇔(p ∨ q) → r (conditional as disjunction).

4. Is (¬p ∧ (P ∨ q)) → q is a tautology.


Solution.
(¬p ∧ (p ∨ q)) → q
⇔(¬p ∧ p) ∨ (¬p ∧ q) → q (Distributive law)
⇔F ∨ (¬p ∧ q) → q [p ∧ ¬p ⇔ F ]
⇔(¬p ∧ q) → q [p ∨ F ⇔ p]
⇔¬(¬p ∧ q) ∨ q [p → q ⇔ ¬p ∨ q]
⇔(p ∨ ¬q) ∨ q (De Morgan’s law)
⇔p ∨ q ∨ ¬q (Associative law)
⇔p ∨ T [p ∨ ¬p ⇔ T ]
⇔T.
Therefore, the given statement is a tautology.
Logics and Proofs 9

5. Show that the propositions p → q and ¬p∨q are logically equivalent.


Solution.
From the truth table, p → q and ¬p ∨ q are equivalent.

Truth Table of p → q and ¬p ∨ q


p q ¬p p → q ¬p ∨ q
T T F T T
T F F F F
F T T T T
F F T T T

1.6 Normal Forms


1.6.1 Principal Disjunctive Normal Form or Sum of
Products Canonical Form
Consider two statements P and Q. Consider a possible formula using
conjunction as follows: P ∧ Q, ¬P ∧ Q, P ∧ ¬Q, ¬P ∧ ¬Q (Duplication is
not allowed and only distinct formulas are considered). We call the above
terms as “minterms”.
For a given formula, an equivalent formula consisting of disjunction of
minterms only is known as Principal Disjunctive Normal Form (PDNF)
or Sum of Products Canonical Form.

Procedure I:
For every truth value T in the truth table of the given formula, select the
minterm which also has the value T for the same combination of the truth
values of P and Q. The disjunction of these minterms will then be equivalent
to the given formula. From the table below, we observe that

Truth Table showing Disjunctions of P and Q


P Q P ∧ Q ¬P ¬Q ¬P ∧ Q P ∧ ¬Q ¬P ∧ ¬Q P → Q P ∨ Q
T T T F F F F F T T
T F F F T F T F F T
F T F T F T F F T T
F F F T T F F T T F

P → Q ⇔ (P ∧ Q) ∨ (¬P ∧ Q) ∨ (¬P ∧ ¬Q)


P ∨ Q ⇔ (P ∧ Q) ∨ (P ∧ ¬Q) ∨ (¬P ∧ Q).
10 Discrete Mathematical Structures

Procedure II:
This is explained in the following example:
P ∨ Q ⇔ ¬P ∧ (Q ∨ ¬Q) ∨ (Q ∧ (P ∨ ¬P )) [since A ∧ T ⇔ A]
⇔ (¬P ∧ Q) ∨ (¬P ∧ ¬Q) ∨ (Q ∧ P ) ∨ (Q ∧ ¬P ) [Distributive laws]
⇔ (¬P ∧ Q) ∨ (¬P ∧ ¬Q) ∨ (P ∧ Q) [P ∨ P ⇔ P ]
Note:
1. The number of minterms appearing in the normal form is the same
as the number of entries with the truth value T in the truth table of
the given formula. Thus, every formula which is not a contradiction
has an equivalent PDNF.
2. If a formula is a tautology, then all the minterms will appear in its
PDNF.
3. To show that two formulas are equivalent, obtain PDNFs of the two
formulas. If the normal forms are identical, then both the formulas
are equivalent.
4. Minterms of three variables are P ∧ Q ∧ R, P ∧ Q ∧ ¬R, P ∧ ¬Q ∧ R,
¬P ∧Q∧R, P ∧¬Q∧¬R, ¬P ∧Q∧¬R, ¬P ∧¬Q∧¬R, ¬P ∧¬Q∧¬R.

1.6.2 Principal Conjunctive Normal Form or Product of


Sum Canonical Form
For a given number of variables, the “maxterms” consist of disjunctions in
which each variable or its negation, but not both, appear only once. For a
given formula, an equivalent formula consisting of conjunctions of maxterms
only is known as its Principal Conjunctive Normal Form (PCNF) or
Product of Sum of Canonical Form.
Note:
If the PDNF or PCNF of a given formula A consisting of n variables is known,
then the PDNF or PCNF of ¬A will consist of the disjunction (or conjunction)
of the remaining minterms (or maxterms) which do not appear in the PDNF
or PCNF of A. From A ⇔ ¬¬A, one can obtain the PDNF or PCNF of A by
repeated applications of De Morgan’s laws to the PDNF or PCNF of ¬A.

1.6.3 Solved Problems


1. Obtain the PDNF and PCNF of (P ∧ Q) ∨ (¬P ∧ R).
Solution.
(P ∧ Q) ∨ (¬P ∧ R)
⇔((P ∧ Q) ∧ T ) ∨ ((¬P ∧ R) ∧ T ) [since A ∧ T ⇔ A]
⇔((P ∧ Q) ∧ (R ∨ ¬R)) ∨ ((¬P ∧ R) ∧ (Q ∨ ¬Q))
[since P ∨ ¬P ⇔ T ]
Logics and Proofs 11

⇔(P ∧ Q ∧ R) ∨ (P ∧ Q ∧ ¬R) ∨ (¬P ∧ Q ∧ R) ∨ (¬P ∧ ¬Q ∧ R)


[Distributive laws]
is the required PDNF.
The remaining minterms are
P ∧ ¬Q ∧ R, P ∧ ¬Q ∧ ¬R, ¬P ∧ Q ∧ ¬R, ¬P ∧ ¬Q ∧ ¬R.
The required PCNF is
¬((P ∧ ¬Q ∧ R) ∨ (P ∧ ¬Q ∧ ¬R) ∨ (¬P ∧ Q ∧ ¬R)
∨ (¬P ∧ ¬Q ∧ ¬R))
⇔ (¬P ∨ Q ∨ ¬R) ∧ (¬P ∨ Q ∨ R) ∧ (P ∨ ¬Q ∨ R) ∧ (P ∨ Q ∨ R).
2. Obtain the PCNF and PDNF of (¬P → R) ∧ (Q ↔ P ) by using
equivalences.
Solution.
(¬P → R) ∧ (Q ↔ P )
⇔(P ∨ R) ∧ ((Q → P ) ∧ (P → Q))
[since P → Q ⇔ ¬P ∨ Q, P ↔ Q ⇔ (P → Q) ∧ (Q → P )]
⇔(P ∨ R) ∧ ((¬Q ∨ P ) ∧ (¬P ∨ Q))
⇔((P ∨ R) ∨ (Q ∧ ¬Q)) ∧ ((¬Q ∨ P ) ∨ (R ∨ ¬R)) ∧ ((¬P ∨ Q)
∨ (R ∧ ¬R))
⇔(P ∨ Q ∨ R) ∧ (P ∨ ¬Q ∨ R) ∧ (P ∨ ¬Q ∨ R) ∧ (P ∨ ¬Q ∨ ¬R)
∧ (¬P ∨ Q ∨ R) ∧ (¬P ∨ Q ∧ ¬R)
⇔(P ∨ Q ∨ R) ∧ (P ∨ ¬Q ∨ R) ∧ (P ∨ ¬Q ∨ ¬R) ∧ (¬P ∨ Q ∨ R)
∧ (¬P ∨ Q ∨ ¬R)
which is the required PDNF.
The remaining minterms are
P ∨ Q ∨ ¬R, ¬P ∨ ¬Q ∨ R, ¬P ∨ ¬Q ∨ ¬R.
The required PCNF is
¬((P ∨ Q ∨ ¬R) ∧ (¬P ∨ ¬Q ∨ R) ∧ (¬P ∨ ¬Q ∨ ¬R))
⇔(¬P ∧ ¬Q ∧ R) ∨ (P ∧ Q¬R) ∨ (P ∧ Q ∧ R).
3. Find the PDNF of (Q ∨ (P ∧ R)) ∧ ¬((P ∨ R) ∧ Q).
Solution.
(Q ∨ (P ∧ R)) ∧ ¬((P ∨ R) ∧ Q)
⇔(Q ∨ (P ∧ R)) ∧ ((¬P ∧ ¬R) ∨ ¬Q) [by De Morgan’s laws]
⇔Q ∧ (¬P ∧ ¬R) ∨ (Q ∧ ¬Q) ∨ (P ∧ R ∧ ¬P ∧ ¬R) ∨ (P ∧ R ∧ ¬Q)
[using Distributive law]
12 Discrete Mathematical Structures

⇔(¬P ∧ Q ∧ ¬R) ∨ F ∨ (F ∧ R) ∨ (P ∧ ¬Q ∧ R)
⇔(¬P ∧ Q ∧ ¬R) ∨ (P ∧ ¬Q ∧ R).

4. Prove that ((P ∨Q)∧¬(¬P ∧(¬Q∨¬R))))∨(¬P ∧¬Q)∨(¬P ∧¬R)


is a tautology.
Solution.
Consider

¬(¬P ∧ (¬Q ∨ ¬R)) ⇔ P ∨ ¬(¬Q ∨ ¬R) [De Morgan’s law]


⇔ P ∨ (Q ∧ R) [De Morgan’s law]
⇔ (P ∨ Q) ∧ (P ∨ R) [De Morgan’s law]. (1.1)

Now, consider

(¬P ∧ ¬Q) ∨ (¬P ∧ ¬R)


⇔(¬(P ∨ Q) ∨ ¬(P ∨ R) [De Morgan’s law]
⇔¬((P ∨ Q) ∧ (P ∨ R)) [De Morgan’s law]. (1.2)

From (1.1) and (1.2), we obtain

((P ∨ Q) ∧ (P ∨ Q) ∧ (P ∨ R)) ∨ ¬((P ∨ Q) ∧ (P ∨ R))


⇔((P ∨ Q) ∧ (P ∨ R)) ∨ ¬((P ∨ Q) ∧ (P ∨ R))
⇔T.

Hence, the given statement formula is a tautology.


5. Prove that (P → Q) ∧ (Q → R) ⇒ (P → R).
Solution.
It is enough to prove (P → Q)∧(Q → R) ⇒ (P → R) is a tautology.
Let S = (P → Q) ∧ (Q → R) ⇒ (P → R).
Since the last column is T for all eight combinations as shown in
the table below, the given statement formula is a tautology.

Truth Table showing (P → Q) ∧ (Q → R) ⇒ (P → R)


P Q R P → Q Q → R P → R (P → Q) ∧ (Q → R) S
T T T T T T T T
T T F T F F F T
T F T F T T F T
F T T T T T T T
T F F F T F F T
F T F T F T F T
F F T T T T T T
F F F T T T T T
Logics and Proofs 13

6. Show that P ∨(Q∧R) and (P ∨Q)∧(P ∨R) are logically equivalent.


Solution.
From columns (6) and (8), we have P ∨(Q∧R) and (P ∨Q)∧(P ∨R)
are logically equivalent.

Truth Table showing the given Formulas are Logically Equivalent


P Q R P ∨ Q P ∨ R (P ∨ Q) ∧ (P ∨ R) Q ∧ R P ∨ (Q ∧ R)
T T T T T T T T
T T F T T T F T
T F T T T T F T
F T T T T T T T
T F F T T T F T
F T F T F F F F
F F T F T F F F
F F F F F F F F

7. Show that the propositions P → Q and ¬P ∨ Q are logically


equivalent.
Solution.
From columns (3) and (5) in the table below, we have P → Q and
¬P ∨ Q are logically equivalent.
Truth Table showing the given
Formulas are Logically Equivalent
P Q P → Q ¬P ¬P ∨ Q
T T T F T
T F F F F
F T T T T
F F T T T

1.7 Inference Theory


Given a set of premises H1 , H2 , . . . , Hm and a conclusion C, we want to show
whether C logically follows from H1 , H2 , . . . , Hm .
That is, we want to show (H1 ∧ H2 ∧ · · · ∧ Hm ) → C is a tautology.

Procedure 1.
We look for those rows of H1 , H2 , . . . , Hm which have a truth value T . If for
every such row C also has a truth value T , then the conclusion C logically
follows from H1 , H2 , . . . , Hm .
Example 1: H1 : P , H2 : P → Q, C : Q.
In the first row of the truth table,
14 Discrete Mathematical Structures

Truth Table Showing P, P → Q ⇒ Q


P Q P →Q
T T T
T F F
F T T
F F T

H1 : P is True, H2 : P → Q is True. Also, C : Q is True.


Therefore, P, P → Q ⇒ Q.

1.7.1 Rules of Inference


1. Rule P: A premise can be introduced at any point of derivation.
2. Rule T: A formula can be introduced provided it is tautologically
implied by previously introduced formulas in the derivation.
3. Rule CP: If S can be derived from R and a set of premises, then
R → S can be derived from the set of premises alone.

Set of premises

R S
S

Rule CP

We use the following tables of implications and equivalences.


Implications Table
I1 : P ⇒ P ∨ Q
I2 : Q ⇒ P ∨ Q
I3 : P ∧ Q ⇒ P
I4 : P ∧ Q ⇒ Q
I5 : P, P → Q ⇒ Q
I6 : ¬Q, P → Q ⇒ ¬P
Logics and Proofs 15

I7 : ¬P, P ∨ Q ⇒ Q
I8 : P → Q, Q → R ⇒ P → R
I9 : P, Q ⇒ P ∨ Q
I10 : Q ⇒ P → Q
I11 : P ∨ Q, Q → R ⇒ ¬P → R
I12 : ¬P ⇒ P → Q.
Equivalences Table
E1 : ¬¬P ⇔ P
E2 : P → Q ⇔ ¬P ∨ Q
E3 : P → Q ⇔ ¬Q → ¬P
E4 : (P Q) ⇔ (P → Q) ∧ (Q → P )
E5 : P → (Q → R) ⇔ (P ∧ Q) → R
E6 : ¬(P ∧ Q) ⇔ ¬P ∨ ¬Q.
Note:
1. Rule CP means rule of Conditional Proof.
2. Rule CP is also called the deduction theorem.
3. In general, whenever conclusion is of the form R → S (in terms of
conditional), we should apply Rule CP . In such case, R is taken as
an additional premise, and S can be derived from the given premises
and R.

1.7.2 Solved Problems


1. Show that R∧(P ∨Q) is a valid conclusion from the premises P ∨Q,
Q → R, P → M , and ¬M .
Solution.
Given premises are P ∨ Q, Q → R, P → M , ¬M
Conclusion: R ∧ (P ∨ Q).
{1} (1) P →M Rule P
{2} (2) ¬M Rule P
{1, 2} (3) ¬P Rule T [¬Q, P → Q ⇒ ¬P ]
{4} (4) P ∨Q Rule P
{4} (5) ¬P → Q Rule T [P → Q ⇔ ¬P ∨ Q]
{1, 2, 4} (6) Q Rule T [P, P → Q ⇒ Q]
{7} (7) Q→R Rule P
{1, 2, 4, 7} (8) R Rule T [P, P → Q ⇒ Q]
{1, 2, 4, 7} (9) R ∧ (P ∨ Q) Rule T [P, Q ⇒ P ∧ Q].
16 Discrete Mathematical Structures

2. Prove that the premises P → Q, Q → R, R → S, S → ¬R, and


P ∧ S are inconsistent.
Solution.
{1} (1) P →Q Rule P
{2} (2) Q→R Rule P
{1, 2} (3) P →R Rule T [P → Q, Q → R ⇒ P → R]
{4} (4) S → ¬R Rule P
{4} (5) R → ¬S Rule T [P → Q ⇔ ¬Q → ¬P ]
{1, 2, 4} (6) P → ¬S Rule T [P → R, R → ¬S ⇒ P → ¬S]
{1, 2, 4} (7) ¬P ∨ ¬S Rule T [P → Q ⇔ ¬P ∨ Q]
{1, 2, 4} (8) ¬(P ∧ S) Rule T [De Morgan’s law]
{9} (9) P ∧S Rule P
{1, 2, 4, 9} (10) (P ∧ S)∧ Rule T [P, Q ⇒ P ∧ Q]
¬(P ∧ S)
which is false. Therefore, the given set of premises are inconsistent.
3. Show that (p → q) ∧ (r → s), (q → t) ∧ (s → u), ¬(t ∧ u), and
(p → r) ⇒ ¬p.
Solution.
{1} (1) (p → q) ∧ (r → s) Rule P
{1} (2) p→q Rule T [P ∧ Q ⇒ P ]
{1} (3) r→s Rule T [P ∧ Q ⇒ Q]
{4} (4) (q → t) ∧ (s → u) Rule P
{4} (5) q→t Rule T [P ∧ Q ⇒ P ]
{4} (6) s→u Rule T [P ∧ Q ⇒ Q]
{1, 4} (7) p→t Rule T [P → Q,
Q → R ⇒ P → R]
{1, 4} (8) r→u Rule T [P → Q,
Q → R ⇒ P → R]
{9} (9) p→r Rule P
{1, 4, 9} (10) p→u Rule T [P → Q,
Q → R ⇒ P → R]
{1, 4, 9} (11) ¬u → ¬p Rule T
{1, 4} (12) ¬t → ¬p Rule T
{1, 4, 9} (13) (¬t ∨ ¬u) → ¬p Rule T [P → Q,
R → Q ⇒ (P ∨ R) → Q]
{1, 4, 9} (14) ¬(t ∧ u) → ¬p Rule T [De Morgan’s law]
Logics and Proofs 17

{15} (15) ¬(t ∧ u) Rule P


{1, 4, 9, 15} (16) ¬p Rule T [P, P → Q ⇒ Q].

4. Show that the hypotheses, “It is not sunny this afternoon and it is
colder than yesterday”, “We will go swimming only if it is sunny”,
“If we do not go swimming, then we will take a canoe trip”, and “if
we take a canoe trip, then we will be home by sunset”, lead to the
conclusion, “We will be home by sunset”.
Solution.
Let A: It is sunny
B: It is colder than yesterday
C: We will go swimming
D: We will take a canoe trip
E: We will be home by sunset.
Then the given premises are
(1) ¬A ∧ B (2) A → C (3) ¬C → D (4) D → E.
The conclusion is C.

{1} (1) ¬A ∧ B Rule P


{1} (2) ¬A Rule T [P ∧ Q ⇒ P ]
{2} (3) A→C Rule P
{1, 2} (4) ¬C Rule T [¬P, P → Q ⇒ ¬Q]
{5} (5) ¬C → D Rule P
{1, 2, 5} (6) D Rule T [P, P → Q ⇒ Q]
{7} (7) D→E Rule P
{1, 2, 5, 7} (8) E Rule T [P, P → Q ⇒ Q].

5. Prove that the following argument is valid.


p → ¬q, r → q, r ⇒ ¬p.
Solution.
{1} (1) r Rule P
{2} (2) r→q Rule P
{1, 2} (3) q Rule T [P, P → Q ⇒ Q]
{4} (4) p → ¬q Rule P
{1, 2, 4} (5) ¬p Rule T [P → ¬Q, q ⇒ ¬P ].

6. Using indirect method of proof, derive p → ¬s from the premises


p → (q ∨ r), q → ¬p, s → ¬r, and p.
18 Discrete Mathematical Structures

Solution.
Let us include ¬(p → ¬s) as an additional premise and prove this
problem by the method of contradiction.
Now, ¬(p → ¬s) = ¬(¬p ∨ ¬s) = p ∧ s.
Therefore, the additional premise is p ∧ s.

{1} (1) p∧s Additional premise


{2} (2) p → (q ∨ r) Rule P
{3} (3) p Rule P
{2, 3} (4) q∨r Rule T [P, P → Q ⇒ Q]
{1, 2, 3} (5) s Rule T [P ∧ Q ⇒ Q]
{6} (6) s → ¬r Rule P
{1, 2, 3, 6} (7) ¬r Rule T [P, P → Q ⇒ Q]
{1, 2, 3, 6} (8) q Rule T [P, P → Q ⇒ Q]
{9} (9) q → ¬p Rule P
{1, 2, 3, 6, 9} (10) ¬p Rule P
{1, 2, 3, 6, 9} (11) p ∧ ¬p Rule T [P, Q ⇒ P ∧ Q]

which is false. Therefore, by the method of contradiction, p → ¬s


follows.
7. Show that R → S can be derived from the premises P → (Q → S),
¬R ∨ P , and Q.
Solution.

{1} (1) R Assumed premise


{2} (2) ¬R ∨ P Rule P
{2} (3) R→P Rule T [P → Q ⇔ ¬P ∨ Q]
{1, 2} (4) P Rule T [P, P → Q ⇒ Q]
{5} (5) P → (Q → S) Rule P
{1, 2, 5} (6) Q→S Rule P [P, P → Q ⇒ Q]
{7} (7) Q Rule P
{1, 2, 5, 7} (8) S Rule T [P, P → Q ⇒ Q]
{1, 2, 5, 7} (9) R→S Rule CP.

8. Prove that A → ¬D is a conclusion from the premises A → B ∨ C,


B → ¬A, and D → ¬C by using conditional proof.
Solution.
{1} (1) A Assumed premise
{2} (2) A→B∨C Rule P
Logics and Proofs 19

{1, 2} (3) B∨C Rule T [P, P → Q ⇒ Q]


{1, 2} (4) ¬B → C Rule T
{5} (5) B → ¬A Rule P
{5} (6) A → ¬B Rule T [P → Q ⇔ ¬Q → ¬P ]
{1, 2, 5} (7) A→C Rule T
[P → Q, Q → R ⇒ P → R]
{8} (8) D → ¬C Rule P
{8} (9) C → ¬D Rule T [P → Q ⇔ ¬P ]
{1, 2, 5, 8} (10) A → ¬D Rule CP.

1.8 Indirect Method of Proof


1.8.1 Method of Contradiction
In order to show that a conclusion C follows logically from the premises
H1 , H2 , . . . , Hm , we assume that C is false and consider ¬C as an additional
premise. If the new set of premises gives contradicting value, then the
assumption ¬C is true does not hold simultaneously with H1 ∧ H2 ∧ · · · ∧ Hm
being true.
Therefore, C is true whenever H1 ∧ H2 ∧ · · · ∧ Hm is true. Thus, C logically
follows from the premises H1 ∧ H2 ∧ · · · ∧ Hm .

1.8.2 Solved Problems



1. Show that 2 is irrational.
Solution.

Suppose 2 is irrational.
p
Therefore, 2 = q for p, q ∈ Z, q 6= 0, p and q have no common
divisor.
p2
Therefore, q2 = 2 =⇒ p2 = 2q 2 .
Since p2 is an even integer, p is an even integer.
Let p = 2m for some integer m.
∴ p2 = 2q 2 =⇒ (2m)2 = 2q 2
=⇒ q 2 = 2m2 .
Since q 2 is an even integer, q is an even integer.
Let q = 2n for some integer n.
20 Discrete Mathematical Structures

Hence, p and q are even integers.


Hence, they can have a common factor 2 which is a contradiction
to our assumption.

∴ 2 is irrational.
2. Prove that the following argument is valid:
“If 7 is less than 4, then 7 is not a prime number; 7 is not less than
4; therefore 7 is a prime number”.
Solution.
Let A: 7 is less than 4.
B: 7 is a prime number.
The given premises are (i) A → ¬B (ii) ¬A, and the conclusion
is B.

{1} (1) ¬A Rule P


{2} (2) A → ¬B Rule P
{1, 2} (3) ¬(¬B) Rule T [¬P, P → Q ⇒ ¬Q]
{1, 2} (4) B

Therefore, the statements are valid.


3. Prove that the sum of an irrational number and a rational number
is irrational.
Solution.
Let a be a rational number and b be an irrational number.
Assume a + b = c is a rational number.
∴ b = c + (−a) is a rational number since the sum of two
rational numbers is a rational number, which is a contradiction to
our assumption.
Hence, the sum of an irrational number and a rational number is
an irrational number.
4. Show that the statement “Every positive integer is the sum of the
squares of three integers” is false.
Solution.
Any number of the form 4m + 7, where m is non-negative integer,
cannot be written as the sum of the squares of three integers.
For example, 7, 11, 15, 19, . . . cannot be written as the sum of
squares of three integers.
∴ “Every positive integer is the sum of the squares of three
integers” is a false statement.
Logics and Proofs 21

1.9 Method of Contrapositive


In order to prove that H1 ∧ H2 ∧ · · · ∧ Hm ⇒ C, if we prove
¬C ⇒ ¬(H1 ∧H2 ∧· · ·∧Hm ), then the original problem follows. This procedure
is called the method of contrapositive.

1.9.1 Solved Problems


1. Prove that if 3n + 2 is odd, then n is odd.
Solution.
Let us prove this problem by the method of contrapositive.
Assume that n is even. Then n = 2k, for some integer k.
Now,
3n + 2 = 3(2k) + 2 = 2(3k + 1)
=⇒ 3n + 2 is an even number which contradicts that 3n + 2 is
odd.
∴ n is odd.

2. Prove that if n is an integer and n3 + 5 is odd, then n is even.


Solution.
Given: n3 + 5 is odd.
To prove: n is even.

Assume that n is odd.


∴ n = 2k + 1 for some integer k.
n3 + 5 = (2k + 1)3 + 5
= 8k 3 + 12k 2 + 6k + 6
= 2 4k 3 + 6k 2 + 3k + 3


=⇒ n3 + 5 is an even number which is a contradiction.


∴ n is even.

1.10 Various Methods of Proof


1.10.1 Trivial Proof
In an implication p → q, if we can establish that q is true, then regardless of
the truth value of p, the implication p → q will be true.
Hence, to construct a trivial proof of p → q, we need to show that the
truth value of q is true.
22 Discrete Mathematical Structures

1.10.2 Vacuous Proof


If the hypothesis p of an implication p → q is false, then p → q is true for any
proposition q.
Example:
Prove the proposition p(0) where p(n) is the proposition: “If n is a positive
integer greater than 1, then n2 > n”.
Solution.
Let p(0): If 0 is a positive integer greater than 1, then 02 > 0.
Since 0 is not a positive integer greater than 1, the proposition is true.

1.10.3 Direct Proof


Suppose, the hypothesis p is true. Then, the implication p → q can be proved
if we can prove that q is true by using the rules of inference and some other
theorems.
Example:
Prove that the sum of two odd integers is even.
Solution.
Let a and b be two odd integers.
∴ a = 2m + 1, for some integer m.
b = 2n + 1, for some integer n.
∴ a + b = 2m + 2n + 2 = 2(m + n + 1) =⇒ a + b is even.
∴ Sum of two odd integers is even.

1.11 Predicate Calculus


Consider the following statement:
Amruta is a student. Suppose “Amruta” is taken as a and “is a student” as
S, then the above statement can be symbolically written as S(a). Here, “is a
student” is a predicate, and “Amruta” is a subject.
Any statement “r is P ” can be written as P (r). This can be extended to
two place predicate. For example, Amruta is taller than Arvindh. Similarly, a
statement with three place predicate can be extended as “Amruta is younger
than Boomika but elder than Arvindh”.
If a statement function of one variable is defined to be an expression
consisting of a predicate symbol and an individual variable, such a statement
function becomes a statement when the variable is replaced by the name of
the object. Example for statement function of two variables:
A(x, y): x is taller than y.
G(a, r): Amruta is taller than Arvindh where a is Amruta and r is Arvindh.
Logics and Proofs 23

1.11.1 Quantifiers
Consider the following example:
All apples are red. This can be understood as “for any x, if x is an apple, then
x is red”.
If we denote A(x): x is an apple and R(x): x is red,
then we can write the above statement as

(x)(A(x) → R(x)).

Here (x) is called “Universal Quantifier”. We use universal quantifier for those
statements of the form “All P are Q”.
Now, consider the following example:
Some men are clever. This can be written as “There exists x; if x is a man,
then he is clever”.
If we denote M (x): x is a man and C(x): x is clever,
then we can write the above statement as

(∃x)(M (x) ∧ C(x)).

Here (∃x) is called “Existential Quantifier”. Existential quantifier is used for


those statements which are of the form “Some P are Q”.

1.11.2 Universe of Discourse, Free and Bound Variables


Consider the following statement:
All men are giants. This can be symbolically written as

(x)(M (x) → G(x))

where M (x): x is a man and G(x): x is giant.


In the above example, if we restrict the class as the class of men, then
the symbolic representation will be (x)G(x). Such a restricted class is called
“Universe of Discourse”.
In any formula, the part containing (x)P (x) or ∃xP (x) is called the x
bound part of the formula. Any variable appearing in an x bound part of
the formula is called bound variable. Otherwise, it is called free. Any formula
immediately following (x) or (∃x) is called the scope of the quantifier.

Example:
(x)P (x) ∧ Q(x)
In this, all x in P (x) is bound, whereas the x in Q(x) is free. The scope of (x)
is P (x).
24 Discrete Mathematical Structures

1.11.3 Solved Problems


1. Symbolise the expression “x is the father of the mother of y”.
Solution.
Let P (x): x is a person
F (x, y): x is the father of y
M (x, y): x is the mother of y.
Let z be the mother of y. Hence, the given statement can be written
as
“x is the father of z, and z is the mother of y”.
In symbolic notation,

(∃z)(P (z) ∧ F (x, z) ∧ M (z, y)).

2. Given P = {2, 3, 4, 5, 6}, state the truth value of the statement


(∃x ∈ P )(x + 3 = 10).
Solution.
For x = 2, 3, 4, 5, 6, no x satisfies x + 3 = 10.
∴ (∃x ∈ P )(x + 3 = 10) is false.
3. Write the symbolic form, and negate the following statements:
(i) Everyone who is healthy can do all kinds of work.
(ii) Some people are not admired by everyone.
(iii) Everyone should help his neighbours, or his neighbours will not
help him.
Solution.
(i) Let H(x): x is healthy
W (x): x can do all kinds of work.
Symbolic form of the statement is

(x)(H(x) → W (x)).

Negation of this expression is

¬((x)(H(x) → W (x))
=⇒ ¬((x)(¬H(x) ∨ W (x)))
=⇒ (∃x)(H(x) ∧ ¬W (x)).

That is, someone is healthy and cannot do all kinds of work.


Logics and Proofs 25

(ii) Let A(x): x is admired.


Then, the given statement can be written as “for some x, it is
not a case that x is admired by everyone”.
Symbolic form is
(∃x)(¬A(x)).
Negation of the above statement is

¬((∃x)¬A(x)) ⇒ (∀x)A(x).

That is, all people are admired by everyone.


(iii) This statement can be rewritten as
“For all x, x is a person, x should help his neighbours, or his
neighbours will not help him”.
Let P (x): x is a person
H(x): x helps his neighbour.
The symbolic form is

(x)((P (x) → H(x)) ∨ (H(x) → ¬P (x))).

Negation of the above statement is

(∃x)((H(x) → P (x)) ∧ (¬P (x) → H(x))).

4. Find a counterexample, if possible to these universally quantified


statements, in which the universe of discourse for all variables
consists of all integers.
(i) ∀x ∀y (x2 = y 2 → x = y).
(ii) ∀x ∀y (xy ≥ x).

Solution.
(i) Suppose x = −3 and y = 3, then x2 = y 2 = 9.
But x 6= y.
∴ ∀x ∀y (x2 = y 2 → x = y) is false.
(ii) Suppose x = −3 and y = 3, then xy = −9.
−9 < −3 =⇒ xy < x.
∴ ∀x ∀y(xy > x) is false.
5. Establish this logical equivalence, where A is a proposition not
involving any quantifiers. Show that (∀xp(x)) ∧ A ≡ ∀x(p(x) ∧ A)
and (∃x p(x)) ∧ A ≡ ∃x(p(x) ∧ A).
Solution.
(a) Consider
(∀xp(x)) ∧ A ≡ ∀x(p(x) ∧ A). (1.3)
26 Discrete Mathematical Structures

Case (i): Suppose A is true.


Since P ∧T = P , equation (1.3) =⇒ (∀xp(x)) ∧ T = (∀x)p(x).
Therefore, both sides are the same, and hence equation (1.3)
is logically equivalent.
Case (ii): Suppose A is false.
Since P ∧ F = F , the left-hand side of equation (1.3) is false.
Also, for every x, p(x) ∧ A is false. Hence, right-hand side of
equation (1.3) is false. Therefore, both sides are the same,
and hence, it is logically equivalent.
(b) Now, consider

(∃xp(x) ∧ A) = ∃x(p(x) ∧ A). (1.4)

Case (i): Suppose A is true.


Since P ∧T = P , equation (1.4) becomes ∃xp(x) = (∃x)p(x),
and hence, equation (1.4) is logically equivalent.
Case (ii): Suppose A is false.
Since P ∧ F = F , the left-hand side of equation (1.4) is false.
Also, for every x, p(x) ∧ A is false. Therefore, ∃x(p(x) ∧ A)
is false. Hence, the two sides of equation (1.4) are the same,
and hence equation (1.4) is logically equivalent.
6. Show that ∃xP (x) ∧ ∃xQ(x) and ∃x(P (x) ∧ Q(x)) are not logically
equivalent.
Solution.
We prove the two statements are not logically equivalent by using
the following counterexample.
Let us assume that the universe of discourse is the set of integers.
Let P (x): x is a positive integer
Q(x): x is a negative integer.
Then, ∃xP (x) ∧ ∃xQ(x) is true.
But the truth value of ∃x(P (x) ∧ Q(x)) is false.
∴ Given two statements are not logically equivalent.
7. Use quantifiers and predicate to express the fact that lim f (x) does
x→a
not exist.
Solution.
Limit exists means “f (x) approaches L (where L ∈ R) as
x approaches a, if given  > 0, ∃ δ > 0 such that
|f (x) − L| <  whenever 0 < |x − a| < δ 00 .
Logics and Proofs 27

Limit does not exist means “for all real numbers L, lim f (x) 6= L”.
x→a
The above statement can be expressed as

¬(∀  > 0, ∃ δ > 0, ∀x(0 < |x − a| < δ) → |f (x) − L| < ))


=⇒ ∀ L, ∃ > 0, ∀ δ > 0, (∃ x)() < |x − a| < δ ∧ |f (x) − L| < )
[∵ ¬(p → q) = p ∧ ¬q].

8. Let H = {−1, 0, 1, 2} denote the universe of discourse.


If p(x, y) : x + y = 1, find the truth value of (∃ x)(∃ y) p(x, y).
Solution.
When x = −1, ∃y = 2 such that − 1 + 2 = 1.
When x = 0, ∃y = 1 such that 0 + 1 = 1.
When x = 1, ∃y = 0 such that 1 + 0 = 1.
When x = 2, ∃y = −1 such that 2 − 1 = 1.

∴ (∀x)(∃y) p(x, y) is true.


9. What are the negations of the statements ∀x(x2 > x) and
∃x(x2 = 2)?
Solution.
(i) Given statement is ∀x(x2 > x).
Its negation is ∃x(x2 ≤ x).
(ii) Given statement is ∃x(x2 = 2).
Its negation is ∀x(x2 6= 2).
10. Write the negation of the statement ∃x (∀y) p(x, y).
Solution.
Given statement is (∃x) (∀y) p(x, y).
Its negation is ¬((∃x) (∀y) p(x, y)) =⇒ (∀x) (∃y)¬p(x, y).
11. Consider the statement “Given any positive integer, there is a
greater positive integer”. Symbolize this statement using and
without using the set of positive integers as the universe of discourse.
Solution.
Let G(x, y) : x is greater than y.
If we use the universe of discourse as the set of positive integers,
then we can write (x) (∃y) G(y, x).
If we do not impose the restriction on the universe of discourse and
if we write P (x) : x is a positive integer, then we can write as

(x) (P (x) → (∃y) (P (y) ∧ G(y, x))).


28 Discrete Mathematical Structures

12. Indicate free and bound variables. Also indicate the scope of the
quantifier in
(i) (x) (P (x) ∧ R(x)) ⇒ (x) (P (x)) ∧ Q(x)).
(ii) ((x)(P (x) Q(x) ∧ ∃(x) R(x))) ∧ S(x).

Solution.
(i) All occurrences of x in P (x) ∧ R(x) are bound occurrences. The
occurrence of x in xP (x) is bound. The occurrence of x in Q(x)
is free. The scope of (x) is P (x) ∧ R(x) and P (x).
(ii) All occurrences of x in P (x) Q(x) ∧ (∃ x) R(x) are bound,
and the occurrence of x in S(x) is free. The scope of (x) is
P (x) Q(x) ∧ (∃x) R(x), and the scope of (∃x) is R(x).

1.11.4 Inference Theory for Predicate Calculus


We have seen implication table and equivalence table already. Those rules
can be extended here also. For example, P, P → Q ⇒ Q can be extended as
P (x), P (x) → Q(x) ⇒ Q(x).
In addition, we use the following rules:
1. Universal Specification (US): (x)A(x) ⇒ A(y)
2. Universal Generalization (UG): A(y) ⇒ (x)A(x)
3. Existential Specification (ES): (∃x)A(x) ⇒ A(y)
4. Existential Generalization (EG): A(y) ⇒ (∃x)A(x).
Remark.
Let the universe of discourse be denoted by S = {a1 , a2 , . . . , an }. Then,

(x)A(x) = A(a1 ) ∧ A(a2 ) ∧ · · · ∧ A(an )


(∃x)A(x) = A(a1 ) ∨ A(a2 ) ∨ · · · ∨ A(an ).

Consider now ¬(x)A(x)


⇔¬(A(a1 ) ∧ A(a2 ) ∧ · · · ∧ A(an ))
⇔¬A(a1 ) ∨ ¬A(a2 ) ∨ · · · ∨ ¬A(an )
⇔(∃x)¬A(x).

Similarly, ¬((∃x)A(x)) ⇔ (x)¬A(x)).

1.11.5 Solved Problems


1. Show that
(x)(P (x) → Q(x)) ∧ (x)(Q(x) → R(x)) ⇒ (x)(P (x) → R(x)).
Logics and Proofs 29

Solution.
{1} (1) (x)(P (x) → Q(x)) Rule P
{2} (2) P (y) → Q(y) Rule U S
{3} (3) (x)(Q(x) → R(x)) Rule P
{3} (4) Q(y) → R(y) Rule U S
{1, 3} (5) P (y) → R(y) Rule T
[∵ P → Q, Q → R ⇒ P → R]
{1, 3} (6) (x)(P (x) → R(x)) Rule U G.

2. Show that ∀x(P (x) ∨ Q(x)) ⇒ ∀x P (x) ∨ ∃x Q(x) using indirect


method.
Solution.
We use the method of contradiction. Assume ¬((x)P (x)∨(∃x)Q(x))
as an additional premise.

{1} (1) ¬((x)P (x) ∨ (∃x)Q(x)) Assumed premise


{1} (2) (∃x)¬P (x) ∧ (x)¬Q(x) Rule T [De Morgan’s law]
{1} (3) (∃x)¬P (x) Rule T [P ∧ Q ⇒ P ]
{1} (4) (x)¬Q(x) Rule T [P ∧ Q ⇒ Q]
{1} (5) ¬P (y) Rule ES
{1} (6) ¬Q(y) Rule U S
{1} (7) ¬P (y) ∧ ¬Q(y) Rule T [P, Q ⇒ P ∧ Q]
{1} (8) ¬(P (y) ∨ Q(y)) Rule T [De Morgan’s law]
{9} (9) (x)(P (x) ∨ Q(x)) Rule P
{9} (10) P (y) ∨ Q(y) Rule U S
{1, 9} (11) (P (y) ∨ Q(y))∧ Rule T [P, Q ⇒ P ∧ Q]
¬(P (y) ∨ Q(y))

which is false.
Therefore, by the method of contradiction, we have

∀x(P (x) ∨ Q(x)) ⇒ ∀x P (x) ∨ ∃x Q(x).

3. Show that ∀xP (x) ∧ ∃xQ(x) is equivalent to ∀x∃y(P (x) ∧ Q(y)).


Solution.
{1} (1) ∀xP (x) ∧ ∃xQ(x) Rule P
{1} (2) ∀xP (x) Rule T [P ∧ Q ⇒ P ]
{1} (3) P (m) Rule U S
{1} (4) ∃xQ(x) Rule T [P ∧ Q ⇒ Q]
30 Discrete Mathematical Structures

{1} (5) Q(n) Rule ES


{1} (6) P (m) ∧ Q(n) Rule T
{1} (7) ∃y(P (m) ∧ Q(y)) Rule EG
{1} (8) (∀x)∃y(P (x) ∧ Q(y)) Rule U G

Hence, ∀xP (x) ∧ ∃xQ(x) and ∀x∃y(P (x) ∧ Q(y)) are logically
equivalent.
4. Show that (x)(P (x) → Q(x)) ⇒ (x)P (x) → (x)Q(x).
Solution.
We use contrapositive method to prove this problem.

{1} (1) ¬((x)P (x) → (x)Q(x)) Assumed premise


{1} (2) (x)P (x) ∧ ¬((x)Q(x)) Rule T [¬(P → Q) ⇔ P ∧ ¬Q]
{1} (3) (x)P (x) Rule T [P ∧ Q ⇒ P ]
{1} (4) ¬((x)Q(x)) Rule T [P ∧ Q ⇒ Q]
{1} (5) (∃x)¬Q(x) Rule T [apply ¬]
{1} (6) P (y) Rule U S
{1} (7) ¬Q(y) Rule ES
{1} (8) P (y) ∧ ¬Q(y) Rule T [P, Q ⇒ P ∧ Q]
{1} (9) ¬(P (y) → Q(y)) Rule T [P ∧ ¬Q ⇔ ¬(P → Q)]
{1} (10) (∃x)¬(P (x) → Q(x)) Rule EG
{1} (11) ¬((x)(P (x) → Q(x))) Rule T [apply ¬].

∴ By the method of contrapositive, we have

(x)(P (x) → Q(x)) ⇒ (x)P (x) → (x)Q(x).

5. Show that ∃x(P (x) ∧ Q(x)) ⇒ (∃x)P (x) ∧ (∃x)Q(x). Is the converse
true?
Solution.
{1} (1) (∃x)(P (x) ∧ Q(x)) Rule P
{1} (2) P (y) ∧ Q(y) Rule ES
{1} (3) P (y) Rule T [P ∧ Q ⇒ P ]
{1} (4) Q(y) Rule T [P ∧ Q ⇒ Q]
{1} (5) ∃xP (x) Rule EG
{1} (6) ∃xQ(x) Rule EG
{1} (7) ∃xP (x) ∧ ∃xQ(x) Rule T [P, Q ⇒ P ∧ Q].
Logics and Proofs 31

Converse is also true, since


{1} (1) (∃x) P (x) ∧ (∃x)Q(x) Rule P
{1} (2) (∃x) P (x) Rule T [P ∧ Q ⇒ P ]
{1} (3) (∃x)Q(x) Rule T [P ∧ Q ⇒ Q]
{1} (4) P (y) Rule ES
{1} (5) Q(y) Rule ES
{1} (6) P (y) ∧ Q(y) Rule T [P, Q ⇒ P ∧ Q]
{1} (7) (∃x)(P (x) ∧ Q(x)) Rule EG.

6. Verify the validity of the following argument. Every living thing is


a plant or an animal. John’s gold fish is alive, and it is not a plant.
All animals have hearts. Therefore, John’s gold fish has a heart.
Solution.
Let L(x): x is a living thing
P (x): x is a plant
A(x): x is an animal
H(x): x has a heart.
Given premises with their symbolic forms are
(i) Every living thing is a plant or an animal.
(x)(L(x) → P (x) ∨ A(x)).
(ii) John’s gold fish is alive, and it is not a plant.
L(y) ∧ ¬P (y).
(iii) All animals have hearts.
(x)(A(x) → H(x)).
Conclusion is H(y).
{1} (1) (x)(L(x) → P (x) ∨ A(x)) Rule P
{1} (2) L(y) → P (y) ∨ A(y) Rule P
{3} (3) L(y) ∧ ¬P (y) Rule P
{4} (4) L(y) Rule T [P ∧ Q ⇒ P ]
{5} (5) ¬P (y) Rule T [P ∧ Q ⇒ Q]
{1, 3} (6) P (y) ∨ A(y) Rule T [P, P → Q ⇒ Q]
32 Discrete Mathematical Structures

{1, 3} (7) ¬P (y) → A(y) Rule T [P → Q ⇒ ¬P ∨ Q]


{8} (8) (x)(A(x) → H(x)) Rule P
{8} (9) A(y) → H(y) Rule U S
{1, 3, 8} (10) ¬P (y) → H(y) Rule T
[P → Q, Q → R ⇒ P → R]
{1, 3, 8} (11) H(y) Rule T [P → Q, P ⇒ Q].

Hence, the given statements are valid statements.

1.12 Additional Solved Problems


1. Let P (x) denote the statement x ≤ 4. Write the truth values of
P (2) and P (4).
Solution.
P (x) : x ≤ 4
P (2) : 2 ≤ 4 is True
P (6) : 6 ≤ 4 is False.

2. Give the converse and the contrapositive of the implication: “If it


is raining, then I get wet”.
Solution.
Let P : It is raining
Q: I get wet.
Given statement is P → Q.
Its inverse is
Q → P : If I get wet, then it is raining.
Its contrapositive is
¬Q → ¬P : If I do not get wet, then it is not raining.
3. Show that R ∨ S follows logically from the premises C ∨ D,
C ∨ D → ¬H, ¬H → (A ∧ ¬B), and (A ∧ ¬B) → (R ∨ S).
Solution.
{1} (1) C ∨ D → ¬H Rule P
{2} (2) ¬H → (A ∧ ¬B) Rule P
{1, 2} (3) C ∨ D → A ∧ ¬B Rule T
{4} (4) (A ∧ ¬B) → (R ∨ S) Rule P
{1, 2, 4} (5) C ∨D →R∨S Rule T
Logics and Proofs 33

{6} (6) C ∨D Rule P


{1, 2, 4, 6} (7) R∨S Rule T [P → Q, P ⇒ Q].

4. Show that the following premises are inconsistent.


(i) If Jack misses many classes through illness, then he fails high
school.
(ii) If Jack fails high school, then he is uneducated.
(iii) If Jack reads a lot of books, then he is not uneducated.
(iv) Jack misses many classes through illness and reads a lot of
books.
Solution.

E : Jack misses many classes


S : Jack fails high school
A : Jack reads a lot of books
H : Jack is uneducated

The given premises are

E → S, S → H, A → ¬H, and E ∧ A.

{1} (1) E→S Rule P


{2} (2) S→H Rule P
{1, 2} (3) E→H Rule T
{4} (4) A → ¬H Rule P
{4} (5) H → ¬A Rule T
{1, 2, 4} (6) E → ¬A Rule T
{1, 2, 4} (7) ¬E ∧ ¬A Rule T
{1, 2, 4} (8) ¬(E ∧ A) Rule T
{9} (9) E∧A Rule P
{1, 2, 4, 9} (10) (E ∧ A) ∧ (¬(E ∧ A)) Rule T

which is false. Hence, the given premises are inconsistent.


5. Write the dual of (P ∧ Q) ∨ T .
Solution.
(P ∨ Q) ∧ F .
6. Negate the following statements.
(i) Ottawa is a small town.
(ii) Every city in Canada is clean.
34 Discrete Mathematical Structures

Solution.
(i) Ottawa is not a small town.
(ii) Every city in Canada is not clean.
7. Construct the truth table for P ∧ (P ∨ Q).
Solution.
The truth table is shown below.

Truth Table for P ∧ (P ∨ Q)


P Q P ∨ Q P ∧ (P ∨ Q)
T T T T
T F T T
F T T F
F F F F

8. Write the following in symbolic form:


If John takes calculus or Peter takes analytical geometry, then
Mohan will take English.
Solution.
Let J: John takes calculus
P : Peter takes analytical geometry
M : Mohan will take English.
Then the symbolic form is (J ∨ P ) → M .
9. Symbolise the expression: “All the world loves a lover”.
Solution.
Let P (x): x is a person
L(x): x is a lover
R(x, y): x loves y.
The symbolic form is (x)(P (x) → (y)(P (y) ∧ L(y) → R(x, y)).
10. Obtain the PDNF and PCNF of

P → ((P → Q) ∧ ¬(¬Q ∨ ¬P )).


Solution.

Let A ⇔ P → ((P → Q) ∧ ¬(¬Q ∨ ¬P ))


⇔ ¬P ∨ ((¬P ∨ Q) ∧ (Q ∧ P ))
↔ ¬P ∨ ((¬P ∧ (Q ∧ P )) ∨ (Q ∧ (Q ∧ P ))
⇔ ¬P ∨ F ∨ (P ∧ Q)
⇔ ¬P ∨ (P ∧ Q)
Logics and Proofs 35

⇔ ∨(P ∧ Q)
⇔ (¬P ∧ (Q ∨ ¬Q)) ∨ (P ∧ Q)
⇔ (¬P ∧ Q) ∨ (¬P ∧ ¬Q) ∨ (P ∧ Q)

which is the required PDNF. Its PCNF is

¬¬A ⇔ ¬P ∨ Q.

11. Obtain the PCNF of S : (¬P → R) ∧ (Q P ). Hence, obtain


PDNF.
Solution.

S : (¬P → R) ∧ (Q P)
⇔ (¬P → R) ∧ ((Q → P ) ∧ (P → Q))
⇔ (P ∨ R) ∧ (¬Q ∨ P ) ∧ (¬P ∨ Q)
⇔ ((P ∨ R) ∨ F ) ∧ ((¬Q ∨ P ) ∨ F ) ∧ ((¬P ∨ Q) ∨ F )
⇔ ((P ∨ R) ∨ (Q ∧ ¬Q)) ∧ ((¬Q ∨ P ) ∨ (R ∧ ¬R))
∧ ((¬P ∨ Q) ∨ (R ∧ ¬R))
⇔ (P ∨ Q ∨ R) ∧ (P ∨ ¬Q ∨ R) ∧ (P ∨ ¬Q ∨ R)
∧ (P ∨ ¬Q ∨ ¬R) ∧ (¬P ∨ Q ∨ R) ∧ (¬P ∨ Q ∨ ¬R)
⇔ (P ∨ Q ∨ R) ∧ (P ∨ ¬Q ∨ R) ∧ (P ∨ ¬Q ∨ ¬R)
∧ (¬P ∨ Q ∨ R) ∧ (¬P ∨ Q ∨ ¬R)

which is the required PCNF. Now,

¬S : (P ∨ Q¬R) ∧ (¬P ∨ ¬Q ∨ R) ∧ (¬P ∨ ¬Q¬R)


¬¬S :¬((P ∨ Q¬R) ∧ (¬P ∨ ¬Q ∨ R) ∧ (¬P ∨ ¬Q¬R))
⇔ (¬P ∧ ¬Q ∧ R) ∨ (P ∧ Q ∧ R) ∨ (P ∧ Q ∧ R)

which is the required PDNF.


12. What is the contrapositive statement of “The home town wins
whenever it is raining”?
Solution.
Let P : It is raining
Q: The home town wins.
Given statement is P → Q.
Its contrapositive statement is
¬Q → ¬P : If the home town does not win, then it is not
raining.
36 Discrete Mathematical Structures

13. Give the symbolic form of “Some men are giants”.


Solution.
The given statement can be written as
“there is an x such that x is a man and x is giant”.
Let M (x): x is a man
G(x): x is a giant.
∴ The symbolic form is

(∃x)(M (x) ∧ G(x)).

14. Find the PCNF of (P ∨ R) ∧ (P ∨ ¬Q). Also find its PDNF, without
using truth table.
Solution.

Let A ⇔ (P ∨ R) ∧ (P ∨ ¬Q)
⇔ ((P ∨ R) ∨ F ) ∧ ((P ∨ ¬Q) ∨ F )
⇔ ((P ∨ R) ∨ (Q ∧ ¬Q)) ∧ ((P ∨ ¬Q) ∨ (R ∧ ¬R))
⇔ (P ∨ Q ∨ R) ∧ (P ∨ ¬Q ∨ R) ∧ (P ∨ ¬Q ∨ R)
∧ (P ∨ ¬Q ∧ ¬R)
⇔ (P ∨ Q ∨ R) ∧ (P ∨ ¬Q ∨ R) ∧ (P ∨ ¬Q ∨ ¬R)

which is the required PCNF. Now,

¬A : (P ∨ Q ∨ ¬R) ∧ (¬P ∨ Q ∨ R) ∧ (¬P ∨ Q ∨ ¬R)


∧ (¬P ∨ ¬Q ∨ R) ∧ (¬P ∨ ¬Q ∨ ¬R)
¬¬A :¬ (P ∨ Q ∨ ¬R) ∧ (¬P ∨ Q ∨ R) ∧ (¬P ∨ Q ∨ ¬R)

∧ (¬P ∨ ¬Q ∨ R) ∧ (¬P ∨ ¬Q ∨ ¬R)
⇔ (¬P ∧ ¬Q ∧ R) ∨ (P ∧ ¬Q ∧ ¬R) ∨ (P ∧ ¬Q ∧ R)
∨ (P ∧ Q ∧ ¬R) ∨ (P ∧ Q ∧ R)

which is the required PDNF.


15. Show that using rule CP, ¬P ∨ Q, ¬Q ∨ R, R → S ⇔ P → S.
Solution.

{1} (1) P Assumed premise


{2} (2) ¬P ∨ Q Rule P
{2} (3) P →Q Rule T [P → Q ⇔ ¬P ∨ Q]
{1, 2} (4) Q Rule T [P, P → Q ⇒ Q]
{5} (5) ¬Q ∨ R Rule P
{5} (6) Q→R Rule T [P → Q ⇔ ¬P ∨ Q]
Logics and Proofs 37

{1, 2, 5} (7) R Rule T [P, P → Q ⇒ Q]


{8} (8) R→S Rule P
{1, 2, 5, 8} (9) S Rule T [P, P → Q ⇒ Q]
{1, 2, 5, 8} (10) P →S Rule CP.

16. Show that (¬P ∧ (¬Q ∧ R)) ∨ (Q ∧ R) ∨ (P ∧ R) ⇔ R without using


truth table.
Solution.

¬P ∧ (¬Q ∧ R) ⇔ (¬P ∧ ¬Q) ∧ R (Associative law)


⇔ ¬(P ∨ Q) ∧ R (De Morgan’s law). (1.5)

Now,

(Q ∧ R) ∨ (P ∧ R) ⇔ (Q ∨ P ) ∧ R (Distributive law)
⇔ (P ∨ Q) ∧ R (Commutative law). (1.6)

From (1.5) and (1.6), we have

(¬P ∧ (¬Q ∧ R)) ∨ (Q ∧ R) ∨ (P ∧ R)


⇔ (¬(P ∨ Q) ∧ R) ∨ ((P ∨ Q) ∧ R)
⇔ (¬(P ∨ Q) ∨ (P ∨ Q)) ∧ R (Distributive law)
⇔T ∧R
⇔ R.

∴ (¬P ∧ (¬Q ∧ R)) ∨ (Q ∧ R) ∨ (P ∧ R) ⇔ R.


2
Combinatorics

2.1 Introduction
In this chapter, we discuss about the technique of mathematical induction
which is used for proving many standard results over natural numbers. Then,
we discuss about the basis of counting, pigeonhole principle, permutations
and combinations, and recurrence relation. These concepts are useful
in the analysis of certain discrete time systems, analysis of algorithms,
error-correcting code, etc. At the end of the chapter, we discuss about
generating function which is used to solve linear recurrence relations.

2.2 Mathematical Induction


Mathematical induction is a method of finding the truth from a general
statement for particular cases.
A statement may be true with reference to more than hundred cases yet
we cannot conclude it to be true in general. It is possible to disprove the
statement by a counter-example. A statement need not be accepted to be
true. Such a statement inferred from a particular case is called a conjecture.
If the conjecture is a statement involving natural numbers, we can use the
principle of the mathematical induction to prove the same.

2.2.1 Principle of Mathematical Induction


For a given statement involving a natural number n, if we can show that
(1) The statement is true for n = 1 or n = n0 ,
(2) The statement is true for n = m + 1 under the assumption that the
statement is true for n = m(m ≥ n0 ),
then we can conclude that the statement is true for all natural numbers.
Remark 2.2.1 In the above principle, (1) is usually referred to as the basis
of induction, and (2) is usually referred to as the induction step. Also, the
assumption that the statement is true for n = m in (2) is usually referred to
as the induction hypothesis.
39
40 Discrete Mathematical Structures

2.2.2 Procedure to Prove that a Statement P (n) is True


for all Natural Numbers
Step 1. We must prove that P (1) is true.
Step 2. By assuming P (m) is true, we must prove that P (m + 1) is also
true.
In the sequel, we apply the principle of mathematical induction to prove
statements involving natural numbers.

2.2.3 Solved Problems


n(n + 1)
1. Prove that 1 + 2 + 3 + · · · + n = by induction principle.
2
Solution.
n(n + 1)
Let P (n) : 1 + 2 + 3 + · · · + n = .
2
1(1 + 1)
(1) P (1) : 1 = = 1 is true.
2
(2) Assume that P (m) is true.
m(m + 1)
That is, 1 + 2 + 3 + · · · + m = .
2
(3) Now,

m(m + 1)
1 + 2 + 3 + · · · + m + (m + 1) = +m+1
2
m(m + 1) + 2(m + 1)
=
2
(m + 1)(m + 2)
= .
2
∴ By mathematical induction, the given statement is true for all n.
n(n + 1)(2n + 1)
2. Show that 12 + 22 + 32 + · · · + n2 = , n ≥ 1 by
6
mathematical induction.
Solution.
n(n + 1)(2n + 1)
Let P (n) : 12 + 22 + 32 + · · · + n2 = .
6
1(1 + 1)(2 + 1)
(1) P (1) : 12 = = 1 is true.
6
(2) Assume P (m) is true.
m(m + 1)(2m + 1)
That is, 12 + 22 + 32 + · · · + m2 = .
6
(3) Now,

12 + 22 + 32 + · · · + m2 + (m + 1)2
Combinatorics 41
m(m + 1)(2m + 1)
= + (m + 1)2
6
m(m + 1)(2m + 1) + 6(m + 1)2
=
6
(m + 1)[m(2m + 1) + 6(m + 1)]
=
6
(m + 1)(2m2 + m + 6m + 6)
=
6
(m + 1)(2m2 + 7m + 6)
=
6
(m + 1)(2m2 + 4m + 3m + 6)
=
6
(m + 1)[2m(m + 2) + 3(m + 2)]
=
6
(m + 1)(m + 2)(m + 3)
=
6
(m + 1)[(m + 1) + 1][2(m + 1) + 1]
= .
6
∴ By mathematical induction, the given statement is true for
all n ≥ 1.
n2 (n + 1)2
3. Prove that 13 + 23 + 33 + · · · + n3 = , n ∈ N.
4
Solution.
n2 (n + 1)2
Let P (n) : 13 + 23 + 33 + · · · + n3 = .
4
12 (1 + 1)2
(1) P (1) : 12 = is true.
4
(2) Assume P (m) is true.
m2 (m + 1)2
That is, 13 + 23 + 33 + · · · + m3 = .
4
(3) Now,
m2 (m + 1)2
13 + 23 + 33 + · · · + m3 + (m + 1)3 = + (m + 1)3
4
m2 (m + 1)2 + 4(m + 1)3
=
4
(m + 1)2 [m2 + 4(m + 1)]
=
4
(m + 1)2 (m2 + 4m + 4)
=
4
(m + 1)2 (m + 2)2
=
4
(m + 1)[(m + 1) + 1]2
= .
4
42 Discrete Mathematical Structures

∴ By mathematical induction, the statement is true for all n ∈ N.


Pn
4. Prove that i=1 (2i − 1)2 = n2 , for all n ∈ N.

(or)

Prove that the sum of the first n odd integers is n2 for all integers n.
Solution.
Let P (n) : 1 + 3 + 5 + · · · + (2n − 1) = n2 .
(1) P (1) : 1 = 12 is true.
(2) Assume P (m) is true.
That is, 1 + 3 + 5 + · · · + (2m − 1) = m2 .
(3) Now,

1 + 3 + 5 + · · · + (2m − 1) + [2(m + 1) − 1] = m2 + [2(m + 1) − 1]


= m2 + [2(m + 1) − 1]
= m2 + 2m + 2 − 1
= m2 + 2m + 1
= (m + 1)2 .

∴ By mathematical induction, the given statement is true for all


n ∈ N.
(or)
(1) P (1) : 1 = 12 is true.
(2) Assume P (m) is true.
Pm
That is, i=1 (2i − 1) = m2 .
(3) Now,
m+1
X m
X
(2i − 1) = (2i − 1) + [2(m + 1) − 1]
i=1 i=1
= m2 + 2m + 2 − 1
= m2 + 2m + 1
= (m + 1)2 .

∴ By mathematical induction, the given statement is true for all


n ∈ N.
5. Using mathematical induction, prove that
n(3n + 1)
2 + 5 + 8 + · · · + (3n − 1) = .
2
Solution.
n(3n + 1)
Let P (n) : 2 + 5 + 8 + · · · + (3n − 1) = .
2
Combinatorics 43
1[3 · 1 + 1]
(1) P (1) : 2 = = 2 is true.
2
(2) Assume P (m) is true.
m(3m + 1)
That is, 2 + 5 + 8 + · · · + (3m − 1) = .
2
(3) Now,

2 + 5 + 8 + · · · + (3m − 1) + [3(m + 1) − 1]
m(3m + 1)
= + (3m + 3 − 1)
2
m(3m + 1) + 2(3m + 2)
=
2
3m2 + m + 6m + 4
=
2
3m2 + 7m + 4
=
2
3m2 + 3m + 4m + 4
=
2
3m(m + 1) + 4(m + 1)
=
2
(m + 1)(3m + 4)
=
2
(m + 1)[3(m + 1) + 1]
= .
2
∴ By mathematical induction, the given statement is true.
6. Prove that for n ≥ 0, 1 + 2 + 4 + · · · + 2n = 2n+1 − 1.
Solution.
Let P (n) : 1 + 2 + 4 + · · · + 2n = 2n+1 − 1.
(1) P (1) : 1 + 2 = 21+1 − 1 is true.
(2) Assume P (m) is true.
That is, 1 + 2 + 4 + · · · + 2m = 2m+1 − 1.
(3) Now,

1 + 2 + 4 + · · · + 2m + 2m+1 = (2m+1 − 1) + 2m+1


= 2 · 2m+1 − 1
= 2m+2 − 1
= 2(m+1)+1 − 1.

∴ By mathematical induction, the given statement is true.


7. Prove that if n ≥ 1, then

1(1!) + 2(2!) + · · · + n(n!) = (n + 1)! − 1.


44 Discrete Mathematical Structures

Solution.
Let P (n) : 1(1!) + 2(2!) + · · · + n(n!) = (n + 1)! − 1.
(1) P (1) : 1(1!) = (1 + 1)! − 1 is true.
(2) Assume P (m) is true.
That is, 1(1!) + 2(2!) + · · · + m(m!) = (m + 1)! − 1.
(3) Now,
1(1!) + 2(2!) + · · · + m(m!) + (m + 1)[(m + 1)!]
= [(m + 1)! − 1] + (m + 1)[(m + 1)!]
= (m + 1)! + (m + 1)(m + 1)! − 1
= (m + 1)![1 + (m + 1)] − 1
= (m + 1)!(m + 2) − 1
= (m + 2)! − 1.
∴ By mathematical induction, the given statement is true for all
n ≥ 1.
8. Use mathematical induction to show that
1 1 1 1 n
+ + + ··· + = , for all n ≥ 1.
1·2 2·3 3·4 n(n + 1) n+1
Solution.
1 1 1 1 n
Let P (n) : + + + ··· + = .
1·2 2·3 3·4 n(n + 1) n+1
1 1
(1) P (1) : = is true.
1·2 1+1
(2) Assume P (m) is true.
1 1 1 1 m
That is, + + + ··· + = .
1·2 2·3 3·4 m(m + 1) m+1
(3) Now,
1 1 1 1 1
+ + + ··· + +
1·2 2·3 3·4 m(m + 1) (m + 1)(m + 2)
m 1
= +
m + 1 (m + 1)(m + 2)
m(m + 2) + 1
=
(m + 1)(m + 2)
m2 + 2m + 1
=
(m + 1)(m + 2)
(m + 1)2
=
(m + 1)(m + 2)
m+1
= .
(m + 1) + 1
∴ By mathematical induction, the given statement is true for all n.
Combinatorics 45

9. Use mathematical induction to prove that n3 − n is divisible by 3


whenever n is a positive integer.
Solution.
Let P (n) : n3 − n is divisible by 3.
(1) P (1) : 11 − 1 = 0 is divisible by 3.
(2) Assume P (m) is true.
That is, m3 − m is divisible by 3.
(3) Now,

(m + 1)3 − (m + 1) = (m3 + 3m2 + 3m + 1) − (m + 1)


= (m3 − m) + 3m2 + 3m
= (m3 − m) + 3(m2 + m).

Since both terms in this sum are divisible by 3, it follows that


(m + 1)3 − (m + 1) is also divisible by 3.
∴ By mathematical induction, n3 − n is divisible by 3 for all n ∈ N.
10. Use mathematical induction to show that n3 + 2n is divisible by 3
whenever n is a non-negative integer.
(or)
3
Prove that for all n ≥ 1, n + 2n is a multiple of 3.
Solution.
Let P (n) : n3 + 2n is a multiple of 3.
(1) P (1) : 13 + 2 · 1 = 3 is a multiple of 3.
(2) Assume P (m) is true.
That is, m3 + 2m is a multiple of 3.
(3) Now,

(m + 1)3 + 2(m + 1)
= m3 + 3m2 + 3m + 1 + 2m + 2
= (m3 + 2m) + 3(m2 + m + 1).

Since both terms on the right-hand side are divisible by 3, it


follows that (m + 1)3 + 2(m + 1) is a multiple of 3.
Hence, by mathematical induction, n3 + 2n is a multiple of 3 for all
n ∈ N.
46 Discrete Mathematical Structures

11. Use mathematical induction to show that 8n − 3n is a multiple of 5.


Solution.
Let P (n) : 8n − 3n is a multiple of 5.
(1) P (1) : 8 − 3 = 5 is a multiple of 5.
(2) Assume P (m) is true. That is, 8m − 3m is a multiple of 5.
(3) Now,

8m+1 − 3m+1 = 8m (8) − 3m+1


= 8m (5 + 3) − 3m+1
= 5 · 8m + 3 · 8m − 3m+1
= 5 · 8m + 3(8m − 3m ).

Since both terms on the right-hand side are multiples of 5, it


follows that 8m+1 − 3m+1 is also a multiple of 5.
∴ By mathematical induction, 8n − 3n is a multiple of 5.
12. Use mathematical induction to show that n2 − 7n + 12 is
non-negative whenever n is an integer greater than 3.
Solution.
Let P (n) : n2 − 7n + 12 be non-negative.
(1) P (4) : 42 − 7(4) + 12 = 16 − 28 + 12 = 0, which is non-negative.
(2) Assume P (m) is true.
That is, m2 − 7m + 12 is non-negative.
(3) Now,

(m + 1)2 − 7(m + 1) + 12
= m2 + 2m + 1 − 7m − 7 + 12
= (m2 − 7m + 12) + (2m − 6).

The first term is non-negative, and the second term is also


non-negative for m > 3. Hence, (m + 1)2 − 7(m + 1) + 12 is
non-negative.
∴ By mathematical induction, n2 − 7n + 12 is non-negative for
n > 3.
13. Use mathematical induction to prove the inequality n < 2n for all
positive integers n.
Solution.
Let P (n) : n < 2n .
(1) P (1) : 1 < 21 is true.
(2) Assume P (m) is true.
That is, m < 2m .
Combinatorics 47

(3) Now,

m + 1 < 2m + 1
< 2m + 2m since 1 < 2m
< 2 · 2m
< 2m+1 .

∴ By mathematical induction, the result is true for all n ∈ N.


14. Prove that n + 10 ≤ 2n for all n ∈ N and n ≥ 4.
Solution.
Let P (n) : n + 10 ≤ 2n , n ≥ 4.
(1) P (4) : 4 + 10 ≤ 24 is true.
(2) Assume P (m) is true.
That is, m + 10 ≤ 2m .
(3) Now,

(m + 1) + 10 = (m + 10) + 1
≤ 2m + 1
≤ 2m + 2m since 1 < 2m
≤ 2 · 2m
≤ 2m+1 .

∴ By the principle of mathematical induction, the result is true for


all n ∈ N and n ≥ 4.
Note: In the above problem, the result is false for n = 1, 2 and 3.
15. Prove that 2n < n! for n ≥ 4 and n ∈ N.
Solution.
Let P (n) : 2n < n! for n ≥ 4.
(1) P (4) : 24 < 4! is true (since 16 < 24).
(2) Assume P (m) is true.
That is, 2m < m!
(3) Now,

2m+1 = 2m · 2
< 2m (m + 1) since 2 < m + 1 for m ≥ 4
< m!(m + 1) by assumption
< (m + 1)!

∴ The result is true for all n ≥ 4 by the principle of mathematical


induction.
48 Discrete Mathematical Structures

16. Suppose there are n people in a room (n ≥ 1) and all shake


(n − 1)n
hands with one another. Prove that handshakes will have
2
occurred.
Solution.
(n − 1)n
Let P (n) : handshakes.
2
(1 − 1)1
(1) P (1) : = 0. There is no handshake when n = 1.
2
(2) Assume P (m) is true.
(m − 1)m
That is, there are handshakes.
2
(3) Now, if one more person enters the room, he will shake hands
with m people. So,

(m − 1)m m2 − m + 2m
+m=
2 2
2
m +m
=
2
m(m + 1)
=
2
[(m + 1) − 1](m + 1)
= .
2
∴ By the principle of mathematical induction, the result follows.
17. Use mathematical induction to prove that 3n + 7n − 2 is divisible
by 8, for n ≥ 1.
Solution.
Let P (n) : 3n + 7n − 2 is divisible by 8.
(1) P (1) : 31 + 71 − 2 = 8 is divisible by 8, which is true.
(2) Assume P (m) is true.
That is, 3m + 7m − 2 is divisible by 8.
(3) Now,

3m+1 + 7m+1 − 2 = 3 · 3m + 7 · 7m − 2
= 3 · 3m + (3 + 4) · 7m − 2
= 3 · 3m + 3 · 7m + 4 · 7m − 6 + 4
= 3(3m + 7m − 2) + 4(7m + 1).

Since both terms on right-hand side are divisible by 8,


3m+1 + 7m+1 − 2 is also divisible by 8.
∴ By the principle of mathematical induction, the result follows.
Combinatorics 49

18. Show that an − bn is divisible by a − b.


Solution.
Let P (n) : an − bn be divisible by a − b.
(1) P (1) : a1 − b1 is divisible by a − b.
(2) Assume P (m) is true.
That is, am − bm is divisible by a − b.

⇒ am − bm = k(a − b)

⇒ am = bm + k(a − b). (2.1)


(3) Now,

am+1 − bm+1 = am · a − bm · b [using (2.1)]


= ak(a − b) + abm − b · bm
= (a − b)ak + (a − b)bm
= (a − b)[ak + bm ], which is a multiple of (a − b).

Hence, am+1 − bm+1 is divisible by a − b.


∴ By the principle of mathematical induction, an − bn is divisible
by a − b for all n ≥ 1.
19. Using mathematical induction, prove that
2 + 22 + 23 + · · · + 2n = 2n+1 − 2.
Solution.
Let P (n) : 2 + 22 + 23 + · · · + 2n = 2n+1 − 2.
(1) P (1) : 21 = 21+1 − 2 = 2 is true.
(2) Assume P (m) is true.
That is, 2 + 22 + 23 + · · · + 2m = 2m+1 − 2 is true.
(3) Now,

2 + 22 + 23 + · · · + 2m + 2m+1 = 2m+1 − 2 + 2m+1


= 2 · 2m+1 − 2
= 2m+2 − 2.

∴ By the principle of mathematical induction, the result is true for


all n.
50 Discrete Mathematical Structures

20. Use mathematical induction to show that n! ≥ 2n+1 ; n = 5, 6, . . .


Solution.
Let P (n) : n! ≥ 2n+1 ; n = 5, 6 . . .
(1) P (5) : 5! ≥ 25+1 is true since 120 ≥ 64.
(2) Assume P (m) is true.
That is, m! ≥ 2m+1 ; m = 5, 6 . . . (2.2)
(3) Multiplying both sides of (2.2) by 2, we have
2(m!) ≥ 2 · 2m+1
=⇒ (m + 1)m! ≥ 2m+2
=⇒ (m + 1)! ≥ 2m+2 .
∴ By mathematical induction,

n! ≥ 2n+1 ; n = 5, 6, . . .
1 1 1 n
21. Show that + + ··· + = .
1·2 2·3 n(n + 1) n+1
Solution.
1 1 1 n
Let P (n) : + + ··· + = .
1·2 2·3 n(n + 1) n+1
1 1
(1) P (1) : = is true.
1·2 1+1
(2) Assume P (m) is true.
1 1 1 m
That is, + + ··· + = .
1·2 2·3 m(m + 1) m+1
(3) Now,
1 1 1 1
+ + ··· + +
1·2 2·3 m(m + 1) (m + 1)(m + 2)
m 1
= +
m + 1 (m + 1)(m + 2)
m(m + 2) + 1
=
(m + 1)(m + 2)
m2 + 2m + 1
=
(m + 1)(m + 2)
(m + 1)2
=
(m + 1)(m + 2)
m+1
=
m+2
m+1
= .
(m + 1) + 1
∴ By the principle of mathematical induction, the result follows.
Combinatorics 51

22. Using mathematical induction, prove that if n ≥ 1, then


1 · 1! + 2 · 2! + 3 · 3! + · · · + n · n! = (n + 1)! − 1, n ≥ 1.
Solution.
Let P (n) : 1 · 1! + 2 · 2! + 3 · 3! + · · · + n · n! = (n + 1)! − 1.
(1) P (1) : 1 · 1! = (1 + 1)! − 1 is true.
(2) Assume P (m) is true.
That is, 1 · 1! + 2 · 2! + 3 · 3! + · · · + m · m! = (m + 1)! − 1.
(3) Now,
1 · 1! + 2 · 2! + 3 · 3! + · · · + m · m! + (m + 1)(m + 1)!
= [(m + 1)! − 1] + (m + 1)(m + 1)!
= (m + 1)![1 + m + 1] − 1
= (m + 1)!(m + 2) − 1
= (m + 2)! − 1.

∴ By mathematical induction,
1 · 1! + 2 · 2! + 3 · 3! + · · · + n · n! = (n + 1)! − 1, n ≥ 1.

Pn 3n+1 − 1
23. Using mathematical induction, prove that k=0 3k = .
2
Solution.
3n+1 − 1
Let P (n) : 30 + 31 + · · · + 3n = .
2
31 − 1 2
(1) P (0) : 30 = = is true.
2 2
(2) Assume P (m) is true.
3m+1 − 1
That is, 30 + 31 + · · · + 3m = .
2
(3) Now,
3m+1 − 1
30 + 31 + · · · + 3m + 3m+1 = + 3m+1
2
3m+1 − 1 + 2 · 3m+1
=
2
3 · 3m+1 − 1
=
2
3m+2 − 1
=
2
3(m+1)+1
= .
2
Pn 3n+1 − 1
∴ By mathematical induction, k=0 3k = is true.
2
52 Discrete Mathematical Structures

24. Using mathematical induction, prove that


1 1 1 1 √
√ + √ + √ + · · · + √ > n for n ≥ 2.
1 2 3 n

Solution.
1 1 1 1 √
Let P (n) : √ + √ + √ + · · · + √ > n for n ≥ 2.
1 2 3 n
1 1 √
(1) P (2) : √ + √ = 1.707 > 2 = 1.414 is true.
1 2
(2) Assume P (m) is true.
1 1 1 1 √
That is, √ + √ + √ + · · · + √ > m for m ≥ 2.
1 2 3 m
(3) Now,
1 1 1 1 1
√ + √ + √ + ··· + √ + √
1 2 3 m m+1
√ 1
> m+ √
m+1
√ √
m m+1+1
> √
m+1
p
m(m + 1) + 1
> √
m+1

m·m+1
> √ (∵ m + 1 > m)
m+1

m2 + 1
> √
m+1
m+1
>√
m+1

> m + 1.

∴ By the principle of mathematical induction,


1 1 1 1 √
√ + √ + √ + · · · + √ > n for n ≥ 2.
1 2 3 n

25. Using mathematical induction, prove that n3 + (n + 1)3 + (n + 2)3


is divisible by 9, for n ≥ 1.
Solution.
Let P (n) : n3 + (n + 1)3 + (n + 2)3 be divisible by 9, for n ≥ 1.
(1) P (1) : 13 + (1 + 1)3 + (1 + 2)3 = 1 + 8 + 27 = 36 is divisible
by 9.
Combinatorics 53

(2) Assume P (m) is true.


That is, m3 + (m + 1)3 + (m + 2)3 is divisible by 9, for m ≥ 1.
(3) Now,

(m + 1)3 + [(m + 1) + 1]3 + [(m + 1) + 2]3


= (m + 1)3 + (m + 2)3 + (m + 3)3
= (m + 1)3 + (m + 2)3 + m3 + 9m2 + 27m + 27
[m3 + (m + 1)3 + (m + 2)3 ] + 9(m2 + 3m + 3).

Both the terms on the right-hand side are divisible by 9, and


hence, the terms on the left-hand side are also divisible by 9.
∴ By the principle of mathematical induction, the result follows.
26. Show that 32n + 4n+1 is divisible by 5, for n ≥ 0.
Solution.
Let P (n) : 32n + 4n+1 be divisible by 5, for n ≥ 0.
(1) P (0) : 30 + 41 = 5 is divisible by 5.
(2) Assume P (m) is true.

That is, 32m + 4m+1 is divisible by 5.


⇒ 32m + 4m+1 = 5k (where k is an integer)
2m m+1
⇒ 3 = 5k − 4 . (2.3)

(3) Now,

32(m+1) + 4m+1+1 = 32m · 32 + 4m+2


= (5k − 4m+1 ) · 32 + 4m+1 · 4 [using (2.3)]
m+1 2 m+1
= 5k · 9 − 4 ·3 +4 ·4
m+1 m+1
= 5k · 9 − 4 ·9+4·4
m+1
= 5k · 9 − 5 · 4
= 5(9k − 4m+1 ) which is a multiple of 5.

∴ By the principle of mathematical induction, 32n + 4n+1 is


divisible by 5 for n ≥ 0.
27. Using mathematical induction, prove that
n 1 1 1
H2n ≥ 1 + where Hk = 1 + + + ··· + .
2 2 3 k
54 Discrete Mathematical Structures

Solution.
1 1 1 n
Let P (n) : H2n = 1 + + + · · · + ≥1+ .
2 3 2n 2
0
(1) P (0) : H20 = H1 = 1 ≥ 1 + .
2
That is, 1 ≥ 1 is true.
(2) Assume P (m) is true.
1 1 1 m
That is, P (m) : H2m = 1 + + + ··· + m ≥ 1 + .
2 3 2 2
(3) Now,
1 1 1 1 1 1
H2m+1 = 1 + + + ··· + m + m + + · · · + m+1
2 3 2 2 + 1 2m + 2 2
1 1 1
= H2m + + m + · · · + m+1
2m +1 2 +2 2
 m 1 1 1
≥ 1+ + m + m + · · · + m+1
2 2 +1 2 +2 2
 m  1 
≥ 1+ + 2m · m+1 since there are 2m terms
2 2
1 
each not less than m+1
2
 m  1
≥ 1+ +
2 2
 
m+1
≥1+ .
2
n
∴ By the principle of mathematical induction, H2n ≥ 1 + .
2
28. Using mathematical induction, prove that

n(2n − 1)(2n + 1)
12 + 32 + 52 + · · · + (2n − 1)2 = .
3
Solution.
n(2n − 1)(2n + 1)
Let P (n) : 12 + 32 + 52 + · · · + (2n − 1)2 = .
3
1(2 − 1)(2 + 1)
(1) P (1) : 12 = = 1 is true.
3
(2) Assume P (m) is true.
m(2m − 1)(2m + 1)
That is, P (n) : 12 + 32 + 52 + · · · + (2m − 1)2 = .
3
Combinatorics 55

(3) Now,

12 + 32 + 52 + · · · + (2m − 1)2 + [(2(m + 1) − 1]2


m(2m − 1)(2m + 1)
= + (2m + 1)2
3
1
= [m(2m − 1)(2m + 1) + 3(2m + 1)]2
3
2m + 1
= [m(2m − 1) + 3(2m + 1)]
3
2m + 1
= [2m2 − m + 6m + 3]
3
2m + 1
= (2m2 + 5m + 3)
3
2m + 1
= (2m2 + 2m + 3m + 3)
3
2m + 1
= [2m(m + 1) + 3(m + 1)]
3
2m + 1
= (2m + 3)(m + 1)
3
1
= (m + 1)(2m + 1)(2m + 3)
3
1
= (m + 1)[2(m + 1) − 1][2(m + 1) + 1].
3
∴ By the principle of mathematical induction,
n(2n − 1)(2n + 1)
12 + 32 + 52 + · · · + (2n − 1)2 = .
3

29. Use mathematical induction to show that n3 − n is divisible by 3,


for n ∈ N.
Solution.
Let P (n) : n3 − n be divisible by 3.
(1) P (1) : 13 − 1 = 0 is divisible by 3.
(2) Assume P (m) is true.
That is, m3 − m is divisible by 3.
(3) Now,

(m + 1)3 − (m + 1) = m3 + 3m2 + 3m + 1 − m − 1
= m3 + 3m2 + 2m
= m3 − m + 3m2 + 2m + m
= (m3 − m) + 3(m2 + m).
56 Discrete Mathematical Structures

Since both m3 − m and 3(m2 + m) are divisible by 3,


(m + 1)3 − (m + 1) is also divisible by 3.
Hence, by mathematical induction, n3 − n is divisible by 3.

2.2.4 Problems for Practice


Using the principle of mathematical induction,
1. Prove that 2 + 22 + 23 + · · · + 2n = 2n+1 − 2.
2. Show that the sum of first n even integers is n2 + n.
4n3 − n
3. Prove that 12 + 32 + 52 + · · · + (2n − 1)2 = .
3
1 1 1 1 1
4. Prove that + 2 + 3 + · · · + n = 1 − n .
2 2 2 2 2
5. Show that n4 − 4n2 is divisible by 3 for all n ∈ N.
n(n + 1)(n + 2)
6. Prove that 1 · 2 + 2 · 3 + · · · + n(n + 1) = .
3
n(3n − 1)
7. Prove that 1 + 4 + 7 + · · · + (3n − 2) = .
2
1 1 1 1 n
8. Show that + + + ··· + = .
1·3 3·5 5·7 (2n − 1)(2n + 1) 2n + 1
9. Prove that the sum of the cubes of three consecutive integers is
divisible by 9.
10. Show that 22n − 1 is divisible by 3 for all n ∈ N.
11. Prove that 5n − 4n − 1 is exactly divisible by 16 for n ∈ N.
12. Show that 11n − 4n is divisible by 7 for n ∈ N.
13. Show that n + 1 < n2 for n ≥ 2.
14. Show that 2n < 3n for all n ∈ N.
15. Show that 2n < n3 for n ≥ 10.
16. Prove that n3 − n is divisible by 6.
17. Show that 52n − 25n is divisible by 7.
18. Prove that n5 − n is divisible by 5.
19. Show that 10n+1 + 10n + 1 is divisible by 3.
20. Show that n2 < 2n for n ≥ 4.
21. Show that 2n ≥ (2n + 1) for n ≥ 3.
22. Prove that

1· 2 · 3 + 2 · 3 · 4 + 3 · 4 · 5 + · · · + n(n + 1)(n + 2)
n(n + 1)(n + 2)(n + 3)
= , n ≥ 1.
4
Combinatorics 57

2.2.5 Strong Induction


Strong induction is another form of mathematical induction which is very
useful in proofs. In this form, we use the basic step similar to the principle
of mathematical induction, and for inductive step, we assume that P (k) is
true for k = 1, 2, . . . , m and then show that P (m + 1) is true based on
this assumption. Strong induction is also called as the second principle of
mathematical induction.

Procedure for strong induction


Basis step: P (1) has to be proved as true.
Inductive step: [P (1) ∧ P (2) ∧ · · · ∧ P (m)] → P (m + 1) has to be proved as
true by assuming P (k) is true for k = 1, 2, . . . , m.

Solved Example:
Show that if n is an integer greater than 1, then n can be written as the
product of primes.
Solution.
Let P (n) : n be written as the product of primes.
Basic step: P (2) is true since 2 = 1 × 2, product of primes.
Inductive step:
Assume that P (k) is true for all positive integers k with k ≤ m. To
complete the inductive step, it must be shown that P (m + 1) is true under
this assumption.
Two cases arise, namely
(i) when (m + 1) is prime
(ii) when (m + 1) is composite.
Case (i) : If (m + 1) is prime, it is obvious that P (m + 1) is true.
Case (ii) : If (m + 1) is composite, then it can be written as a product of
two positive integers a and b with 2 ≤ a < b ≤ m + 1. By the
induction hypothesis, both a and b can be written as the product
of primes. Thus, if (m + 1) is composite, it can be written as the
product of primes, namely those primes in the factorisation of a
and those in the factorisation of b.

2.2.6 Well-Ordering Property


The validity of mathematical induction follows from the following fundamental
axioms about the set of integers.
Every non-empty set of non-negative integers has a least element.
The well-ordering property can often be used directly in the proof.
58 Discrete Mathematical Structures

Solved Example:
What is wrong with this “proof” by strong induction?
Theorem: For every non-negative integer n, 5n = 0.
Proof.
Basis step: 5 · 0 = 0.
Induction step: Suppose that 5j = 0 for all non-negative integers j with
0 ≤ j ≤ m. Write m + 1 = i + j where i and j are natural numbers less than
m + 1. By the induction hypothesis,

5(m + 1) = 5(i + 1) = 5i + 5j = 0 + 0 = 0.

2.3 Pigeonhole Principle


If n pigeonholes are occupied by n + 1 or more pigeons, then at least one
pigeonhole is occupied by more than one pigeon.
Example 1. Suppose a department contains 13 professors. Then, two
of the professors (pigeons) were born in the same month
(pigeonhole).
Example 2. Suppose a laundry bag contains many red, white, and
blue socks. Then, one needs to only grab four socks
(pigeons) to be sure of getting a pair with the same colour
(pigeonhole).
Example 3. Find the minimum number of elements that one needs to take
from the set S = {1, 2, . . . , 9} to be sure that two of the
numbers add up to 10. Hence, the pigeonholes are the five
sets {1, 9}, {2, 8}, {3, 7}, {4, 6}, {5}. Thus, any choice of six
elements (pigeons) of S will guarantee that two of the numbers
add up to 10.

2.3.1 Generalized Pigeonhole Principle


If n pigeonholes are occupied by kn + 1 or more pigeons, where k is a positive
integer, then at least one pigeonhole is occupied by k + 1 or more pigeons.

2.3.2 Solved Problems


1. Find the minimum number of students in a class to be sure that
three of them were born in the same month.
Combinatorics 59

Solution.
Here, n = 12 months are the pigeonholes and k + 1 = 3 or k = 2.
Hence, among any kn + 1 = 25 students (pigeons), three of them
were born in the same month.
2. Suppose a laundry bag contains many red, white, and blue socks.
Find the minimum number of socks that one needs to choose in
order to get two pairs (four socks) of the same colour.
Solution.
There are n = 3 colours (pigeonholes) and k + 1 = 4 or k = 3. Thus,
among any kn + 1 = 10 socks (pigeons), four of them have the same
colour.
3. Assume there are n distinct pairs of shoes in a closet. Show that if
you choose n + 1 single shoes at random from the closet, you are
certain to have a pair.
Solution.
The n distinct pairs constitute n pigeonholes. The n+1 single shoes
correspond to n + 1 pigeons. Therefore, there must be at least one
pigeonhole with two shoes, and thus you will certainly have drawn
at least one pair of shoes.
4. Assume there are three men and five women at a party. Show that
if these people are lined up in a row, at least two women will be
next to each other.
Solution.
Consider the case where the men are placed so that no two men are
next to each other and not at either end of the line. In this case, the
three men generate four potential locations (pigeonholes) to place
women (at either end of the line and two locations between men
within the line). Since there are five women (pigeons), at least one
slot will contain two women who must, therefore, be next to each
other. If the men are allowed to be placed next to each other or
at the end of the line, there are even fewer pigeonholes and, once
again, at least two women will have to be placed next to each other.
5. Find the minimum number of students needed to guarantee that five
of them belong to the same class (Freshman, Sophomore, Junior,
Senior).
Solution.
Here, n = 4 classes are the pigeonholes and k+1 = 5 or k = 4. Thus,
among any kn + 1 = 17 students (pigeons), five of them belong to
the same class.
6. A student must take five classes from three areas of study. Numerous
classes are offered in each discipline, but the student cannot take
60 Discrete Mathematical Structures

more than two classes in any given area. Using pigeonhole principle,
show that the student will take at least two classes in one area.
Solution.
The three areas are the pigeonholes, and the student must take five
classes (pigeons). Hence, the student must take at least two classes
in one area.
7. Let L be a list (not necessarily in alphabetical order) of the 26
letters in the English alphabet (which consists of 5 vowels, A, E, I,
O, U, and 21 consonants).
(i) Show that L has a sublist consisting of four or more consecutive
consonants.
(ii) Assuming L begins with a vowel, say A, show that L has a
sublist consisting of five or more consecutive consonants.

Solution.
(i) The five letters partition L into n = 6 sublists (pigeonholes) of
consecutive consonants. Here, k + 1 = 4 and so k = 3. Hence,
nk + 1 = 6(3) + 1 = 19 < 21. Hence, some sublist has at least
four consecutive consonants.
(ii) Since L begins with a vowel, the remaining vowels partition
L into n = 5 sublists. Here, k + 1 = 5 and so k = 4. Hence,
kn + 1 = 21. Thus, some sublist has at least five consecutive
consonants.
8. Find the minimum number n of integers to be selected from
S = {1, 2, . . . , 9} so that
(i) the sum of two of the n integers is even
(ii) the difference of two of the n integers is 5.

Solution.
(i) The sum of two even integers or of two odd integers is
even. Consider the subsets {1, 3, 5, 7, 9} and {2, 4, 6, 8} of S as
pigeonholes. Hence, n = 3.
(ii) Consider the five subsets {1, 6}, {2, 7}, {3, 8}, {4, 9}, {5} of S
as pigeonholes. Then, n = 6 will guarantee that two integers
will belong to one of the subsets and their difference will be 5.

2.3.3 Another Form of Generalized Pigeonhole Principle


If m pigeons
  occupy n holes (m > n), then at least one hole has more than
m−1
+ 1 pigeons.
n
Here, [x] denotes the greatest integer less than or equal to x, which is a
real number.
Combinatorics 61

2.3.4 Solved Problems


1. Show that among 100 people, at least nine of them were born in
the same month.
Solution.
Here, number of pigeons = m = number of people = 100.
number of holes = n = number of months = 12.
Then by generalized pigeonhole principle, at least
     
m−1 100 − 1 99
+1= +1= + 1 = 8 + 1 = 9 were born
n 12 2
in the same month.
2. Show that if seven colours are used to paint 50 bicycles, at least
eight bicycles will be the same colour.
Solution.
Here, number of pigeons = m = number of bicycles = 50.
number of holes = n = number of colours = 7.
Then by generalized pigeonhole principle, at least
   
m−1 50 − 1
+1= + 1 = 7 + 1 = 8 bicycles will have the
n 7
same colour.
3. Show that if 25 dictionaries in a library contain a total of 40235
pages, then one of the dictionaries must have at least 1,614 pages.
Solution.
Here, number of pigeons = m = number of bicycles = 50.
number of holes = n = number of colours = 7.
Then by generalized pigeonhole principle, at least
     
m−1 40325 − 1 40324
+1= +1= + 1 = 1613 + 1 = 1614
n 25 25
pages.
4. Prove that in any group of six people, there must be at least three
mutual friends or at least three mutual enemies.
Solution.
Let those six people be A, B, C, D, E and F. Fix A. The remaining
five people can be accommodated into two groups, namely
(i) Friends of A and
(ii) Enemies of A.
Now, by generalized
 pigeonhole
 principle, at least one of the groups
5−1
must contain + 1 = 3 people.
2
62 Discrete Mathematical Structures

(i) If any two of these three people (B, C, D) are friends, then
these two together with A form three mutual friends.
(ii) If no two of these three people are friends, then these three
people (B, C, D) are mutual enemies. In either case, we get the
required conclusion.
If the group of enemies of A contains three people, by the above
similar argument, we get the required conclusion.
5. If we select ten points in the interior of an equilateral triangle of
side 1, show that there must be at least two points whose distance
1
apart is less than .
3
Solution.
Let ABC be the given equilateral triangle. Let D and E be the
points of trisection of the side AB, F and G be the points of
trisection of the side BC, and H and I be the points of trisection
of AC, so that the triangle ABC is divided into nine equilateral
1
triangles each of side .
3
A

1
D H
3
2 4
E I
6 8
5 7 9

B F G C
Equilateral triangle of side 1 unit

Here, number of pigeons = m = number of interior points = 10.


number of holes = n = number of triangles = 9.
Then by generalized pigeonhole principle, at least one triangle
contains  
10 − 1
+ 1 = 2 interior points.
2
1
Since each triangle is of length , the distance between any two
3
1
interior points of any sub-triangle cannot exceed .
3
6. Find the minimum number of students needed to guarantee that
five of them belong to the same subject, having majors as English,
Maths, Physics, and Chemistry.
Combinatorics 63

Solution.
Number of pigeonholes = Number of subjects = n = 4
Let k be the number of students (pigeons) in each subject.
Now, k + 1 = 5 ⇒ k = 4.
Therefore, the total number of students = kn + 1 = 4(4) + 1 = 17.
7. Show that if any 11 numbers from 1 to 20 are chosen, then 2 of
them will add up to 21.
Solution.
Construct the following sets with two numbers that add up to 21.

A1 = {1, 20}, A2 = {2, 19}, A3 = {3, 18}, A4 = {4, 17},


A5 = {5, 16}, A6 = {6, 15}, A7 = {7, 14}, A8 = {8, 13},
A9 = {9, 12}, A10 = {10, 11}.

By pigeonhole principle, if any 11 numbers from 1 to 20 are chosen,


then we must have to select all 2 elements from at least 1 set from
the above 10 sets, which will give the sum as 21.
8. If we select any group of 1,000 students on campus, show that at
least 3 of them must have the same birthday.
Solution.
The maximum number of days in a year is 366.
Here, number of students = number of pigeons = m = 1,000.
Number of days in a year = number of holes = n = 366.
By generalized pigeonhole principle, at least
   
m−1 1000 − 1
+1 = + 1 = 2 + 1 = 3 students must have the
n 366
same birthday.
9. How many students must be in a class to guarantee that at least
two students receive the same score on the final exam, if exam is
graded on a scale from 0 to 100 points.
Solution.
There are 101 possible scores as 0, 1, 2, . . . , 100. By pigeonhole
principle, we have 102 students. Hence, there must be at least two
students with the same score.
Therefore, the class must contain minimum 102 students.
10. Show that among (n + 1) positive integers not exceeding 2n, there
must be an integer that divides one of the other integers.
Solution.
Let the (n + 1) integers be a1 , a2 , . . . , an+1 .
64 Discrete Mathematical Structures

Each of these numbers can be expressed as an odd multiple of a


power of 2.
That is, ai = 2ki × mi ,
where ki is a non-negative integer, mi is an odd number;
i = 1, 2, . . . , n + 1.
Here, pigeon = odd positive integers m1 , m2 , . . . , mn+1 less than 2n.
Pigeonhole = n odd positive integers less than 2n.
Therefore, by pigeonhole principle, two of the integers must be
equal. Let it be mi = mj .
Now, ai = 2ki and aj = 2kj mj
ai 2ki
⇒ = kj (∵ mi = mj ).
aj 2
Case (i): If ki < kj , then 2ki divides 2kj , and hence ai divides aj .
Case(ii): If ki > kj , then aj divides ai .
11. Prove that in an equilateral triangle whose sides are of length 1 unit,
if any five points are chosen, then at least two of them lie in a
1
triangle whose sides apart is less than .
2
Solution.
Let D, E, and F be midpoints of the sides AB, BC, and AC,
respectively, so that triangle ABC is divided into four equilateral
1
triangles each of side .
2

1
D F

3
2 4

B E C
Equilateral triangle of side 1 unit
Combinatorics 65

Now, number of pigeon = m = number of interior points = 5.


Number of pigeonholes = n = number of triangles = 4.
By pigeonhole principle, at least one triangle has more than one
point (or maximum two points).
1
Since each triangle side is , the distance between two interior
2
1
points of any subtriangle is less than .
2
12. Show that among 13 children, there are at least 2 children who were
born in the same month.
Solution.
Assume the 13 children as pigeons and 12 months (from January to
December) as the pigeonholes. Then, by the pigeonhole principle,
there will be at least two children who were born in the same month.
13. Show that if any four numbers from 1 to 6 are chosen, then two of
them will add up to 7.
Solution.
The following sets contain two numbers whose sum is 7.

A1 = {1, 6}, A2 = {2, 5}, A3 = {3, 4}.

The numbers from 1 to 6 can be splitted into 3 sets above who sum
add up to 7. Hence if any four numbers from 1 to 6 are chosen,
then two of them will belong to any one of the above 3 sets whose
sum is 7.
14. Show that among any group of five (not necessarily consecutive)
integers, there are two with the same remainder when divided by 4.
Solution.
Take any group of five integers. When these are divided by 4,
each has some remainder. Since there are five integers and four
possible remainders when an integer is divided by 4, the pigeonhole
principle implies that given five integers, at least two have the same
remainder.
15. A bag contains 12 pairs of socks (each pair is in different colour). If
a person draws the socks one by one at random, determine at most
how many draws are required to get at least one pair of matched
socks.
Solution.
Let n denote the number of draws. For n ≤ 12, it is possible that
the socks drawn are of different colours, since there are 12 colours.
For n = 13, all socks cannot have different colours, and at least two
must have the same colour. Here 13 is the number of pigeons and
66 Discrete Mathematical Structures

12 colours are 12 pigeonholes. Hence, at most 13 draws are required


to have at least one pair of socks of the same colour.
16. Show that for every integer n there is a multiple of n that has only
0’s and 1’s in its decimal expansion.
Solution.
Let n be a positive integer.
Consider the n + 1 integers 1, 11, 111, . . . There are n possible
remainders when an integer is divided by n. Since there are n + 1
integers in this list, by the pigeonhole principle, there must be two
with the same remainder when divided by n.
The larger number of these integers minus the smaller one is a
multiple of n, which has a decimal expansion consisting entirely of
0’s and 1’s.
17. Prove the statement: If m = kn+1 pigeons (where k ≥ 1) occupy
n pigeonholes, then at least one pigeonhole must contain k + 1 or
more pigeons.
Solution.
Let us assume that the conclusion of the given statement is false.
Then, every pigeonhole contains k or less number of pigeons. Then,
the total number of pigeons would be nk. This is a contradiction.
Hence, the assumption made is wrong, and the given statement is
true.
18. Let n1 , n2 , . . . , nr be r objects. Show that if n1 +n2 +· · ·+nr −r +1
objects are placed in r boxes, then for some i = 1, 2, . . . , r, the ith
box contains at least ni objects.
Solution.
Assume that the conclusion part of the given statement is false.
Here n1 , n2 , . . . , nr are pigeons, r boxes are pigeonholes. Then, for
holes containing nj−1 or less number of pigeons, j = 1, 2, . . . , m.
Then, the total number of pigeons would be less than or equal to

(n1 − 1) + (n2 − 1) + · · · + (nr − 1) = n1 + n2 + · · · + nr − r = m − 1.

This is a contradiction, since the number of pigeons is equal to m.


Hence, the assumption made is wrong, and the given statement is
true.
19. Seven members of a family have totally Rs. 2886 in their pockets.
Show that at least one of them must have at least Rs. 416 in his
pocket.
Solution.
Assume “members” as pigeonholes and “rupees” as pigeons.
2886 pigeons are to be assigned to seven pigeonholes.
Combinatorics 67

By generalized pigeonhole principle, at least


   
m−1 2886 − 1
+1= + 1 = 416 rupees in one member’s
n 7
pocket.
20. If nine books are to be kept in four shelves, there must be at least
one shelf which contains at least three books.
Solution.
Assume “books” as pigeons and “shelves” as pigeonholes.
Nine pigeons are to be assigned to four shelves.
By generalized pigeonhole principle, at least
   
m−1 9−1
+1= + 1 = 3 books in one shelf.
n 4
21. How many people must you have to guarantee that at least nine of
them will have birthdays in the same day of the week.
Solution.
Assume “days in a week” as pigeonholes and “people” as pigeons.
We have to find the number of people (pigeons) to be assigned to
seven pigeonholes.
By generalized pigeonhole principle (given at least nine of them will
have birthdays in the same week),
 
m−1
+1=9
n
 
m−1
+1=9
7
m−1+7
=9
7
m+6
=9
7
m = 57.

Hence, there must be 57 people to guarantee that at least nine of


them will have birthdays in the same day of the week.
22. Show that if 30 dictionaries in a library contain a total of 61327
pages, then one of the dictionaries must have at least 2045 pages.
Solution.
Assume “pages” as pigeons and “dictionaries” as pigeonholes.
61327 pages (pigeons) are to be assigned to 30 dictionaries
(pigeonholes).
By generalized pigeonhole principle, one dictionary must contain
68 Discrete Mathematical Structures
   
m−1 61327 − 1
+1= + 1 = 2045 pages.
n 30
23. What is the maximum number of students required in a
mathematics class to be sure that at least six will receive the same
grade, if there are five possible grades A, B, C, D, and F?
Solution.
The minimum number of students needed to ensure that at least
six students
  receive the same grade is the smaller integer N such
N
that = 6. The smallest such integer is N = 5 · 5 + 1 = 26.
5
If you have only 25 students, it is possible for there to be five who
have received each grade so that no six students have received the
same grade.
Therefore, 26 is the minimum number of students needed to ensure
that at least six students will receive the same grade.
24. How many persons must be chosen in order that at least five of
them will have birthdays in the same calendar month?
Solution.
Let n be the required number of persons. Since the number of
months over which the birthdays are distributed is 12, the minimum
number of persons who have their birthdays in the  same month is,
m−1
by the generalized pigeonhole principle, equal to + 1. This
12
number is 5.
 
m−1
That is, + 1 = 5 or m = 49.
12
Hence, the number of persons is at least 49.
25. Find the least number of ways of choosing three different numbers
from 1 to 10 so that all choices have the same sum.
Solution.
From the numbers 1 to 10, we can choose three different numbers
in 10C3 = 120 ways.
The smallest possible sum that we get from a choice is
1 + 2 + 3 = 6, and the largest sum is 8 + 9 + 10 = 27. Thus, the
sums vary from 6 to 27 (both inclusive), and these sums are 22 in
number.
Accordingly, there are 120 choices (pigeons) and 22 sums
(pigeonholes).
Therefore, the least number of choices assigned
 to the same sum
120 − 1
is, by the generalized pigeonhole principle, + 1 = 6.
22
26. Show that if any five numbers from 1 to 8 are chosen, then two of
them will have their sum equal to 9.
Combinatorics 69

Solution.
Consider the following sets:

A1 = {1, 8}, A2 = {2, 7}, A3 = {3, 6}, A4 = {4, 5}.

These are the only sets containing two numbers from 1 to 8, whose
sum is 9.
Since every number from 1 to 8 belongs to one of the above sets,
each of the five numbers chosen must belong to one of the sets.
Since there are only four sets, two of the five chosen numbers have
to belong to the same set (by the pigeonhole principle).
These two numbers have their sum equal to 9.

2.3.5 Problems for Practice


1. If m is an odd positive integer, then prove that there exists a positive
integer n such that m divides 2n − 1.
2. A man hiked for 10 hours and covered a total distance of 45 km.
It is known that he hiked 6 km in the first hour and only 3 km in
the last hour. Show that he must have hiked at least 9 km within a
certain period of two consecutive hours.
3. Consider a tournament in which each of n players plays against
every other player and each player wins at least once. Show that
there are at least two players having the same number of wins.
4. Show that any set of seven distinct integers includes two integers,
x and y, such that either x + y or x − y is divisible by 10.
5. What is the minimum number of students, each of whom comes
from one of the 50 states, who must be enrolled in a university to
guarantee that there are at least 100 who come from the same state?
6. Show that if any eight positive integers are chosen, two of them will
have the same remainder when divided by 7.
7. A drawer contains a dozen brown socks and a dozen black socks, all
unmatched. A man takes socks out at random in the dark.
(i) How many socks must he take out to be sure that he has at
least two socks of the same colour?
(ii) How many socks must he take out to be sure that he has at
least two black socks?
8. There are 38 different time periods during which classes at a
university can be scheduled. If there are 677 different classes, how
many different rooms will be needed?
9. Construct a sequence of 16 positive integers that has no increasing
or decreasing subsequence of five terms.
70 Discrete Mathematical Structures

10. Suppose there are 26 students and seven cars to transport them.
Then, show that at least one car must have four or more passengers.
11. Show that in any set of eleven integers, there are two whose
difference is divisible by 15.
12. Show that in any room of people who have been doing handshaking,
there will always be at least two people who have shaken hands the
same number of times.
13. Show that if nine colours are used to paint 100 houses, at least 12
houses will be of the same colour.
14. Show that if any five integers from 1 to 8 are chosen, then at least
two of them will have a sum 9.
15. Prove that if any 30 people are selected, then we may choose a
subset of five so that all five were born on the same day of the
week.
16. Show that in any set of 11 integers, there are two whose difference
is divisible by 15.
17. A drawer contains ten black and ten white socks. What is the last
number of socks one must pull out to be sure to get a matched pair?
18. In a group of 13 children, show that there must be at least two
children who were born in the same month.
19. Prove that every set of 37 positive integers contains at least two
integers that leave the same remainder upon division by 36.
20. Let A be some fixed ten element set of {1, 2, 3, . . . , 50}. Show that
A possesses two different five element subsets, the sum of whose
elements are equal.

2.4 Permutation
Any arrangement of a set of n objects in a given order is called a permutation
of the objects (taken all at a time). An arrangement of any r ≤ n of these
objects in a given order is called an r-permutation or a permutation of the n
objects taken r at a time.
For example, consider the set of letters: a, b, c, and d. Then,
(i) bdca, dcba, and acdb are permutations of the four letters (taken all
at a time).
(ii) bad adb, cbd, and bca are permutations of the four letters taken
three at a time.
Combinatorics 71

(iii) ad, cb, da, and bd are permutations of the four letters taken two at
a time.
The number of permutations of n objects taken r at a time is denoted by
nPr or P (n, r) or Pn,r or Prn or (n)r .
We shall use nPr or P (n, r).
Example:
Find the number of permutations of six objects, say, A, B, C, D, E, and F
taken three at a time. In other words, find the number of three-letter words
using only the given six letters without repetition.
Solution.
Let the general three-letter words be represented by the following three boxes:

The first letter can be chosen in six different ways. Following this, the second
letter can be chosen in five different ways, and, following this, the last letter
can be chosen in four different ways. Write each number in its appropriate
box as follows:
6 5 4

Therefore, by the fundamental principle of counting, there are 6 × 5 × 4 = 120


possible three-letter words without repetition from the six letters, or there are
120 permutations of six objects taken three at a time.
6P3 = P (6, 3) = 120.
Formula for nP r :

n!
nPr =
(n − r)!
Remark:
(i) When r = n, then nPn = n!
(ii) There are n! permutations of n objects (taken all at a time). For
example, there are 3! = 1 × 2 × 3 = 6 permutations of the three
letters a, b, and c. They are abc, acb, bac, bca, cab, and cba.

2.4.1 Permutations with Repetitions


The number of permutations of n objects of which n1 is alike, n2 is alike,. . . ,
nr is alike is
n!
P (n; n1 , n2 , . . . , nr ) = .
n1 !n2 ! . . . nr !
72 Discrete Mathematical Structures

2.4.2 Solved Problems


1. How many seven-letter words can be formed using the letters of the
word “BENZENE”?
Solution.
There are three E’s and two N’s in the given word.
Therefore, here n = 7, n1 = 3, n2 = 2.
n!
P (n; n1 , n2 ) =
n1 !n2 !
7! 7×6×5×4×3×2×1
P (7; 3, 2) = = = 420.
3! × 2! 6×2
2. How many different signals, each consisting of eight flags hung in a
vertical line, can be formed from a set of four indistinguishable red
flags, three indistinguishable white flags, and a blue flag?
Solution.
Here n = 8, n1 = 4, n2 = 3.
n!
P (n; n1 , n2 ) =
n1 !n2 !
8!
P (8; 4, 3) = = 280.
4! × 3!
3. There are four bus lines between A and B, and three bus lines
between B and C. In how many ways can a man travel
(i) by bus from A to C by way of B?
(ii) round-trip by bus from A to C by way of B?
(iii) round-trip by bus from A to C by way of B, if he does not want
to use a bus line more than once?
Solution.
(i) There are four ways to go from A to B and three ways to go
from B to C. Hence, there are 4 × 3 = 12 ways to go from A to
C by way of B.
(ii) There are 12 ways to go from A to C by way of B and 12
ways to return. Hence, there are 12 × 12 = 144 ways to travel
round-trip.
(iii) The men will travel from A to B to C to B to A. Enter these
letters connecting arrows as follows:

A −→ B −→ C −→ B −→ A.

The man can travel four ways from A to B and three ways from
B to C, but he can only travel two ways from C to B and three
Combinatorics 73

ways from B to A since he does not want to use a bus line


more than once. Enter these numbers above the corresponding
arrows as follows:
4 3 2 3
A −→ B −→ C −→ B −→ A.

Therefore, there are 4 × 3 × 2 × 3 = 72 ways to travel round-trip


without using the same bus line more than once.
4. Suppose repetitions are not permitted.
(i) How many three-digit numbers can be formed from the six
digits 2, 3, 5, 6, 7, and 9?
(ii) How many of these numbers are less than 400?
(iii) How many are even?
Solution.
In each case, draw three boxes to represent an arbitrary
number, and then write in each box the number of digits that can
be placed there.
(i) The box on the left can be filled in six ways. Following this,
the middle box can be filled in five ways. Lastly, the box on the
right can be filled in four ways: 6 5 4 . Therefore, there
are 6 × 5 × 4 = 120 numbers.
(ii) The box on the left can be filled only in two ways by 2 or 3,
since each number must be less than 400. The middle box can
be filled in five ways. Lastly, the box on the right can be filled
in four ways. Therefore, there are 2 × 5 × 4 = 40 numbers.
(iii) The box on the right can be filled in only two ways by two or
six, since the numbers must be even. The box on the left can
be filled in five ways, and lastly, the middle box can be filled in
four ways: 5 4 2 . Therefore, there are 5 × 4 × 2 = 40
numbers.
5. Find the number of ways in which a party of seven persons can
arrange themselves:
(i) in a row of seven chairs
(ii) around a circular table.
Solution.
(i) The seven persons can arrange themselves in a row in
7 × 6 × 5 × 4 × 3 × 2 × 1 = 7! ways.
(ii) One person can sit at any place in the circular table. The other
six persons can then arrange themselves in
6 × 5 × 4 × 3 × 2 × 1 = 6! ways around the table.
74 Discrete Mathematical Structures

Remark. This is an example of a circular permutation. In general,


n objects can be arranged in a circle in
(n − 1) × (n − 2) × · · · × 3 × 2 × 1 = (n − 1)! ways.
6. Find the number of distinct permutations that can be formed from
all the letters of the word:
(i) RADAR
(ii) UNUSUAL.
Solution.
5!
(i) = 30, since there are five letters of which two are R and
2! × 2!
two are A.
7!
(ii) = 840, since there are seven letters of which three are U.
3!
7. In how many ways can four mathematics books, three history books,
three chemistry books, and two sociology books be arranged on a
shelf so that all books of the same subject are together?
Solution.
First, the books must be arranged on the shelf in four units
according to subject matter: . The box on
the left can be filled by any of the four subjects, the next by any
three remaining subjects, the next by any two remaining subjects,
and the box on the right by the last subject: 4 3 2 1 .
Therefore, there are 4 × 3 × 2 × 1 = 4! ways to arrange the books
on the shelf according to subject matter.
Now, in each of the above cases, the mathematics books can be
arranged in 4! ways, the history books in 3! ways, the chemistry
books in 3! ways, and the sociology books in 2! ways. Thus,
altogether, there are 4! × 4! × 3! × 3! × 2! = 41472 arrangements.
8. Find n if
(i) P (n, 2) = 72
(ii) P (n, 4) = 42P (n, 2)
(iii) 2P (n, 2) + 50 = P (2n, 2).
Solution.
(i) P (n, 2) = n(n − 1) = n2 − n

∴ n2 − n = 72 ⇒ n2 − n − 72 = 0
⇒ (n − 9)(n + 8) = 0
⇒ n = 9, −8
⇒ n = 9 since n is positive.
Combinatorics 75

(ii) P (n, 4) = n(n − 1)(n − 2)(n − 3) and P (n, 2) = n(n − 1)


Therefore, n(n − 1)(n − 2)(n − 3) = 42n(n − 1).
If n 6= 0, n 6= 1,
(n − 2)(n − 3) = 42
⇒ n2 − 5n + 6 = 42
⇒ n2 − 5n − 36 = 0
⇒ (n − 9)(n + 4) = 0
⇒ n = 9 since n is positive.

(iii) P (n, 2) = n(n − 1) = n2 − n


P (2n, 2) = 2n(2n − 1) = 4n2 − 2n
Therefore,
2(n2 − n) + 50 = 4n2 − 2n
⇒ 2n2 − 2n + 50 = 4n2 − 2n
⇒ 2n2 = 50
⇒ n2 = 25
⇒ n = 5 since n must be positive.

9. In how many ways can six persons occupy three vacant seats?
Solution.
Total number of ways = P (6, 3) = 6 × 5 × 4 = 120 ways.
10. How many permutations of the letters A, B, C, D, E, F, G, H
contain the string ABC?
Solution.
Since the letters A, B, and C must occur as block, we can find the
answer by finding number of permutations of six objects, namely
the block ABC and individual letters D, E, F, G, and H.
Therefore, there are 6! = 720 permutations of the letters A, B,
C, D, E, F, G, H in which ABC occurs.
11. If P (12, r) = 1320, find r.
Solution.
P (12, r) = 12 × 11 × 10 . . . r factors
⇒ 1320 = 12 × 11 × 10 . . . r factors
⇒ r = 3.

12. In how many of the permutations of ten things taken four at a


time will
(i) one thing always occur
(ii) one thing never occur.
76 Discrete Mathematical Structures

Solution.
(i) We can keep aside the particular thing which will always occur;
the number of permutations of nine things taken three at a time
is P (9, 3). Now, this particular thing can take up any one of the
4 places, and 50 can be arranged in four ways.
Therefore, the total number of permutations
= P (9, 3) × 4 = 9 × 8 × 7 × 4 = 2016.
(ii) If we are keeping the particular thing aside as never to occur,
the number of permutations of nine things (10 − 1 = 9) taken
four at a time is P (9, 4) = 9 × 8 × 7 × 6 = 3026.
13. In how many ways can six boys and four girls be arranged in a
straight line so that no two girls are ever together.
Solution.
The arrangement may be done in two operations.
(i) First, we fix the positions of six boys. Their positions are
indicated by B1 , B2 , . . . , B6 . That is,

X B1 X B2 X B3 X B4 X B5 X B6 .

This can be done in 6! ways.


(ii) If the positions of girls are fixed at places including those at the
two ends as shown by the crosses, the four girls will never come
together. In any one of these arrangements, there are seven
places for four girls, and so the girls can sit in 7P4 ways.
∴ The number of ways of seating six boys and four girls
= 7P4 × 6! = 7 × 6 × 5 × 4 × 6 × 5 × 4 × 3 × 2 × 1 = 604800.
14. There are six books on Economics, three on Commerce and two on
History. In how many ways can these be placed on a shelf of books
if the same subjects are to be together?
Solution.
Six Economics books can be arranged in 6P6 ways or 6! ways. Three
commerce books can be arranged in 3P3 ways or 3! ways. Two
history books can be arranged in 2P2 ways or 2! ways.
The three subject books, Economics, Commerce, and History
books, can be arranged in 3P3 ways or 3! ways.
∴ The total number of required arrangements
= 6! × 3! × 2! × 3! ways = 51840 ways.
15. Suppose there are six boys and five girls.
(i) In how many ways can they sit in a row?
(ii) In how many ways can they sit in a row if the boys and girls
each sit together?
(iii) In how many ways can they sit in a row if the girls are to sit
together and the boys are not to sit together?
Combinatorics 77

(iv) How many seating arrangements are there with no two girls
sitting together?
Solution.
(i) There are 6 + 5 = 11 persons, and they can sit in 11P11 ways.
That is, 11P11 = 11! ways.
(ii) The boys among themselves can sit in 6! ways, and girls among
themselves can sit in 5! ways.
They can be considered as two units and can be permuted in
2! ways.
Thus, the required seating arrangements can be done in
= 2! × 6! × 5! ways = 2 × 720 × 120 ways = 172800 ways.
(iii) The boys can sit in 6! ways and girls in 5! ways. Since girls have
to sit together, they are considered as one unit. Among the six
boys, either 0 or 1 or 2 or 3 or 4 or 5 or 6 have to sit to the left
of the girls’ units. Of these seven ways, 0 and 6 cases have to
be omitted as the boys do not sit together.
Thus, the required number of arrangements
= 5 × 6! × 5! ways = 5 × 720 × 120 ways = 432000 ways.
(iv) The boys can sit in 6! ways. There are seven places where the
girls can be placed. Thus, the total arrangements are
= 7P5 × 6! ways
7!
= × 720
2!
= 2520 × 720
= 1814400 ways.
16. Find the number of ways in which five boys and five girls can be
seated in a row if the boys and girls are to have alternate seats.
Solution.
Case (i): Boys can be arranged among themselves in 5! ways.

B B B B B
There are six places for girls. Hence, there are 6P5 × 5!
arrangements.
Case (ii): Girls can be arranged in 5! ways.

G G G G G
There are six places for boys. Hence, there are 6P5 × 5! ways.
Hence, taking two cases into account, there are 2 × 6P5 × 5!
arrangements in total.
∴ There are 2 × 120 × 6 = 240 ways.
17. How many permutations of {a, b, c, d, e, f, g}
(i) end with a
(ii) begin with c
78 Discrete Mathematical Structures

(iii) begin with c and end with a


(iv) have c and a occupying the end places?
Solution.
(i) The last position can be filled in only one way.
The remaining six letters can be arranged in 6! ways.
∴ The total number of permutations ending with a
= 6! × 1 = 720 ways.
(ii) The first position can be filled in only one way.
The remaining six letters can be arranged in 6! ways.
∴ The total number of permutations starting with c
= 1 × 6! ways = 720 ways.
(iii) The first position can be filled in only one way, and the last
position can be filled in only one way.
The remaining five letters can be arranged in 5! ways.
∴ The total number of permutations begin with c and end
with a is = 1 × 5! × 1 ways = 120 ways.
(iv) c and a occupy end positions in 2! ways, and the remaining five
letters can be arranged in 5! ways.
∴ The total number of permutations
= 5! × 2! ways = 240 ways.
18. How many bit strings of length 10 contain
(i) exactly four 1’s
(ii) at most four 1’s
(iii) at least four 1’s
(iv) an equal number of 0’s and 1’s?

Solution.
(i) A bit string of length 10 can be considered to have ten positions
and should be filled with four 1’s and six 0’s.
10!
∴ Required number of bit strings = = 210.
4! × 6!
(ii) Required number of bit strings

10! 10! 10! 10! 10!


= + + + + = 386.
0! × 10! 1 × 9! 2! × 8! 3! × 7! 4! × 6!

(iii) Required number of bit strings

10! 10! 10! 10! 10! 10! 10!


= + + + + + + = 848.
4! × 6! 5! × 5! 6! × 4! 7! × 3! 8! × 2! 9! × 1! 10! × 0!

10!
(iv) Required number of bit strings = = 252.
5! × 5!
Combinatorics 79

19. Suppose that there are 9 faculty members in the mathematics


department and 11 in the computer science department. How
many ways are there to select a committee to develop a discrete
mathematics course in a school if the committee is to consist of
three faculty members from the mathematics department and four
from the computer science department?
Solution.
By the product rule, the answer is the product of the number
of 3-combinations of a set with nine elements and the number of
4-combinations of a set with 11 elements. The number of ways to
select the committee
9! 11!
= 9C3 × 11C4 = × = 84 × 330 = 27720.
3! × 6! 4! × 7!
20. How many possibilities are there for the win, place, and show (first,
second, and third) positions in a horse race with 12 horses if all
orders of finish are possible?
Solution.
The number of ways to pick the three winners is the number of
ordered selections of three elements from 12.
∴ The required number of possibilities = 12P3 = 12 × 11 × 10 = 1320.

2.4.3 Problems for Practice


1. How many automobile license plates can be made if each plate
contains two different letters followed by three different digits? Solve
the problem if the first digit cannot be 0.
2. There are six roads between A and B and four roads between B and
C. Find the number of ways in which one can drive
(i) from A to C by way of B
(ii) round-trip from A to C by way of B
(iii) round-trip from A to C by way of B without using the same
road more than once.
3. Find the number of ways in which six people can ride a toboggan
if one of a subset of three must drive.
4. (i) Find the number of ways in which five persons can sit in a row.
(ii) How many ways are there if two of the persons insist on sitting
next to one another?
(iii) Solve (i) assuming they sit around a circular table.
(iv) Solve (ii) assuming they sit around a circular table.
5. Find the number of ways in which five large books, four medium-size
books, and three small books can be placed on a shelf so that all
books of the same size are together.
80 Discrete Mathematical Structures

6. (i) Find the number of permutations that can be performed from


the letters of the word ELEVEN.
(ii) How many of them begin and end with E?
(iii) How many of them have three E’s together?
(iv) How many begin with E and end with N?
7. (i) In how many ways can three boys and two girls sit in a row?
(ii) In how many ways can they sit in a row if the boys and girls
are each to sit together?
(iii) In how many ways can they sit in a row if just the girls are to
sit together?
8. Show that
(i) P (n, 0) + P (n, 1) + P (n, 2) + · · · + P (n, n) = 2n .
(ii) P (n, 0) − P (n, 1) + P (n, 2) − P (n, 3) + · · · + P (n, n) = 0.
9. How many bit strings of length 10 contain at least three 1’s and at
least three 0’s?
10. How many ways are there for eight men and five women to stand
in a line so that no two women stand next to each other?
11. The English alphabet contains 21 consonants and five vowels. How
many strings of six lowercase letters of the English alphabet contain
(i) exactly one vowel
(ii) exactly two vowels
(iii) at least one vowel
(iv) at least two vowels?
12. A committee of 11 members sit at a round table. In how many ways
can they be seated if the “president” and “secretary” choose to sit
together?
13. In an examination, six papers are set of which two are mathematics.
In how many ways can the examination be arranged if the
mathematics papers are not to be together?
14. In how many ways can eight people sit around a table?
15. How many numbers are there in all which consist of five digits?
16. How many odd numbers of three digits can be formed with 1, 2, 3, 4,
and 5?

2.5 Combination
Suppose we have a collection of n objects. A combination of these n objects
taken r at a time is any selection of r of the objects where order does not
Combinatorics 81

count. In other words, an r-combination of a set of n objects is any subset of


r elements.
For example, the combinations of the letters a, b, c, d taken three at a time
are
{a, b, c}, {a, b, d}, {a, c, d}, {b, c, d}
or simply abc, abd, acd, bcd, respectively.
It can be noted that the following combinations are equal:
abc, acb, bac, bca, cab and cba.
That is, each denote the same set {a, b, c}.
The number of combinations of n objects taken r at a time is denoted by
C(n, r). The symbols nCr , Cn,r and Crn can also be used.
Formula for nC r :
n!
nCr =
r!(n − r)!

2.5.1 Solved Problems


1. How many committees of three can be formed from eight people?
Solution.
Each committee is a combination of eight people taken three at a
time. Therefore, the number of committees that can be formed is
8!
8C3 = = 56.
3! × 5!
2. A farmer buys three cows, two pigs, and four hens from a man who
has six cows, five pigs, and eight hens. How many choices does the
farmer have?
Solution.
The farmer can choose the cows in 6C3 ways, the pigs in 5C2 ways,
and the hens in 8C4 ways.
Hence, altogether he can choose the animals in
6C3 × 5C2 × 8C4 = 20 × 10 × 70 = 14000 ways.
3. In how many ways can a committee consisting of three men and
two women be chosen from seven men and five women?
Solution.
The three men can be chosen from the seven men in 7C3 ways, and
the two women can be chosen from the five women in 5C2 ways.
Hence, the committee can be chosen in 7C3 × 5C2 = 350 ways.
4. How many committees of five with a given chairperson can be
selected from 12 persons?
Solution.
The chairperson can be chosen in 12 ways, and, following this, the
other four on the committee can be chosen from the 11 remaining
82 Discrete Mathematical Structures

in 11C4 ways. There are 12 × 11C4 = 12 × 330 = 3960 such


committees.
5. A bag contains six white marbles and five red marbles. Find the
number of ways in which four marbles can be drawn from the bag if
(i) they can be any colour
(ii) two must be white and two red
(iii) they must all be of the same colour.
Solution.
(i) The four marbles (of any colour) can be chosen from the 11
marbles in 11C4 = 330 ways.
(ii) The two white marbles can be chosen in 6C2 ways, and the
two red marbles can be chosen in 5C2 ways. Thus, there are
6C2 × 5C2 = 150 ways of drawing two white marbles and two
red marbles.
(iii) There are 6C4 = 15 ways of drawing four white marbles and
5C4 = 5 ways of drawing four red marbles. Thus, there are
15 + 5 = 20 ways of drawing four marbles of the same colour.
6. In how many ways can a set of five letters be selected from the
English alphabet?
Solution.
The number of ways to select five letters from 26 alphabets is
26C5 = 65780.
7. How many bit strings of length n contain exactly r 1’s?
Solution.
The positions of r 1’s in a bit string of length n form r-combination
of the set {1, 2, . . . , n}. Hence, there are nCr bit strings of length n
that contain exactly r 1’s.
Note:
(i) nCn = nC0 = 1.
(ii) nCr = nCn−r .
(iii) nCr = nP
r! .
r

(iv) nCx = nCy ⇒ n = x + y or x = y.


8. Find the value of n if 20Cn+2 = 20C2n−1 .
Solution.

Given: 20Cn+2 = 20C2n−1


⇒ n + 2 = 2n − 1 (∵ ncx = ncy ⇒ x = y)
⇒ n = 3.
Combinatorics 83

9. How many ways are there to form a committee, if the committee


consists of 3 educationalists and 4 socialists, if there are 9
educationalists and 11 socialists.
Solution.
Three educationalists can be chosen from nine educationalists in
9C3 ways.
Four socialists can be chosen from 11 socialists in 11C4 ways.
Hence, by product rule, the number of ways to select the
committee
= 9C3 × 11C4
9! 11!
= × = 27720 ways.
3! × 6! 4! × 7!
10. A team of 11 players is to be chosen from 15 members. In how many
ways can this be done if
(i) one particular player is always included
(ii) two such players have to be always included?
Solution.
(i) Let one player be fixed. The remaining players are 14. Out of
these 14 players, we have to select ten players in 14C10 = 1001
ways.
(ii) Let two players be fixed. The remaining players are 13. Out of
these 13 players, we have to select nine players in 13C9 = 715
ways.
11. Find the number of diagonals that can be drawn by joining the
angular points of a heptagon.
Solution.
A heptagon has seven angular points and seven sides. The join of
two angular points is either a side or a diagonal.
7×6
The number of lines joining the angular points = 7C2 = = 21.
1×2
But the number of sides = 7.
∴ The number of diagonals = 21 − 7 = 14.
12. There are five questions in a question paper. In how many ways can
a boy solve one or more questions?

Solution.
The boy can dispose of each question in two ways. He may either
solve it or leave it. Thus, the number of ways of disposing all the
questions = 25 .
84 Discrete Mathematical Structures

But this includes the case in which he has left all the questions
unsolved.
∴ The total number of ways of solving the paper = 25 − 1 = 31.
13. Find the value of r if 20Cr = 20Cr+2 .
Solution.

20Cr = 20Cr+2
⇒ 20Cr = 20C20−(r+2)
⇒ r = 20 − (r + 2) (∵ r = r + 2 ⇒ 2 = 0 is not possible)
⇒ 2r = 18
⇒ r = 9.

14. If nC5 = 20 · nC4 , find n.


Solution.
nC5 = 20 × C4
n(n − 1)(n − 2)(n − 3)(n − 4) n(n − 1)(n − 2)(n − 3)
= 20 ×
1×2×3×4×5 1×2×3×4
n−4
= 20
5
n − 4 = 100
n = 104.

15. From a committee consisting of six men and seven women, in how
many ways can we select a committee of
(i) three men and four women
(ii) four members that has at least one woman
(iii) four persons that has at most one man
(iv) four persons of both genders
(v) four persons in which Mr and Mrs Joseph are not included.

Solution.
(i) Three men can be selected from six men in 6C3 ways.
Four women can be selected from seven women in 7C4 ways.
∴ By product rule, the committee of three men and four
women can be selected in 6C3 × 7C4 = 700 ways.
(ii) For the committee of at least one woman, we have the following
possibilities:
Combinatorics 85

(i) One woman and three men


(ii) Two women and two men
(iii) Three women and one man
(iv) Four women and zero men.
Hence, the selection can be done in
= 7C4 × 6C3 + 7C2 × 6C2 + 7C3 × 6C1 + 7C4 × 6C0 = 700
ways.
(iii) For the committee of at most one man, we have the following
possibilities:
(i) One man and three women
(ii) Zero men and four women.
Hence, the selection can be done in
= 6C1 × 7C3 + 6C0 × 7C4 = 245 ways.
(iv) For the committee of both genders, we have the following
possibilities:
(i) One man and three women
(ii) Two men and two women
(iii) Three men and one woman
which can be done in 6C1 × 7C3 + 6C2 × 7C2 + 6C3 × 7C1 = 665
ways.
(v) Since the committee does not consist of Mr. and Mrs. Joseph,
we have five men and six women in the committee.
Now, we can select 4 members from 11 members in
11C4 = 330 ways.

2.5.2 Problems for Practice


1. A woman has 11 close friends.
(i) In how many ways can she invite five of them to dinner?
(ii) In how many ways if two of the friends are married and will
not attend separately?
(iii) In how many ways if two of them are not on speaking terms
and will not attend together?
2. A woman has 11 close friends of whom six are also women.
(i) In how many ways can she invite three or more to a party?
(ii) In how many ways can she invite three or more of them if she
wants the same number of men and women (including herself)?
3. A student is to answer 10 out of 13 questions in an exam.
(i) How many choices does he have?
(ii) How many if he must answer the first two questions?
86 Discrete Mathematical Structures

(iii) How many if he must answer the first or second question not
both?
(iv) How many if he must answer exactly three out of the first five
questions?
(v) How many if he must answer at least three of the first five
questions?
4. How many diagonals are there in a polygon of ten sides?
5. A committee is to consist of two men and three women. How many
different committees are possible if five men and seven women are
eligible.
6. How many different groups can be selected for playing tennis out
of four ladies and three gentlemen, there being one lady and one
gentleman on each side?
7. From a committee of five women and seven men, in how many ways
can a subcommittee of four be chosen so as to contain one particular
man?
8. In how many ways can a selection be made out of five oranges, eight
apples, and seven plantains?
9. In how many ways can 20 students be divided into four equal
groups?
10. How many bit strings of length 10 have
(i) exactly three 0’s
(ii) at least three 1’s
(iii) more 0’s than 1’s
(iv) an odd number of 0’s?
11. How many bit strings of length 12 contain
(i) exactly three 1’s
(ii) at least three 1’s
(iii) an equal number of 1’s and 0’s?
12. In how many ways can a party of 16 people can be conveyed in two
vehicles, one of which will not hold more than eight and the other
not more than ten?
13. In how many ways can a committee of 8 be chosen from 12 socialists
and 9 conservatives to give a socialist majority with at least 2
conservatives included?
14. A committee of 12 is to be selected from 10 men and 10 women. In
how many ways can the selection be carried out if
(i) there are no restrictions
(ii) there must be equal number of men and women
Combinatorics 87

(iii) there must be an even number of women


(iv) there must be more women than men
(v) there must be at least eight men?

2.5.3 Recurrence Relation


A recurrence relation for the sequence {fn } is a formula that expresses fn
in terms of one or more of the previous terms of the sequence, namely
f0 , f1 , . . . , fn−1 , for all integers n with n ≥ n0 , where n0 is non-negative
integer.
A sequence is called a solution of a recurrence relation if its terms satisfy
the recurrence relation.

2.5.4 Solved Problems


1. Determine whether the sequence {fn } = {3n} is a solution of the
recurrence relation: fn = 2fn−1 − fn−2 , for n = 2, 3, 4, . . .
Solution.
Suppose fn = 3n. Then for n ≥ 2,

fn = 2fn−1 − fn−2
= 2[3(n − 1)] − 3(n − 2) since fn = 3n
= 6n − 6 − 3n + 6 = 3n.

∴ {fn }, where fn = 3n, is a solution of the recurrence relation.


2. Show that the sequence {fn } is a solution of the recurrence relation
fn = −3fn−1 + 4fn−2 if fn = 2(−4)n + 3.
Solution.
fn = −3fn−1 + 4fn−2
= −3 2(−4)n−1 + 3 + 4 2(−4)n−2 + 3
   

= −6(−4)n−1 − 9 + 8(−4)n−2 + 12
= −6(−4)n−1 + 8(−4)n−2 + 3
= −6(−4)n−1 − 2(−4)n−1 + 3
= 2(−4)n + 3.

∴ fn = 2(−4)n + 3 is a solution of the recurrence relation.


Now, we discuss about a class of recurrence relations known as linear
recurrence relations with constant coefficients.
88 Discrete Mathematical Structures

2.5.5 Linear Recurrence Relation


A recurrence relation of the form

a0 fn + a1 fn−1 + a2 fn−2 + · · · + ak fn−k = f (n) (2.4)

where ai ’s are constants, is called a linear recurrence relation with constant


coefficients. The recurrence relation (2.4) is known as a k th -order recurrence
relation, provided both a0 and ak are non-zero.
Note: The phrase “k th -order” means that each term in the sequence depends
only on the k previous terms.
Example 1:
Consider the Fibonacci sequence defined by the recurrence relation
fn = fn−1 + fn−2 , n ≥ 2 and the initial conditions f0 = 0 and f1 = 1. The
recurrence relation is called a second-order relation because fn depends on
the two previous terms of fn .

Example 2:
Consider the recurrence relation f (k) − 5f (k − 1) + 6f (k − 2) = 4k + 10 defined
for k ≥ 2, together with the initial conditions f (0) = 73 and f (1) = 5. Clearly,
it is a second-order linear recurrence relation.

2.5.6 Homogenous Recurrence Relation


th
A k -order linear relation is a homogenous recurrence relation if f (n) = 0 for
all n. Otherwise, it is called non-homogenous.
Example 1:
Consider the recurrence relation C(k)−5C(k−1)+8C(k−2) = 0 together with
the initial conditions C(0) = 5 and C(1) = 2. It is a second-order homogenous
recurrence relation.

Example 2:
Which of the following recurrence relations are homogenous and which of them
are non-homogenous?
(i) fn = fn−2 .
(ii) an = an−1 + an−3 .
(iii) bn = bn−1 + 2.
(iv) s(n) = s(n − 2) + s(n − 4).
Solution.
The relations fn = fn−2 , an = an−1 + an−3 , s(n) = s(n − 2) + s(n − 4) are all
homogenous, and the relation bn = bn−1 + is non-homogenous.
Combinatorics 89

2.5.7 Recurrence Relations obtained from Solutions


Before giving an algorithm for solving a recurrence relation, we will examine
a few recurrence relations that arise from certain closed form expressions. The
procedure is illustrated by the following examples.
1. Form the recurrence relation given fn = 3 · 5n , n ≥ 0.
Solution.
If n ≥ 1, then
fn = 3 · 5n = 3 · 5 · 5n−1
= 5 · 3 · 5n−1
= 5fn−1 .
∴ The recurrence relation is fn = 5fn−1 with f0 = 3.
2. Find the recurrence relation satisfying yn = A(3)n + B(−2)n .
Solution.
Given yn = A(3)n + B(−2)n .
∴ yn+1 = A(3)n+1 + B(−2)n+1 = 3A(3)n − 2B(−2)n
yn+2 = A(3)n+2 + b(−2)n+2 = 9A(3)n + 4B(−2)n
Eliminating A and B from the above equations,
yn 1 1
yn+1 3 −2 = 0
yn+2 9 4
Expanding along column 1,
yn (12 + 18) − yn+1 (4 − 9) + yn+2 (−2 − 3) = 0
or 30yn + 5yn+1 − 5yn+2 = 0
or 6yn + yn+1 − yn+2 = 0
or yn+2 − yn+1 − 6yn = 0
which is the required recurrence relation.
3. Find the recurrence relation satisfying yn = A(3)n + B(−4)n .
Solution.
Given yn = A(3)n + B(−4)n .
∴ yn+1 = 3A(3)n − 4B(−4)n
yn+2 = 9A(3)n + 16B(−4)n
Eliminating A and B from the above equations,
yn 1 1
yn+1 3 −4 = 0
yn+2 9 16
90 Discrete Mathematical Structures

Expanding along column 1,

yn (48 + 36) − yn+1 (16 − 9) + yn+2 (−4 − 3) = 0


or 84yn − 7yn+1 − 7yn+2 = 0
or 12yn − yn+1 − yn+2 = 0
or yn+2 + yn+1 − 12yn = 0

which is the required recurrence relation.


4. Find the recurrence relation satisfying yn = (A + Bn)4n .
Solution.
Given yn = (A + Bn)4n = A(4)n + Bn(4)n .

∴ yn+1 = 4A(4)n + 4B(n + 1)(4)n


yn+2 = 16A(4)n + 16B(n + 2)(4)n

Eliminating A and B from the above equations,

yn 1 n
yn+1 4 4(n + 1) = 0
yn+2 16 16(n + 2)

Expanding along column 1,

yn [64(n + 2) − 64(n + 1)] − yn+1 [16n + 32 − 16n]


+ yn+2 [4n + 4 − 4n] = 0
or 64yn − 32yn+1 + 4yn+2 = 0
or yn+2 − 8yn+1 + 16yn = 0

which is the required recurrence relation.

2.6 Solving Linear Homogenous Recurrence Relations


Consider a linear homogenous recurrence relation of degree k with constant
coefficients
fn = a1 fn−1 + a2 fn−2 + · · · + ak fn−k
where a1 , a2 , . . . , ak are real numbers and ak 6= 0. The basic approach for
solving linear homogenous recurrence relations is to look for solutions of the
form fn = rn , where r is a constant. Note that fn = rn is a solution of the
recurrence relation fn = a1 fn−1 + a2 fn−2 + · · · + ak fn−k if and only if

rn = c1 rn−1 + c2 rn−2 + · · · + ck rn−k .


Combinatorics 91

When both sides of this equation are divided by rn−k and the right-hand side
is subtracted from the left, we obtain

rk − c1 rk−1 − c2 rk−2 − · · · − ck−1 r − ck = 0.

Consequently, the sequence {fn } with fn = rn is a solution if and only if r is


a solution of this last equation.

2.6.1 Characteristic Equation


The characteristic equation of the homogenous k th -order linear recurrence
relation fn + a1 fn−1 + a2 fn−2 + · · · + ak fn−k = 0 is the k th -degree polynomial
equation
rk + a1 rk−1 + a2 rk−2 + · · · + ak−1 rk−1 + ak = 0.
The solutions of this equation are called the characteristic roots of the
recurrence relation.
Examples:
1. The characteristic equation of
Q(k) + 2Q(k − 1) − 3Q(k − 2) − 6Q(k-4) = 0 is r4 + 2r3 − 3r2 − 6 = 0.
Note that the absence of Q(k − 3) term implies that there is no
r4−3 = r term in the characteristic equation.
2. The characteristic equation of T (k) − 7T (k − 2) + 6T (k − 3) = 0
is r3 − 7r + 6 = 0, i.e. r3 − 7r + 6 = 0, and 1, 2, and −3 are the
characteristic roots.

2.6.2 Algorithm for Solving k th -order Homogenous Linear


Recurrence Relations
Step 1:
If fn + a1 fn−1 + a2 fn−2 + · · · + ak fn−k = 0 is a given recurrence relation, then
write its characteristic equation as rk +a1 rk−1 +a2 rk−2 +· · ·+ak−1 r +ak = 0.
Step 2:
Find all the characteristic roots of this equation.
Step 3:
Case (i): If there are k distinct roots, say c1 , c2 , . . . , ck , then the general
solution of the recurrence relation is fn = A1 ck1 + A2 ck2 + · · · + Ak Ckk .
Case (ii): Suppose that c1 is a root of multiplicity m. Then, the corresponding
solution is

fn A1 rm−1 + A2 rm−2 + · · · + Am−2 r2 + Am−1 r + Am cr1 .




Step 4:
Use the boundary conditions to determine the constants A1 , A2 , . . . , Ak .
92 Discrete Mathematical Structures

2.6.3 Solved Problems


1. Solve the Fibonacci sequence {fn } defined by fn = fn−1 + fn−2 for
n ≥ 2 with the initial conditions f0 = 0 and f1 = 1.
Solution.
The characteristic equation of the given recurrence relation is
r2 − r − 1 = 0.
Solving this equation, we get
√ √
1± 1+4 1± 5
r= = .
2 2
√ √
1+ 5 1− 5
∴ c1 = , c2 = .
2 2
The general solution is
fn = A1 cn1 + A2 cn2
where A1 and A2 are constants.
Given:
f0 = 1 ⇒ A1 + A2 = 0. (2.5)

f1 = 1 ⇒ A1 c1 + A2 c2 = 1
√ ! √ !
1+ 5 1− 5
⇒ A1 + A2 = 1. (2.6)
2 2

Solving (2.5) and (2.6), we get


1 1
A1 = √ and A2 = − √ .
5 5
∴ The solution is
" √ !n √ !n #
1 1+ 5 1− 5
fn = √ − .
5 2 2

2. If the recurrence relation is un+1 − 2un = 0, find the closed form


expression (solution) for un .
Solution.
The characteristic equation is
r−2=0
⇒ r = 2.
The general solution is
un = A · 2 n
where A is a constant.
Combinatorics 93

3. Find f (n) if f (n) = 7f (n − 1) − 10f (n − 2), given that f (0) = 4


and f (1) = 17.
Solution.
The characteristic equation is

r2 − 7r + 10 = 0

⇒ r = 2, 5
∴ c1 = 2, c2 = 5.
The general solution is

f (n) = A1 cn1 + A2 cn2


= A1 2n + A2 5n .

Given:
f (0) = 4 ⇒ A1 + A2 = 4. (2.7)
f (1) = 17 ⇒ 2A1 + 5A2 = 17. (2.8)
Solving (2.7) and (2.8), we get

A1 = 3 and A2 = 3.

∴ f (n) = 2n + 3(5)n .

4. Find T (k) if T (k) − 7T (k − 2) + 6T (k − 3) = 0, where T (0) = 8,


T (1) = 6, and T (2) = 22.
Solution.
The characteristic equation is

r3 − 7r + 6 = 0.

1 1 0 −7 6
0 1 1 6
2 1 1 −6 0
0 2 6
1 3 0

The characteristic roots are

c1 = 1, c2 = 2, c3 = −3.

The general solution is

T (k) = A1 ck1 + A2 ck2 + A3 ck3


T (k) = A1 + A2 (2)k + A3 (−3)k .
94 Discrete Mathematical Structures

Given:
T (0) = 8 ⇒ A1 + A2 + A3 = 8. (2.9)
T (1) = 6 ⇒ A1 + 2A2 − 3A3 = 6. (2.10)
T (2) = 22 ⇒ A1 + 4A2 + 9A3 = 22. (2.11)

Solving (2.9), (2.10) and (2.11), we get

A1 = 5, A2 = 2, A3 = 1.

∴ T (k) = 5 + 2(2)k + 1(−3)k


or T (k) = 5+2k+1 +(−3)k .

5. Solve fk − 8fk−1 + 16fk−2 = 0 where f2 = 16 and f3 = 80.


Solution.
The characteristic equation is

r2 − 84 + 16 = 0
⇒ r = 4, 4 (repeated).

The general solution is

fk = (A1 + A2 k)4k .

Given:

f2 = 16 ⇒ (A1 + 2A2 )16 = 16


⇒ A1 + 2A2 = 1. (2.12)

f3 = 80 ⇒ (A1 + 3A2 )64 = 80


⇒ 4(A1 + 3A2 ) = 5
⇒ 4A1 + 12A2 = 5. (2.13)

Solving (2.12) and (2.13), we get


1 1
A1 = , A2 = .
2 4
∴ The solution is
 
1 1
fk = + k 4k = (2 + k)4k−1 .
2 4
Combinatorics 95

6. Find a solution to the recurrence relation


Cn = −3Cn−1 − 3Cn−2 − Cn−3 for n ≥ 3 with initial conditions
C0 = 1, C1 = −2, and C2 = 1.
Solution.
The characteristic equation is

r3 + 3r2 + 3r + 1 = 0
⇒ (r + 1)3 = 0.

∴ r = −1 is a characteristic root of multiplicity 3.


The general solution is

Cn = A1 + A2 n + A3 n2 (−1)n .


Given:
C0 = 1 ⇒ A1 = 1. (2.14)
C1 = −2 ⇒ −(A1 + A2 + A3 ) = −2. (2.15)
C2 = 1 ⇒ A1 + 2A2 + 4A3 = 1. (2.16)
Solving (2.14), (2.15), and (2.16), we get

A1 = 1, A2 = 2, A3 = −1.

∴ The solution is

Cn = 1 + 2n − n2 (−1)n .


2.7 Solving Linear Non-homogenous Recurrence


Relations
The solution of a linear non-homogenous recurrence relation with constant
coefficients is the sum of the two parts, the homogenous solution, which
satisfies the recurrence relation when the right-hand side of the equation is set
to 0, and the particular solution, which satisfies the difference equation with
f (n) on the right-hand side.
There is no general procedure for determining the particular solution of a
difference equation. However, in simple cases, this solution can be obtained
by the method of inspection. To determine the particular solution, we use the
following rules:

Rule 1:
When f (n) is of the form of a polynomial of degree m in n,

k0 + k1 n + k2 n2 + k3 n3 + · · · + km−1 nm−1 + km nm ,
96 Discrete Mathematical Structures

the corresponding particular solution will be of the form

Q0 + Q1 n + Q2 n2 + Q3 n3 + · · · + Qm−1 nm−1 + Qm nm .

Rule 2:
When f (n) is of the form

k0 + k1 n + k2 n2 + · · · + km−1 nm−1 + km nm an ,


the corresponding particular solution is of the form

Q0 + Q1 n + Q2 n2 + · · · + Qm−1 nm−1 + Qm nm an


if a is not a characteristic root of the recurrence relation.


Rule 3:
If a is a characteristic root of multiplicity r − 1, when f (n) is of the form

k0 + k1 n + k2 n2 + · · · + Km−1 nm−1 + K − mnm an ,




the corresponding particular solution is of the form

rn−1 Q0 + Q1 n + Q2 n2 + · · · + Qm−1 nm−1 + Qm nm an .




Note:
The general solution of the recurrence relation is the sum of the homogenous
solution and particular solution. If no initial conditions are given, then you
have finished. If m initial conditions are given, obtain m linear equations in
m unknowns and solve the system, if possible, to get a complete solution.

2.7.1 Solved Problems


1. Solve S(k)−S(k−1)−6S(k−2) = −30 where S(0) = 20, S(1) = −5.
Solution.
The associated homogenous relation is

S(k) − S(k − 1) − 6S(k − 2) = 0.

The characteristic equation is

r2 − r − 6 = 0.

The characteristic roots are r = −2, 3.


The homogenous solution is

A1 (−2)k + A2 (3)k .

Since the right-hand side of

S(k) − S(k − 1) − 6S(k − 2) = −30 (2.17)


Combinatorics 97

is a constant, by Rule 1, the particular solution will be a constant,


say Q. Substituting Q into (2.17), we obtain

Q − Q − 6Q = −30
⇒ Q = 5.

∴ The general solution is

S(k) = A1 (−2)k + A2 (3)k + 5.

Using the initial conditions, we have

S(0) = 20 ⇒ A1 + A2 + 5 = 20. (2.18)

S(1) = −5 ⇒= 2A1 + 3A2 + 5 = −5. (2.19)


Solving (2.18) and (2.19), we get A1 = 11, A2 = 4.
∴ The complete solution is

S(k) = 11(−2)k + 4(3)k + 5.

2. Solve the recurrence relation fn − 5fn−1 + 6fn−2 = 1.


Solution.
The associated homogenous relation is

fn − 5fn−1 + 6fn−2 = 0.

The characteristic equation is r2 − 5r + 6 = 0.


The characteristic roots are r = 2, 3.
The homogenous solution is A2 (2)n + A2 (3)n .
Since the right-hand side of the given relation is 1 (a constant), by
Rule 1, the particular solution will also be a constant, say Q.

Q − 5Q + 6Q = 1
1
⇒ Q= .
2
∴ The complete solution is
1
fn = A1 (2)n + A2 (3)n + .
2

3. Find the particular solution of the recurrence relation

f (n) + 5f (n − 1) + 6f (n − 2) = 3n2 − 2n + 1. (2.20)


98 Discrete Mathematical Structures

Solution.
By rule 2, the particular solution is of the form

Q0 + Q1 n + Q2 n2 . (2.21)

Substituting (2.21) in (2.20), we get

Q0 + Q1 n + Q2 n2 + 5 Q0 + Q1 (n − 1) + Q2 (n − 1)2
  

+ 6 Q0 + Q1 (n − 2) + Q2 (n − 2)2 = 3n2 − 2n + 1
 

which simplifies to

(12Q0 − 17A1 + 29Q2 ) + (12Q1 − 34Q2 )n + 12Q2 n2


= 3n2 − 2n + 1. (2.22)

Comparing both sides of (2.22), we obtain

12Q2 = 3; 12Q1 − 34Q2 = −2; 12Q0 − 17Q1 + 29Q2 = 1

which gives
1 13 71
Q2 = ; Q1 = ; Q0 = .
4 24 288
∴ The particular solution is
71 13 1
+ n + n2 .
288 24 4

4. Solve ar + 5ar−1 = 9 with initial condition a0 = 6.


Solution.
The associated homogenous relation is

ar + 5ar−1 = 0.

The characteristic equation is r + 5 = 0.


The characteristic root is r = −5.
The homogenous solution is A(−5)r .
Since the right-hand side of the given relation is a constant, the
particular solution will also be a constant Q. Substituting in the
relation, we get

Q + 5Q = 9
3
⇒ Q= .
2
∴ The general solution is
3
Ar = A(−5)r + .
2
Combinatorics 99

Given:
3
a0 = 6 ⇒ A + =6
2
9
⇒A= .
2
∴ The complete solution is
9 3
ar = (−5)r + .
2 2

5. Solve the recurrence relation f (n) − 7f (n − 1) + 10f (n − 2) = 6 + 8n


with f (0) = 1 and f (1) = 2.
Solution.
The characteristic equation is

r2 − 7r + 10 = 10
⇒ r = 2, 5.

Homogenous solution is A1 (2)n + A2 (5)n .


By Rule 1, the particular solution is of the form Q0 + Qn1 .
Substituting in the given relation, we obtain

(Q0 + Qn1 ) − 7[Q0 + Q1 (n − 1)] + 10[Q0 + Q1 (n − 2)] = 6 + 8n.

Comparing both sides, we obtain

4Q0 − 13Q1 = 6 and 4Q1 = 8

which yield Q0 = 8 and Q1 = 2.


∴ The particular solution is 8 + 2n.
The general solution is

f (n) − A1 (2)n + A2 (5)n + 8 + 2n.

Given:
f (0) = 1 ⇒ A1 + A2 + 8 = 1. (2.23)
f (1) = 2 ⇒ 2A1 + 5A2 + 10 = 2. (2.24)
Solving (2.23) and (2.24), we get A1 = −9, A2 = 2.
∴ The complete solution is

f (n) = −9(2)n + 2(5)n + 8 + 2n.


100 Discrete Mathematical Structures

6. Find the particular solution of the recurrence relation


an + 5an−1 + 6an−2 = 42(4)n .
Solution.
The characteristic equation is r2 + 5r + 6 = 0.
The characteristic roots are r = −2, −3.
Since 4 is not a characteristic root, by Rule 2, we assume that the
general form of the particular solution is Q · (4)n . Substituting in
the given relation, we obtain

Q · (4)n + 5Q · (4)n−1 + 6Q · (4)n−2 = 42(4)n


⇒ Q · 4n−2 [16 + 20 + 6] = 42(4)n
⇒ Q · 4n−2 (42) = 42(4)n
⇒ Q = 16.

∴ The particular solution is 16(4)n = 4n+2 .


7. Find the particular solution of the recurrence relation
fn + fn−1 = 3n2n .
Solution.
The characteristic equation is r + 1 = 0.
The characteristic root is r = −1.
Since 2 is not a characteristic root, by Rule 2, the general form of
the particular solution is (Q0 + Q1 n)2n .
Substituting in the given relation, we obtain

(Q0 + Q1 n)2n + [Q0 + Q1 (n − 1)]2n − 1 = 3n2n

which simplifies to
1 1 1
Q0 2n + Q1 n2n + Q0 2n + Q1 n2n − Q1 n2n = 3n2n
 2 2 2
3 1 3
⇒ Q0 − Q1 2 + Q1 n2 = 3n2n
n n
2 2 2
3 1
⇒ Q0 − Q1 = 0
2 2
3
and Q1 = 3.
2
2
Solving, we get Q0 = 3 and Q1 = 2.
 
2
∴ The particular solution is + 2n 2n .
3
8. Find the particular solution of the recurrence relation
f (n) − 2f (n − 1) = 3 · 2n .
Combinatorics 101

Solution.
The characteristic equation is r − 2 = 0.
The characteristic equation is r = 2.
Since r = 2 is the characteristic root of multiplicity 1, by Rule 3,
the general form of the particular solution is Qn · 2n .
Substituting in the given relation, we obtain

Qn2n − 2 Q · (n − 1)2n−1 = 3 · 2n
 

⇒ Qn2n − Qn2n + Q2n = 3 · 2n


⇒ Q = 3.

∴ The particular solution is 3n(2n ).


9. Find the general solution of

f (n) − 3f (n − 1) − 4f (n − 2) = 4n . (2.25)
Solution.
The associated homogenous relation is

f (n) − 3f (n − 1) − 4f (n − 2) = 0.

The characteristic equation is r2 − 3r − 4 = 0.


The characteristic roots are r = −1, 4.
The homogenous solution is A1 (−1)n + A2 (4)n . Since 4 is a
characteristic root, by Rule 3, we assume that the general form
of the particular solution is Qn4n .
Substituting in (2.25), we obtain

Qn4n − 3Q · (n − 1)4n−1 − 4Q · (n − 2)4n−2 = 4n


⇒ Qn4n − 3Qn4n−1 + 3Q4n−1 − 4Qn4n−2 + 8Q4n−2 = 4n
⇒ (16Qn − 12Qn + 12Q − 4Qn + 8Q)4n−2 = (16)4n−2
⇒ 20Q = 16
4
⇒ Q= .
5
4
∴ The particular solution is n(4)n .
5
The general solution of the given recurrence relation is
4
f (n) = A1 (−1)n + A2 (4)n + n4n .
5
102 Discrete Mathematical Structures

Remark:
What if the characteristic equation gives rise to complex roots? Here,
our methods are still valid, but the method for expressing the solutions
of the recurrence relations is different. Since an understanding of these
representations require some background in complex numbers, we suggest
that an interested reader refer to a more advanced treatment of recurrence
relations.

2.7.2 Problems for Practice


1. Find the general solution of the following recurrence relations.
(i) fn − 3fn−1 − 10fn−2 = 0
(ii) fn+2 + 6fn+1 + 9fn = 0
(iii) 2fn + 2fn−1 − fn−2 = 0
(iv) fn − 3fn−1 − 4fn−2 .
2. Solve the following recurrence relations.
(i) f (n) − 10f (n − 1) + 9f (n − 2) = 0; f (0) = 3; f (1) = 11
(ii) f (n) − 9f (n − 1) + 18f (n − 2) = 0; f (0) = 1; f (1) = 4
(iii) f (n + 2) − 8f (n + 1) + 16f (n) = 0; f (0) = 0; f (1) = 8
(iv) f (n) − 3f (n − 1) + 3f (n − 2) − f (n − 3) = 0; f (1) = 0;
f (2) = 1; f (3) = 0
(v) f (n + 2) − 2f (n + 1) + f (n) = 0; f (0) = 1; f (1) = 2
(vi) f (n) − 20f (n − 1) + 100f (n − 2) = 0; f (0) = 2; f (1) = 30.
3. Find the recurrence relation satisfying
(i) yn = A(3)n + B(8)n
(ii) yn = (A + Bn)(−2)n
(iii) yn = (A + Bn)(6)n
(iv) yn = A(3)n + B(5)n
(v) yn = 2(3)n .
4. Solve the following set of recurrence relations with the initial
conditions.
(i) yn − 2yn−1 = 6n; y1 = 2
(ii) yn+2 + 2yn+1 − 15yn = 6n + 10; y0 = 1; y1 = − 21
(iii) yn+1 + 2yn = 3 + 4n ; y0 = 2
(iv) yn+2 − 2yn+1 + yn = 1; y0 = 1; y1 = 12
(v) yn+1 + yn = 5; y0 = 1
(vi) yn − 3yn−1 + 2yn−2 − n2 ; y0 = 0; y1 = 0
(vii) yn − 4yn−1 + 4yn−2 = 3n + 2n ; y0 = 1; y1 = 1
(viii) yn − 5yn−1 = 5n ; y0 = 3.
Combinatorics 103

2.8 Generating Functions


In this section, we will show how recurrence relations can be solved using the
powerful generating function method. Generating function is an important
tool in discrete mathematics, and its use is by no means confined to the
solution of recurrence relations.
If a0 , a1 , a2 , . . . , an is a finite sequence of numbers, the generating function
for the an ’s is the polynomial
n
X
G(z) = ak z k = a0 + a1 z + a2 z 2 + · · · + an z n
k=0

where z is an indeterminate (that is, an abstract) symbol. If


a0 , a1 , a2 , . . . , an , . . . is an infinite sequence of numbers, its generating function
is defined to be

X
G(z) = ak z k = a0 + a1 z + a2 z 2 + . . . .
k=0

The symbol z is just the name given to a variable and has no special
significance. For any sequence {an }, we write G(z) to denote the generating
function of {an }. Clearly, given a sequence, we can easily obtain its generating
function and its converse. For example, the generating function of an = αn ,
n ≥ 0 is
α0 + αz + α2 z 2 + α3 z 3 + . . . (2.26)
1
We note that the infinite series (2.26) can be written in closed form as
1 − αz
which is a rather compact way to represent the sequence {an } or (a, α, α2 , . . . ).

2.8.1 Solved Problems


1. Find the generating function for the sequence 1, 1, 1, 1, 1, 1.
Solution.
By definition, the generating function of 1, 1, 1, 1, 1, 1 is
z6 − 1
G(z) = 1 + z + z 2 + z 3 + z 4 + z 5 = .
z−1

2. Let n be a positive integer. Let ak = C(n, k) for k = 0, 1, 2, . . . , n.


Find the generating function for the sequence a0 , a1 , . . . , an .
Solution.
The generating function for this sequence is

G(z) = C(n, 0) + C(n, 1)z + C(n, 2)z 2 + · · · + c(n, n)z n


= (1 + z)n , by binomial theorem.
104 Discrete Mathematical Structures

3. Find the generating function for the infinite sequence 1, α, α2 , α3 , . . . ,


where α is a fixed constant.
Solution.
The generating function for this sequence is
G(z) = 1 + αz + α2 z 2 + α3 z 3 + . . .
1
= .
1 − αz

4. Find the generating function for fn = 3n , n ≥ 0 in closed form.


Solution.
The generating function for this sequence is
G(z) = 1 + 3z + (3z)2 + (3z)3 + . . .
1
= .
1 − 3z

5. Find the generating function (in closed form) of the Fibonacci


sequence {fn } defined by
fn = fn−1 + fn−2 ; f0 = 0; f1 = 1.
Solution.
The generating function is

X
G(z) = f0 + f1 z + f2 z 2 + f3 z 3 + · · · = fn z n . (2.27)
n=0

Consider fn = fn−1 + fn−2 ; n ≥ 0.


Multiplying both sides by z n and summing over all n ≥ 2, we get

X ∞
X ∞
X
fn z n = fn−1 z n + fn−2 z n . (2.28)
n=2 n=2 n=2

Consider the first sum



X
fn z n = f2 z 2 + f3 z 3 + . . .
n=2
= G(z) − f0 − f1 z [using (2.28)] .
Similarly,

X
fn−1 z n = f1 z 2 + f2 z 3 + . . .
n=2
= z(f1 z + f2 z 2 + . . . )
= z[G(z) − f0 ] [using (2.28)]
Combinatorics 105

and

X
fn−2 z n = f0 z 2 + f1 z 3 + f2 z 4 + . . .
n=2
= z 2 (f0 + f1 z + f2 z 2 + . . . )
= z 2 [G(z)] [using (2.28)] .

Substituting these expressions in (2.28), we obtain

G(z) − f0 − f1 z = z[G(z) − f0 ] + z 2 G(z).

Since f0 = 0 and f1 = 1, we get

G(z) − z = zG(z) + z 2 G(z)


⇒ G(z)(1 − z − z 2 ) = z
z
⇒ G(z) =
1 − z − z2
which is the required generating function.
6. Find the generating function of the sequence y0 , y1 , . . . , yn defined
as follows:
yn + 2yn−1 − 15yn−2 = 0 for n ≥ 2 with y0 , y1 = 1. (2.29)
Solution.
The generating function is

X
G(z) = y0 + y1 z + y2 z 2 + · · · = yn z n .
n=2

Multiplying (2.29) by z n and summing over all n ≥ 2, we get



X ∞
X ∞
X
n n
yn z + 2 yn−1 z − 15 yn−2 z n = 0
n=2 n=2 n=2
y0 − y12 + 2 [z{G(z) − y0 }] − 15z 2 G(z) = 0.
 
⇒ G(z) −

Since y0 = 0, y1 = 1, we get

[G(z) − 1] + 2[zG(z) − 15z 2 G(z)] = 0


⇒ G(z)(1 + 2z − 15z 2 ) = z
z
⇒ G(z) =
1 + 2z − 15z 2
z
⇒ G(z) = .
(1 − 3z)(1 + 5z)
106 Discrete Mathematical Structures

2.8.2 Solution of Recurrence Relations Using Generating


Function
We can find the solution to a recurrence relation with its initial conditions
by finding an explicit formula for the associated generating function. This
is illustrated in the following examples. The following are some important
fundamental results useful for solved examples presented below:
P∞
1. (1 + x)n = r=0 n(n−1)...(n−r+1)
r! xr
P∞
2. (1 + x)−n = r=0 (−1)r n(n−1)...(n−r+1)
r!
P∞
3. (1 + x)−1 = r=0 (−1)r xr = 1 + x + x2 + . . .
P∞
4. (1 + x)−2 = r=0 (−1)r (r + 1)xr = 1 − 2x + 3x2 − 4x3 + . . .
P∞
5. (1 − x)−1 = r=0 xr = 1 + x + x2 + . . .

6. (1 − x)−2 = r=0 (r + 1)xr = 1 + 2x + 3x2 + 4x3 + . . .
P
P∞ r 2 3
7. ex = r=0 xr! = 1 + 1! x
+ x2! + x3! + . . .

2.8.3 Solved Problems


1. Using the generating function, solve the recurrence relation
fn = 3fn−1 , for n = 1, 2, 3 . . . and initial condition f0 = 2.
Solution.
Let the generating function be

X
G(z) = fn z n . (2.30)
n=0

Multiplying the given relation by z n and summing for all n ≥ 1, we


obtain
X∞ ∞
X
fn z n = 3 fn−1 z n . (2.31)
n=1 n=1

The first sum



X
fn z n = f1 z + f2 z 2 + · · · = G(z) − f0 [using (2.30)]
n=1

and the second sum



X
fn−1 z n = f0 z + f1 z 2 + f2 z 3 + . . .
n=1
= z[f0 + f1 z + f2 z 2 + . . . ]
= zG(z) [using (2.30)] .
Combinatorics 107

Hence, (2.31) becomes

G(z) − f0 = 3zG(z)
⇒ G(z) − 2 = 3zG(Z)

2 X
⇒ G(z) = = 2(1 − 3z)−1 = 2 (3z)n
1 − 3z n=0

X
= 2 · 3n z n . (2.32)
n=0

Comparing (2.30) and (2.32), we get

fn = 2 · 3n

which is the required solution.


2. Using generating function, solve the recurrence relation yn = 3yn−1 + 2;
n ≥ 1 with y0 = 1.
Solution.
Let the generating function be

X
G(z) = yn z n . (2.33)
n=0

Given:
yn = 3yn−1 + 2. (2.34)
Multiplying both sides of (2.34) by z n and summing for n ≥ 1, we
get
X∞ ∞
X ∞
X
yn z n = 3 yn−1 z n + 2 zn. (2.35)
n=1 n=1 n=1

Consider the first sum



X
yn z n = y1 z + y2 z 2 + · · · = G(z) − y0 [using (2.33)] ,
n=1

the second sum



X
yn−1 z n = y0 z + y1 z 2 + y2 z 3 + . . .
n=1
= z[y0 + y1 z + y2 z 2 + . . . ]
= zG(z) [using (2.33)] ,
108 Discrete Mathematical Structures

and the third sum



X
= z + z2 + z3 + . . .
n=1
= z(1 + z + z 2 + . . . )
= z(1 − z)−1
z
= .
1−z
Hence, (2.35) becomes
2z
G(z) − y0 = 3zG(z) + .
1−z
Using y0 = 1,
2z 1+z
G(z)(1 − 3z) = 1 + = .
1−z 1−z
1+z
⇒ G(z) = .
(1 − z)(1 − 3z)

Using partial fraction,


1+z A B
= +
(1 − z)(1 − 3z) 1 − 3z 1−z
⇒ 1 + z = A(1 − z) + B(1 − 3z).

Put z = 1 =⇒ 2 = −2B =⇒ B = −1.


1 4 2
Put z = =⇒ = A =⇒ A = 2.
3 3 3
2 1
∴ G(z) = − = 2(1 − 3z)−1 − (1 − z)−1
1 − 3z 1−z

X ∞
X
=2 (3z)n − zn
n=0 n=0

X
= (2 · 3n − 1) z n . (2.36)
n=0

Comparing (2.33) and (2.36), we get

yn = 2 · 3n − 1.

3. Using generating function, solve the difference equation

yn+2 − 4yn+1 + 3yn = 0; y0 = 2; y1 = 4.


Combinatorics 109

Solution.
Let the generating function be

X
G(z) = yn z n . (2.37)
n=0

Multiplying the given relation by z n and summing for n ≥ 0, we


get
X∞ ∞
X ∞
X
yn+2 z n − 4 yn+1 z n + 3 yn z n = 0. (2.38)
n=0 n=0 n=0

Consider the first sum



X
yn+2 z n = y2 + y3 z + y4 z 2 + y5 z 3 + . . .
n=2
1
y2 z 2 + y3 z 3 + y4 z 4 + . . .

=
z2
1
= [G(z) − y0 ] [using (2.37)] ,
2
the second sum

X
= y1 + y2 + y3 z 2 + y4 z 3 + . . .
n=0
1
= [y1 z + y2 z 2 + y3 z 3 + . . . ]
z
1
= [G(z) − y0 ] [using (2.37)] ,
z
and the third sum

X
yn z n = G(z) [using (2.37)] .
n=0

∴ (2.38) becomes
1 4
2
[G(z) − y0 − y1 z] − [G(z) − y0 ] + 3G(z) = 0.
z z
Since y0 = 2, y1 = 4,
1 4
[G(z) − 2 − 4z] − [G(z) − 2] + 3G(z) = 0
z2 z
=⇒ [G(z) − 4z − 2] − 4z[G(z) − 2] + 3z 2 G(z) = 0
=⇒ G(z)[1 − 4z + 3z 2 ] = 2 − 4z
2 − 4z 2 − 4z
=⇒ G(z) = = .
1 − 4z + 3z 2 (1 − z)(1 − 3z)
110 Discrete Mathematical Structures

Using partial fractions,


2 − 4z A B
= +
(1 − z)(1 − 3z) 1−z 1 − 3z
=⇒ 2 − 4z = A(1 − 3z) + B(1 − z).

Put z = 1 =⇒ −2 = −2A =⇒ A = 1.
Put z = 13 =⇒ 23 = 32 B =⇒ B = 1.

1 1
∴ G(z) = + = (1 − z)−1 + (1 − 3z)−1
1−z 1 − 3z
X∞ X∞
= zn + (3z)n
n=0 n=0
X∞
= (1 + 3n )z n . (2.39)
n=0

Comparing (2.37) and (2.39), the required solution is

yn = 1 + 3n .

4. Using generating function, solve the difference equation


yn+2 − 6yn+1 + 8yn = 0, y0 = 1, y1 = 4.
Solution.
Let the generating function be

X
G(z) = yn z n . (2.40)
n=0

Multiplying the given equation by z n and summing for n ≥ 0, we


get
X∞ X∞ X∞
n n
yn+2 z − 6 yn+1 z + 8 yn z n = 0. (2.41)
n=0 n=0 n=0

Consider the first sum



X
yn+2 z n = y2 + y3 z + y4 z 2 + y5 z 3 + . . .
n=0
1
y2 z 2 + y3 z 3 + y4 z 4 + . . .

= 2
z
1
= 2 [G(z) − y0 − y1 z] [using (2.40)] ,
z
Combinatorics 111

the second sum



X
yn+1 z n = y1 + y2 z + y3 z 2 + . . .
n=0
1
= (y1 z + y2 z 2 + y3 z 3 + . . . )
z
1
= [G(z) − y0 ] [using (2.40)] ,
z
and the third sum

X
yn z n = G(z) [using (2.40)] .
n=0

Hence, (2.41) becomes


1 6
2
[G(z) − y0 − y1 z] − [G(z) − y0 ] + 8G(z) = 0
z z
=⇒ [G(z) − y0 − y1 z] − 6z[G(z) − y0 ] + 8z 2 G(z) = 0
=⇒ G(z)[1 − 6z + 8z 2 ] = 1 − 2z (∵ y0 = 1, y1 = 4)
1 − 2z 1 − 2z
=⇒ G(z) = =
1 − 6z + 8z 2 (1 − 2z)(1 − 4z)
X∞
−1
= (1 − 4z) = 4n z n . (2.42)
n=0

Comparing (2.40) and (2.42), the solution is yn = 4n .


5. Solve S(k) − 7S(k − 2) + 6S(k − 3) = 0, S(0) = 8, S(1) = 6, and
S(2) = 22.
Solution.
Let the generating function be

X
G(z) = S(k)z k . (2.43)
k=0

Multiplying the given equation by z k and summing for k ≥ 3, we


get

X ∞
X ∞
X
S(k)z k − 7 S(k − 2)z k + 6 S(k − 3)z k = 0. (2.44)
k=3 k=3 k=3

Consider the first sum



X
S(k)z k = S(3)z 3 + s(4)z 4 + s(5)z 5 + . . .
k=3
= G(z) − S(0)S(1)z − S(2)z 2 [using (2.43)] ,
112 Discrete Mathematical Structures

the second sum



X
s(k − 3)z k = S(0)z 3 + S(1)z 4 + S(2)z 5 + . . .
k=3
= z 3 [S(0) + S(1)z + S(2)z 2 + . . . ]
= z 3 G(z) [using (2.43)] ,

and the third sum



X
s(k − 3)z k = S(0)z 3 + S(1)z 4 + S(2)z 5 + . . .
k=3
= z 3 [S(0) + S(1)z + S(2)z 2 + . . . ]
= z 3 G(z) [using (2.43)] .

∴ (2.44) becomes

[G(z) − s(0) − s(1)z − s(2)z 2 ] − 7z 2 [G(z) − s(0)] + 6z 3 G(z) = 0.

Since S(0) = 8, S(1) = 6, S(2) = 22,

[G(z) − 8 − 6z − 22z 2 ] − 7z 2 [G(z) − 8] + 6z 3 G(z) = 0


=⇒ G(z)[1 − 7z 2 + 6z 3 ] = 8 + 6z − 34z 2
8 + 6z − 34z 2
=⇒ G(z) =
1 − 7z 2 + 6z 3
8 + 6z − 34z 2
=⇒ G(z) = .
(1 − z)(1 − 2z)(1 − 3z)
Using partial fractions,
8 + 6z − 34z 2 A B C
= + +
(1 − z)(1 − 2z)(1 − 3z) 1−z 1 − 2z 1 + 3z
=⇒ 8 + 6z − 34z 2 = A(1 − 2z)(1 + 3z)
+ B(1 − z)(1 + 3z) + C(1 − z)(1 − 2z).

Put z = 1 =⇒ −20 = −4A =⇒ A = 5.


1 10 5
Put z = =⇒ = B =⇒ B = 2.
2 4 4
1 20 20
Put z = − =⇒ = C =⇒ C = 1.
3 9 9

5 2 1
∴ G(z) = + +
1−z 1 − 2z 1 + 3z
= 5(1 − z)−1 + 2(1 − 2z)−1 + (1 + 3z)−1
X∞ X∞ X∞
=5 zk + 2 (2z)k + (−3)k z k
k=0 k=0 k=0
Combinatorics 113

X
5 + 2k+1 + (−3)k z k .
 
= (2.45)
k=0

Comparing (2.43) and (2.45), the solution is


S(k) = 5 + 2k+1 + (−3)k .

6. Suppose that a valid code word is an n-digit number in decimal


notation containing an even number of 0’s. Let an denote the
number of valid code words of length n. The sequence {an } satisfies
the recurrence relation an = 8an−1 +10n−1 and the initial condition
a1 = 9. Use generating function to find an explicit formula for an .
Solution.
To make our work with generating function simpler, we extend this
sequence by setting a0 = 1 so that a1 = 8a0 + 100 = 9.
Let the generating function be

X
G(z) = an z n . (2.46)
n=0

Multiplying both sides of an = 8an−1 + 10n−1 by z n and summing


for all n ≥ 1, we get

X ∞
X ∞
X
an z n = 8 an−1 z n + 10n−1 z n . (2.47)
n=1 n=1 n=1

Consider the first sum


X∞
an z n = a1 z + a2 z 2 + a3 z 3 + · · · = G(z) − a0 [using (2.47)] ,
n=1

the second sum


X∞
an−1 z n = a0 z + a1 z 2 + a2 z 3 + . . .
n=1
= z(a0 + az + a2 z 2 + . . . )
= zG(z) [using (2.46)] ,
and the third sum
X∞ ∞
X
10n−1 z n = z (10z)n
n=1 n=1
z
= z(1 − 10z)−1 = .
1 − 10z
∴ (2.47) becomes
z
[G(z) − a0 ] = 8zG(z) + .
1 − 10z
114 Discrete Mathematical Structures

Using a0 = 1,
z 1 − 9z
G(z)[1 − 8z] = 1 + =
1 − 10z 1 − 10z
1 − 9z
=⇒ G(z) = .
(1 − 8z)(1 − 10z)
Using partial fraction,
1 − 9z A B
= +
(1 − 8z)(1 − 10z) 1 − 8z 1 − 10z
=⇒ 1 − 9z = A(1 − 10z) + B(1 − 8z).

 
1 1 5
z = =⇒ − = A 1 −
8 8 4
1 1
=⇒ − = − A
8 4
1
=⇒ A = .
2

1 1 1
z= =⇒ = B
10 10 5
1
=⇒ B = .
2

1 1
∴ G(z) = (1 − 8z)−1 + (1 − 10z)−1
2 2
∞ ∞
1 X
n 1X
= (8z) + (10z)n
2 n=0 2 n=0

X 1 n
= (8 + 10n ) z n . (2.48)
n=0
2

Comparing (2.46) and (2.48), the solution is


1 n
an = (8 + 10n ) .
2

7. Solve S(n) − 2S(n − 1) − 3S(n − 2) = 0, n ≥ 2 with S(0) = 3 and


S(1) = 1 using generating function.
Solution.
Let the generating function be

X
G(z) = S(n)z n . (2.49)
n=0
Combinatorics 115

Multiplying the given equation by z n and summing for n ≥ 2, we


get

X ∞
X ∞
X
S(n)z n − 2 S(n − 1)z n − 3 S(n − 2)z n = 0. (2.50)
n=2 n=2 n=2

Consider the first sum



X
S(n)z n = S(2)z 2 + S(3)z 3 + S(4)z 4 + . . .
n=2
= G(z) − S(0) − S(1)z [using (2.49)] ,

the second sum



X
S(n − 1)z n = S(1)z 2 + S(2)z 3 + S(3)z 4 + . . .
n=2
= z[S(1)z + S(2)z 2 + S(3)z 3 + . . . ]
= z[G(z) − S(0)] [using (2.49)] ,

and the third sum



X
S(n − 2)z n S(0)z 2 + S(1)z 3 + S(2)z 4 + . . .
n=2
= z 2 [S(0) + S(1)z + S(2)z 2 + . . . ]
= z 2 G(z).

∴ (2.50) becomes

[G(z) − S(0) − S(1)z] − 2z[G(z) − S(0)] − 3z 2 G(z) = 0.

Using S(0) = 3 and S(1) = 1, we get

[G(z) − 3 − z] − 2z[G(z) − 3] − 3z 2 G(z) = 0


=⇒ G(z)[1 − 2z − 3z 2 ] = 3 − 5z
3 − 5z 3 − 5z
=⇒ G(z) = 2
= .
1 − 2z − 3z (1 − 3z)(1 + z)

Using partial fractions,


3 − 5z A B
= +
(1 − 3z)(1 + z) 1 − 3z 1+z
=⇒ 3 − 5z = A(1 + z) + B(1 − 3z).

Put z = −1 =⇒ 8 = 4B =⇒ B = 2.
116 Discrete Mathematical Structures
1 4 4
Put z = =⇒ = A =⇒ A = 1.
3 3 3
1 2
∴ G(z) = +
1 − 3z 1+z
= (1 − 3z)−1 + 2(1 + z)−1
X∞ ∞
X
= (3z)n + 2 (−z)n
n=0 n=0
X∞
= [3n + 2(−1)n ] z n . (2.51)
n=0

Comparing (2.49) and (2.51), the solution is

S(n) = 3n + 2(−1)n .

2.8.4 Problems for Practice


1. Find the generating function of the following sequences.
(i) 2, 2, 2, 2, 2, 2, . . .
(ii) an = (−2)n
(iii) 0, 0, 1, 1, 1, 1, . . .
(iv) 1, 0, −1, 0, 1, 0, 0, −1, 0, 1, 0, −1, 0, . . .
2. For the following expressions, identify the sequences having the
expression as a generating function. [Hint: Use partial fractions.]
5 + 2z
(i)
1 − 4z 2
6 − 29z
(ii)
1 − 11z + 30z 2
32 − 22z
(iii)
2 − 3z + z 2
3 + 7z
(iv)
1 + 3z − 4z 2
3 + 5z
(v) .
1 − 2z − 3z 2
3. Find the generating functions for the following sequences satisfying
the given initial conditions.
(i) yn+2 + 2yn+1 − 15yn = 0; y0 =, y1 = 1.
(ii) yn+1 − yn−1 = 0; y0 = 0, y1 = 1.
(iii) yn+2 − 2yn+1 − 3yn = 0; y0 = 0, y1 = 1.
(iv) S(n) − 2S(n − 1) − 3S(n − 2) = 0; S(0) = 3, S(1) = 1.
(v) S(k) + 3S(k − 1) − S(k − 2) = 0; S(0) = 3, S(1) = 2.
Combinatorics 117

4. Using generating function, solve the following recurrence relations.


(i) yn = 7yn−1 ; y0 = 5.
(ii) yn = 3yn−1 + 4n−1 ; y0 = 1.
(iii) yn + 2yn−1 − 15yn−2 = 0; y0 = 0, y1 = 1.
(iv) yn+2 − 8yn+1 + 16yn = 0; y0 = 0, y1 = 8.
(v) yn+2 − 2yn+1 + yn = 0; y0 = 2, y1 = 1.
(vi) yn+2 − yn+1 − 6yn = 0; y0 = 2, y1 = 1.
(vii) yn+2 − 5yn+1 + 6yn = 0; y0 = 1, y1 = 3.
5. Use a generating function to find an explicit formula for the
Fibonacci numbers.

2.9 Inclusion–Exclusion Principle


Let A and B be any finite sets. Then,

n(A ∪ B) = n(A) + n(B) − n(A ∩ B).

In other words, to find the number n(A ∪ B) of elements in the union A ∪ B,


we add n(A) and n(B), and then we subtract n(A ∩ B); that is, we “include”
n(A) and n(B), and we “exclude” n(A ∩ B). This follows from the fact that,
when we add n(A) and n(B), we have counted the elements of A ∩ B twice.
This principle holds for any number of sets. We first state it for three sets.
Theorem 2.9.1 For any finite sets A, B, and C, we have

n(A ∪ B ∪ C)
= n(A) + n(B) + n(C) − n(A ∩ B)
− n(B ∩ C) − n(A ∩ C) + n(A ∩ B ∩ C).

That is, we “include” n(A), n(B), n(C), we “exclude” n(A ∩ B), n(B ∩ C),
n(A ∩ C), and we “include” n(A ∩ B ∩ C).
Example:
Find the number of mathematics students at a college taking at least one of
the languages French, German, and Russian given the following data:

65 study French, 20 study French and German,


45 study German, 25 study French and Russian,
42 study Russian, 15 study German and Russian,
and 8 study all three languages.
118 Discrete Mathematical Structures

We want to find n(F ∪ G ∪ R), where F , G, and R denote the sets of students
studying French, German, and Russian, respectively.
By the inclusion–exclusion principle,

n(F ∪ G ∪ R) = n(F ) + n(G) + n(R) − n(F ∩ G)


− n(F ∩ R) − n(G ∩ R) + n(F ∩ G ∩ R)
= 65 + 45 + 42 − 20 − 25 − 15 + 8
= 100.

∴ 100 students study at least one of the languages.

Note:

Principle of inclusion–exclusion can also be denoted as


(i) |A ∪ B| = |A| + |B| − |A ∩ B| if A and B are not disjoint sets
(ii) |A ∪ B| = |A| + |B| if A and B are disjoint sets
where |A| = n(A) = cardinality of A = number of elements in A.

2.9.1 Solved Problems


1. Find the number of positive integers not exceeding 100 that are
divisible by 7 or by 11.
Solution.
Let A be the set of positive integers not exceeding 100 that are
divisible by 7.
Let B be the set of positive integers not exceeding 100 that are
divisible by 11.
Then, A∪B is the set of positive integers not exceeding 100 that are
divisible by either 7 or 11, and A ∩ B is the set of positive integers
not exceeding 100 that are divisible by both 7 and 11.
We knowthat among the positive integers
 not exceeding 100, there
100 100
are integers divisible by 7 and integers divisible by 11.
7 11
Since 7 and 11 are relatively prime, the integers divisible by both 7
and 11 are those divisible by 7 and 11.
 
100
There are positive integers not exceeding 100 that are
7 × 11
Combinatorics 119

divisible by both 7 and 11.

∴ |A ∩ B| = |A| + |B| − |A ∩ B|
     
100 100 100
= + −
7 11 7 × 11
= 14 + 9 − 1 = 22.

2. Among the first 1000 positive integers, determine the integers which
are not divisible by 5, nor by 7, nor by 9.
Solution.
Let A = set of integers divisible by 5
B = set of integers divisible by 7
C = set of integers divisible by 9.
     
1000 1000 1000
∴ |A| = = 200; |B| = = 142; C = = 111.
5 7 9
   
1000 1000
|A ∩ B| = = = 28
LCM (5, 7) 35
   
1000 1000
|B ∩ C| = = = 15
LCM (7, 9) 63
   
1000 1000
|A ∩ C| = = = 22
LCM (5, 9) 45
     
1000 1000 1000
|A ∩ B ∩ C| = = = = 3.
LCM (5, 7, 9) 5×7×9 315

The number of integers divisible by 5, 7, and 9 is

|A ∪ B ∪ C| = |A| + |B| + |C| − |A ∩ B| − |B ∩ C| − |A ∩ C| + |A ∩ B ∩ C|


= 200 + 142 + 111 − 28 − 15 − 22 + 3
= 391.
The number of integers not divisible by 5 nor 7 nor 9
= Total number of integers - integers divisible by 5, 7, and 9
= 1000 − 391
= 609.
3. In a survey of 300 students, 64 had taken a Mathematics course,
94 had taken an English course, 58 had taken a Computer course,
28 had taken both Mathematics and Computer courses, 26 had
taken both English and Mathematics courses, 22 had taken both
120 Discrete Mathematical Structures

English and Computer courses, and 14 had taken all three courses.
How many students were surveyed who had taken none of the three
courses?
Solution.
Given: |M | = 64; |E| = 94; |C| = 58;
|M ∩ C| = 28; |M ∩ E| = 26; |E ∩ C| = 22; |M ∩ C ∩ E| = 14.

|M ∪ E ∪ E| = |M | + |E| + |C| − |M ∩ E|
− |M ∩ C| − |E ∩ C| + |M ∩ E ∩ C|
= 64 + 94 + 58 − 26 − 28 − 22 + 14
= 154.

∴ Students who had taken none of the courses = 300−154 =146.


4. How many solutions does x1 + x2 + x3 = 13 have, where x1 , x2 , and
x3 are non-negative integers with x1 < 6, x2 < 6, and x3 < 6?
Solution.
To apply the principle of inclusion–exclusion, let a solution have
property P1 if x ≥ 6, property P2 if x2 ≥ 6, and property P3 if
x3 ≥ 6. The number of solutions satisfying the inequalities x1 < 6,
x2 < 6, and x3 < 6 is

N (P10 P20 P30 ) = N − N (P1 ) − N (P2 ) − N (P3 ) + N (P1 P2 )


+ N (P1 P3 ) + N (P2 P3 ) − N (P1 P2 P3 ) (2.52)
where N = total number of solutions
= C(3 + 13 − 1, 13) = C(15, 13) = 105.
N (P1 ) = number of solutions with x1 ≥ 6
= C(3 + 7 − 1, 7) = C(9, 7) = 36
N (P2 ) = number of solutions with x2 ≥ 6
= C(3 + 7 − 1, 7) = C(9, 7) = 36
N (P3 ) = number of solutions with x3 ≥ 6
= C(3 + 7 − 1, 7) = C(9, 7) = 36
N (P1 P2 ) = number of solutions with x1 ≥ 6 and x2 ≥ 6
= C(3 + 1 − 1, 1) = C(3, 1) = 3
N (P1 P3 ) = number of solutions with x1 ≥ 6 and x3 ≥ 6
= C(3 + 1 − 1, 1) = C(3, 1) = 3
N (P2 P3 ) = number of solutions with x2 ≥ 6 and x3 ≥ 6
= C(3 + 1 − 1, 1) = C(3, 1) = 3
N (P1 P2 P3 ) = number of solutions with x ≥ 6, x2 ≥ 6, and x3 ≥ 6 = 0.
Combinatorics 121

Inserting these quantities into the formula N (P10 P20 N30 ) shows that
the number of solutions with x1 ≤ 6, x2 ≤ 6, and x3 ≤ 6 equals
[implying from (2.52)]
N (P10 P20 N30 ) = 105 − 36 − 36 − 36 + 3 + 3 + 3 − 0 = 6.

5. A survey of 500 students from a school produced the following


information. 200 play volleyball, 120 play hockey, 60 play both
volleyball and hockey. How many are not playing either volleyball
or hockey?
Solution.
Let A be the set of students who play volleyball.
Let B be the set of students who play hockey.
Given: |A| = 200, |B| = 120, |A ∩ B| = 60.
By the principle of inclusion–exclusion, the number of students
playing either volleyball or hockey is
|A ∪ B| = |A| + |B| − |A ∩ B|
= 200 + 120 − 60
= 260.
∴ The number of students not playing either volleyball or hockey
= 500 − 260 = 240.
6. A total of 1232 students have taken a course in Russian, 879 have
taken a course in German, and 114 have taken a course in French.
Further 103 have taken a course in both Russian and German, 23
have taken a course in Russian and French, and 14 have taken a
course in German and French. If 2092 students have taken at least
one of the courses Russian, German, and French, how many students
have taken a course in all three languages.
Solution.
Let A be the set of students who have taken a course in Russian.
Let B be the set of students who have taken a course in German.
Let C be the set of students who have taken a course in French.
Given: |A| = 1232, |B| = 879, |C| = 114,
|A ∩ B| = 103, |A ∩ C| = 23, |B ∩ C|14, |A ∩ B ∩ C| = 2092.
By the principle of inclusion–exclusion, we have
|A ∪ B ∪ C| = |A| + |B| + |C| − |A ∩ B| − |B ∩ C|
− |A ∩ C| + |A ∩ B ∩ C|
=⇒ 2092 = 1232 + 879 + 114 − 103 − 23 − 14 + |A ∩ B ∩ C|
=⇒ |A ∩ B ∩ C| = 7.
122 Discrete Mathematical Structures

∴ There are seven students who have taken a course in Russian,


German, and French.
7. In a survey of 100 students, it was found that 30 studied
Mathematics, 54 studied Statistics, 25 studied Operations Research,
1 studied all the three subjects, 20 studied Mathematics and
Statistics, 3 studied Mathematics and Operations Research, and
15 studied Statistics and Operations Research.
(i) How many students studied none of these subjects?
(ii) How many students studied only Mathematics?
Solution.
Let A denote the set of students who studied Mathematics.
Let B denote the set of students who studied Statistics.
Let C denote the set of students who studied Operations Research.
Given: |A| = 30, |B| = 54, C = 25,
|A ∩ B| = 20, |A ∩ C| = 3, |B ∩ C| = 15, |A ∩ B ∩ C| = 1.
(i) By the principle of inclusion–exclusion, the number of students
who studied any one of the subjects is

|A ∪ B ∪ C| = |A| + |B| + |C| − |A ∩ B| − |B ∩ C|


− |A ∩ C| + |A ∩ B ∩ C|
= 30 + 54 + 25 − 2003 − 15 + 1 = 72.

∴ Number of students who studied none of these subjects


= 100 − 72 = 28.
(ii) Number of students who studied only Mathematics and
Statistics
= |A ∩ B| − |A ∩ B ∩ C| = 20 − 1 = 19.
Number of students who studied only Mathematics and
Operations Research
= |A ∩ C| − |A ∩ B ∩ C| = 3 − 1 = 2.
∴ Number of students who studied only Mathematics
= 30 − 19 − 2 − 1 = 8.
8. How many positive integers not exceeding 1000 are divisible by 7 or
11?
Solution.
Let A denote the set of positive integers not exceeding 1000 that
are divisible by 7.
Let B denote the set of positive integers not exceeding 1000 that
are divisible by 11. Then,
Combinatorics 123
     
1000 1000 1000
|A| = = 142, |B| = = 90, |A ∩ B| = = 12.
7 11 7 × 11
The number of positive integers not exceeding 1000 that are
divisible by either 7 or 11 is |A ∪ B|.
By the principle of inclusion–exclusion,

|A ∪ B| = |A| + |B| − |A ∩ B|
= 142 + 90 − 12
= 220.
9. Determine n such that 1 ≤ n ≤ 100 and it is not divisible by 5 or 7.
Solution.
Let A denote the number n, 1 ≤ n ≤ 100, which is divisible by 5.
Let B denote the number n, 1 ≤ n ≤ 100, which is divisible by 7.
Then,
     
100 100 100
|A| = = 20, |B| = = 14, |A ∩ B| = = 2.
5 7 5×5
By the principle of inclusion–exclusion, the number n, 1 ≤ n ≤ 100,
which is divisible by either 5 or 7 is |A ∪ B|.

|A ∪ B| = |A| + |B| − |A ∩ B|
= 20 + 14 − 2
= 32.

∴ The number n, 1 ≤ n ≤ 100, which is not divisible by 5 and 7 is


= 100 − 32 = 68.
10. A survey among 100 students shows that of the three ice cream
flavours vanilla, chocolate, and strawberry, 50 students like vanilla,
43 like chocolate, 28 like strawberry, 13 like vanilla and chocolate,
11 like chocolate and strawberry, 12 like strawberry and vanilla, and
5 like all of them. Find the number of students surveyed who like
the following flavours:
(i) chocolate but not strawberry
(ii) chocolate and strawberry but not vanilla
(iii) vanilla or chocolate but not strawberry.
Solution.
Let A denote the set of students who like vanilla.
Let B denote the set of students who like chocolate.
Let C denote the set of students who like strawberry.
Since five students like all flavours, |A ∩ B ∩ C| = 5.
124 Discrete Mathematical Structures

Twelve students like both strawberry and vanilla =⇒ |A ∩ C| = 12.


But five of them like chocolate also =⇒ |A ∩ C| − 5 = 12 − 5 = 7.
Six of them like vanilla =⇒ |B ∩ C| − 6 = 12 − 6 = 6.
Out of 28 students who like strawberry,
we have already accounted for 7 + 5 + 6 = 18.
∴ The remaining ten students belong to the set C − (A ∪ B).
Similarly,

|A − (B ∪ C)| = 30 and |B − (A ∪ C)| = 24.

Hence, we have accounted for 90 of the 100 students. The remaining


ten students like outside the region A ∪ B ∪ C. Now,
(i) |B − C| = 24 + 8 = 32.
∴ 32 students like chocolate but not strawberry.
(ii) |(B ∩ C) − A| = 6.
∴ Six students like both chocolate and strawberry but not
vanilla.
(iii) |(A ∪ B) − C| = 30 + 8 + 24 = 62.
∴ 62 students like vanilla or chocolate, but not strawberry.
11. Find the number of integers between 1 and 250 that are not divisible
by any of the integers 2, 3, 5, and 7.
Solution.
Let A denote the set of integers between 1 and 250 that are divisible
by 2.
Let B denote the set of integers between 1 and 250 that are divisible
by 3.
Let C denote the set of integers between 1 and 250 that are divisible
by 5.
Let D denote the set of integers between 1 and 250 that are divisible
by 7.
Now,    
250 250
|A| = = 125, |B| = = 83
2 3
   
250 250
|C| = = 5, |D| = = 35.
5 7

Number of integers between 1 and 250 that are divisible by 2 and 3


 
250
= |A ∩ B| = = 41
2×3
Combinatorics 125

Number of integers between 1 and 250 that are divisible by 2 and 5


 
250
= |A ∩ C| = = 25.
2×5
Similarly,
 
250
|A ∩ D| = = 17
2×7
 
250
|B ∩ C| = = 16
3×5
 
250
|B ∩ D| = = 11
3×7
 
250
|C ∩ D| = = 7.
5×7

Number of integers between 1 and 250 that are divisible by 2, 3, and 5


 
250
= |A ∩ B ∩ C| = = 8.
2 × 3 × 50
Similarly,
 
250
|A ∩ B ∩ D| = =5
2×3×7
 
250
|A ∩ C ∩ D| = =3
2×5×7
 
250
|B ∩ C ∩ D| = =2
3×5×7
 
250
|A ∩ B ∩ C ∩ D| = = 1.
2×3×5×7
The number of integers between 1 and 250 that are divisible by 2,
3, 5, and 7 is |A ∪ B ∪ C ∪ D|.
By principle of inclusion–exclusion,

|A ∪ B ∪ C ∪ D| = |A| + |B| + |C| + |D| − |A ∩ B| − |B ∩ C|


− |C ∩ D| − |A ∩ C| − |B ∩ D| − |A ∩ D|
+ |A ∩ B ∩ C| + |A ∩ B ∩ D| + |A ∩ C ∩ D|
+ |B ∩ C ∩ D| − |A ∩ B ∩ C ∩ D|
= 125 + 83 + 50 + 35 − 41 − 25 − 17 − 16 − 11 − 7
+8+5+3+2−1
= 193.
126 Discrete Mathematical Structures

∴ Number of integers between 
1 and 250 that are not = 250 − 193 = 57.

divisible by 2, 3, 5, and 7

12. A survey shows that 57% of Indians like coffee whereas 75% like
tea. What can you say about the percentage of Indians who like
both coffee and tea.
Solution.
Let A denote the set of Indians who like coffee.
Let B denote the set of Indians who like tea.
Assume the total population is 100.
∴ |A| = 57; |B| = 75. Now,

|A ∪ B| = |A| + |B| − |A ∩ B|
= 57 + 75 − |A ∩ B|
= 132 − |A ∩ B|.

Since |A ∪ B| ≤ 100, it follows that

|A ∩ B| ≥ 32. (2.53)

Since A ∩ B ⊆ A and A ∩ B ⊆ B, we have

|A ∩ B| ≤ |A| and |A ∩ B| ≤ |B|.

∴ |A ∩ B| ≤ 57 and |A ∩ B| ≤ 75.

=⇒ |A ∩ B| ≤ 57. (2.54)

From (2.53) and (2.54), the percentage of Indians who like both
coffee and tea lies between 32 and 57.
13. Out of 100 students in a college, 38 play tennis, 57 play cricket, 31
play hockey, 9 play cricket and hockey, 10 play hockey and tennis,
and 12 play tennis and cricket. How many play
(i) all three games
(ii) just one game
(iii) tennis and cricket but not hockey.
Assume that each student plays at least one game.
Solution.
Let T, C, and H denote the set of students playing tennis, cricket,
and hockey, respectively (Figure 2.1).
Given: |T | = 38, |C| = 57, |H| = 31
Combinatorics 127
T C

21
7 41

5
5 4

17

FIGURE 2.1
Venn diagram

|T ∩ C| = 12, |T ∩ H| = 10, |C ∩ H| = 9, |T ∪ C ∪ H| = 100.


Number of students who play all three games = |T ∩ C ∩ H|.
By principle of inclusion–exclusion, we have

|T ∩ C ∩ H| = |T | + |C| + |H| − |T ∩ C| − |C ∩ H|
− |T ∩ H| + |T ∩ C ∩ H|
=⇒ 100 = 38 + 57 + 31 − 12 − 9 − 10 + |T ∩ C ∩ H|
=⇒ |T ∩ C ∩ H| = 100 − 126 + 31 = 5.

∴ Number of students who play all three games = 5.


From the given data, we have

Number of students
playing just one game = Number of students playing tennis only
+ Number of students playing cricket only
+ Number of students playing hockey only
= 21 + 41 + 17 = 79.

Number of students playing tennis and cricket but not hockey


= |T ∩ C| − |T ∩ C ∩ H| = 12 − 5 = 7.
14. How many integers between 1 and 100 are
(i) not divisible by 7, 11, or 13
(ii) divisible by 3 but not by 7?
128 Discrete Mathematical Structures

Solution.
Let A, B, and C be the set of integers between 1 and 100 that are
divisible by 7, 11, and 13, respectively.
     
100 100 100
∴ |A| = = 14; |B| = ; |C| = = 7;
7 11 13
   
100 100
|A ∩ B| = = 1; |A ∩ C| = = 1;
7 × 11 7 × 13
   
100 100
|B ∩ C| = = 0; = 0.
11 × 13 7 × 11 × 13
Number of integers between 1 and 100 that are divisible by 7, 11,
or 13 is |A ∪ B ∪ C|.
By principle of inclusion–exclusion, we have

|A ∪ B ∪ C| = |A| + |B| + |C| − |A ∩ B| − |B ∩ C|


− |A ∩ C| + |A ∩ B ∩ C|
= 14 + 9 + 7 − 1 − 0 − 1 + 0
= 28.

(i) Number of integers between 1 and 100 that are not divisible by
7, 11, or 13 = 100 − 28 = 72.
(ii) Let U and V denote the set of integers between 1 and 100 that
are divisible by 3 and 7, respectively.
     
100 100 100
|U | = = 33; |V | = = 14; |U ∩ V | = = 4.
3 7 3×7

∴ Number of integers divisible by 3 but not by 7 = |U | − |U ∩ V |


= 33 − 4
= 29.

15. Find the number of integers between 1 and 100 that are divisible by
(i) 2, 3, 5, or 7
(ii) 2, 3, 5 but not by 7.
Solution.
Let A, B, C, and D denote the set of positive integers between 1
and 100 that are divisible by 2, 3, 5, and 7, respectively.
   
100 100
∴ |A| = = 50; |B| = = 33;
2 3
   
100 100
|C| = = 20; |D| = = 14;
5 7
   
100 100
|A ∩ B| = = 16; |A ∩ C| = = 10;
2×3 2×5
Combinatorics 129
   
100 100
|A ∩ D| = = 7; |B ∩ C| = = 7;
2×7 3×5
   
100 100
|B ∩ D| = = 4; |C ∩ D| = = 2;
3×7 5×7
   
100 100
|A ∩ B ∩ C| = = 3; |A ∩ B ∩ D| = = 2;
2×3×5 2×3×7
   
100 100
|A ∩ C ∩ D| = = 1; |B ∩ C ∩ D| = = 0;
2×5×7 3×5×7
 
100
|A ∩ B ∩ C ∩ D| = = 0.
2×3×5×7
(i) By the principle of inclusion–exclusion, we have

|A ∪ B ∪ C ∪ D|
= |A| + |B| + |C| + |D| − |A ∩ B| − |A ∩ C| − |A ∩ D|
− |B ∩ C| − |B ∩ D| + |A ∩ B ∩ C| + |A ∩ B ∩ D|
+ |A ∩ C ∩ D| + |B ∩ C ∩ D| − |A ∩ B ∩ C ∩ D|
= 50 + 33 + 20 + 14 − 16 − 10 − 7 − 7 − 4 − 2
+3+2+1+0−0
= 117 − 46 + 6
= 123 − 46
= 77.

(ii) The number of integers between 1 and 100 that are divisible by
2, 3, 5 but not by 7
= |A ∩ B ∩ C| − |A ∩ B ∩ C ∩ D| = 3 − 0 = 3.
16. How many prime numbers not exceeding 100 are there? Or
determine a prime number n, where 1 ≤ n ≤ 100.
Solution.
To find the number of primes not exceeding 100, first note that a
composite integer not exceeding 100 must have a prime factor not
exceeding 10.
The primes not exceeding 100 are 2, 3, 5, and 7 and the numbers
that are divisible by none of 2, 3, 5, or 7.
Let P1 be the property that an integer is divisible by 2.
Let P2 be the property that an integer is divisible by 3.
Let P3 be the property that an integer is divisible by 5.
Let P4 be the property that an integer is divisible by 7.
Number of primes not exceeding = 4 + N (P10 P20 P30 P40 ).
130 Discrete Mathematical Structures

Now,
N (P10 P20 P30 P40 ) = 99 − N (P1 ) − N (P2 ) − N (P3 ) − N (P4 )
+ N (P1 P2 ) + N (P1 P3 ) + N (P1 P4 )
+ N (P2 P3 ) + N (P2 P4 ) + N (P3 P4 )
− N (P1 P2 P3 ) − N (P1 P2 P4 ) − N (P1 P3 P4 )
− N (P2 P3 P4 ) − N (P1 P2 P3 P4 )
(∵ there are 99 integers > 1 and not exceeding 100).
       
100 100 100 100
N (P10 P20 P30 P40 ) = 99 − − − −
2 3 5 7
     
100 100 100
+ + +
2×3 2×5 2×7
     
100 100 100
+ + +
3×5 3×7 5×7
     
100 100 100
− − −
2×3×5 2×3×7 2×5×7
   
100 100
− +
3×5×7 2×3×5×7
= 99 − 50 − 33 − 20 − 14 + 16 + 10 + 7 + 6
+4+2−3−2−1−0+0
= 21.
∴ There are 4 + 21 = 25 primes.
17. How many solutions does x1 + x2 + x3 = 11 have, where x1 , x2 , and
x3 are non-negative integers with x1 ≤ 3, x2 ≤ 4, and x3 ≤ 6?
Solution.
Let P1 be the property that x1 > 3.
Let P2 be the property that x2 > 4.
Let P3 be the property that x3 > 6.
Now, the number of solutions satisfying the inequalities x1 ≤ 1,
x2 ≤ 4, and x3 ≤ 6 is N (P10 P20 P30 ).
By principle of inclusion–exclusion, we have
N (P10 P20 P30 )N − N (P1 ) − N (P2 ) − N (P3 )
+ N (P1 P2 ) + N (P1 P3 ) + N (P2 P3 ) − N (P1 P2 P3 ).
Now,
N = Total number of solutions
= C(3 + 11 − 1, 11) = 78
Combinatorics 131

[since the number of r-combinations from a set with n elements


when repetitions are allowed is (n+r −1)Cr ways or C(n+r −1, r)].

N (P1 ) = Number of solutions with x1 ≥ 4


= C(3 + 7 − 1, 7) = C(9, 7) = 36.
N (P2 ) = Number of solutions with x2 ≥ 5
= C(3 + 6 − 1, 6) = C(8, 6) = 28.
N (P3 ) = Number of solutions with x3 ≥ 7
= C(3 + 4 − 1, 4) = C(6, 4) = 15.
N (P1 P2 ) = Number of solutions with x1 ≥ 4 and x2 ≥ 5
= C(3 + 2 − 1, 2) = C(4, 2) = 6.
N (P1 P3 ) = Number of solutions with x1 ≥ 4 and x3 ≥ 7
= C(3 + 0 − 1, 0) = C(2, 0) = 1.
N (P2 P3 ) = Number of solutions with x2 ≥ 5 and x3 ≥ 7 = 0.
N (P1 P2 P3 ) = Number of solutions with x1 ≥ 4, x2 ≥ 5 and x3 ≥ 7 = 0.

∴ N (P10 P20 P30 ) = 78 − 36 − 28 − 15 + 6 + 1 + 0 + 0 = 6.


Hence, the equation x1 + x2 + x3 = 11 with respect to the given
conditions has six solutions.

2.9.2 Problems for Practice


1. Find the number of elements in A1 ∪ A2 ∪ A3 if there are 100
elements in each set if
(i) the sets are pairwise disjoint
(ii) there are 50 common elements in each pair of sets and no
element in all three sets
(iii) the sets are equal.
2. How many elements are in the union of four sets if the sets have
50, 60, 70, and 80 elements, respectively, each pair of sets has five
elements in common, each trio of the sets has one common element,
and no element is in all four sets?
3. How many bit strings of length 8 do not contain six consecutive 0’s?
4. How many solutions does the equation x1 +x2 +x3 = 16 have where
x1 , x2 , and x3 are non-negative integers less than 6?
5. How many ways are there to distribute six different toys to three
different children such that each child gets at least one toy?
6. How many ways can the digits 0, 1, 2, 3, 4, 5, 6, 7, 8, 9 be arranged
so that no even digit is in its original position?
132 Discrete Mathematical Structures

7. By using principle of inclusion–exclusion, find how many positive


integers between 1 and 250 are
(i) divisible by 2, 3, 5, or 7
(ii) not divisible by 2, 3, 5, and 7.
8. How many positive integers not exceeding 1000 are divisible by 7
or 11?
9. How many elements are in A1 ∪ A2 if there are 12 elements in A1 ,
18 elements in A2 , and
(i) |A1 ∩ A2 | = 1
(ii) |A1 ∩ A2 | = 6?
10. There are 2504 Computer Science students at a college. Of these,
1876 have taken a course in Pascal, 999 have taken a course in
Fortran, and 345 have taken course in C. Further, 876 have taken
courses in both Pascal and Fortran, 231 have taken courses in both
Fortran and C, and 290 have taken courses in both Pascal and C.
If 189 of these students have taken courses in Fortran, Pascal, and
C, how many of these 2504 students have not taken a course in any
of these 3 programming languages?
11. How many permutations of ten digits either begin with the three
digits 987, contain the digits 45 in the fifth and sixth positions, or
end with the three digits 123?
12. Find the probability that when four numbers from 1 to 100,
inclusive, are picked at random with no repetitions allowed, either
all are odd, all are divisible by 3, or all are divisible by 5.
13. Find the number of primes less than 200 using the principle of
inclusion–exclusion.
14. How many derangements are there of a set with seven elements?
15. How many positive integers less than 10000 are not the second or
higher powers of an integer?
16. How many integers between 1 and 300 are divisible by
(i) at least one of 3, 5, 7
(ii) 3 and 5 but not by 7
(iii) 5 but not by 3 and 7.
17. A survey of households in the United States reveals that 96% have at
least one television set, 98% have telephone service, and 95% have
telephone service and at least one television set. What percentage
of households in the United States have neither telephone service
nor a television set?
Combinatorics 133

18. How many students are enrolled in a course either in calculus,


discrete mathematics, data structures, or programming languages
at a school if there are 507, 292, 312, and 344 students in these
courses, respectively; 14 in both calculus and data structures; 213
in both calculus and programming languages; 211 in both discrete
mathematics and data structures; 43 in both discrete mathematics
and programming languages; and no student may take calculus
and discrete mathematics, or data structures and programming
languages, concurrently?
19. Find the number of positive integers not exceeding 100 that are
either odd or the square of an integer.
20. Suppose in a bushel of 100 apples, there are 20 that have worms
in them and 15 that have bruises. Only those apples with neither
worms or bruises can be sold. If there are ten bruised apples that
have worms in them, how many of the 100 apples can be sold?
21. In how many ways can seven different jobs be assigned to four
different employees so that each employee is assigned at least one
job and the most difficult job is assigned to the best employee?
3
Graphs

3.1 Introduction
Graphs are mathematical discrete structures which have major role
in computer science (algorithms and computation), electrical engineering
(communication networks and coding theory), operations research (scheduling),
and in many fields of engineering and also in sciences such as chemistry,
biochemistry (genomics), biology, linguistics, sociology, and other fields. For
instance, graphs are encoded to represent the relationship between objects.
Many real-world situations can conveniently be described by means of a
diagram consisting of a set of points together with lines joining certain pairs
of these points. For example, the points could represent people, with lines
joining pairs of friends; or the points might be communication centres, with
lines representing communication links. Notice that in such diagrams, one is
mainly interested in whether or not two given points are joined by a line; the
manner in which they are joined is immaterial. A mathematical abstraction
of situations of this type gives rise to the concept of a graph.
In this chapter, we focus on the terminology of graphs, its various
types, connectivity of graphs, Eulerian path, and Hamiltonian path. Graph
theory is introduced as an abstract mathematical system. The most common
representation of a graph is by means of a diagram, in which the vertices are
represented as points and each edge as a line segment joining its end vertices.

3.2 Graphs and Graph Models


Definition 3.2.1 Graph: A graph G = (V, E, φ) consists of a non-empty set
V = {v1 , v2 , . . . } called the set of vertices of the graph and E = {e1 , e2 , . . . }
called the set of edges of the graph, and φ is a mapping from the set of edges
E to the set of ordered or unordered pair of elements of vertices.

135
136 Discrete Mathematical Structures
v4 e3 v3

e4
e2

v1 e1 v2
Example of a graph

Definition 3.2.2 Self-loop: An edge having same vertex as both its end
vertices is called a self-loop.
Example 3.2.3 Here e is a self-loop.
v3

v1 v2

Example of a self-loop

Definition 3.2.4 Parallel Edges: If more than one edge has the same pair
of end vertices, then the edges are called parallel edges.

v3

e1
e2

e3
v1 v2

e4

Example of parallel edges

Example 3.2.5 In the graph in


(i) e1 , e2 are incident with v1 .
(ii) e2 , e3 are incident with v5 .
Here, e3 and e4 are parallel edges.
Graphs 137

Definition 3.2.6 If vi is an end vertex of some edge ej , then ej is said to be


incident at vi .

v1 e1 v2

v3
e2
e4

v5 e3 v4

Example of incident edges on vertices


Example 3.2.7

The edge e1 is incident at the vertex v1 . The edge e3 is incident at the vertex v5 .

Definition 3.2.8 Adjacent Edges and Vertices: Two non-parallel edges


are said to be adjacent if they are incident on a common vertex.
Two vertices are said to be adjacent if they are the end vertices of the same
edge.

Example 3.2.9

Here, e1 is adjacent to e2 and e3 , and v1 is adjacent to v2 and v3 .

v3

e1
e2

e3
v1 v2

Adjacent vertices and edges

Definition 3.2.10 Simple Graph: A graph which has neither self-loops nor
parallel edges is called a simple graph as shown below.

Examples of simple graphs

Definition 3.2.12 Isolated Vertex: A vertex having no edge incident on it


is called an isolated vertex as shown below.
138 Discrete Mathematical Structures
v2 v3

v1 v4
Example of isolated vertex

Definition 3.2.14 Directed Graph: A graph in which every edge is directed


is called a directed graph or digraph as shown below.

Example of directed graph

Definition 3.2.16 Undirected Graph: A graph in which every edge is


undirected is called an undirected graph.

Example of undirected graph

Definition 3.2.18 Mixed Graph: A graph in which some edges are directed
and some are undirected is called a mixed graph.
Definition 3.2.19 Multigraph: A graph which contains some parallel edges
is called a multigraph.
Definition 3.2.20 Pseudo Graph: A graph in which loops and parallel
edges are allowed is called a pseudograph.

3.3 Graph Terminology and Special Types of Graphs


Definition 3.3.1 Degree of a vertex: The number of edges incident at the
vertex vi in an undirected graph is called the degree of the vertex vi . But a
loop at a vertex contributes twice to the degree of that vertex. The degree of
the vertex vi is denoted by deg(vi ).
Graphs 139

Definition 3.3.2 In-degree and Out-degree of a vertex: In a directed


graph, the in-degree of a vertex vi is denoted by deg − (vi ) and defined by the
number of edges with vi as their terminal vertex.
The out-degree of a vertex vj is denoted by deg + (vj ) and defined as the
number of edges with their initial.

Theorem 3.3.3 HandshakingP Theorem: Let G = (V, E) be an undirected


graph with e edges. Then, v∈V deg(v) = 2e.
Proof.
Since every edge is incident with exactly two vertices, every edge contributes
two to the sum of degrees of the vertex.
Therefore,
P all the edges e contribute 2e to the sum of degrees of the vertex.
∴ v∈V = 2e.

Theorem 3.3.4 The maximum number of edges in a simple graph with n


n(n − 1)
vertices is .
2
Proof.
From handshaking theorem, we have
X
deg(v) = 2e,
v∈V

where e is the number of edges with n vertices in the graph G. That is,

deg(v1 ) + deg(v2 ) + · · · + deg(vn ) = 2e. (3.1)

We know that the maximum degree of each vertex in the graph G can be
(n − 1).

∴ (3.1) =⇒ (n − 1) + (n − 1) + . . . ton terms = 2e


=⇒ n(n − 1) = 2e
n(n − 1)
=⇒ e = .
2
Hence, the maximum number of edges in any simple graph with n vertices is
n(n − 1)
.
2
Theorem 3.3.5 A simple graph with at least two vertices has at least two
vertices of same degree.
Proof.
Let G be a simple graph with n ≥ 2 vertices.
The graph G has no loop and parallel edges.
Hence, the degree of each vertex is ≤ n − 1.
Suppose that all the vertices of G are of different degrees.
Following degrees 0, 1, 2, . . . , n − 1 are possible for n vertices of G.
140 Discrete Mathematical Structures

Let u be the vertex with degree 0. Then, u is an isolated vertex.


Let v be the vertex with degree n − 1. Then, v has n − 1 adjacent vertices.
Since v is not an adjacent vertex of itself, every vertex of G other than u
is an adjacent vertex of G other than u.
Hence, u cannot be an isolated vertex; this contradiction proves that a
simple graph contains two vertices of same degree.
Note: The converse of the above theorem is not true.
Theorem 3.3.6 The number of odd degree vertices is always even.
Proof.
Let G = (V, E) be any graph with n number of vertices and e number of
degrees.
Let v1 , v2 , . . . , vk be the vertices of odd degree and v10 , v20 , . . . , vm
0
be the
vertices of even degree.
To prove, k is even. P
We know that v∈V deg(v) = 2|E| = 2e.
Pk Pm
=⇒ i=1 deg(vi ) + j=1 deg(vj0 ) = 2e.
Pm
Clearly, j=1 deg(vj0 ) and 2e are even numbers.
Pk
That is, i=1 deg(vi ) = an even number.
Since each term deg(vi ) is odd, the number of terms in the left-hand side
sum must be even.
=⇒ k is even.
Hence, the theorem is proved.

3.3.1 Solved Problems


1. Can a simple graph exist with 15 vertices each of degree 5?
Solution.
We know that the number of odd degree vertices is even.
Hence, the number of odd degree vertices to be odd is not possible.
We cannot say that a simple graph exists with 15 vertices each of
degree 5.
2. How many vertices does a regular graph of degree 4 with ten edges
have?
Solution.
We know that
X
deg(v) = 2e
v∈V
n × 4 = 2 × 10
20
n=
4
n = 5.

∴ The number of vertices is five.


Graphs 141

3. Is there a simple graph corresponding to the following degree


sequences?
(i) (1, 1, 2, 3)
(ii) (2, 2, 4, 6)
Solution.
(i) There are odd number (3) of degree vertices 1, 1, and 3.
Hence, there does not exist a graph corresponding to this degree
sequence.
(ii) The number of vertices in the graph is four, and the maximum
degree of a vertex is 6 which is not possible as the maximum
degree cannot be one less than the number of vertices.
4. Show that in a group, there must be two people who know the same
number of other people in the group.
Solution.
Construct the simple graph model in which V is the set of people
in the group, and there is an edge associated with (u, v) if u and
v know each other. Then, the degree of vertex v is the number of
people v knows.
We know that there are two vertices with the same degree.
Therefore, there are two people who know the same number of other
people in the group.
5. Show that the degree of a vertex of a simple graph G of n vertices
cannot exceed n − 1.
Solution.
Let v be a vertex of G because G is simple and no multiple edges or
loops are allowed in G. Thus, v can be adjacent to at most all the
remaining n − 1 vertices of G. Hence, v may be of maximum degree
n − 1 in G.
Then, 0 ≤ deg(v) ≤ n − 1, or all v ∈ V .
6. How many edges are there in a graph with ten vertices each of
degree 6?
Solution.
Sum of the degrees of the ten vertices is

(6) × (10) = 60
=⇒ 2e = 60
=⇒ e = 30.

7. Show that the sum of degrees of all the vertices in a graph G is even.
Solution.
Each edge contributes two degrees in a graph.
142 Discrete Mathematical Structures

Also, each edge contributes one degree to each of the vertices on


which it is incident. Hence, if there are N edges in G, then

2N = deg(v1 ) + deg(v2 ) + · · · + deg(vN ).

Thus, 2N is always even.

Definition 3.3.7 Regular Graph: If every vertex of a simple graph has the
same degree, then the graph is called a regular graph.
If every vertex in a regular graph has degree k, then the graph is called
k-regular graph.

Note:
1. Every null graph is regular of degree 0.
2. The complete graph Kn is of degree n − 1.
3. If a graph G has n vertices and is regular of degree k, then G has
rn
edges.
2
Definition 3.3.8 Complete Graph: A simple graph with n vertices is said
to be a complete graph if the degree of every vertex is n − 1.

(or)

In a graph G, if every vertex v is adjacent to all other vertices, then G is


called a complete graph.
The complete graph with n vertices is denoted by Kn .

K3 K4 K5

Examples of complete graphs

Example 3.3.9

Definition 3.3.10 Subgraph: A graph H = (V 0 , E 0 ) is called a subgraph of


a graph G = (V, E) if V 0 ⊆ V and E 0 ⊆ E.

Definition 3.3.11 Cycle Graph: A cycle graph of order n is a connected


graph whose edges form a cycle of length n and is denoted by Cn .
Graphs 143

Example 3.3.12

C6
Example of cycle graph
Note:
1. In a graph, a cycle that is not a loop must have length at least
three, but there may be cycles of length two in a multigraph.
2. A simple digraph having no cycles is called a cyclic graph.
3. A cyclic graph cannot have any loops.
4. The cycle Cn , n ≥ 3, consists of n vertices 1, 2, . . . , n and edges
{1, 2}, {2, 3}, . . . , {n − 1, n}.
Definition 3.3.13 Wheel Graph: A wheel graph of order n is obtained by
joining a new vertex called “Hub” to each vertex of a cycle graph of order
n − 1, denoted by Wn .

Hub

W5

Example of wheel graph


Example 3.3.14
Note: We obtain the wheel Wn when we add an additional vertex to the cycle
Cn , for n ≥ 3, and connect this new vertex to each of the n vertices in Cn , by
new edges.
Definition 3.3.15 Bipartite Graph: A graph G is said to be bipartite if
its vertex set V can be partitioned into two disjoint non-empty sets V1 and V2
such that V1 ∪ V2 and every edge in E has one end in V1 and the other end
in V2 .
Example 3.3.16 Let V = V1 ∪ V2 where
V1 = {u1 , u3 , u5 , u7 } and V2 = {u2 , u4 , u6 , u8 }.
Then, G is a bipartite graph shown below.
144 Discrete Mathematical Structures

u1 u2

u3 u4

u5 u6

u7 u8

V1 V2
Example of bipartite graph

Definition 3.3.17 Complete Bipartite Graph: A bipartite graph G with


the bipartition V1 and V2 is called complete bipartite if every vertex in V1 is
adjacent to every vertex in V2 .
A complete bipartite graph that may be partitioned into sets A and B as
above such that |A| = a and |B| = b is denoted by Ka,b .

Example 3.3.18 The graph K3,3 is a complete bipartite graph.

V1 V2 V3

V4 V5 V6

K3,3
Example of complete bipartite graph

Definition 3.3.19 Star Graph: Any graph that is K1,n is called a star
graph.

Example 3.3.20 The graph K1,6 is a star graph.

K1,6
Example of star graph
Graphs 145

3.3.2 Graph Colouring


The assignment of colours to the vertices of G, one colour to each vertex,
so that adjacent vertices are assigned different colours is called the proper
colouring of G or simply vertex colouring.
If G has n colouring, then G is said to be n-colourable.
Theorem 3.3.21 A simple graph is bipartite if and only if it is possible to
assign one of two different colours to each vertex of the graph so that no two
adjacent vertices are assigned the same colour.
Proof.
Let G = (V, E) be a bipartite simple graph. Then, V = V1 ∪ V2 , where V1 and
V2 are disjoint sets and every edge in E connects a vertex in V1 and a vertex
in V2 .
If we assign one colour to each vertex in V1 and a second colour to each
vertex in V2 , then no two adjacent vertices are assigned the same colour.
Suppose that it is possible to assign colours to the vertices of the graph
using just two colours.
=⇒ No two adjacent vertices are assigned the same colour.
Let V1 be the set of vertices assigned one colour and V2 be the set of vertices
assigned the other colour. Then, V1 and V2 are disjoint and V = V1 ∪ V2 .
That is, every edge connects a vertex in V1 and a vertex in V2 since no
two adjacent vertices are either both in V1 or both in V2 . Consequently, G is
bipartite.

3.3.3 Solved Problems


1. What is the degree sequence of Kn , where n is positive integer?
Explain your answer.
Solution.
Each of the n vertices is adjacent to each of the other n − 1 vertices,
so the degree sequence is n − 1, n − 1,. . . , n − 1 (n terms).
2. Determine whether each of the following sequences is a graph. For
those that are, draw a graph having the given degree sequence.
(i) 5, 4, 3, 2, 1
(ii) 3, 2, 2, 1, 0
(iii) 1, 1, 1, 1, 1
Solution.
(i) No, since the sum of degrees = 5 + 4 + 3 + 2 + 1 = 15 which is
odd.
(ii) Yes.
(iii) No, since the sum of degrees = 1 + 1 + 1 + 1 + 1 = 5 which is
odd.
146 Discrete Mathematical Structures

Graph of the given sequence

3. How many vertices and edges are there in Kn ?


Solution.
n(n − 1)
Kn has n vertices and edges.
2
4. Find the degree sequence of each of the following graphs.
(i) K4
(ii) K5
(iii) K2
Solution.
(i) 3, 3, 3, 3
(ii) 4, 4, 4, 4, 4
(iii) 1, 1
5. How many vertices and edges do the following graphs have?
(i) Cn
(ii) C8
(iii) Also find the degree sequence of C4 .
Solution.
(i) n vertices and n edges
(ii) Eight vertices and eight edges
(iii) 2, 2, 2, 2.
6. Show that C6 is a bipartite graph?
Solution.
The vertex set of C6 can be partitioned into the two sets
V1 = {v1 , v3 , v5 } and V2 = {v2 , v4 , v6 }, and every edge of C6
connects a vertex in V1 and a vertex in V2 . Hence, C6 is a bipartite
graph.
Graphs 147
v1 v2

v3
v6

v5 v4
Graph of C6

7. Is K3 bipartite?
Solution.
No, the complete graph K3 is not bipartite as shown below.

v1

v3
v2
Graph of K3

If we divide the vertex set of K3 into two disjoint sets, one of the
two sets must contain two vertices. If the graph is bipartite, these
two vertices should not be connected by an edge, but in K3 each
vertex is connected to every other vertex by an edge.
∴ K3 is not bipartite.
8. How many vertices and edges are there in a complete bipartite graph
Km,n ?
Solution.
There are m + n vertices and mn edges.
9. Find the degree sequence of the graph K2,3 .
Solution.
3, 3, 2, 2, 2.
10. For which values of m and n is Km,n regular?
Solution.
A complete bipartite graph Km,n is not regular if m 6= n.
=⇒ If m = n, then Km,n is regular.
11. Prove that a graph which contains a triangle cannot be bipartite.
Solution.
At least two of the three vertices must lie in one of the bipartite
sets because these two are joined by edge; thus, the graph cannot
be bipartite.
148 Discrete Mathematical Structures

12. Show that if G is a bipartite simple graph with v vertices and e


v2
edges, then e ≤ .
4
Solution.
Let G be a complete bipartite graph with v vertices.
Let v1 and v2 be the number of vertices in the partitions V1 and
V2 of vertex set of G.
Since G is complete bipartite, each vertex in V1 is joined to each
vertex in V2 by exactly one edge.
Thus, G has v1 v2 edges when v1 + v2 = v.
But we know the maximum value of v1 v2 subject to v1 + v2 = v
v2
is .
4
v2
Thus, the maximum number of edges in G is .
4
2
v
That is, e ≤ .
4
13. Show that the graph G is bipartite.

V2
V1
V3

V7

V6 V4

V5
Graph G

Solution.
Graph G is bipartite since its vertex set is the union of two disjoint
sets {v1 , v2 , v3 } and {v4 , v5 , v6 , v7 } and each edge connects a vertex
in one of these subsets to a vertex in the other subset.
14. Draw the complete bipartite graphs K2,3 , K3,3 , K3,5 , and K2,6 .
Solution.
The complete bipartite graphs are shown below.
Graphs 149

K2,3
K3,3

K3,5 K2,6
Given complete bipartite graphs

15. How many subgraphs with at least one vertex does K3 have?
Solution. 17

3.4 Representing Graphs and Graph Isomorphism


We can represent a simple graph in the form of edge list or in the form of
adjacency lists which may be useful in computer programming.
Definition 3.4.1 Adjacency matrix of a simple graph: Let G = (V, E)
be a simple graph with n vertices {v1 , v2 , . . . , vn }. Its adjacency matrix is
denoted by A = (aij ) and defined by
(
1, if vi and vj are adjacent
aij =
0, otherwise.

Example 3.4.2
v1

v5 v2

v4 v3
Example of a graph with adjacency matrix
150 Discrete Mathematical Structures

The adjoint matrix for the graph in the figure above is

 v1 v2 v3 v4 v5 
v1 0 1 0 0 1
v2  1 0 1 1 1 
A=  
v3  0
 1 0 1 1 
v4  0 1 1 0 1 
v5 1 1 1 1 0

Definition 3.4.3 Incidence Matrix: Let G = (V, E) be a graph with n


vertices v1 , v2 , . . . , vn and m edges e1 , e2 , . . . , em . Then, the n × m matrix
B = (bij ) where
(
1, if the edge ej is incident on vi
bij =
0, otherwise.

Example 3.4.4

v1 e1 v2

e4 e2
e5

v4 v3
e3
Example of a graph with incidence matrix

The incidence matrix for the graph in the figure above is

e1 e2 e3 e4 e5
 
v1 1 0 0 1 1
 1 1 0 0 0 
B= v2  
 0 1 1 0 1 
v3  
v4 0 0 1 1 0

Observations about the incidence matrix:


1. Since every edge is incident on exactly two vertices, each column of
B has exactly two 1’s.
2. The number of 1’s in each row is equal to the degree of the
corresponding vertex.
3. A row with all 0’s represents an isolated vertex.
4. Parallel edges in a graph produce identical columns in its incidence
matrix.
Graphs 151

Definition 3.4.5 Isomorphic Graphs: The simple graphs G1 = (V1 , E1 )


and G2 = (V2 , E2 ) are isomorphic if there is a one-to-one and onto function
f from V1 to V2 with the property that a and b are adjacent in G1 if and only
if f (a) and f (b) are adjacent in G2 , for all a and b in V1 . Such a function f
is called an isomorphism.
In other words, two graphs G1 and G2 are isomorphic if there is a function
f : V (G1 ) −→ V (G2 ) from the vertices of G1 to the vertices of G2 such that
(i) f is one-to-one
(ii) f is onto and
(iii) for each pair of vertices u and v of G1
[u, v] ∈ E(G1 ) ⇔ [f (u), f (v)] ∈ E(G2 ).

Any function f with the above three properties is called an isomorphism from
G1 to G2 .

Example 3.4.6 Consider the graphs G1 and G2 in the following figure. Let
f : G1 −→ G2 be a function with f (u1 ) = v1 , f (u2 ) = v4 , f (u3 ) = v3 ,
f (u4 ) = v2 . Then, f is a one-to-one and onto function between G1 and G2 .
Here, f preserves the adjacency. The adjacent vertices in G1 are u1 and u2 ,
u1 and u3 , u2 and u4 , and u3 and u4 , and each of the pairs f (u1 ) = v1 and
f (u2 ) = v4 , f (u1 ) = v1 and f (u3 ) = v3 , f (u2 ) = v4 and f (u4 ) = v2 , and
f (u3 ) = v3 and f (u4 ) = v2 are adjacent in G2 . Hence, the graphs G1 and G2
are isomorphic.
u1 u2 v1 v2

u3 u4 v3 v4
G1 G2

Example of isomorphic graphs

3.4.1 Solved Problems


1. Write the adjacency matrix of the following graph.
u1 u2

u3 u4
Given graph
152 Discrete Mathematical Structures

Solution.
The adjacency matrix is

u u2 u3 u4
 1 
u1 0 1 1 1
 1 0 0 1 
A= u2  
 1 0 0 1 
u3  
u4 1 1 1 0

2. Write the incidence matrix for the following graph.


e2
v1 v2 e4 v3

e1 e3
e7 e6 e5
v4

v5
e8
Given graph
Solution.
The incidence matrix is

e1 e2 e3 e4 e5 e6 e7 e8 
1 1 1 0 0 0 0 0
v1  0 1 1 1 0 1 1 0 
v2  
B=  0 0

0 1 1 0 0 0 

v3  
v4  0 0
 0 0 0 0 0 0 
v5 0 0 0 0 1 1 0 0

3. What is the sum of the entries in a row of the incidence matrix for
an undirected graph?
Solution.
Sum is 2 if the edge e is not a loop and 1 if the edge e is a loop.
4. Check whether the two graphs are isomorphic or not.

G1 G2
Given graphs
Graphs 153

Solution.
In the graph G1 , vertices of degree 2 are not adjacent, while in
the graph G2 , vertices of degree 2 are adjacent. Since isomorphism
preserves adjacency of vertices, the graphs are not isomorphic.
5. Prove that the graphs G1 and G2 are isomorphic.

v1
u1

v5 v2
u3 u2

u4 v4 v3
u5
G1 G2

Given graphs

Solution.
The two graphs have the same number of vertices, same number of
edges, and same degree sequence. Consider the function f defined by
f (u1 ) = v1 , f (u2 ) = v3 , f (u3 ) = v4 , f (u4 ) = v2 , f (u5 ) = v5 .
Then, the adjacency matrices of the two graphs corresponding to
f are
u1 u2 u3 u4 u5
u1 0 1 1 0 0
u2  1 0 0 0 1 
A(G1 ) =  
u3  1 0 0 1 0 
 
u4  0 0 1 0 1 
u5 0 1 0 1 0

 v1 v2 v3 v4 v5 
v1 0 1 1 0 0
v2  1 0 0 0 1 
A(G2 ) =  
v3  1
 0 0 1 0 
v4  0 0 1 0 1 
v5 0 1 0 1 0
Therefore, A(G1 ) = A(G2 ). Hence, G1 and G2 are isomorphic to
each other.
6. Prove that any two simple connected graphs with n vertices all of
degree 2 are isomorphic.
Solution.
We know that the total degree of a graph is given by
n
X
deg(vi ) = 2|E|.
i=1
154 Discrete Mathematical Structures

Then, |V | = number of vertices = n


|E| = number of edges.
Further,
Pn the degree of each vertex is 2.
∴ i=1 2 = 2|E|
=⇒ n = |E|
∴ Number of edges = number of vertices. Hence, the graphs are
cycle graphs. Therefore, they are isomorphic.
7. Can a simple graph with seven vertices be isomorphic to its
complement?
Solution.
A graph with seven vertices can have a maximum number of edges
7(7 − 1) 7×6
= = = 21 edges.
2 2
But 21 edges cannot be splitted into two equal integers. Therefore,
a graph and its complement cannot have equal number of edges.
Hence, a graph with seven vertices cannot be isomorphic to its
complement.
8. Let G be a simple graph, all of whose vertices have degree 3 and
|E| = 2|V | − 3. What can be said about G?
Solution.
|V |
X
deg(vi ) = 2|E|
i=1
3(|V | − 1 + 1) = 2|E|
3|V | = 2|E| = 2(2|V | − 3) = 4|V | − 6
=⇒ |V | = 6.
The number of vertices in G is six. Hence, it can be concluded that
G is isomorphic to K3,3 .
9. Show that isomorphism of simple graphs is an equivalence relation.
Solution.
(i) Reflexive: G is isomorphic to itself by the identity function.
Hence, isomorphism is reflexive.
(ii) Symmetric: Suppose that G is isomorphic to H. Then,
there exists a one-to-one correspondence f from G to H that
preserves adjacency and non-adjacency. From this, f −1 is a one-
to-one correspondence from H to G that preserves adjacency
and non-adjacency. Hence, isomorphism is symmetric.
(iii) Transitive: If G is isomorphic to H and H is isomorphic to
K, then there are one-to-one correspondences f and g from
G to H and from H to K that preserve adjacency and non-
adjacency. It follows that g ◦ f is a one-to-one correspondence
Graphs 155

from G to K that preserves adjacency and non-adjacency.


Hence, isomorphism is transitive.
From the above (i)–(iii), isomorphism is an equivalence relation.

3.4.2 Problems for Practice


1. Write the adjacency matrix of the following graph.

a b

c d

2. Draw the directed graph for the following adjacency matrix.


 
0 0 1 1
0 0 1 0
 
1 1 0 1
1 1 1 0

3. Draw an undirected graph for the following adjacency matrix.


 
1 2 0 1
2 0 3 0
 
0 3 1 1
1 0 1 0

4. Find the adjacency matrix of the given directed graph.

v1
v2

v4
v3
156 Discrete Mathematical Structures

5. Check whether the graphs with the following adjacency matrices


are isomorphic.
   
0 0 1 0 1 1
0 0 1 , 1 0 0
1 1 0 1 0 0

6. Determine whether the graphs with the following adjacency


matrices are isomorphic.
   
0 1 1 0 0 1 0 1
1 0 0 1 1 0 0 0
 , 
1 0 0 1 0 0 0 1
0 1 1 0 1 0 1 0

7. Find whether the following graphs are isomorphic to each other.

u1 u2 v1 v2

v4
u3 u4 v3

G1 G2

8. Find a pair of non-isomorphic graphs with the same degree sequence


such that one graph is bipartite but the other graph is not bipartite.
9. What is the product of the incidence matrix and its transpose for
an undirected graph?
10. Draw the graph represented by the following adjacency matrix.
 
1 2 1
2 0 0
0 2 2

3.5 Connectivity
Definition 3.5.1 Walk: A walk is defined as a finite alternating sequence
of vertices and edges, beginning and ending with vertices such that each edge
is incident with the vertex preceding and following it. (No edge appears more
than once, and vertex may be repeated.)
Graphs 157

Example 3.5.2 Consider the graph G. In this graph, v1 e1 v4 e2 v3 e6 v5 is a


walk.
v1 e4
v2 v5

e3
e1 e5 e6

v4 v3
e2
Example of a walk

Remarks:
1. A walk is also referred to as an edge train or a chain.
2. No edge appears more than once in a walk.
3. Every walk is a subgraph of G.

Definition 3.5.3 Terminal Vertex: In a walk, the vertex that begins and
ends the walk is called its terminal vertex.
For example, in the walk v1 e3 v3 e5 v2 in the figure above, the terminal
vertices are v1 and v2 .

Definition 3.5.4 Closed Walk: A walk with same end vertices is called a
closed walk.
In the example above, v1 e3 v3 e5 v2 e4 v1 is a closed walk.

Definition 3.5.5 Open Walk: A walk which is not closed is called an open
walk.
In the graph above, v1 e3 v3 e5 v2 is an open walk.

Definition 3.5.6 Path or Simple Path or Elementary Path: An open


walk in which no vertex appears more than once is called a path or simple path
or an elementary path.
For example, in the graph above, v1 e3 v3 e6 v5 is a path, but
v1 e3 v3 e5 v2 e4 v1 e1 v4 is not a path, since v1 is repeated twice.

Definition 3.5.7 Circuit or Cycle or Circular Path or Elementary


Cycle: A closed walk in which no vertex appears more than once is called a
circuit or a cycle or a circular path or an elementary cycle.
In the graph above, v1 e3 v3 e5 v2 e4 v1 is a circuit.

Remarks:
1. A path does not intersect itself.
2. A self-loop can be included in a walk but not a path.
158 Discrete Mathematical Structures

3. Every circuit is called as a non-intersecting walk.


4. Every self-loop is a circuit, but every circuit is not a self-loop.
5. Every vertex in a circuit is of degree 2.

3.5.1 Connected and Disconnected Graphs


Definition 3.5.8 Connected Graph: A graph G is a connected graph if
there is at least one path between every pair of vertices in G. Otherwise G is
a disconnected graph.

Example 3.5.9 Consider the connected graph below.


v1 e4
v2 v5

e3
e1 e5 e6

v4 v3
e2
Connected graph

Definition 3.5.10 Component or Block: A disconnected graph consists of


two or more connected graphs. Each of these connected subgraphs is called a
component or block.

Example 3.5.11

Disconnected graph with two components

Theorem 3.5.12 A graph G is disconnected iff its vertex set V can be


partitioned into two non-empty disjoint subsets V1 and V2 such that there
exists no edge in G whose one end vertex is in the subset V1 and the other in
the subset V2 .
Proof.
Let G be a disconnected graph.
Let us fix a vertex a in G. Let V1 be the set of all vertices such that they
are joined by paths to a.
Graphs 159

Now, consider V2 = V − V1 . Clearly, V2 is not empty (since G is


disconnected, V1 does not contain all vertices of G). Also, no vertex in V1
is joined to any vertex in V2 by an edge.
Conversely, suppose the partition exists.
Consider any two arbitrary vertices a and b of G such that a ∈ V1 and
b ∈ V2 .
=⇒ There exists no path between a and b.
Otherwise, there would be at least one edge whose one end vertex would
be in V1 and other end in V2 .
=⇒ G is not connected.
Hence, the theorem is proved.

Theorem 3.5.13 If a graph (connected or disconnected) has exactly two


vertices of odd degree, then there must be a path joining these two vertices.
Proof.
Let G be a graph with exactly two odd degree vertices. Let them be v1 and v2 .
Case (i): Let G be a connected graph.
=⇒ There exists at least one path between v1 and v2 .
Case (ii): Let G be a disconnected graph.
=⇒ There exist at least two components.
Without loss of generality, we can assume G contains two
components g1 and g2 .
If both v1 and v2 lie in the same component, then there exists
a path joining between v1 and v2 .
Suppose v1 lies in g1 and v2 lies in g2 . Then, g1 is a graph
containing exactly one odd degree vertex, and g2 is a graph
having exactly one odd degree vertex, which is a contradiction
to the theorem that “the number of odd degree vertices is
always even”.
Therefore, v1 and v2 must be in the same component.
Hence, the theorem is proved.
Theorem 3.5.14 A simple graph with n vertices and k components can have
(n − k)(n − k + 1)
at most edges.
2
Proof.
Let G be a simple graph with n vertices and k components, namely
g1 , g2 , . . . , gk .
Let ni be the number of vertices in the ith components gi , for all
i = 1, 2, . . . , k.
Clearly, Pkn1 + n2 + · · · + nk = n
=⇒ i=1 ni = n, and ni ≥ 1.
We know that the maximum number of edges in the ith component is
ni (ni − 1)
.
2
160 Discrete Mathematical Structures

∴ The maximum number of edges in G


= the number of edges in g1 + the number of edges in g2
+ · · · + the number of edges in gk
= n1 (n1 − 1)/2 + n2 (n2 − 1)/2 + · · · + nk (nk − 1)/2
k k k k
!
X X X X
n2i − ni /2 = n2i −

= ni (ni − 1)/2 = ni /2
i=1 i=1 i=1 i=1
k
!
X
= n2i − n /2. (3.2)
i=1

Let us consider
k
X k
X k
X
(ni − 1) = ni − 1
i=1 i=1 i=1
= (n − k).

Squaring on both sides, we get


" k #2
X
(ni − 1) = (n − k)2
i=1
(or) [(n1 − 1) + (n2 − 1) + · · · + (nk − 1)]2 = (n − k)2
[(n1 − 1)2 + (n2 − 1)2 + · · · + (nk − 1)2 ]
+ (positive cross terms) = (n − k)2
(or) [(n21 + 1 − 2n1 ) + (n22 + 1 − 2n2 ) + · · · + (n2k + 1 − 2nk )]
+ (positive cross terms) = (n − k)2
(or) (n21 + n22 + · · · + n2k ) + (1 + 1 + · · · + 1)
| {z }
k times
− 2(n1 + n2 + · · · + nk ) + (positive cross terms) = (n − k)2
k
X k
X
(or) n2i − k − 2 ni + (positive cross terms) = (n − k)2
i=1 i=1
k
X k
X
(or) n2i + k − 2 ni ≤ (n − k)2 (by omitting the positive cross terms)
i=1 i=1
k
X
(or) n2i + k − 2n ≤ (n − k)2
i=1
k
X
(or) n2i ≤ (n − k)2 − k + 2n. (3.3)
i=1
Graphs 161
Therefore, the maximum number of edges in G
k
1X 2 n
= n − [using (3.2)]
2 i=1 i 2
≤ [(n − k)2 − k + 2n − n]/2 [using (3.3)]
2
= [(n − k) − k + n]/2
= (n − k)(n − k + 1)/2.

Hence, the theorem is proved.

3.6 Eulerian and Hamiltonian Paths


Definition 3.6.1 Eulerian Circuit: An Eulerian circuit in a graph G is a
simple circuit containing every edge of G.

Definition 3.6.2 Eulerian Trail: A trail in G is called an Eulerian trail if


it includes each edge of G exactly once.

Definition 3.6.3 Eulerian Path: An Eulerian path in G is a simple path


containing every edge of G.

Definition 3.6.4 Eulerian Graph: A closed walk which contains all edges
of the graph G is called an Euler line, and the graph containing at least one
Euler line is called an Eulerian graph.

Example 3.6.5 The graphs are Eulerian graphs.

Star of David Mohammed's Semi Stars


Examples of Eulerian graphs

Theorem 3.6.6 A given connected graph G is an Eulerian graph if and only


if all vertices of G are of even degree.
Proof.
Suppose G is an Eulerian graph.
162 Discrete Mathematical Structures

=⇒ G contains an Euler line.


=⇒ G contains a closed walk covering all edges.
To prove: All vertices of G are of even degree.
In tracing the closed walk, every time the walk meets a vertex v, it goes
through two new edges incident on v with one we “entered” and other “exited”.
Since it is a closed walk, this is true for all vertices (intermediate and terminal
vertices). Thus, the degree of every vertex is even.
Conversely, suppose that all vertices of G are of even degree.
To prove: G is an Eulerian graph.
That is, to prove G contains an Euler line.
Construct a closed walk starting at an arbitrary vertex v and going through
the edges of G such that no edge is repeated. Since each vertex is of even
degree, we can exit from each and every vertex where we enter, and the tracing
can stop only at the vertex v. Name the closed walk as h.
Case (i): If h covers all edges of G, then h becomes an Euler line, and hence
G is an Eulerian graph.
Case (ii): If h does not cover all edges of G, then remove all edges of h from
G and obtain the graph G0 . Since both G and G0 have vertices which are of
even degree, every vertex in G0 is also of even degree.
Since G is connected, h will touch G0 in at least one vertex v 0 . Starting
from v 0 , we can again construct a new walk h0 in G0 . This will terminate only
at v 0 , since every vertex in G0 is also of even degree.
Now, this walk h0 combined with h forms a closed walk, starts and ends
at v, and has more edges than h.
This process is repeated until we obtain a closed walk covering all edges
of G. Thus, G is an Eulerian graph.
Theorem 3.6.7 A connected multigraph with at least two vertices has an
Eulerian circuit if and only if each of its vertices has even degree.
Proof.
Given: A connected multigraph G has an Eulerian path but not an Eulerian
circuit.
To prove: G has exactly two vertices of odd degree.
Suppose that a connected multigraph has an Eulerian path from a to b but
not an Eulerian circuit. The first edge of the path contributes 1 to the degree
of a. A contribution of 2 to the degree of a is made every time the path passes
through a. The last edge in the path contributes 1 to the degree of b.
Every time the path goes through b, there is a contribution of 2 to its
degree. Consequently, both a and b have odd degree. Every other vertex has
even degree, because the path contributes 2 to the degree of a vertex whenever
it passes through it.
Theorem 3.6.8 A connected multigraph has an Eulerian path but not an
Eulerian circuit if and only if it has exactly two vertices of odd degree.
Proof.
Given: The graph has exactly two vertices of odd degree.
Graphs 163

To prove: G has an Eulerian path.


Suppose that a graph has exactly two vertices of odd degree, say a and b.
Consider the larger graph made up of the original graph with the addition of
an edge {a, b}.
Every vertex of this larger graph has even degree, so there is an Eulerian
circuit. The removal of the new edge produces an Eulerian path in the original
graph.
Chinese postman problem:
If a postman can find an Eulerian path in the graph that represents the
streets the postman needs to cover, this path produces a route that traverses
each street of the route exactly once. If no Eulerian path exists, some streets
will have to be traversed more than once. This problem is known as the Chinese
postman problem.
Theorem 3.6.9 A directed multigraph having no isolated vertices has an
Eulerian path but not an Eulerian circuit if and only if the graph is weakly
connected and the in-degree and out-degree of each vertex are equal for all but
two vertices, one that has in-degree larger than its out-degree by 1 and the
other that has out-degree larger than its in-degree by 1.
Proof.
If there is an Eulerian path, as we follow, each vertex except the starting
and ending vertices must have equal in-degree and out-degree, since whenever
we come to a vertex along an edge, we leave it along another edge. The starting
vertex must have out-degree 1 larger than its in-degree, since we use one edge
leading out of this vertex and whenever we visit it again, we use one edge
leading into it and one leaving it.
Similarly, the ending vertex must have in-degree 1 greater than its
out-degree. Since the Eulerian path with directions erased produces a path
between any two vertices, in the underlying undirected graph, the graph is
weakly connected.
Conversely, suppose the graph meets the degree conditions stated. If we
add one more edge from the vertex of deficient out-degree to the vertex X with
equal in-degree and out-degree. Because the graph is still weakly connected,
by this new graph has an Eulerian circuit. Now, delete the added edge to
obtain the Eulerian path.

3.6.1 Hamiltonian Path and Hamiltonian Circuits


Definition 3.6.10 Hamiltonian Path: A simple path in a graph G that
passes through every vertex exactly once is called a Hamiltonian path. That
is, the simple path x0 , x1 , x2 , . . . , xn−1 , xn in the graph G = (V, E) is
a Hamiltonian path if V = {x0 , x1 , x2 , . . . , xn−1 , xn } and xi 6= xj for
0 ≤ i < j ≤ n.
Definition 3.6.11 A simple circuit in a graph G that passes through
every vertex exactly once is called a Hamiltonian circuit. And the simple
164 Discrete Mathematical Structures

circuit x0 , x1 , . . . , xn−1 , xn , x0 (with n > 0) is a Hamiltonian circuit if


x0 , x1 , . . . , xn−1 , xn is a Hamiltonian path.
Definition 3.6.12 Unicursal Graph: An open walk that includes all edges
of G without retracing any edge is called a unicursal line or an open Euler
line.
A connected graph which has a unicursal line is called a unicursal graph.
Example 3.6.13 In the graph, the unicursal line is v2 e1 v4 e2 v3 e3 v2 e4 v1 .

v1 v4

e4 e1 e2

v2 v3
e3
Example of unicursal graph
Remark:
1. Adding an edge between the initial and final vertices of unicursal
line, we obtain an Euler line.
2. A connected graph is unicursal, if it has exactly two odd degree
vertices.
Theorem 3.6.14 In a connected graph G, with exactly 2k odd degree vertices,
there exist k edge-disjoint subgraphs such that they together contain all edges
of G and that each is a unicursal graph.
Proof.
Let the odd degree vertices of the given graph be named as v1 , v2 , . . . , vk and
ω1 , ω2 , . . . , ωk in any arbitrary order.
Add k edges (new edges) to G between the pair of vertices (v1 , ω1 ),
(v2 , ω2 ),. . . , (vk , ωk ) to form a new graph G0 .
In the resultant graph G0 , every vertex is of even degree.
=⇒ G0 is an Eulerian graph.
=⇒ G0 contains an Euler line, say P .
If we remove k newly added edges from P , that will split P into k walks
each of a unicursal line to itself.
That is, the first removal will leave a single unicursal line. The second
removal will split that into two unicursal lines, and each successive removal
will split a unicursal line into two unicursal lines, until there are k of them.
Hence, the theorem is proved.

3.6.2 Solved Problems


1. Does the graph given below have a Hamiltonian path? If so, find
such a path. If it does not, give an argument to show why no such
path exists.
Graphs 165
a d

c f

b e
Given graph
Solution.
a–b–c–f –d–e is a Hamiltonian path.
2. Does the graph below have a Hamiltonian path. If so, find such a
path. If it does not, give an argument to show why no such path
exists.
a b

d
e f
Given graph
Solution.
f –e–d–a–b–c is a Hamiltonian path.
3. Does the graph given below have a Hamiltonian path? If so, find
such a path. If it does not, give an argument to show why no such
path exists.

b
a c

j
i k

d h
o p q

n l
m

e g
f
Given graph

Solution.
No Hamiltonian path exists. There are eight vertices of degree 2,
and only two of them can be end vertices of a path. For each of the
other six, their two incident edges must be in the path. It is easy
to see that if there is to be a Hamiltonian path, exactly one of the
inside corner vertices must be an end and that this is impossible.
166 Discrete Mathematical Structures

4. Show that Kn has a Hamiltonian circuit whenever n ≥ 3.


Solution.
We can form a Hamiltonian circuit in Kn beginning at any vertex.
Such a circuit can be built by visiting vertices in any order we
choose, as long as the path begins and ends at the same vertex and
visits each of the other vertices exactly once.
It is possible since there are edges in Kn between any two vertices.
5. Give an example of a graph that has an Eulerian circuit and a
Hamiltonian circuit, which are distinct.
Solution.
The graph having an Eulerian circuit and a Hamiltonian circuit
which are distinct as shown below. The Eulerian circuit is
a–c–b–c–d–b–a.
b

a d

c
Graph with Eulerian and Hamiltonian circuits

The Hamiltonian circuit is a–b–d–c–a.


6. Give an example of a graph which has a Hamiltonian circuit but
not an Eulerian circuit.
Solution.
The graph having a Hamiltonian circuit but not an Eulerian circuit
as shown below.
a b

c d
Graph with Hamiltonian circuit but no Eulerian circuit

The Hamiltonian circuit is a–b–d–c–a.


There is no Eulerian circuit.
7. Give an example of a graph which has an Eulerian circuit but not
a Hamiltonian circuit.
Graphs 167

Solution.
The graph having an Eulerian circuit but not a Hamiltonian circuit
as shown below.
b d

a
c

e f

Graph with Eulerian circuit but no Hamiltonian circuit


The Eulerian circuit is a–e–b–e–c–d–f –c–b–a.
There is no Hamiltonian circuit.
8. Show that a bipartite graph with an odd number of vertices does
not have a Hamiltonian circuit.
Solution.
Suppose that G = (V, E) is a bipartite graph with V = V1 ∪ V2 ,
where no edge connects a vertex in V1 and a vertex in V2 .
Suppose that G has a Hamiltonian circuit. Such a circuit must
be of the form a1 , b1 , a2 , b2 , . . . , ak , bk , where ai ∈ V1 and bi ∈ V2
for i = 1, 2, . . . , k. Since the Hamiltonian circuit visits each vertex
exactly once, except for v1 , where it begins and ends, the number
of vertices in the graph equals 2k, an even number.
Hence, a bipartite graph with an odd number of vertices cannot
have a Hamiltonian circuit.
9. For which values of m and n does the complete bipartite graph
Km,n have a Hamiltonian circuit?
Solution. m = n ≥ 2.
n−1
10. In a complete graph with n vertices, show that there are
2
edge-disjoint Hamiltonian circuits, if n is an odd number ≥ 3.
Solution.
A complete graph G of n vertices has n(n − 1)/2 edges, and a
Hamiltonian circuit in G consists of n edges.
Therefore, the number of edge-disjoint Hamiltonian circuits in G
cannot exceed (n − 1)/2. That is, there are (n − 1)/2 edge-disjoint
Hamiltonian circuits, when n is odd.
The subgraph (of a complete graph of n vertices) in the figure below
is a Hamiltonian circuit. Keeping the vertices fixed on a circle, rotate
the polygonal pattern clockwise by 360/(n − 1), 2 · 360/(n − 1),
168 Discrete Mathematical Structures

3 · 360/(n − 1),. . . , (n − 3)/2 · 360/(n − 1) degrees. Observe that


each rotation produces a Hamiltonian circuit that has no edge in
common with any of the previous ones.
Thus, we have (n − 3)/2 new Hamiltonian circuits, all edge-
disjoint from the one as shown below and also edge-disjoint among
themselves.
5

3 n-2

n
2 1

n-1
4

n-3
Required graph for the problem

11. Explain Konigsberg bridge problem. Represent the problem by


means of a graph. Does the problem have a solution?
Solution.
There are two islands A and B formed by a river. They are
connected to each other and to the river banks C and D by means
of seven bridges. The problem is to start from any one of the four
land areas A, B, C, D, walk across each bridge exactly once, and
return to the starting point.

D
A

Konigsberg bridge and its graph

The situation is represented by a graph, with vertices representing


the land areas and the edges and the bridges.
Graphs 169

This problem is the same as that of drawing the graph without


lifting the pen from the paper and without retracing any line.
In other words, the problem is to find whether there is an Eulerian
circuit in the graph. But a connected graph has an Eulerian circuit
if and only if each of its vertices is of even degree.
In the present case, all the vertices are of odd degree. Hence,
Konigsberg bridge problem has no solution.

3.6.3 Problems for Practice


1. Can someone cross all the bridges shown in this map exactly once
and return to the starting point?

2. In Kaliningrad (the Russian name for Konigsberg), there are two


additional bridges, besides the seven that were present in the 18th
century. These new bridges connect regions B and C and regions
B and D, respectively. Can someone cross all nine bridges in
Kaliningrad exactly once and return to the starting point?
3. Show that a directed multigraph having no isolated vertices has an
Eulerian circuit if and only if the graph is weakly connected and
the in-degree and out-degree of each vertex are equal.
4. For which values of n do the following graphs have an Eulerian
circuit?
(a) Kn (b) Cn (c) Wn (d) Qn

3.6.4 Additional Problems for Practice


1. Can you draw a graph of five vertices with degree sequence 1, 2, 3,
4, 5?
2. Show that there does not exist a graph with five vertices with
degrees 1, 3, 4, 2, 3, respectively.
3. How many edges are there in a graph with ten vertices each of
degree 5?
170 Discrete Mathematical Structures

4. Let G be a graph with ten vertices. If four vertices have degree 4


and six vertices have degree 5, then find the number of edges of G.
5. Draw
 the graph
 represented by the given adjacency matrix:
0 1 0 1
1 0 1 0
0 1 0 1 .
 

1 0 1 0
6. How do you find the number of different paths of length r from i to
j in a graph G with adjacency matrix A?
7. Is the directed graph given below strongly connected? Why or
why not?
1 2

4 3

8. Define a bipartite graph.


9. Draw the complete graph K5 .
10. Define isomorphism of directed graphs.
11. What do strongly connected components of a telephone call graph
represent?
12. Define Hamiltonian path.
13. Give an example for a graph which is
(i) Eulerian and Hamiltonian
(ii) neither Eulerian nor Hamiltonian.
14. Describe a discrete structure based on a graph that can be used to
model airline routes and their flight times.
15. Show that a simple graph G with n vertices is connected if it has
more than (n − 1)(n − 2)/2 edges.
16. Show the isomorphism of simple graphs is an equivalence relation.
17. Derive an algorithm for constructing Eulerian path in directed
graphs.
18. Are simple graphs with the following adjacency matrices isomorphic?
   
0 1 0 0 0 1 0 1 0 0 0 1
1 0 1 0 1 0 1 0 1 0 0 1
   
0 1 0 1 0 1 0 1 0 1 1 0
 , 
0 0 1 0 1 0 0 0 1 0 1 0
   
0 1 0 1 0 1 0 0 1 1 0 1
1 0 1 0 1 0 1 1 0 0 1 0
Graphs 171

19. Examine whether the two graphs G and G0 associated with the
following adjacency matrices are isomorphic.
   
0 1 0 1 0 0 0 1 0 0 1 0
1 0 1 0 0 1 1 0 1 0 0 0
   
0 1 0 1 0 0 0 1 0 1 0 1
1 0 1 0 1 0 , 0 0 1
   
   0 1 0

0 0 0 1 0 1 1 0 0 1 0 1
0 1 0 0 1 0 0 0 1 0 1 0

20. Discuss the various graph invariants preserved by isomorphic


graphs.
21. If G is a self-complementary graph, then prove that G has n ≡ 0 or
1 (mod 4) vertices.
22. If G is a connected simple graph with n vertices with n ≥ 3, such
n
that the degree of every vertex in G is at least , then prove that
2
G has Hamiltonian cycle.
23. In a round robin tournament, the team 1 beats team 2, team 3, and
team 4; team 2 beats team 3 and team 4; and team 3 beats team
4. Model this outcome with a directed graph.
24. Show that the number of vertices of odd degree in an undirected
graph is even.
25. If a graph, either connected or disconnected, has exactly two vertices
of odd degree, prove that there is a path joining these two vertices.
26. Find an Eulerian path or Eulerian circuit if it exists in each of the
following two graphs.

A B B C

A
(i) (ii) D

C D E F E

27. Determine whether the following graphs G and H are isomorphic.


Give reason.
u2 v1

u1 u3 v5 v2

u5 v4 v3
u4

G H
172 Discrete Mathematical Structures

28. Which of the following simple graphs have a Hamiltonian circuit,


or if not, a Hamiltonian path?
G1 G2 G3
a b g
a b a b

e c d c
d c e f

29. Prove that a simple graph is bipartite if and only if it is possible


to assign one of two different colours to each vertex of the graph so
that no two adjacent vertices are assigned the same colour.
30. Show that every connected graph with n vertices has at least n − 1
edges.
31. Show that there does not exist a graph with five vertices with
degrees 1, 3, 4, 2, 3, respectively.
32. Can you draw a graph of five vertices with degree sequence 1, 2, 3,
4, 5?
33. Draw a graph that is Eulerian but not Hamiltonian.
33. Let G be a graph with ten vertices. If four vertices have degree 4
and six vertices have degree 5, then find the number of edges of G.
34. How many edges are there in a graph with ten vertices each of
degree 5?
35. Give an example for a graph which is
(i) Eulerian and Hamiltonian
(ii) neither Eulerian nor Hamiltonian.
36. How do you find the number of different paths of length r from i to
j in a graph G with adjacency matrix A?
37. Establish the isomorphism of the following pairs of graphs.
v6
u6

v1 v2 v3 v4 v5 u1 u2 u3 u4 u5
4
Algebraic Structures

4.1 Introduction
An algebraic system can be described as a set of objects together with some
operations. These operations will impose a certain structure on the set. In this
chapter, we study the axiomatic set theory, semigroups, groups, and monoids
which are the basic tools of discrete mathematics.

4.2 Algebraic Systems


Definition 4.2.1 Binary Operation: Let A be any set. A mapping
f : A × A −→ A is called a binary operation.

Definition 4.2.2 Algebraic System or Algebra: A set together with a


number of (binary) operations on the set is called an algebraic system or an
algebra.

Properties of Binary Operations:


Let G be a set.
(i) Closure Property: A binary operation ? : G × G −→ G is said to
be closed, if for all a, b ∈ G, an element a ? b = x ∈ G.
(ii) Associative Property: a ? (b ? c) = (a ? b) ? c, for all a, b, c ∈ G.
(iii) Existence of Identity: There exists an element e ∈ G such that
e ? a = a ? e = a, for all a ∈ G. The element e is called the identity
element.
(iv) Existence of Inverse: For a ∈ G, there exists an element b ∈ G
such that a ? b = b ? a = e. The element b is called the inverse of a,
and it is denoted by b = a−1 .
(v) Commutativity: For all a, b ∈ G, if a ? b = b ? a, then ? is
commutative.

173
174 Discrete Mathematical Structures

(vi) Distributive Properties: Let • be any other binary operation on


G. Then,

a ? (b • c) = (a ? b) • (a ? c) (Left distributive law)


(b • c) ? a = (b ? a) • (c ? a) (Right distributive law)

for all a, b, c ∈ G.
(vii) Cancellation Property: For a, b, c ∈ G and a 6= 0,

a ? b = a ? c =⇒ b = c.

Definition 4.2.3 Algebraic Structure: The operations on a set G define


a structure on the elements of G. Then, the algebraic system G is called an
algebraic structure.

Example 4.2.4 Let R be the set of real numbers. Consider the algebraic
system (R, +, ×) where + and × are the operations of addition and
multiplication on R.

4.2.1 Semigroups and Monoids


Definition 4.2.5 Semigroup: A non-empty set S together with the binary
operation ? : S × S −→ S is said to be a semigroup if ? satisfies the
following conditions, namely, the closure property and the associative property.
We denote the semigroup as (S, ?).

Example 4.2.6 Let N = {1, 2, 3, . . . } be the set of natural numbers. Then,


(N, +) and (N, •) are semigroups under the binary operations of addition and
multiplication, respectively.

Definition 4.2.7 Monoid: A semigroup (M, ?) with an identity element e,


with respect to the operation ? is called a monoid. In other words, a non-empty
set M together with the binary operation ? : M × M −→ M is said to be a
monoid if ? satisfies the closure property, associative property, and the identity
property.

Example 4.2.8 Let Z+ = {0, 1, 2, 3, . . . } be the set of all non-negative


integers. Then, (Z+ , +) is a semigroup as well as a monoid.

Definition 4.2.9 A semigroup (or monoid) (S, ?) is said to be commutative


or abelian if a ? b = b ? a, for all a, b ∈ S.

Example 4.2.10 The set of integers, the set of reals, the set of complex
numbers are abelian semigroups (abelian monoids) under the usual operations
of addition and multiplication.

Definition 4.2.11 Idempotent element: Let (G, ?) be a group. An element


a ∈ G is said to be idempotent if a ? a = a.
Algebraic Structures 175

Definition 4.2.12 Congruence Relation: Let (X, ?) be an algebraic


system and E an equivalence relation on X. The relation E is called a
congruence relation on (X, ?) if E satisfies the substitution property with
respect to the operation ?.

Remark 4.2.13 Substitution Property: Let (X, ?) be an algebraic system


in which ? is a binary operation on X. Let us assume that E is an equivalence
relation on X.
The equivalence relation E is said to have the substitution property with
respect to the operation ? if and only if for any x1 , x2 ∈ X,

(x1 Ex01 ) ∧ (x2 Ex02 ) = (x1 ? x2 )E(x2 ? x02 )

where x01 , x02 ∈ X.

4.2.2 Solved Problems


1. Show that intersection of any two congruence relations on a set A is
again a congruence relation on A.
Solution.
Let E1 and E2 be two congruence relations on (A, ?).
=⇒ (a1 E1 a01 ) ∧ (a2 E1 a02 ) = (a1 ? a2 )E1 (a01 ? a02 )
and (a1 E2 a01 ) ∧ (a2 E2 a02 ) = (a1 ? a2 )E2 (a01 ? a02 ).
Let E = E1 ∩ E2 .
To prove: E is a congruence relation on A

(a1 Ea01 ) ∧ (a2 Ea02 ) = [a1 (E1 ∩ E2 )a01 ] ∧ [a2 (E1 ∩ E2 )a02 ]
= (a1 E1 a01 ) and (a1 E1 a01 ) ∧ (a2 E2 a02 ) and (a2 E2 a02 )
= (a1 E1 a01 ) ∧ (a2 E1 a02 ) and (a1 E2 a01 ) ∧ (a2 E2 a02 )
= (a1 ? a2 )E1 (a01 ? a02 ) and (a1 ? a2 )E2 (a01 ? a02 )
= (a1 ? a2 )(E1 ∩ E2 )(a01 ? a02 )
= (a1 ? a2 )E(a01 ? a02 ).

Hence, E is a congruence relation on A.


2. Show that a semigroup with more than one idempotent cannot be
a group. Give an example of a semigroup which is not a group.
Solution.
Let (S, ?) be a semigroup.
Let a, b are two idempotents.
∴ a ? a = a and b ? b = b.
Let us assume that (S, ?) is group. Then, each element has an
inverse. Hence,
(a ? a) ? a−1 = a ? (a ? a−1 ) (by associative property).
176 Discrete Mathematical Structures

Now,

(a ? a) ? a−1 = a ? a−1 = e since a ? a = a. (4.1)

Also,
a ? (a ? a−1 ) = a ? e = a. (4.2)
From (4.1) and (4.2), we get a = e.
Similarly, we can prove that b = e.
But in a group we cannot have two identities, and hence (S, ?)
cannot be a group.
This contradiction is due to an assumption that (S, ?) has two
idempotents.
Example: Let S = {a, b, c} under the operation ?. The composition
table of (S, ?) is shown in the following table.
Composition Table of (S, ?)
? a b c
a a c a
b c b a
c b a c

(S, ?) is a semigroup which is not a group.

3. Give an example of a semigroup which is not a monoid.


Solution.
Let D = {. . . , −4, −2, 0, 2, 4, . . . }
(D, ·) is a semigroup but not a monoid since its multiplicative
identity is 1, but 1 ∈/ D.
4. Give an example of a monoid which is not a group.
Solution.
(Z+ , ·) is a monoid which is not a group, since for all a ∈ Z+ ,
1
/ Z+ .

a
5. What do you call a homomorphism of a semigroup into itself?
Solution.
A homomorphism of a semigroup into itself is called a semigroup
endomorphism.
6. If (Z, +) and (E, +) are two semigroups, where Z is the set of all
integers and E is the set of all even integers, show that the two
semigroups (Z, +) and (E, +) are isomorphic.
Algebraic Structures 177

Solution.
First, we define a function g : Z −→ E by g(a) = 2a, for all a ∈ Z.
To prove g is one-to-one:
Suppose g(a1 ) = g(a2 ), where a1 , a2 ∈ Z.
Then, 2a1 = 2a2 =⇒ a1 = a2 .
Therefore, g is one-to-one.
To prove g is onto:
Suppose b is an even integer.
b
Let a = . Then, a ∈ Z and
2  
b b
g(a) = g = 2 · = b.
2 2
That is, every element b ∈ E has a preimage in Z.
Therefore, g is onto.
To prove g is homomorphism:
Let a, b ∈ Z.
g(a + b) = 2(a + b)
= 2a + 2b
= g(a) + g(b).
Hence, (Z, +) and (E, +) are isomorphic semigroups.
7. If ? is a binary operation on the set of R of real numbers defined
by a ? b = a + b + 2ab,
(i) show that (R, ?) is a semigroup.
(ii) find the identity element if it exists.
(iii) which elements has inverse and what are they?
Solution.
(i)
(a ? b) ? c = (a + b + 2ab) + c + 2(a + b + 2ab)c
= a + b + c + 2(ab + bc + ca) + 4abc
and
a ? (b ? c) = a + (b + c + 2bc) + 2a(b + c + 2bc)
= a + b + c + 2(ab + bc + ca) + 4abc.
Hence, (a ? b) ? c = a ? (b ? c).
Therefore, ? is associative.
(ii) If the identity element exists, let it be e. Then for any a ∈ R,
a?e=a
or a + e + 2ae = a
or e(1 + 2a) = 0.
Therefore, e = 0, since 1 + 2a 6= 0, for any a ∈ R.
178 Discrete Mathematical Structures

(iii) Let a−1 be the inverse of an element a ∈ R. Then, a ? a−1 = e.


That is, a + a−1 + 2a · a−1 = 0
a−1 · (1 + 2a) = −a.
a
Therefore, a−1 = − .
1 + 2a
1 a
Hence, if a 6= , then a−1 = − .
2 1 + 2a
8. Show that a semigroup with more than one idempotent cannot be
a group. Give an example of a semigroup which is not a group.
Solution.
Let (S, ?) be a semigroup.
Let a, b ∈ S be two idempotents. Then,
a ? a = a and b ? b = b.
Let us assume that (S, ?) is a group. Then, each element has its
inverse. Now, by associative property, we have

(a ? a) ? a−1 = a ? (a ? a−1 ).
(a ? a) ? a−1 = a ? a−1 = e. (4.3)
−1
a ? (a ? a ) = a ? e = a. (4.4)

From (4.3) and (4.4), we get a = e.


Similarly, we can prove that b = e.
But in a group, we cannot have two identities and hence (S, ?)
cannot be a group. This contradiction is due to the assumption that
(S, ?) has two idempotents.
Example: Let S = {a, b, c} under the operation ?. The composition
table of (S, ?) is given inthe following table.
Composition Table of (S, ?)
? a b c
a a c a
b c b a
c b a c
(S, ?) is a semigroup but not a group.
9. Let (N, +) be the semigroup of natural numbers and (S, ?) be a
semigroup where S = {e, 0, 1} with the operation ? given in the
following table.
Composition Table of (S, ?)
? e 0 1
e e 0 1
0 0 0 0
1 1 0 1
Algebraic Structures 179

A mapping g : N −→ S is defined by g(0) = 1 and g(j) = 0 for


j 6= 0. Is g is a semigroup homomorphism?
Solution.
Though both (N, +) and (S, ?) are monoids with identities 0 and e,
respectively, g is not a monoid homomorphism because g(0) 6= e.
∴ g is a semigroup homomorphism.
10. If ? is the operation defined on S = Q × Q, the set of ordered pair
of rational numbers and given by (a, b) ? (x, y) = (ax, ay + b), show
that (S, ?) is a semigroup. Is it commutative? Also, find the identity
element of S.
Solution.
Given
(a, b) ? (x, y) = (ax, ay + b). (4.5)
To prove: (S ,?) is a semigroup, that is, to prove ? is associative.

[(a, b) ? (x, y)] ? (c, d) = (ax + ay + b) ? (c, d) [using (4.5)]


= (acx, adx + ay + b) [using (4.5)]. (4.6)

(a, b) ? [(x, y) ? (c, d)] = (a, b) ? (cx, dx + y)


= (acx, adx + ay + b). (4.7)

From (4.6) and (4.7), ? is associative on S.


To prove (S ,?) is not commutative:

(x, y) ? (a, b) = (ax, bx + y). (4.8)

(a, b) ? (x, y) = (ax, ay + b). (4.9)

From (4.8) and (4.9), (a, b) ? (x, y) 6= (x, y) ? (a, b).


Hence, (S, ?) is not commutative.
To find the identity element of (S ,?):
Let (e1 , e2 ) be the identity element of (S, ?). Then, for all
(a, b) ∈ S, we have (a, b) ? (e1 , e2 ) = (a, b)
=⇒ (ae1 , ae2 + b) = (a, b).
=⇒ ae1 = a and ae2 + b = b.
=⇒ e1 = 1 and ae2 = 0 or e2 = 0.
Therefore, (1, 0) is the identity element of (S, ?).
11. Is it true that a semigroup homomorphism preserves identity?
Justify your answer. Or prove by an example that semigroup
homomorphism need not preserve an identity.
Solution.
To prove that semigroup homomorphism need not preserve
an identity:
180 Discrete Mathematical Structures

Let W = {0, 1, 2 . . . }. Then, (W, +) is a semigroup homomorphism


with identity element 0. Let S = {e, 0, 1} and ? be the operation on
S given by the table below.

Composition Table of (S, ?)


? e 0 1
e e 0 1
0 0 0 0
1 1 0 1

Then, (S, ?) is a semigroup with identity e.


Now, define a mapping g : W −→ S by g(0) = 1 and g(i) = 0 for
i 6= 0.
We can see that g(a + b) = g(a) ? g(b), for all a, b ∈ W . Thus, g
is a semigroup homomorphism. But g(0) = 1 6= e. Thus, g does not
preserve the identity.
12. Find all semigroups of Z6 , ×6 where Z6 = {[0], [1], [2], [3], ][4], [5]}.
Solution.
The composition table is given below.

Composition Table of (Z6 , ×6 )


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

The semigroups are

{[0]}, {[0], [1]}, {[1]},


{[1], [2], [4]}, {[0], [1], [2], [4]}, {[2], [4]},
{[0], [3], [4]}, {[1], [5]}, {[0], [1], [5]},
{[0], [4]}, {[0], [1], [4]}, {[0], [2], [3]},
{[0], [1], [2], [3]}.

13. Prove that (Z5 , ×5 ) is a commutative monoid, where ×5 is the


multiplication modulo 5.
Solution.
Z5 = {[0], [1], [2], [3], [4]}.
The composition table is given below.
Algebraic Structures 181
Composition Table of (Z5 , ×5 )
×5 [0] [1] [2] [3] [4]
[0] [0] [0] [0] [0] [0]
[1] [0] [1] [2] [3] [4]
[2] [0] [2] [4] [1] [3]
[3] [0] [3] [1] [4] [2]
[4] [0] [4] [3] [2] [1]

(i) Closure Property:


From the table above, it is clear that Z5 is closed under ×5 .
(ii) Associative Property:
It is also clear that the associative property holds from the
table above. That is, [a] ×5 ([b] ×5 [c]) = ([a] ×5 [b]) ×5 [c], for
all [a], [b], [c] ∈ Z5 .
(iii) Existence of Identity:
[1] is the identity element since [a] ×5 [1] = [a], for all [a] ∈ Z5 .
(iv) Commutative Property:
Clearly, from the table above, [a] ×5 [b] = [b] ×5 [a], for all
[a], [b] ∈ Z5 .
Hence, (Z5 , ×5 ) is a commutative monoid.
14. Let (M, ?, eM ) be a monoid and a ∈ M . If a is invertible, then show
that its inverse is unique.
Solution.
Let b and c be inverse elements of a ∈ M such that
a ? b = b ? a = e and a ? c = c ? a = e.
Now, b = b ? e = b ? (a ? c) = (b ? a) ? c = e ? c = c.
Therefore, its inverse is unique.
15. Show that the set N of natural numbers is a semigroup under the
operation x ? y = max{x, y}. Is it a monoid?
Solution.
Let N = {1, 2, 3, . . . }. Define the operation x ? y = max{x, y} for
x, y ∈ N.
Clearly, (N, ?) is closed because x ? y = max{x, y} ∈ N and ? is
associative since

(x ? y) ? z = max{x ? y, z}
= max{max{x, y}, z}
= max{x, y, z}
= max{x, max{y, z}} = max{x, y ? z} = x ? (y ? z).

Therefore, (N, ?) is a semigroup.


16. Prove that monoid homomorphism preserves invertibility and
monoid epimorphism preserves a zero element (if it exists).
182 Discrete Mathematical Structures

Solution.
Let (M, ?, eM ) and (T, ∆, eT ) be any two monoids, and let
g : M −→ T be a monoid homomorphism. If a ∈ M is invertible,
let a−1 be the inverse of a in M . We will now show that g(a−1 ) will
be an inverse of g(a) in T .
a ? a−1 = a−1 ? a = eM (by definition of inverse)
So, g(a ? a−1 ) = g(a−1 ? a) = g(eM ).
Hence, g(a)∆g(a−1 ) = g(a−1 )∆g(a) = g(eM ). (since g is a
homomorphism)
But g(eM ) = eT . (since g is a monoid homomorphism)
Therefore, g(a)∆g(a−1 ) = g(a−1 )∆g(a) = eT .
This means g(a−1 ) is an inverse of g(a). That is, g(a) is invertible.
Thus, the property of invertibility is preserved under monoid
homomorphism.
Assume g is a monoid epimorphism. Let t = g(b) ∈ T . Then
t∆g(z) = g(b)∆g(z) = g(b ? z) = g(z)
and g(z)∆t = g(z)∆g(b) = g(z ? b) = g(z).
Therefore, g(z) is the zero element of T .
17. The operation ? is defined by a ? b = a + b − ab, on the set Q
of all rational numbers. Show that under this operation, Q is a
commutative monoid.
Solution.
(i) Closure Property:
Since a + b − ab is a rational number for all rational numbers
a, b, the given operation ? is a binary operation on Q.
(ii) Associative Property:
For all a, b, c ∈ Q,

(a ? b) ? c = (a + b − ab) ? c
= (a + b − ab) + c − (a + b − ab)c
= a + b − ab + c − ac − bc + abc
= a + (b + c − bc) − a(b + c − bc)
= a ? (b + c − bc)
= a ? (b ? c).
Hence, ? is associative.
(iii) Existence of Identity:
For any a ∈ Q,
a?0=a+0−a·0=a
and 0 ? a = 0 + a − 0 · a = a.
Hence, 0 is the identity element in Q under the operation ?.
Algebraic Structures 183

(iv) Commutative Property:


From the definition of the operation ?, it is clear that ? is
commutative.
Hence, under the operation ?, Q is a commutative monoid with 0
as the identity element.
18. Let V = {a, b} and A be the set of all sequences on V including ∧
beginning with a. Show that (A, ◦, ∧) is a monoid.
Solution.
Let V = {a, b} and A be the set of all sequences on V including
∧ beginning with a. Then, A = {∧, a, ab, aa, ab, aba, abb, . . . }. Let
◦ be a concatenation operation on the sequences in A. Clearly for
any two elements α, β ∈ A, α ◦ β = αβ also belongs to A, and hence
(A, ◦) is closed.
Also ◦ is associative because

(α ◦ β) ◦ γ = αβγ
= α ◦ (βγ)
= (α ◦ β ◦ γ).

∧ is the identity since ∧ ◦ α = α ◦ ∧ = α, for all α ∈ A.


Therefore, (A, ◦, ∧) is a monoid.
19. Let V = {a, b}. Show that (V ? , ◦, ∧) is an infinite monoid.
Solution.
While defining the alphabet and set of strings V ? , we proved that
(V ? , ◦, ∧) is a monoid, where ∧ is an empty string. So, it is enough
to show that V ? is an infinite set. As a is an element of V , a, aa, aaa,
aaaa,. . . ; b, bb, bbb, bbbb,. . . ; and ab, abb, abbb, . . . are the elements
of V ? , and hence V ? contains infinitely many strings including an
empty set.

4.2.3 Groups
Definition 4.2.14 Group: A non-empty set G together with a binary
operation ?, that is (G, ?), is called a group if ? satisfies the following
conditions:

(i) Associative: For every a, b, c ∈ G, a ? (b ? c) = (a ? b) ? c.


(ii) Existence of Identity: There exists an element e ∈ G called
the identity element such that a ? e = e ? a = a, for all a ∈ G.
(iii) Existence of Inverse: There exists an element a−1 ∈ G called
the inverse of a such that a ? a−1 = a−1 ? a = e, for each a ∈ G.
184 Discrete Mathematical Structures

Example 4.2.15 The set of all integers Z with the addition operation is a
group.

Example 4.2.16 The set of all non-zero real numbers R? under the
multiplication operation is a group.

Definition 4.2.17 Abelian Group or Commutative Group: A group


(G, ?) is said to be an abelian group or commutative group if a ? b = b ? a, for
all a, b ∈ G.
Otherwise, it is a non-abelian group.
The set of all integers Z with the addition operation is an abelian group.

Properties of Groups
1. The identity of a group is unique.
2. The left and right cancellation laws are true.
(i) a ? b = a ? c =⇒ b = c (left cancellation law) and
(ii) b ? a = c ? a =⇒ b = c (right cancellation law).
3. The inverse of any element in a group is unique.
−1
4. If a is an element of a group G, then a−1 = a.
5. For any two elements a, b in a group G, (a ? b)−1 = b−1 ? a−1 .
6. In a group, the solution for the equations a ? x = b and y ? b = a
exists, and it is unique.

Theorem 4.2.18 Every row or column in the composition table of a group


(G, ?) is a permutation of the elements of G.
Proof.
Initially, we shall show that no row or column in the composition table can
have an element of G more than once.
Let us assume the contrary. Suppose that the row corresponding to an
element a ∈ G has two entries which are both k. That is, assume that
a ? b1 = a ? b2 = k, where k, b1 , b2 ∈ G and b1 6= b2 . Then by cancellation
law, we have b1 = b2 which is a contradiction. A similar result holds for any
column.
Next we will show that every element of G appears in each row and column
of the table of composition. Consider the row corresponding to the element
a ∈ G, and let b be an arbitrary element of G. Since b = a ? (a−1 ? b), “b” must
appear in the row corresponding to the element a ∈ G. The same argument
applies to every column of the table.
Thus, we obtain that no two rows or columns are identical. Hence, every
row of the composition table is obtained by a permutation of the elements G
and that each row is a distinct permutation. The same result applies to the
columns of the composition table.
Algebraic Structures 185

Theorem 4.2.19 In a group (G, ?), an element a ∈ G such that a2 = e,


a 6= e if and only if a = a−1 .
Proof.
Let us assume that a = a−1 .
Then, a2 = a ? a = a ? a−1 = e.
Conversely, assume that a2 = e with a 6= e.
That is, a?a=e
−1
a ? a ? a = a−1 ? e
e ? a = a−1
a = a−1 .

Theorem 4.2.20 In a group (G, ?), (a−1 )−1 = a, for all a ∈ G.


Proof.
Let a−1 be the inverse of a.
a ? a−1 = a−1 ? a = e
=⇒ a is the inverse of a−1 .
That is, (a−1 )−1 = a.

Definition 4.2.21 Permutation Group or Symmetric Group: The set


Pn of all permutations of n elements is a permutation group or a symmetric
group under the composition of functions.
That is, Pn = {f /f is a one-to-one and onto mapping from S to S} is a
group under the composition operation of functions, where S is any non-empty
set.

Example 4.2.22 The set P3 of all permutations on S = {1, 2, 3} is a finite


non-abelian group of order six with respect to composition of mappings.
The composition table for P3 is given in the table below where,

Composition Table of P3
◦ f1 f2 f3 f4 f5 f6
f1 f1 f2 f3 f4 f5 f6
f2 f2 f1 f6 f5 f4 f3
f3 f3 f5 f1 f6 f2 f4
f4 f4 f6 f5 f1 f3 f2
f5 f5 f3 f4 f2 f6 f1
f6 f6 f4 f2 f3 f1 f5

     
1 2 3 1 2 3 1 2 3
f1 = , f2 = , f3 =
1 2 3 3 2 1 2 3 1
     
1 2 3 1 2 3 1 2 3
f4 = , f5 = , f6 = .
3 1 2 2 1 3 1 3 2
186 Discrete Mathematical Structures

4.2.4 Solved Problems


n a 0 o
1. Show that G = : a 6= 0 ∈ R is an abelian group under
0 0
matrix multiplication.
Solution.
(i) Closure Property:
  
a 0 b 0
Let A = ,B = ∈ G.
0 0 0 0
ab 0
Then, AB = ∈ G since ab ∈ R, for all a, b ∈ R.
0 0
(ii) Commutative Property:
AB = BA is  true forall A, B ∈ G, since
ab 0
AB = BA = [∵ ab = ba is true in R].
0 0
(iii) Associative Property:
Matrix multiplication is associative always. That is,
A(BC) = (AB)C, for all A, B, C ∈ G.
(iv) Existence
 of Identity:
1 0
E= ∈ G is the identity in G, since
0 0
   
a 0 1 0 a 0
AE = = = A, for all A ∈ G.
0 0 0 0 0 0
(v) Existence
 ofInverse: 1 
a 0 −1 0
If A = ∈ G, then A = a ∈ G is the inverse of
0 0 0 0
 
1 0 1
A, since AA−1 = (∵ a 6= 0 ∈ R =⇒ 6= 0 ∈ R).
0 0 a
Hence, G is an abelian group under matrix multiplication.
n 
a a
o
2. Examine whether G = : a 6= 0 ∈ R is a commutative
a a
group under matrix multiplication, where R is the set of all real
numbers.
Solution.
(i) Closure Property:
  
a a b b
Let A = ,B = ∈ G. Then,
a a b b
 
2ab 2ab
AB = ∈ G since 2ab ∈ R, for all a, b ∈ R.
2ab 2ab
(ii) Commutative Property:
AB = BA is true for allA, B ∈ G, since
2ab 2ab
AB = BA = [∵ 2ab = 2ba is true in R].
2ab 2ab
Algebraic Structures 187

(iii) Associative Property:


Matrix multiplication is associative always. That is,
A(BC) = (AB)C, for all A, B, C ∈ G.
(iv) Existence of! Identity:
1 1
2 2
E= 1 1
∈ G is the identity in G, since
2 2
  1 1!  
a a 2 2 a a
AE = = = A, for all A ∈ G.
a a 1 1 a a
2 2
(v) Existence of Inverse: !
  1 1
a a 4a 4a
If A = ∈ G, then A−1 = ∈ G is the inverse
a a 1 1
! 4a 4a
1 1
2 2
of A, since AA−1 = 1 1
(∵ a 6= 0 ∈ R =⇒ 1
4a 6= 0 ∈ R).
2 2
Hence, G is a commutative group under matrix multiplication.
n 1 0 −1 0 1 0  −1 0  o
3. Prove that G = , , ,
0 1 0 1 0 −1 0 −1
forms an abelian group under matrix multiplication.
Solution.
       
1 0 −1 0 1 0 −1 0
Let A = ,B = ,C = ,D = .
0 1 0 1 0 −1 0 −1
The composition table is shown in the table below.

Composition Table of (G, ·)


· A B C D
A A B C D
B B A D C
C C D A B
D D C B A

(i) Closure Property:


Clearly, from the table above, we have xy ∈ G for all x, y ∈ G.
Hence, closure property is satisfied.
(ii) Commutative Property:
We observe from the table above that xy = yx, for all x, y ∈ G.
Hence, commutative property holds.
(iii) Associative Property:
Matrix multiplication is associative always. That is x(yz) = (xy)z,
for all x, y, z ∈ G.
(iv) Existence
 of Identity:
1 0
A= is the identity element in G since
0 1
188 Discrete Mathematical Structures

AA = A, AB = BA = B, AC = CA = C, and
AD = DA = D.
(v) Existence of Inverse:
From Table 4.2, all elements in G are self-inverses.
That is, inverse of A is A, inverse of B is B, inverse of C is C,
inverse of D is D, since AA = A, BB = A, CC = A,
DD = A.
Hence, G forms an abelian group under matrix multiplication.
4. Show that (Q+ , ?) is an abelian group, where ? is defined by
ab
a ? b = , ∀ a, b ∈ Q+ .
2
Solution.
(i) Closure Property:
ab
It is clear that for all a, b ∈ Q+ , a ? b ∈ Q+ , since ∈ Q+ .
2
Hence, closure property is satisfied.
(ii) Commutative Property:
a ? b = b ? a is true for all a, b ∈ Q+ , since
ab ab ba
a?b=b?a= [∵ = is true in Q+ ].
2 2 2
(iii) Associative Property:
a bc
 
bc abc
a ? (b ? c) = a ? = 2 = .
2 2 4
  ab
ab c abc
(a ? b) ? c = ?c= 2 = .
2 2 4
Therefore, a ? (b ? c) = (a ? b) ? c, for all a, b, c ∈ Q+ .
Hence, associative property is satisfied.
(iv) Existence of Identity:
e = 2 ∈ Q+ is the identity element, since
a·2
a?e=a?2= = a, for all a ∈ Q+ .
2
(v) Existence of Inverse:
4
a−1 = ∈ Q+ is the inverse of a ∈ Q+ , since
a
4 a · a4 4a
a ? a−1 = a ? = = = 2.
a 2 2a
Hence, Q+ is an abelian group under the operation ? defined in the
problem.
5. Prove that the identity element of a group is unique.
Solution.
Let (G, ?) be a group.
Let e1 and e2 be two identity elements in G.
Algebraic Structures 189

Then,

e1 ? e 2 = e1 [∵ e2 is the identity]
e1 ? e 2 = e2 [∵ e1 is the identity].

Thus, e1 = e2 .
Hence, the identity is unique.
6. Prove that the identity element is the only idempotent element of
a group.
Solution.
Let (G, ?) be a group.
Since e ? e = e, e is the idempotent element.
Let a be any idempotent element of G.
Then, a ? a = a.
Also, e ? a = a [∵ e is the identity element].
It follows that a ? a = e ? a.
By the right cancellation law, we have a = e, and so e is the only
idempotent element.
7. Prove that if every element in a group is its own inverse, then the
group must be abelian. Or prove that for any group (G, ?), if a2 = e
with a 6= e, then G is abelian.
Solution.
Given a = a−1 for all a ∈ G.
Let a, b ∈ G. Then, a = a−1 and b = b−1 .

Now, (a ? b) = (a ? b)−1
= b−1 ? a−1
= b ? a.
=⇒ G is abelian.

8. Prove for any element a in a group G, the inverse is unique.


Solution.
Let a be any element of a group G.
If possible, let a0 and a00 be two inverses of a. Then
a ? a 0 = a0 ? a = e
a ? a00 = a00 ? a = e.
Now, a = a0 = a0 ? (a ? a00 ) = (a0 ? a) ? a00 = e ? a00 = a00 .
0

Hence, the inverse is unique.


9. Prove that in a group (G, ?), (a ? b)−1 = b−1 ? a−1 .
Solution.

(a ? b)(b−1 ? a−1 ) = a ? (b ? b−1 ) ? a−1


190 Discrete Mathematical Structures

= a ? e ? a−1
= a ? a−1 = e
and
(b−1 ? a−1 ) ? (a ? b) = b−1 ? a−1 ? a ? b
= b−1 ? e ? b
= b−1 ? b = e.
Hence, (a ? b)−1 = b−1 ? a−1 .
10. If a and b are any two elements of a group (G, ?), then show that
G is abelian if and only if (a ? b)2 = a2 ? b2 .
Solution.
Necessary Part:
Given that (G, ?) is an abelian group.
=⇒ For all a, b ∈ G, a ? b = b ? a. (4.10)
2 2 2
To prove: (a ? b) = a ? b .
(a ? b)2 = (a ? b) ? (a ? b)
= a ? (b ? a) ? b
= a ? (a ? b) ? b [using (4.10)]
= (a ? a) ? (b ? b)
= a2 ? b2 .
Sufficient Part:

Given: (a ? b)2 = a2 ? b2 . (4.11)


To prove: a ? b = b ? a.
(4.11) =⇒ (a ? b)2 = a2 ? b2
=⇒ (a ? b) ? (a ? b) = (a ? b) ? (b ? b)
=⇒ a ? [b ? (a ? b)] = a ? [a ? (b ? b)]
=⇒ b ? (a ? b) = a ? (b ? b) (using left cancellation law)
=⇒ (b ? a) ? b = (a ? b) ? b (using associative property)
=⇒ b ? a = a ? b (using right cancellation law)
=⇒ G is abelian.

11. Show that every group of order four is abelian.


Solution.
Let (G, ?) be a group of order four where G = {e, a, b, c}. Since G
is of even order, there exists at least one element, say a such that
a−1 = a. Then, two cases arise:
Algebraic Structures 191

(i) b−1 = b, c−1 = c


(ii) b−1 = c, c−1 = b.
Case (i): e−1 = e, a−1 = a, b−1 = b, c−1 = c.
That is, every element has its own inverse.
Then, (G, ?) is abelian.
Case (ii): a−1 = a, b−1 = c, c−1 = b.
Therefore, a2 = e, b ? c = e, c ? b = e.
Since (G, ?) is a group, its elements will appear in a row (column)
only once.
Since a, e appear in the second row and b appears in the third
column, c will appear as (2, 3)th element.
∴ (2, 4)th element is b
(3, 3)th element is a
(3, 2)th element is c
(4, 2)th element is b
(4, 4)th element is a.
Composition Table of (G, ?)
? e a b c
e e a b c
a a e c b
b b c a e
c c b e a

12. Show that the set S = {[1], [5], [7], [11]} is a group with respect to
multiplication modulo 12.
Solution.
The composition table of S with respect to ×12 is given in the
table below: Here, 5 ×12 7 = 35, which on division by 12 gives the

Composition Table of (S, ×12 )


×12 [1] [5] [7] [11]
[1] [1] [5] [7] [11]
[5] [5] [1] [11] [7]
[7] [7] [11] [1] [5]
[11] [11] [7] [5] [1]

remainder 11, 11 ×12 7 = 77, which on division by 12 gives the


remainder 5, etc.
Hence, S is a group, in which [1] is the identity and each element
of S is its own inverse.
  
cos α − sin α
13. Show the set of matrices G = , a ∈ R forms a
sin α cos α
group under matrix multiplication.
192 Discrete Mathematical Structures

Solution.
(i) Closure Property:
   
cos α − sin α cos β − sin β
Let Aα = ∈ G and Aβ = ∈ G.
sin α cos α sin β cos β
Then
  
cos α − sin α cos β − sin β
Aα Aβ =
sin α cos α sin β cos β
 
cos α cos β − sin α sin β −(cos α sin β + sin α cos β)
=
sin α cos β + cos α sin β cos α cos β − sin α sin β
 
cos(α + β) − sin(α + β)
= = Aα+β ∈ G. (4.12)
sin(α + β) cos(α + β)

(ii) Associative Property:


We know that matrix multiplication is associative.
(iii) Existence
 of Identity:
1 0
I0 = is the identity in G, since Aα I0 = I0 Aα = Aα for
0 1
Aα ∈ G.
(iv) Existence of Inverse:
A−α is the inverse of Aα for each Aα ∈ G,
since Aα A−α = Aα+(−α) = A0 = I0 , using (4.12)
Hence, G forms a group under matrix multiplication.

4.2.5 Subgroups
Definition 4.2.23 Subgroup: A non-empty subset H of a group G is said
to be a subgroup of G, if H itself is a group under the same operation defined
on G and with the same identity element.

Example 4.2.24 The set of all integers Z is a subgroup of the set of all real
numbers R under usual addition. That is, (Z, +) is a subgroup of (R, +).

Theorem 4.2.25 The necessary and sufficient condition is that a non-empty


subset H of a group (G, ?) is a subgroup iff for any a, b ∈ H, a ? b−1 ∈ H.

Proof.
Necessary condition:
Assume that H is a subgroup of G.
Since H itself is a group, we have a, b ∈ H =⇒ a ? b ∈ H (using closure
property). Also, b ∈ H =⇒ b−1 ∈ H (using inverse property).
∴ a, b ∈ H =⇒ a, b−1 ∈ H =⇒ a ? b−1 ∈ H.
Sufficient condition:
Let a ? b−1 ∈ H, for all a, b ∈ H and H is a subset of G.
We have to prove H is a subgroup of G.
Algebraic Structures 193

(i) Existence of Identity:


Let a ∈ H =⇒ a ? a−1 ∈ H ⊆ G
=⇒ e ∈ H.
∴ e is the identity element of H.
(ii) Existence of Inverse:
Let e ∈ H, a ∈ H
=⇒ e ? a−1 ∈ H ⊆ G
=⇒ a−1 ∈ H.
∴ Every element of H has an inverse in H.
(iii) Closure Property:
Let b ∈ H =⇒ b−1 ∈ H.

∴ For a, b ∈ H =⇒ a, b−1 ∈ H
−1
=⇒ a ? b−1 ∈H⊆G
=⇒ a ? b ∈ H.

∴ H is closed under the operation ?.


(iv) Associative Property:
Given that H ⊆ G.
=⇒ The elements of H are also the elements of G. Since ? is
associative in G, it must also be associative in H.
Therefore, H itself is a group under the operation ? in G.

4.2.6 Cyclic Groups


Definition 4.2.26 Order of a group: Let (G, ?) be a group. The number
of elements in G is called the order of the group G and is denoted by O(G).

Note: If O(G) is finite, then G is called a finite group, otherwise it is called


an infinite group.
Definition 4.2.27 Cyclic Group: A group (G, ?) is said to be cyclic if there
exists an element a ∈ G such that any x ∈ G can be written as either x = an
or x = na, where n is some integer.
This element a is called the generator of the cyclic group G, that is, the
cyclic group generated by a, and we denote it by G = <a>.

Example 4.2.28 The multiplicative group, G = {1, −1, i, −i}, (i being the
complex number) is cyclic.
We can write 1 = i4 , −1 = i2 , i = i3 . That is all the elements of G can be
expressed as integral powers of the element i.
Therefore, G is a cyclic group generated by i. Since i is the generator of G,
i−1 is also a generator of G.
Hence, G is a cyclic group, and its generators are i and i−1 .
194 Discrete Mathematical Structures

Theorem 4.2.29 Every cyclic group is abelian.

Proof.
Let (G, ?) be a cyclic group generated by an element a ∈ G, that is G = <a>.
Then, for any two elements x, y ∈ G, we have x = an , y = am , where m, n
are integers. Therefore,

x ? y = an ? am
= am+n
= am ? an
= y ? x.

Thus, (G, ?) is abelian.


Theorem 4.2.30 Let (G, ?) be a finite group generated by an element a ∈ G.
If G is of order n, that is, O(G) = n, then an = e so that
G = {a, a2 , . . . , an = e}. Further, n is the least positive integer for which
an = e.
Proof.
Let us assume that, for some positive integer m < n, am = e.
Since G is cyclic, any element of G can be written as ak , for some k ∈ Z.
By division algorithm, we have k = mq + r, where q ∈ Z and 0 ≤ r ≤ m.
Therefore,

ak = amq+r
= amq ? ar
q
= (am ) ? ar
= eq ? a r
= e ? ar
= ar .

Hence, every element of G can be expressed as ar , for some 0 ≤ r ≤ m.


Therefore, G has at most m distinct elements. That is, O(G) = m < n,
which is a contradiction.
Hence, am = e, for m < n is not possible.
We now proceed to show that the elements a, a2 , a3 , . . . , an are all distinct
where an = e.
If possible, let ai = aj , for i < j ≤ n. Therefore,

ai ? a−j = aj ? a−j
=⇒ ai−j = aj−j = e, where i − j < n,

which is again a contradiction.


Hence, ai 6= aj , for i, j ≤ n.
Hence, the theorem is proved.
Algebraic Structures 195

Theorem 4.2.31 Every subgroup of a cyclic group is cyclic.

Proof.
Let G be a finite cyclic group of order n with generator a. That is,

g = {e, a, a2 , . . . , an−1 }.

Let H be a subgroup of G. Then, elements of H are of the form ak with


1 ≤ k < n. Let t be the smallest positive integer such that at ∈ H. We shall
prove that H = <at >. Indeed, let am ∈ H. By the division algorithm, there
exist unique integers q and r such that m = tq + r where 0 ≤ r < t. It follows
that am = (at )q ar or ar = am (at )−q . But am ∈ H and at ∈ H. Then by
closure, ar ∈ H. Since t is the smallest positive integer such that at ∈ H, we
must have r = 0. Hence, am = (at )q or q m ∈ <at >. Clearly, <at >⊆ H since
at ∈ H and H is a group.
Theorem 4.2.32 Every group of prime order is cyclic and hence is abelian.
Proof.
Let G be a group with O(G) = p, a prime.
Let a 6= e ∈ G and H = <a> be the cyclic group of G generated by a.
By Lagrange’s theorem, O(H)|p. So, O(H) = 1 or p.
Since O(H) 6= 1 (as a 6= e and a, e ∈ H, O(H) ≥ 2), we have O(H) = p.
So, G = H = <a> is a cyclic group.
Since every cyclic group is abelian, G is abelian.

4.2.7 Homomorphisms
Definition 4.2.33 Homomorphism: Let (G, ?) and (H, ∆) be any two
groups. A mapping f : G −→ H is said to be a homomorphism if
f (a ? b) = f (a)∆f (b), for a, b ∈ G.

Example 4.2.34 Let G = (Z, +) and H = (nZ, +) be two groups (for a fixed
integer n). The mapping f : G −→ H defined by f (m) = nm for m ∈ Z is a
homomorphism from G into H.

Definition 4.2.35 Kernel of a Homomorphism: Let f : G −→ G0 be a


group homomorphism. The set of elements of G which are mapped into e0 (the
identity of G0 ) is called the kernel of f and is denoted by ker(f ). That is,

ker(f ) = {x ∈ G/f (x) = e0 , where e0 is the identity of G0 }.

Theorem 4.2.36 The kernel of a homomorphism f from a group G into a


group G0 is a subgroup of G.

Proof.
Let f : (G, ?) −→ (G0 , ?0 ) be any homomorphism.

Then, ker(f ) = {x ∈ G/f (x) = e0 , where e0 is the identity of G0 }.


196 Discrete Mathematical Structures

Since f (e) = e0 is true always, at least e ∈ ker(f ). Therefore, ker(f ) is a


non-empty subset of G.
Let a, b ∈ ker(f ) with f (a) = e0 and f (b) = e0 . Therefore,

f (a ? b−1 ) = f (a) ?0 f (b−1 )(since f is a homomorphism)


= f (a) ?0 (f (b))−1
= e0 ?0 e0
= e0
=⇒ a ? b−1 ∈ ker(f ).

That is, a, b ∈ ker(f ) =⇒ a ? b−1 ∈ ker(f ). Hence, ker(f ) is a subgroup of G.


Definition 4.2.37 Endomorphism: A homomorphism f of a group into
itself is called an endomorphism.

Definition 4.2.38 Isomorphism: A mapping f from a group G to a group


G0 is said to be an isomorphism if f is a one-to-one and onto homomorphism.

Theorem 4.2.39 Cayley’s Representation Theorem: Every finite group


of order n is isomorphic to a permutation group of degree n.
Proof.
Let G be any finite group of order n. For each a ∈ G, define a function
fa : G −→ G such that fa (x) = ax, for every x ∈ G.
Clearly, this function fa is bijective (one-to-one and onto).
Consider G1 = {fa /a ∈ G}.
This G1 becomes a group under the composition operation of functions.
That is, (G1 , ◦) is the permutation group of order n.
Now, define a function Φ : G −→ G1 such that Φ(a) = fa , for all a ∈ G.
Claim 1: Φ is a homomorphism

Φ(b) = fab
= fa ◦ fb [since fab (x) = abx = a(bx) = fa (bx) = fa ◦ fb (x)]
= Φ(a) ◦ Φ(b).

Claim 2: Φ is bijective
Clearly, Φ is one-to-one, since

Φ(a) = Φ(b)
=⇒ fa = fb
=⇒ fa (x) = fb (x), for every x ∈ G
=⇒ ax = bx
=⇒ a = b.

Also, for every fa ∈ G, we have a ∈ G such that Φ(a) = fa .


Algebraic Structures 197

Therefore, Φ is onto.
Hence, Φ is bijective. Thus, Φ : G −→ G1 becomes as an isomorphism.
Hence, every finite group of order n is isomorphic to a permutation group
of degree n.
Theorem 4.2.40 Any cyclic group of order n is isomorphic to the additive
group of residue classes of integers modulo n.
Proof.
Let G = {a, a2 , . . . , an = e} be a cyclic group of order n generated by a.
We know that (Zn , +n ) is the additive group of residue classes modulo n
=⇒ Zn = {[1], [2], . . . , [n] = [0]}.
Let f : G −→ Zn be defined by f (ar ) = [r], for all ar ∈ G.
For all [r] ∈ Zn , there exists ar ∈ G such that f (ar ) = [r]
=⇒ f is onto.
For r 6= s, [r] 6= [s] and hence f (ar ) 6= f (as )
=⇒ f is one-to-one.
For all ar , as ∈ G, f (ar · as ) = f (ar+s ) = [r + s] = [r] + [s]
= f (ar ) +n f (as )
=⇒ f is a homomorphism.
Hence, (G, ·) is isomorphic to (Zn , +n ).

4.2.8 Cosets and Normal Subgroups


Definition 4.2.41 Left and Right Cosets: Let (H, ?) be a subgroup of a
group (G, ?).
(i) For any a ∈ G, the set a ? H defined by

a ? H = {a ? h/h ∈ H}

is called the left coset of H in G determined by the element a ∈ G.


(ii) For any a ∈ G, the set H ? a defined by

H? = {h ? a/h ∈ H}

is called the right coset of H in G determined by the element a ∈ G.

Example 4.2.42 Consider the multiplicative group G = {1, −1, i, −i} and a
subgroup H = {1, −1}. Clearly, iH, −iH, 1H, and −1H are the left cosets.

Definition 4.2.43 Index of a subgroup in a group: Let (H, ?) be a


subgroup of a group (G, ?). Then, the number of different left (or right) cosets
of H in G is called the index of H in G, and it is denoted by iG (H).

Some important results:


(i) If G is abelian, then a ? H = H ? a, for all a ∈ G.
198 Discrete Mathematical Structures

(ii) If H is a subgroup of G and e ∈ H, then e ? H = H ? e = H.


(iii) Any two left or right cosets of H in G are either disjoint or identical.
(iv) The union of all distinct left (or right) cosets of H in G is equal
to G.

Theorem 4.2.44 Let (H, ?) be a subgroup of a group (G, ?). The set of left
cosets of H in G forms a partition of G. Also, every element of G belongs to
one and only one left coset of H in G.

Proof.
To prove: Every element of G belongs to one and only one left coset of H
in G.
Let H be a subgroup of a group G. Let a ∈ G. Then, aH = H if and only
if a ∈ H.
Suppose a ∈ G and aH = H. Then,

aH = H =⇒ ae ∈ H =⇒ a ∈ H (since H is a subgroup and e ∈ H).

Conversely, assume that a ∈ H.


Then ah ∈ H, for all h ∈ H. So,

aH ⊆ H. (4.13)

Given any y ∈ H, a−1 y ∈ H, and y = a(a−1 y) ∈ H. So, y ∈ aH, for all y ∈ H.


That is,
H ⊆ aH. (4.14)
From (4.13) and (4.14), we have

H = aH.

Hence, every element of G belongs to one and only one left coset of H in G.
To prove: The set of left cosets of H in G forms a partition of G.
Let a, b ∈ G and H be a subgroup of G.
If aH ∩ Ha 6= φ, then let c ∈ aH ∩ Ha.
Since c ∈ aH, we have cH = aH.
Let H be a subgroup of a group G. Let a, b ∈ G if b ∈ aH; then bH = aH.
Since c ∈ bH, we have cH = bH. So

aH = cH = bH.

Thus, if aH ∩ bH 6= φ, then aH = bH.


Therefore, any two distinct left cosets are disjoint. Hence, the set of all
(distinct) left cosets of H in G forms a partition of G.
Theorem 4.2.45 Lagrange’s Theorem: The order of each subgroup of a
finite group is a divisor of the order of the group.
Algebraic Structures 199

Proof.
Let G be a finite group and H be a subgroup of G.
Let O(G) = n and O(H) = m. Let us consider all left cosets of H in G.
Each coset has exactly m elements.
ah1 = ah2 =⇒ h1 = h2 , for all a ∈ G.
By result (iii), namely, G is decomposed into say r mutually disjoint
subsets, each of order m.
Therefore, n = rm. That is, O(G) = rO(H).
Thus, O(H) divides O(G).
Note: The converse of Lagrange’s theorem is not true in general. That is, if
n is a divisor of a group G, then it does not necessarily follow that G has a
subgroup of order n.
Theorem 4.2.46 If (G, ?) is a finite group of order n, then for any a ∈ G,
we must have an = e, where e is the identity of the group G.
Proof.
Let O(G) = n. Let a ∈ G.
Then, the order of the subgroup <a> is the order of the element a.
If O(<a>)= m, then am = e, and by Lagrange’s theorem, we get m|n.
Let n = mk. Then, an = amk = (am )k = ek = e.
Definition 4.2.47 Normal Subgroup: A subgroup (H, ?) of a group (G, ?)
is said to be a normal subgroup of G if for every x ∈ G and for every h ∈ H,
xhx−1 ∈ H or xHx−1 ⊆ H.

Example 4.2.48 Consider the group (Z, +). Clearly, (3Z, +) is a normal
subgroup of (Z, +).

Definition 4.2.49 Quotient Group or Factor Group: Let N be a normal


subgroup of a group (G, ?) and the set of all right cosets of N in G be denoted by

G/N = {N a|a ∈ G}.

Now, define ⊗ as binary operation on G/N as

N a ⊗ N b = N (a ? b).

Then, (G/N, ⊗) will form a group called quotient group or factor group.

Theorem 4.2.50 The kernel of a homomorphism is a normal subgroup.


Proof.
Let f : (G, ?) −→ (G0 , ?0 ) be any homomorphism.
Then, ker(f ) = {x ∈ G/f (x) = e0 (where e0 is the identity element of G0 )}
is a subgroup of G by Theorem 4.2.36.
Let x ∈ ker(f ) and let g ∈ G.
=⇒ f (x) = e0 .
200 Discrete Mathematical Structures

Consider

f (g ? x ? g −1 ) = f (g) ?0 f (x ? g −1 )
= f (g) ?0 [f (x) ?0 f (g −1 )]
= f (g) ?0 [e0 ?0 f (g −1 )]
= f (g) ?0 f (g −1 )
= f (g ? g −1 )
= f (e) = e0 .

Thus, f (g ? x ? g −1 ) = e0 .
Therefore, g ? x ? g −1 ∈ ker(f ).
Hence, ker(f ) is a normal subgroup.
Theorem 4.2.51 Let (H, ?) be a subgroup of a group (G, ?). Then, (H, ?) is
a normal subgroup if and only if a ? h ? a−1 = H, for all a ∈ G.
Proof.
Let H be a normal subgroup of G.
Then by definition, a ? H = H ? a, for all a ∈ G. Hence,
a ? H ? a−1 = a ? (a−1 ? H)
= (a ? a−1 ) ? H
=e?H
= H.

Conversely, let a−1 ? H ? a = H, for all a ∈ G.


That is, a ? (a−1 ? H ? a) = a ? H
or (a ? a−1 ) ? (H ? a) = a ? H
or e ? (H ? a) = a ? H
or H ? a = a ? H.
Thus, H is a normal subgroup.
Theorem 4.2.52 Let (G, ?) be a group.
Let H = {a|a ∈ G & a ? b = b ? a, ∀ b ∈ G}. Then, H is a normal subgroup.
Proof.
H = {a|a ∈ G & a ? b = b ? a, ∀ b ∈ G}.
Since e ? a = a ? e = a, ∀a ∈ G, we have e ∈ H.
Therefore, H is non-empty.
Let x, y ∈ H. Then
a ? x = x ? a, ∀ x ∈ G, and a ? y = y ? a, ∀ y ∈ G.
Claim: H is a normal subgroup. Consider

a ? (x ? y) = (a ? x) ? y
= (x ? a) ? y
= x ? (a ? y)
Algebraic Structures 201

= x ? (y ? a)
= (x ? y) ? a
=⇒ x ? y ∈ H.

Let a ∈ H. Then, a ? x = x ? a, ∀x ∈ G. Then

a−1 ? (a ? x) = a−1 ? (x ? a)
=⇒ x = a−1 ? (x ? a)
=⇒ x ? a−1 = a−1 ? (x ? a) ? a−1
= (a−1 ? x) ? (a ? a−1 )
= a−1 ? x
=⇒ x ? a−1 = a−1 ? x, ∀x ∈ G
=⇒ a−1 ∈ H.

Thus, H is a subgroup.
To prove: H is normal.
Let x ∈ H, g ∈ G.
Then, a ? x = x ? a, ∀a ∈ G.
Then, g ? x ? g −1 = x ? g ? g −1
=⇒ x ∈ H.
Thus, g ? x ? g −1 ∈ H =⇒ H is normal.
Theorem 4.2.53 N is a normal subgroup of a group G if and only if
gN g −1 = N , for every g ∈ G (or gN = N g). Show that the number of
right and left cosets are equal in normal subgroups and every left coset is a
right coset.
Proof.
Let N be a normal subgroup of G.
Let x ∈ gN g −1 =⇒ x = gng −1 , for some n ∈ N .
Therefore, x = gng −1 ∈ N (∵ N is a normal subgroup).
Hence, gN g −1 ⊆ N .
Now, g −1 N g = g −1 N (g −1 )−1 ⊆ N , since g −1 ∈ G, and g −1 ng ∈ N .
Therefore, N = g(g −1 N g)g −1 ∈ gN g −1
Therefore, N ⊆ gN g −1 .
Hence, N = gN g −1 .
Conversely, let N g −1 = N , for every g ∈ G.
That is, gN g −1 is the set of all gng −1 , for n ∈ N .
Clearly, gN g −1 ⊆ N .
Therefore, N is a normal subgroup.
We get if N is a normal subgroup, then gN g −1 = N or gN = N g, that
is, the left and right cosets are equal.
Therefore, the right and left cosets are equal in number in normal
subgroups, and every left coset is a right coset.
202 Discrete Mathematical Structures

Fundamental Theorem of Group Homomorphism


Let H be a normal subgroup of a group G. Let G/H be the set of all left
cosets of H in G. That is, G/H = {aH/a ∈ G}. Let us define an operation
“·” as follows.
For any a, b ∈ G, (ab)H = (aH)(bH).
This is a binary operation under which G/H becomes a group. It is called
the quotient group or factor group.
Let f : G −→ G/H be defined as f (a) = aH, for any a ∈ G.
Then, for any a, b ∈ G,

f (ab) = (ab)H = (aH) · (bH) = f (a) · f (b).

Therefore, f is a homomorphism of G into G/H. It is called the natural


homomorphism or canonical homomorphism.
Theorem 4.2.54 Let g be a homomorphism of a group G into a group G0 .
Let K be the kernel of g and R be the image set of g in G0 . Then, G/K is
isomorphic to R.
Proof.
We have already shown that K is a normal subgroup of G. Therefore, there
exists a canonical homomorphism f : G −→ G/K given by f (a) = aK, for
any a ∈ G.
Now, let us define a mapping h : G/K −→ R such that h(aK) = g(a).
The image set of h is the same as the image set of g, and hence h is onto.
Further, for any a, b ∈ G such that aK = bK, we have ak1 = bk2 , for some
k1 , k2 ∈ K.
Therefore, g(ak1 ) = g(a)g(k1 ) = g(a)e0 = g(a)
and g(bk2 ) = g(b)g(k2 ) = g(b) = e0 = g(b)
so that aK = bK =⇒ g(a) = g(b).
Therefore, h(aK) = g(a) = g(b) = h(bK).
Also, f (a) = f (b).
Hence, h is one-to-one and onto.
Further, h(aKbK) = h(abK) == g(ab) = g(a)g(b) = h(aK)h(bK).
Hence, h is an isomorphism of G/K to R.

4.2.9 Solved Problems


1. Prove that the intersection of any two subgroups of a group (G, ?)
is again a subgroup of (G, ?).
Solution.
Let H and K be subgroups of G.
Let a ∈ H =⇒ a, b ∈ H and a, b ∈ K
=⇒ a ? b−1 ∈ H and a ? b−1 ∈ K (as H and K are subgroups)
=⇒ a ? b−1 ∈ H ∩ K.
Hence, H ∩ K is a subgroup of G.
Algebraic Structures 203

2. Is the union of two subgroups of a group a subgroup of G? Justify


your answer.
Solution.
The union of two subgroups of a group G need not be a subgroup
of G.
For example, we know (Z, +) is a group.
Let H = 3Z = {0, ±3, ±6, . . . }
Let K = 2Z = {0, ±2, ±4, . . . }
=⇒ H and K are subgroups of (Z, +)
=⇒ 3 ∈ 3Z ⊆ 3Z ∪ 2Z = H ∪ K
=⇒ 2 ∈ 2Z ⊆ 2Z ∪ 3Z = H ∪ K.
But 3 + 2 = 5 ∈
/ 2Z ∪ 3Z.
∴ H ∪ K is not a subgroup of (Z, +).
3. If H1 and H2 are subgroups of a group (G, ?), then prove that
H1 ∪H2 is a subgroup of G if and only if either H1 ⊆ H2 or H2 ⊆ H1 .
Solution.
Given H1 and H2 are two subgroups of (G, ?) and H1 ⊆ H2 or
H2 ⊆ H1 .
If H1 ⊆ H2 , then H1 ∪ H2 = H2 which is a subgroup of G.
If H2 ⊆ H1 , then H1 ∪ H2 = H1 which is a subgroup of G.
Conversely, suppose H1 6⊂ H2 and H2 6⊂ H1 .
Then, there exist a ∈ H1 and a ∈ / H2 , and there exist b ∈ H2 and
b∈
/ H1 .
Now a, b ∈ H1 ∪ H2 .
Since H1 ∪ H2 is a subgroup, it follows that a ? b ∈ H1 ∪ H2 .
Hence, a ? b ∈ H1 or a ? b ∈ H2 .
Case (i): If a ? b ∈ H1 , then a−1 ? (a ? b) ∈ H1 .
That is, b ∈ H1 , which is a contradiction.
Case(ii): If a ? b ∈ H2 . Then, (a ? b) ? b−1 ∈ H2 .
That is, a ∈ H2 , which is a contradiction.
Thus, either H1 ⊆ H2 or H2 ⊆ H1 .
4. Find all the subgroups of (Z9 , +9 ).
Solution.

Z9 = {[0], [1], [2], [3], [4], [5], [6], [7], [8]}.

The binary operation is addition modulo 9 (or +9 ).


Consider the subsets

H1 = {[0], [2], [4], [6], [8]}, H2 = {[0], [3], [6]},


H3 = {[0], [4], [8]}, H4 = {[0], [5]}.

The improper subgroups of (Z9 , +9 ) are ({[0]}, +9 ) and (Z9 , +9 ).


204 Discrete Mathematical Structures

+9 [0] [3] [6]


[0] [0] [3] [6]
H2 is closed since [3] [3] [6] [0]
[6] [6] [0] [3]

+9 [0] [2] [4] [6] [8]


[0] [0] [2] [4] [6] [8]
[2] [2] [4] [6] [8] [1]
H1 is closed since [4] [4] [6] [8] [1] [3]
[6] [6] [8] [1] [3] [5]
[8] [8] [1] [3] [5] [7]

+9 [0] [4] [8]


[0] [0] [4] [8]
H3 is closed since [4] [4] [8] [3]
[8] [8] [3] [7]

+9 [0] [5]
H4 is closed since [0] [0] [5]
[5] [5] [1]

The above composition tables show that H1 , H2 , H3 , and H4 are


closed under +9 . Therefore, the possible subgroups of (Z9 , +9 ) are
(H1 , +9 ), (H2 , +9 ), (H3 , +9 ), and (H4 , +9 ).
5. Find the left cosets of {[0], [3]} in the addition modular group
(Z6 , +6 ).
Solution.
Let Z6 = {[0], [1], [2], [3], [4], [5]} be a group and H = {[0], [3]} be a
subgroup of Z6 under +6 (addition mod 6).
The left cosets of H are
[0] +6 H = {[0], [3]} = H
[1] +6 H = {[1], [4]}
[2] +6 H = {[2], [5]}
[3] +6 H = {[3], [6]} = {[3], [0]} = {[0], [3]} = H
[4] +6 H = {[4], [7]} = {[4], [1]} = [1] +6 H
[5] +6 H = {[5], [8]} = {[5], [2]} = [2] +6 H.
∴ [0] +6 H = [3] +6 H = H
and [1] +6 H = [4] +6 H, [2] +6 H = [5] +6 H
are the distinct left cosets of H in Z6 .
Algebraic Structures 205

6. Find the left cosets of {[0], [2]} in the group (Z4 , +4 ).


Solution.
Let Z4 = {[0], [1], [2], [3]} be a group and H = {[0], [2]} be a
subgroup of Z4 under +4 (addition mod 4).
The left cosets of H are
[0] + H = {[0], [2]} = H
[1] + H = {[1], [3]}
[2] + H = {[2], [4]} = {[2], [0]} = {[0], [2]} = H
[3] + H = {[3], [5]} = {[3], [1]} = {[1], [3]} = [1] + H.
Therefore, [0] + H = [2] + H = H and [1] + H = [3] + H are the
two distinct left cosets of H in Z4 .
7. Let G = {1, a, a2 , a3 } (a4 = 1) be a group and H = {1, a2 } be a
subgroup of G under multiplication. Find all the cosets of H.
Solution.
The right cosets of H in G are
H1 = {a, a2 } = H
Ha = {a, a3 }
Ha2 = {a2 , a4 } = {a2 , a} = H
and Ha3 = {a3 , a5 } = {a3 , a} = Ha.
Therefore, H1 = H = Ha2 = {1, a2 } and Ha = Ha3 = {a, a3 }
are two distinct right cosets of H in G. Similarly, we can find the
left cosets of H in G.
8. Prove that any two infinite cyclic groups are isomorphic to each
other.
Solution.
Let G1 = <a> and G2 = <b> be two cyclic groups of infinite order.
G1 = {an |n is an integer} and G2 = {bn |n is an integer}.
Define a map f : G1 −→ G2 by f (an ) = bn .
Let x, y ∈ G1 . Then, x = an , y = am for some integers n and m.
Now,
f (x)f (y) = f (an )f (am ) = bn bm = bn+m = f (an+m )
= f (an am ) = f (xy).
Hence, f is a homomorphism.
If f (x) = f (y), then f (an ) = f (am ) =⇒ bn = bm .
Then, bn−m = e0 in G2 . As G2 is an infinite cyclic group generated
by b, there is no non-zero integer k such that bk = e0 . Hence,
from bn−m = e0 , we have n − m = 0, or n = m, and hence
x = an = am = y. Thus, f is one-to-one.
206 Discrete Mathematical Structures

Let z ∈ G2 . Then, z = bn for some integer n. Now, take x = an .


Then, f (x) = f (an ) = bn = z. So, the map f is onto.
Now, f is one-to-one and onto homomorphism. Hence, it is an
isomorphism.
9. Prove that the group homomorphism preserves the identity element.
Solution.
Let (G, ?) and (H, ◦) be two groups. Letf : G −→ H be a group
homomorphism.
Let eG be the identity element of G.
Let eH be the identity element of H.
To prove: f (eG ) = eH .
Consider

f (eG ) = f (eG ? eG ) (since eG is the identity in G)


= f (eG ) ◦ f (eG ) (since f is a homomorphism)
Multiplying bothsides by f (eG )−1 on the right side, we get
f (eG ) ◦ f (eG )−1 = f (eG ) ◦ f (eG ) ◦ f (eG )−1
eH = f (eG ).

Hence, the group homomorphism preserves the identity element.


10. Prove that the group homomorphism preserves the inverse element.
Solution.
Let (G, ?) and (H, ◦) be two groups. Let f : G −→ H be a group
homomorphism.
Let eG be the identity element of G.
Let eH be the identity element of H.
To Prove: [f (x)]−1 = f (x−1 ), for all x ∈ G.
It is sufficient to prove that f (x) ◦ f (x−1 ) = eH .
Now, for all x ∈ G, we can write
f (x) ◦ f (x−1 ) = f (x ? x−1 ) (since f is a homomorphism)
=⇒ f (x) ◦ f (x−1 ) = f (eG ) = eH .
Hence, the group homomorphism preserves the inverse
element.
11. If f : G −→ G0 is a group homomorphism from (G, ?) to (G0 , ∆),
then prove that for any a ∈ G, f (a−1 ) = [f (a)]−1 .
Solution.
For all a, a−1 ∈ G, we have

f (a ? a−1 ) = f (a)∆f (a−1 )


or f (e) = f (a)∆f (a−1 )
or e0 = f (a)∆f (a−1 ). (4.15)
Algebraic Structures 207

Similarly,

f (a−1 ∆a) = f (a−1 )∆f (a)


or f (e) = f (a−1 )∆f (a)
or e0 = f (a−1 )∆f (a). (4.16)

From (4.15) and (4.16), we get


f (a)∆f (a−1 ) = f (a−1 )∆f (a)
=⇒ f (a−1 ) is the inverse of f (a).
That is, f (a−1 ) = [f (a)]−1 .
12. Let G be a group and a ∈ G. Let f : G −→ G be given by f (x) =
axa−1 , for all x ∈ G. Prove that f is an isomorphism of G onto G.
Solution.
To show f is a homomorphism:
If x, y ∈ G, then

f (x)f (y) = (axa−1 )(aya−1 )


= ax(−1 a)ya−1
= axya−1
= a(xy)a−1
= f (xy).

Therefore, f is a homomorphism.
To show f is one-to-one:
If f (x) = f (y), then axa−1 = aya−1 . Hence, by left cancellation
law, we have xa−1 = ya−1 ; again by right cancellation law, we get
x = y. Therefore, f (x) = f (y) =⇒ x = y. Hence, f is one-to-one.
To show f is onto:
Let y ∈ G; then a−1 ya ∈ G and

f (a−1 ya) = a(a−1 ya)a−1


= (aa−1 )y(aa−1 )
= y.

Therefore, f (x) = y, for some x ∈ G.


Thus, f is an isomorphism.
13. Prove that the intersection of two normal subgroups is a normal
subgroup.
Solution.
Let H and K be any two normal subgroups of a group G. We have
to prove that H ∩ K is a normal subgroup of G.
208 Discrete Mathematical Structures

Since H and K are subgroups of G, e ∈ H and e ∈ H. Hence,


e ∈ H ∩ K. Thus, H ∩ K is a non-empty set.
Let a, b ∈ H ∩ K.
Claim: ab−1 ∈ H ∩ K.
Since a, b ∈ H ∩ K, both a and b belong to H and K.
Since H and K are subgroups of G, ab−1 ∈ H and ab−1 ∈ K, so
that ab−1 ∈ H ∩ K.
Hence, H ∩ K is a subgroup of G, by a criterion for subgroup.
To prove: H ∩ K is normal.
Let x ∈ H ∩ K and g ∈ H.
Since x ∈ H ∩ K, x ∈ H and x ∈ K.
Since x ∈ H, g ∈ G =⇒ gxg −1 ∈ K (as H is normal).
Similarly, x ∈ K, g ∈ G =⇒ gxg −1 ∈ K (as K is normal).
Hence, x ∈ H ∩ K and g ∈ G =⇒ gxg −1 ∈ H ∩ K.
Thus, H ∩ K is a normal subgroup of G.
14. Prove that every subgroup of an abelian group is a normal subgroup.
Solution.
Let (G, ?) be an abelian group and (N, ?) be a subgroup of G.
Let g be any element in G, and let n ∈ N .
Now,
g ? n ? g −1 = (n ? g) ? g −1 [∵ G is abelian]
−1
= n ? (g ? g )
=n?e
= n ∈ N.

Therefore, for all g ∈ G and n ∈ N , g ? n ? g −1 ∈ N .


Hence, (N, ?) is a normal subgroup.

4.2.10 Permutation Functions


Definition 4.2.55 A bijection from a set A to itself is called a permutation
of A.

Example 4.2.56 Let A = R, and let f : A −→ A be defined by f (a) = 2a+1.


Since f is one-to-one and onto, it follows that f is a permutation of A.

Example 4.2.57 Let A = {1, 2, 3}. Then, all the permutations of A are
     
1 2 3 1 2 3 1 2 3
1A = , p1 = , p2 = ,
1 2 3 1 3 2 2 1 3
     
1 2 3 1 2 3 1 2 3
p3 = , p4 = , p5 = .
2 3 1 3 1 2 3 2 1
Algebraic Structures 209

Remark 4.2.58 In the above example, we can write the permutations as


ordered pairs. For example,
p4 = {(1, 3), (2, 1), (3, 2)} and p−1 4 = {(3, 1), (1, 2), (2, 3)}.
Or, if the first component of each ordered pair is written in increasing
order, then we have
p−1
4 = {(1,  2), (2, 3),
 (3, 1)}.
−1 1 2 3
Thus, p4 = = p3 .
2 3 1
Remark 4.2.59 The function p2 takes 1 to 2, and p3 takes 2 to 3, so p3 ◦ p2
takes 1 to 3. Also, p2 takes 2 to 1, and p3 takes 1 to 2, so p3 ◦ p2 takes 2 to
2. Finally, p2 takes 3 to 3, and p3 takes 3 to 1, so p3 ◦ p2 takes 3 to 1. Thus,
 
1 2 3
p3 ◦ p2 = .
3 2 1
Remark 4.2.60 The process of forming p3 ◦p2 is shown below. It can be noted
that p3 ◦ p2 = p5 .
   
←− 1 2 3  
 ↑ ◦ ↓
1 2 3   = 1 2 3 = p5
 
p3 ◦ p2 =  ↓  2
 1 3 3 2 1
2 3 1 ←−

Theorem 4.2.61 If A = {a1 , a2 , . . . , an } is a set containing n elements, then


there are n! = n · (n − 1) · · · 2 · 1 permutations of A.

Definition 4.2.62 Cyclic Permutation: Let b1 , b2 , . . . , br be r distinct


elements of the set A = {a1 , a2 , . . . , an }. The permutation p : A −→ A is
defined by

p(b1 ) = b2
p(b − 2) = b3
.. ..
. .
p(br−1 ) = br
p(br ) = b1 .

p(x) = x if x ∈ A, x ∈/ {b1 , b2 , . . . , br } is called a cyclic permutation of length


r, or simply a cycle of length of r, and will be denoted by (b1 , b2 , . . . , br ).

Example 4.2.63 Let A = {1, 2, 3, 4, 5}. The cycle (1, 3, 5) denotes the
permutation  
1 2 3 4 5
.
3 2 5 4 1
Definition 4.2.64 Disjoint Cycles: Two cycles of a set A are said to be
disjoint if no element of A appears in both cycles.
210 Discrete Mathematical Structures

Example 4.2.65 Let A = {1, 2, 3, 4, 5, 6}. Then, the cycles (1, 2, 5) and
(3, 4, 6) are disjoint, whereas the cycles (1, 2, 5) and (2, 4, 6) are not.

Theorem 4.2.66 A permutation of a finite set that is not the identity or a


cycle can be written as a product of disjoint cycles of length ≥ 2.

Definition 4.2.67 Transposition: A cycle of length 2 is called a


transposition. That is, a transposition is a cycle p = (ai , aj ), where p(ai ) = aj
and p(aj ) = ai .

Remark 4.2.68 Note that if p = (ai , aj ) is a transposition of A, then


p ◦ p = 1A , the identity permutation of A.
Every cycle can be written as a product of transpositions. In fact,
(b1 , b2 , . . . , br ) = (b1 , br ) ◦ (b1 , br−1 ) ◦ · · · ◦ (b1 , b3 ) ◦ (b1 , b2 ).
This case can be verified by induction on r as follows.
Basis Step:
If r = 2, then the cycle is just (b1 , b2 ), which already has the proper form.
Induction Step:
We use P (m) to find P (m + 1). Let (b1 , b2 , . . . , bm , bm+1 ) be a cycle of length
m + 1. Then, (b − 1, b − 2, . . . , bm , bm+1 ) = (b1 , bm+1 ) ◦ (b1 , b2 , . . . , bm ) as may
be verified by computing the composition.
Using P (m), (b1 , b2 , . . . , bm ) = (b1 , bk ) ◦ (b1 , bm−1 ) ◦ · · · ◦ (b1 , b2 ).
Thus, by substitution,
(b1 , b2 , . . . , bm+1 ) = (b1 , bm+1 ) ◦ (b1 , bm ) ◦ · · · ◦ (b1 , b3 ) ◦ (b1 , b2 ).
This completes the induction step. Thus, by the principle of mathematical
induction, the result holds for every cycle. For example,
(1, 2, 3, 4, 5) = (1, 5) ◦ (1, 4) ◦ (1, 3) ◦ (1, 2).

Corollary 4.2.69 Every permutation of a finite set with at least two elements
can be written as a product of transpositions.

Theorem 4.2.70 If a permutation of a finite set can be written as a product


of an even number of transpositions, then it can never be written as a product
of an odd number of transpositions, and the converse is also true.

Definition 4.2.71 Even and Odd Permutations: A permutation of a


finite set is called even if it can be written as a product of an even number of
transpositions, and it is called odd if it can be written as a product of an odd
number of transpositions.

Remark 4.2.72 From the definition of even and odd permutations, we have
the following:
(a) The product of two even permutations is even.
(b) The product of two odd permutations is even.
(c) The product of an even and an odd permutation is odd.
Algebraic Structures 211

Theorem 4.2.73 Let A = {a1 , a2 , . . . , an } be a finite set with n elements,


n! n!
n ≥ 2. Then, there are even permutations and odd permutations.
2 2
Proof.
Let An be the set of all even permutations of A, and let Bn be the set of all
odd permutations. We shall define a function f : An −→ Bn , which we shall
show is one-to-one and onto, and this will show that An and Bn have the same
number of elements.
Since n ≥ 2, we can choose a particular transposition q0 of A, say
q0 = (an−1 , an). We now define the function f : An −→ Bn by

f (p) = q0 ◦ p, p ∈ An .

Observe that if p ∈ An , then p is an even permutation, so q0 ◦ p is an odd


permutation, and thus f (p) ∈ Bn .
Suppose now that p1 and p2 are in An and f (p1 ) = f (p2 ). Then

q0 ◦ p1 = q0 ◦ p2 . (4.17)

We now compose each side of equation (4.17) with q0 :

q0 ◦ (q0 ◦ p1 ) = q0 ◦ (q0 ◦ p2 );

so by the associative property,

(q0 ◦ q0 ) ◦ p1 = (q0 ◦ q0 ) ◦ p2 , or since q0 ◦ q0 = 1A ,


1A ◦ p1 = 1A ◦ p2
p1 = p2 .

Thus, f is one-to-one.
Now, let q ∈ Bn . Then, q0 ◦ q ∈ An , and
f (q0 ◦ q) = q0 ◦ (q0 ◦ q) = (q0 ◦ q0 ) = 1A ◦ q = q,
which means that f is an onto function. Since f : An −→ Bn is one-to-one
and onto, we conclude that An and Bn have the same number of elements.
Note that An ∩ Bn = φ, since no permutation can be both even and odd. Also,
by theorem, |An ∪ Bn | = n!
n! = |An ∪ Bn | = |An | + |Bn | − |An ∩ Bn | = 2|An |.
n!
Hence, we have |An | = |Bn | = .
2

4.2.11 Solved Problems


1. Let A = {1, 2, 3, 4, 5, 6}. Compute (4, 1, 3, 5) ◦ (5, 6, 3) and
(5, 6, 3) ◦ (4, 1, 3, 5).
212 Discrete Mathematical Structures

Solution.
We have
   
1 2 3 4 5 6 1 2 3 4 5 6
(4, 1, 3, 5) = and (5, 6, 3) = .
3 2 5 1 4 6 1 2 5 4 6 3
   
1 2 3 4 5 6 1 2 3 4 5 6
Then, (4, 1, 3, 5)◦(5, 6, 3) = ◦
3 2 5 1 4 6 1 2 5 4 6 3
 
1 2 3 4 5 6
=
3 2 4 1 6 5
   
1 2 3 4 5 6 1 2 3 4 5 6
and (5, 6, 3)◦(4, 1, 3, 5) = ◦
1 2 5 4 6 3 3 2 5 1 4 6
 
1 2 3 4 5 6
= .
5 2 6 1 4 3
Observe that
(4, 1, 3, 5) ◦ (5, 6, 3) 6= (5, 6, 3) ◦ (4, 1, 3, 5)
and that neither product is a cycle.
 = {1, 2, 3, 4, 5, 6, 7, 8} be 
2. Let A a set. Then, write the permutation
1 2 3 4 5 6 7 8
p= as a product of disjoint cycles.
3 4 6 5 2 1 8 7
Solution.
We start with 1 and find that p(1) = 3, p(3) = 6, and p(6) = 1, so
we have the cycle (1, 3, 6). Next, we choose the first element of A
that has not appeared in a previous cycle. We choose 2, and we have
P (2) = 4, p(4) = 5, and p(5) = 2, so we obtain the cycle (2, 4, 5).
We now choose 7, the first element of A that has not appeared in
a previous cycle. Since p(7) = 8 and p(8) = 7, we obtain the cycle
(7, 8). We can then write p as a product of disjoint cycles as
p = (7, 8) ◦ (2, 4, 5) ◦ (1, 3, 6).
 
1 2 3 4 5 6 7
3. Is the permutation p = even or odd?
2 4 5 7 6 3 1
Solution.
We first write p as a product of disjoint cycles, obtaining

p = (3, 5, 6) ◦ (1, 2, 4, 7).


Next, we write each of the cycles as a product of transpositions:
(1, 2, 4, 7) = (1, 7) ◦ (1, 4) ◦ (1, 2)
(3, 5, 6) = (3, 6) ◦ (3, 5).
Algebraic Structures 213

Then, p = (3, 6) ◦ (3, 5) ◦ (1, 7) ◦ (1, 4) ◦ (1, 2). Since p is a product


of an odd number of transpositions, it is an odd permutation.
 
1 2 3 4 5 6
4. Show that the permutation is odd, while the
 5 6 2 4 1 3
1 2 3 4 5 6
permutation is even.
6 3 4 5 2 1
Solution.
 
1 2 3 4 5 6
= (1 5)(2 6 3) = (1 5)(2 6)(2 3).
5 6 2 4 1 3
The given permutation can be expressed as the product of an odd
number of transpositions, and hence the permutation is odd. Again
 
1 2 3 4 5 6
= (1 6)(2 3 4 5) = (1 6)(2 3)(2 4)(2 5).
6 3 4 5 2 1
Since it is a product of even number of transpositions, the
permutation is an even permutation.
 
1 2 3 4 5 6
5. Express the permutation as a product of
6 5 2 4 3 1
transpositions.
Solution.
 
1 2 3 4 56
= (1 6)(2 5 3) = (1 6)(2 5)(2 3).
6 5 2 4 31
 
1 2 3 4 5
6. Find the inverse of the permutation .
2 3 1 5 4
Solution.  
1 2 3 4 5
Given .
2 3 1 5 4

5 1 2 3 4
Let the inverse permutation be .
v x y z u
    
1 2 3 4 5 1 2 3 4 5 1 2 3 4 5
Then, =
2 3 1 5 4 x y z u v 1 2 3 4 5
   
1 2 3 4 5 1 2 3 4 5
=⇒ =
y z x v u 1 2 3 4 5
=⇒ y = 1, z = 2, x = 3, v = 4, u = 5.
 
1 2 3 4 5
Hence, the inverse permutation is .
3 1 2 5 4
7. If A = (1 2 3 4 5), B = (2 3)(4 5), find AB.
214 Discrete Mathematical Structures

Solution.
Given A = (1 2 3 4 5), B = (2 3)(4 5).
  
1 2 3 4 5 1 2 3 4 5
AB =
2 3 4 5 1 1 3 2 5 4
 
1 2 3 4 5
=
3 2 5 4 1

= (1 3 5).
8. If A = {1, 2, 3, 4, 5, 6, 7, 8}, then express the following permutations
as a product of disjoint cycles.
 
1 2 3 4 5 6 7 8
(i) p =
6 5 7 8 4 3 2 1
 
1 2 3 4 5 6 7 8
(ii) p = .
2 3 1 4 6 7 8 5
Solution.

(i) p(1) = 6, p(6) = 3, p(3) = 7, p(7) = 2, p(2) = 5, p(5) = 4,


p(4) = 8, p(8) = 1.
Therefore, p = (1 6 3 7 2 5 4 8).
(ii) p(1) = 2, p(2) = 3, p(3) = 1 =⇒ (1 2 3)
p(5) = 6, p(6) = 7, p(7) = 8, p(8) = 5 =⇒ (5 6 7 8).
Therefore, p = (5 6 7 8)(1 2 3).
 
1 2 3 4 5 6
9. Let A = {1, 2, 3, 4, 5, 6} and p = be a
2 4 3 1 5 6
permutation of A.
(i) Write p as a product of disjoint cycles.
(ii) Compute p−1 .
(iii) Compute p2 .
(iv) Find the period of p, that is, the smallest positive integer k
such that pk = 1A .
Solution.
 
1 2 3 4 5 6
(i) Given p = .
2 4 3 1 5 6
Since p(1) = 2, p(2) = 4, and p(4) = 1, we write p = (1, 2, 4) as
the other
 elements are fixed.
  
−1 2 4 3 1 5 6 1 2 3 4 5 6
(ii) p = =p= .
1 2 3 4 5 6 4 1 3 2 5 6
 
1 2 3 4 5 6
(iii) p2 = p ◦ p = .
4 1 3 2 5 6
Algebraic Structures 215
 
1 2 3 4 5 6
(iv) p3 = p2 ◦ p = = 1A .
1 2 3 4 5 6
p4 = p, p5 = p2 , etc.
Therefore, the period of p = 3.
   
1 2 3 4 1 2 3 4
10. If f = and g = are permutations,
3 2 1 4 2 3 4 1
prove that (g ◦ f )−1 = f −1 ◦ g −1 .
Solution.
   
−1 3 2 1 4 −1 1 2 3 4
f = and g = .
1 2 3 4 4 1 2 3
 
−1 −1 1 2 3 4
f ◦g =
4 3 2 1
     
1 2 3 4 1 2 3 4 1 2 3 4
g◦f = ◦ =
2 3 4 1 3 2 1 4 4 3 2 1
 
−1 1 2 3 4
(g ◦ f ) = .
4 3 2 1

Hence, (g ◦ f )−1 = f −1 ◦ g −1 .
   
1 2 3 4 5 6 7 1 2 3 4 5 6 7
11. Let p1 = and p2 = .
7 3 2 1 4 5 6 6 3 2 1 5 4 7
(i) Compute p1 ◦ p2 .
(ii) Compute p−11 .
(iii) Is p1 an even or odd permutation? Explain.
Solution.
   
1 2 3 4 5 6 7 1 2 3 4 5 6 7
(i) p1 ◦ p2 = ◦
7 3 2 1 4 5 6 6 3 2 1 5 4 7
 
1 2 3 4 5 6 7
= .
5 2 3 7 4 1 6
 
1 2 3 4 5 6 7
(ii) p−1
1 = 4 3 2 5 6 7 1 .

(iii) p1 = (1 7 6 5 4) ◦ (2 3)
= (1 4) ◦ (1 5) ◦ (1 6) ◦ (1 7) ◦ (2 3)
= product of odd number of transpositions.
Therefore, p1 is an odd permutation.
12. If x = (1 2 3), y = (2 4 3), and z = (1 3 4), then show that
xyz = 1.
Solution.  
1 2 3 4
Given x = (1 2 3) =
2 3 1 4
216 Discrete Mathematical Structures
 
1 2 3 4
y = (2 4 3) =
1 4 2 3
1 2 3 4
z = (1 3 4) = .
3 2 4 1
     
1 2 3 4 1 2 3 4 1 2 3 4
Therefore, xyz = ◦ ◦
2 3 1 4 1 4 2 3 3 2 4 1
   
1 2 3 4 1 2 3 4
= ◦
4 2 1 3 3 2 1 4
 
1 2 3 4
= = 1.
1 2 3 4

4.2.12 Problems for Practice


1. Which of the following functions f : Z −→ Z are permutations of Z?
(i) f is defined by f (a) = a + 1.
(ii) f is defined by f (a) = (a − 1)2 .
2. Which of the following functions f : R −→ R are permutations
of R?
(i) f is defined by f (a) = a3 .
(ii) f is defined by f (a) = ea .
3. Which of the following functions f : R −→ R are permutations
of R?
(i) f is defined by f (a) = a − 1.
(ii) f is defined by f (a) = a2 .
4. Which of the following functions f : Z −→ Z are permutations of Z?
(i) f is defined by f (a) = a2 + 1.
(ii) f is defined by f (a) = a3 − 3.
5. Let A = {a, b, c, d, e, f, g}. Write each of the following permutations
as a product of disjoint cycles.
 
a b c d e f g
(i)
g d b a c f e
 
a b c d e f g
(ii)
d e a b g f c
6. Let A = {1, 2, 3, 4, 5, 6, 7, 8}. Write each of the following
permutations as a product of transpositions.
(i) (2 1 4 5 8 6)
(ii) (3 1 6) ◦ (4 8 2 5)
7. Code the message “WHERE ARE YOU” by applying the
permutation
(1 7 3 5 11) ◦ (2 6 9) ◦ (4 8 10).
Algebraic Structures 217

8. Decode the message “ATEHAOMOMNTI”, which was encoded


using the permutation
(3 7 1 12) ◦ (2 5 8) ◦ (4 10 6 11 9).
9. Let A = {1, 2, 3, 4, 5, 6, 7, 8}. Determine whether the following
permutations are even or odd.
 
1 2 3 4 5 6 7 8
(i)
4 2 1 6 5 8 7 3
 
1 2 3 4 5 6 7 8
(ii)
7 3 4 2 1 8 6 5
(iii) (6 4 2 1 5)
(iv) (4 8) ◦ (3 5 2 1) ◦ (2 4 7 1)
10. Prove that the product of two even permutations is even.
11. Prove that the product of two odd permutations is even.
12. Prove that the product of an even and odd permutation is odd.
13. Let A = {1, 2, 3, 4, 5}. Let f = (5 2 3) and g = (3 4 1) be
permutations of A. Compute each of the following, and write the
result as the product of disjoint cycles:
(i) f ◦ g
(ii) f −1 ◦ g −1 .

4.2.13 Rings and Fields


Definition 4.2.74 Ring: An algebraic system (S, +·) is called a ring if the
binary operations + and · on S satisfy the following three properties:
1. (S, +) is an abelian group.
2. (S, ·) is a semigroup.
3. The operation · is distributive over +; that is, for any a, b, c ∈ S,
a · (b + c) = a · b + a · c and (b + c) · a = b · a + c · a.

Examples:

1. The set of all integers Z, the set of all rational numbers Q, the set
of all real numbers R are rings under the usual addition and usual
multiplication.
2. The set of all n × n matrices Mn is a ring under the matrix addition
and matrix multiplication.
3. If n is a positive integer, then Zn = {[0], [1], . . . , [n − 1]} is a
ring under +n , the addition modulo n, and ×n , the multiplication
modulo n.
218 Discrete Mathematical Structures

4. Let (R, +, ·) be a ring and X be a non-empty set. Let


A be the set of all functions from X to R. That is,
A = {f |f : X −→ R is a function}. We define ⊕ and · on A as
follows:
(i) if f, g ∈ A, then f ⊕ g : X −→ R is given by

(f ⊕ g)(x) = f (x) + g(x), for all x ∈ X.

(ii) if f, g ∈ A, then f, g : X −→ R is given by

(f · g)(x) = f (x) · g(x), for all x ∈ X.

Definition 4.2.75 Integral Domain: A commutative ring (S, +, •) with


identity and without divisors of zero is called an integral domain.
Definition 4.2.76 Field: A commutative ring (S, +, •) which has more than
one element such that every non-zero element of S has a multiplicative inverse
in S is called a field.
Definition 4.2.77 Subring: A subset R of a ring (S, +, •) is called a subring
if (R, +, •) itself is a ring with the operations + and • restricted to R.
Examples:
1. The set of integers Z is a subring of the ring of all rational
numbers Q.
2. The set of all even integers is a subring of the ring of all integers Z.

Definition 4.2.78 Ring Homomorphism: Let (R, +, •) and (S, ⊕, ) be


rings. A mapping g : R −→ S is called a ring homomorphism from (R, +, •)
to (S, ⊕, ) if for any a, b ∈ R,
g(a + b) = g(a) ⊕ g(b) and
g(a · b) = g(a) g(b).

Examples:

1. The ring Mn of all non-zero matrices is not commutative 


and has

0 1
non-zero divisors. For example, let n = 2; then if A =
     0 0
1 0 0 0 0 1
and B = , then AB = and BA = . So,
0 0 0 0 0 0
AB 6= BA, and A is a non-zero divisor.
2. The ring Q of rational numbers and the ring R of real numbers are
fields.
3. The ring (Z7 , +7 , ×7 ) is a field.
Algebraic Structures 219

4. The ring (Z10 , +10 , ×10 ) is not an integral domain since 5 ×10 2 = 0,
even though 5 6= 0, 2 6= 0 in Z10 .
5. The ring Z of all integers is an integral domain but not a field.

Definition 4.2.79 Commutative Ring: A ring (R, +, ·) is said to be


commutative if a · b = b · a, for all a, b ∈ R.

Theorem 4.2.80 Every finite integral domain is a field.

Proof.
Let (R, +, •) be a finite integral domain.
To prove: (R − {0}, •) is a group, that is, to prove
(i) there exists an element 1 ∈ R such that
1 · a = a · 1 = a, for all a ∈ R (since 1 ∈ R is an identity)
(ii) for every element of 0 6= a ∈ R, there exists an element a−1 ∈ R
such that
a · a−1 = a−1 · a = a.
Let R − {0} = {a1 , a2 , a3 , . . . , an }.
Let a ∈ R − {0}. Then, the elements aa1 , aa2 , . . . , aan are all in R − {0},
and they are all distinct. That is, if a · ai = a · aj , i 6= j, then a · (ai − aj ) = 0.
Since R is an integral domain and a 6= 0, we must have
ai − aj = 0 =⇒ ai = aj , which is a contradiction.
Therefore, R − {0} has exactly n elements, and R is a commutative ring
with cancellation law. Hence, we get
a = a · ai0 , for some i0 (since a ∈ R − {0}).
That is, a · ai0 = ai0 · a (since R is commutative).
Thus, let x = a · ai for some ai ∈ R − {0}, and
y · ai0 = a · ai0 = (ai · a)ai0 = ai · a = a · aj = y.
Therefore, ai0 is unity in R − {0}. We write it as 1.
Since 1 ∈ R − {0}, there exists an element aak ∈ R − {0} such that

aak = 1.

Therefore, ba = b = 1 (let ak = b).


Hence, b is the inverse of a, and the converse is also true.
Hence, (R, +, •) is a field.

Theorem 4.2.81 Every field is an integral domain.


Proof.
Let (F, +, ·) be a field. That is, F is a commutative ring with unity.
To prove F is an integral domain, it is enough to show that it has no zero
divisor.
Let a, b ∈ F such that a · b = 0.
220 Discrete Mathematical Structures

If a 6= 0, then a−1 ∈ F .
Therefore, a · b = 0
=⇒ a−1 · (a · b) = a−1 · 0
=⇒ 1 · b = 0
=⇒ b = 0.
Hence, the theorem is proved.
Note:
The converse of the above Theorem 4.2.81 need not be true.

Theorem 4.2.82 A commutative ring (R, +, ·) is an integral domain if and


only if the cancellation law holds in R. That is,
for a 6= 0, a · b = a · c =⇒ b = c, for all a, b, c ∈ R.

Proof.
Let R be an integral domain and a · b = a · c and a 6= 0, for all a, b, c ∈ R.
We have a · b − a · c = 0 =⇒ a · (b − c) = 0.
Therefore, since R is an integral domain and a 6= 0, b − c = 0. (R has no
zero divisor).
Therefore, b = c. Hence, the cancellation law holds.
Converse Part: Assume that the cancellation law holds in a ring R.
Let a · b = 0, for a 6= 0 and b ∈ R. We have
ab = 0 = a0
=⇒ b = 0.
Thus, ab = 0 in R =⇒ a = 0 or b = 0.
Therefore, R has no zero divisors.
Therefore, R is an integral domain.

4.2.14 Solved Problems


1. Prove that the set Z4 = {[0], [1], [2], [3]} is a commutative ring
with respect to the binary operations addition modulo 4 (+4 ) and
multiplication modulo 4 (×4 ).
Solution.
Tables 4.1 and 4.2 are composition tables for addition modulo 4
(+4 ) and multiplication modulo 4 (×4 ), respectively.
From Tables 4.1 and 4.2, we get the following:

TABLE 4.1
Composition Table for +4
+4 [0] [1] [2] [3]
[0] [0] [1] [2] [3]
[1] [1] [2] [3] [0]
[2] [2] [3] [0] [1]
[3] [3] [0] [1] [2]
Algebraic Structures 221
TABLE 4.2
Composition Table for ×4
×4 [0] [1] [2] [3]
[0] [0] [0] [0] [0]
[1] [0] [1] [2] [3]
[2] [0] [2] [0] [2]
[3] [0] [3] [2] [1]

(i) All the entries in both the tables belong to Z4 .


Therefore, Z4 is closed under the operations +4 and ×4 .
(ii) In both the tables,

Entries in the first row = Entries in the first column


Entries in the second row = Entries in the second column
Entries in the third row = Entries in the third column
Entries in the fourth row = Entries in the fourth column.

Therefore, the operations +4 and ×4 are commutative in Z4 .


(iii) Also, for any a, b, c ∈ Z4 , we have

a +4 (b +4 c) = (a +4 b) +4 c
and a ×4 (b ×4 c) = (a ×4 b) ×4 c
since 0 +4 (1 +4 2) = 0 +4 3 = 3
and (0 +4 1) +4 2 = (1 +4 2) = 3.
∴ (0 +4 1) +4 2 = (0 +4 1) +4 2.
Also, 1 ×4 (2 ×4 3) = 1 ×4 2 = 2
and (1 ×4 2) ×4 3 = 2 ×4 3 = 2.
∴ 1 ×4 (2 ×4 3) = (1 ×4 2) ×4 3.

Thus, the operations +4 and ×4 are associative in Z4 .


(iv) [0] is the additive identity of Z4 , and [1] is the multiplicative
identity of Z4 .
(v) Additive inverses of [0], [1], [2], [3] are, respectively, [0], [3], [2], [1].
Multiplicative inverses of the non-zero elements [1], [2], [3]
are, respectively, [1], [2], [3].
(vi) If a, b, c ∈ Z4 , then
a ×4 (b +4 c) = (a ×4 b) +4 (a ×4 c)
and (a +4 b) ×4 c = (a ×4 c) +4 (b ×4 c).
Thus, the operation ×4 is distributive over +4 in Z4 .
Hence, (Z4 , +4 , ×4 ) is a commutative ring with unity.
2. Show that (Z, +, ×) is an integral domain where Z is the set of all
integers.
222 Discrete Mathematical Structures

Solution.
We know the following:
(Z, +) is an abelian group.
(Z, ×) is a monoid.
The operation × is distributive over +.
(Z, ×) is commutative.
(Z, +, ×) is without zero divisors.
(Z, +, ×) is an integral domain.
3. Give an example of a ring which is not a field.
Solution.
The ring Z of all integers is an integral domain but not a field.

4.2.15 Problems for Practice


1. Discuss a ring and a field with suitable examples.
2. If (R, +, ·) is a ring, then prove that a · 0 = 0, for all a ∈ R, and 0
is the identity element in R under addition.
5
Lattices and Boolean Algebra

5.1 Introduction
In this chapter, we focus on partially ordered sets, lattices, Boolean algebra,
and their properties. These structures are useful in set theory, algebra, sorting,
and searching and in the construction of logical representation for computer
science. The concept of the lattices is a special case of a partially ordered set.
Boolean algebra is a special lattice.

5.2 Partial Ordering and Posets


Definition 5.2.1 Partial Order Relation: A relation R on a non-empty
set P is called a partial order, if R is reflexive, antisymmetric, and transitive.
That is, if R satisfies
(i) xRx, for all x ∈ P (reflexive)
(ii) xRy and yRx =⇒ x = y, for all x, y ∈ P (antisymmetric)
(iii) xRy and yRz =⇒ xRz, for all x, y, z ∈ P (transitive).
Example 5.2.2 Let P be the set of all positive integers. Define the relation
‘R’ such that xRy holds if and only if x ≤ y, for all x, y ∈ P . Clearly, “≤”
relation is reflexive, antisymmetric, and transitive. Hence, “≤” relation on
P is a partial order relation.
Remark 5.2.3 Usually, the partially ordered relation is denoted by “≤”.
Definition 5.2.4 Partially Ordered Set or Poset: A set P with the
partial order relation “≤” is called a partially ordered set or simply a poset.
It is denoted by hP, ≤i.
Example 5.2.5 Consider P = {collection of all subsets of any set}. Clearly,
“⊆” relation (set inclusion) is a partially ordered relation on P .
Definition 5.2.6 Totally Ordered Set: Let hP, ≤i be a partially ordered
set. If for every a, b ∈ P we have either a ≤ b or b ≤ a, then ≤ is called

223
224 Discrete Mathematical Structures

simple ordering or linear ordering on P , and hP, ≤i is called a totally ordered


or simply ordered set or a chain.

Example 5.2.7 The poset hZ, ≤i is totally ordered, since a ≤ b or b ≤ a


whenever a and b are integers.

Definition 5.2.8 Well-ordered Set: A partially ordered set is called well-


ordered if every non-empty subset of it has a least member.

5.2.1 Representation of a Poset by Hasse Diagram


A partially ordered relation “≤” on a set P can be represented by means of
a diagram known as a Hasse diagram. In such a diagram, each element is
represented by a small circle or a dot. The circle for an element x in P is
drawn below the circle for y in P , if x < y, and a line is drawn between x
and y, if y covers x. If x < y but y does not cover x, then x and y are not
connected directly by a single line.
Example 5.2.9 For example, let P = {1, 2, 3, 4} and “≤” be the relation
“less than or equal to”. Then, the Hasse diagram is shown below.

1
Hasse diagram of P

Example 5.2.10 Consider the set X = {2, 3, 6, 12, 24, 36} and the relation
“≤” is defined as x ≤ y if and only if x divides y. The Hasse diagram of the
poset hX, ≤i is shown below.

24 36

12

2 3
Hasse diagram of X
Lattices and Boolean Algebra 225

Note:
1. Hasse diagram is named after the twentieth-century German
mathematician Helmut Hasse.
2. In a digraph, if we apply the following rules, then we get Hasse
diagram.
(i) Each vertex of a poset P must be related to itself. So, the arrows
from vertex to itself are not necessary.
(ii) If a vertex b appears above vertex a and if vertex a is connected
to vertex b by an edge, then we have aRb; so, direction arrows
are not necessary.
(iii) If vertex c is above a and if c is connected to a by a sequence
of edges, then we have aRc.
(iv) The vertices are denoted by points rather than by circles.
Example 5.2.11 Let A = {a, b}. Let B = P (A) = {{φ}, {a}, {b}, {a, b}}.
Then, ⊆ is a relation on A whose digraph and Hasse diagram are given in
Figures 5.1 and 5.2.

{a,b}

{a}

{b}

FIGURE 5.1
Digraph of hB, ⊆i
{a,b}

{a} {b}

FIGURE 5.2
Hasse diagram of hB, ⊆i
226 Discrete Mathematical Structures

5.2.2 Solved Problems


1. Show that the “greater than or equal” relation (≥) is a partial
ordering on the set of integers.

Solution.
(i) Since a ≥ a for every integer a, the relation ≥ is reflexive.
(ii) If a ≥ b and b ≥ a, then a = b. Hence, ≥ is antisymmetric.
(iii) The relation ≥ is transitive since a ≥ b and b ≥ c imply that
a ≥ c.
Hence, ≥ is a partial ordering on the set of integers, and hZ, ≥i is
a poset.
2. Show that the inclusion relation ⊆ is a partial ordering on the power
set of a set S.
Solution.
(i) Since A ⊆ A, whenever A is a subset of S, the relation ⊆ is
reflexive.
(ii) Since A ⊆ B and b ⊆ A imply that A = B, the relation ⊆ is
antisymmetric.
(iii) Since A ⊂ B and B ⊂ C imply that A ⊆ C, the relation ⊆ is
transitive.
Therefore, the relation ⊆ is a partial ordering on P (S), and
hP (S), ⊆i is a poset.
3. Let R be a binary relation on the set of all positive integers such
that R = {(a, b)/a = b2 }. Is R reflexive, symmetric, antisymmetric,
transitive, an equivalence relation, or a partial order relation?
Solution.
(i) R = {(a, b)/a, b are positive integers and a = b2 }.
For R to be reflexive, we should have aRa, for all positive
integers a. But aRa holds only when a = a2 by hypothesis.
Now, a = a2 is not true for all positive integers. In fact, only
when a = 1, we have a = a2 . Hence, R is not reflexive.
(ii) For R to be symmetric, if aRb holds, then we should have bRa.
But aRb implies a = b2 . But a = b2 does not imply b = a2
always for positive integers. For instance, 16 = 42 , but 4 6= 162 .
Hence, aRb does not imply bRa. Hence, R is not symmetric.
(iii) For R to be antisymmetric, for positive integers a, b if aRb and
bRa hold, then a = b. aRb implies a = b2 , and bRa implies
b = a2 . Hence, if a = b2 and b = a2 , then a = b2 = (a2 )2 = a4 ,
that is, a4 − a = 0, that is, a(a3 − 1) = 0. Since a is a positive
Lattices and Boolean Algebra 227

integer, a 6= 0 so that a3 − 1 = 0, that is, a3 = 1 which implies


a = 1. This means b = a2 = 1. Hence, aRb and bRa imply
a = b = 1. Hence, R is antisymmetric.
(iv) For R to be transitive, if aRb holds and bRc holds, then aRc
should hold.
That is, aRb implies a = b2 , and bRc implies b = c2 , so that
a = b2 = c4 . Hence, aRc does not hold.
For example, 256 = 162 and 16 = 42 but 256 6= 42 . Thus, R
is not transitive.
(v) R is not an equivalence relation since an equivalence relation is
reflexive, symmetric, and transitive.
(vi) R is not a partial order relation, since a partial ordering relation
is reflexive, antisymmetric, and transitive.
4. Give examples of a relation which is both a partial ordering relation
and an equivalence relation on a set.
Solution.
Equality and similarity of triangles are examples of a relation which
is both a partial ordering relation and an equivalence relation.
5. Let S be a set. Determine whether there is a greatest element and
a least element in the poset hP (S), ⊆i.
Solution.
The least element is the empty set since φ ⊆ T for any subset T
of S. The set S is the greatest element in this poset. Hence T ⊆ S
whenever T is a subset of S.
6. Is there a greatest element and a least element in the poset hZ+ , |i?
Solution.
The integer 1 is the least element since 1 divides n whenever n is
a positive integer. Since there is no integer that is divisible by all
positive integers, there is no greatest element.
7. Let A be a given finite set and P (A) its power set. Let ⊆ be the
inclusion relation on the elements of P (A). Draw Hasse diagram of
hP (A), ⊆i for

(i) A = {a} (ii) A = {a, b}


(iii) A = {a, b, c} (iv) A = {a, b, c, d}.
228 Discrete Mathematical Structures

Solution.
{a,b,c}
{a} {a,b}

{a,b} {a,c} {b,c}

{a} {b}

{a} {b} {c}

(i) (ii) (iii)


Hasse diagrams of (i), (ii), and (iii)

{a,b,c,d}

{a,b,c} {a,b,d} {a,c,d} {b,c,d}

{a,c} {a,d}
{a,b} {b,c} {c,d}
{b,d}

{a} {c}
{b} {d}

(iv)
Hasse diagram of (iv)
8. Which elements of the poset h{2, 4, 5, 10, 12, 20, 25}, |i are maximal,
and which of them are minimal?
Solution.
The Hasse diagram is shown in Figure 5.3.
From the Hasse diagram in Figure 5.3, this poset shows that the
maximal elements are 12, 20, and 25 and the minimal elements are

12 20

25
4 10

2 5

FIGURE 5.3
Hasse diagram of the given poset
Lattices and Boolean Algebra 229

2 and 5. As this example shows, a poset can have more than one
maximal element and more than one minimal element.
9. Determine whether the posets represented by each of the Hasse
diagrams in the following figure have a greatest element and a least
element.
b c d d e d d

c c
b c

a a b a
a b

(i) (ii) (iii) (iv)


Hasse diagrams of the given posets

Solution.
(i) The least element of the poset with Hasse diagram (i) is a. This
poset has no greatest element.
(ii) The poset with Hasse diagram (ii) has neither a least nor a
greatest element.
(iii) The poset with Hasse diagram (iii) has no least element. Its
greatest element is d.
(iv) The poset with Hasse diagram (iv) has the least element a and
greatest element d.
10. Draw the Hasse diagram of the set of partitions of 5.
Solution.
5

4+1 3+2

3+1+1 2+2+1

2+1+1+1

1+1+1+1+1
Hasse diagram of the set of partitions of 5
230 Discrete Mathematical Structures

5=5
5=4+1
5=3+2
5=3+1+1
5=2+2+1
5=2+1+1+1
5 = 1 + 1 + 1 + 1 + 1.

5.2.3 Problems for Practice


1. Let R be the relation on the set of people such that xRy holds if x
is older than y. Show that R is not a partial ordering.
2. Show that hN, ≤i is a partially ordered set, where N is the set of all
positive integers and ≤ is a relation defined by m ≤ n if and only
if n − m is a non-negative integer.
3. Show that there are only five distinct Hasse diagrams for partially
ordered sets that contain three elements.
4. Give an example of a set X such that hP (X), ⊆i is a totally
ordered set.
5. Let S denote the set of all the partial ordering relations on a set P .
Define a partial ordering relation on S, and interpret this relation
in terms of the elements of P .
6. Let X = {1, 2, 3, 4, 6, 8, 12, 24} and R be a division relation defined
on X. Find the Hasse diagram of the poset hX, Ri.
7. Draw the Hasse diagrams of the following sets under the partial
ordering relation “divides”, and indicate those which are totally
ordered:
{2, 6, 24}, {3, 5, 15}, {1, 2, 3, 6, 12}, {2, 4, 8, 16}, {3, 9, 27, 54}.
8. If R is a partial ordering relation on a set X and A ⊆ X, show that
R ∩ (A × A) is a partial ordering relation on A.
9. Let D30 = {1, 2, 3, 5, 6, 10, 15, 30}, and let the relation R be divisor
on D30 . Find
(i) all the lower bounds of 10 and 15
(ii) the greatest lower bound of 10 and 15
(iii) all upper bounds of 10 and 15
(iv) the least upper bound of 10 and 15.
Also, draw the Hasse diagram.
10. Draw the Hasse diagram of hX, ≤i, where X = {2, 4, 5, 10, 12, 20, 25}
and the relation ≤ be such that x ≤ y if x divides y.
Lattices and Boolean Algebra 231

Definition 5.2.12 Linearly ordered set or Chain: A poset hP, ≤i is called


a linearly ordered set or a chain if every pair of elements in a poset hP, ≤i are
comparable.
Example 5.2.13 Let Z+ be the set of all positive integers. The usual relation
“≤” is a partial order relation on Z+ , since any two integers in Z+ can be
comparable with respect to the relation “≤”. Thus, hZ+ , ≤i is a linearly ordered
set.
Definition 5.2.14 Upper bound and Lower bound: Let S be any subset
of a poset hP, ≤i. An element x ∈ P is called an upper bound of S if y ≤ x,
for all y ∈ S. An element z ∈ P is called a lower bound of S if z ≤ y, for all
y ∈ S.
Example 5.2.15 Let A = {a, b, c} be a given set and ρ(A) be its power set.
Let “⊆” be the relation on ρ(A). Then clearly, hρ(A), ⊆i is a poset.
For the subset S = {{a, b}, {a}, {b}, {c}} ⊆ ρ(A), the upper bounds are {a, b}
and {a, b, c}, and its lower bound is φ.

5.3 Lattices, Sublattices, Direct Product,


Homomorphism of Lattices
Definition 5.3.1 Lattice: A lattice is a poset hL, ≤i in which any subset
{a, b} consisting of two elements has a least upper bound and a greatest lower
bound.
We denote LU B({a, b}) by a ⊕ b and call it as the join of a and b. Similarly,
we denote GLB({a, b}) by a ? b and call it as the meet of a and b.
Example 5.3.2 Let S be a set, and let L = ρ(S). Let “⊆” (set inclusion) be
the relation on L. Clearly, hL, ⊆i is a lattice in which the meet and join are
the same as the operations ∩ and ∪ on sets, respectively.
That is, for any two elements A, B ∈ ρ(S), GLB({A, B}) = A ∩ B and
LU B({A, B}) = A ∪ B.
Definition 5.3.3 Dual Lattices: Let hL, ≤i be a poset, and let hL, ≥i be the
dual poset (the symbol ‘≥’ used for the partial order is ≤0 ).
If hL, ≤i is a lattice, we can show that hL, ≥i is also a lattice. In fact, for
any a, b ∈ L, the LU B(a, b) in hL, ≤i is equal to GLB(a, b) in hL, ≥i.
Similarly, the GLB(a, b) in hL, ≤i is equal to LU B(a, b) in hL, ≥i.
Thus, the dual of hL, ≤i is hL, ≥i and vice-versa.

5.3.1 Properties of Lattices


In the following theorems, let hL, ≤i be a lattice.
232 Discrete Mathematical Structures

Theorem 5.3.4 [Idempotent law] For any a, b, c ∈ L, we have a ? a = a


and a ⊕ a = a.

Proof.
Let a, b, c ∈ L. Then by the definition of GLB of a and b, we have

a?b≤a (5.1)

and if a ≤ a and a ≤ b, then


a ≤ a ? b. (5.2)
Since a ≤ a, from (5.1) and (5.2), we have
a ? a ≤ a and a ≤ a ? a.
By the antisymmetric property, it follows that a = a ? a.
Similarly, we can prove that a ⊕ a = a.
Theorem 5.3.5 [Associative law] The operations of meet and join on
hL, ≤i are associative. That is, for any a, b, c ∈ L, we have the following:
(i) a ? (b ? c) = (a ? b) ? c
(ii) a ⊕ (b ⊕ c) = (a ⊕ b) ⊕ c.
Proof.
To prove: a ? (b ? c) = (a ? b) ? c.
Let a, b, c ∈ L, Then by the definition, we have
(a ? b) ? c ≤ a ? b
and (a ? b) ? c ≤ c.
By the definition of GLB of a and b, we have a ? b ≤ a and a ? b ≤ b.
Hence, by the transitive property of ≤, we have
(a ? b) ? c ≤ a
and (a ? b) ? c ≤ b.
Since (a ? b) ? c ≤ b and (a ? b) ? c ≤ c, we see that (a ? b) ? c is a lower
bound for b and c. From the definition of b ? c, it follows that (a ? b) ? c ≤ b ? c.
Since (a ? b) ? c ≤ a and (a ? b) ? c ≤ b ? c, from the definition of a ? (b ? c),
we have
(a ? b) ? c ≤ a ? (b ? c). (5.3)
Now, a ? (b ? c) ≤ a and a ? (b ? c) ≤ b ? c.
Since b ? c ≤ b, by transitivity, we have a ? (b ? c) ≤ b.
Since a ? (b ? c) ≤ a and a ? (b ? c) ≤ b, we have a ? (b ? c) ≤ a ? b.
Since a ? (b ? c) ≤ b ? c ≤, we have

a ? (b ? c) ≤ (a ? b) ? c. (5.4)

From (5.3), (5.4) and by antisymmetric property, it follows that


a ? (b ? c) = (a ? b) ? c.
Similarly, we can prove that a ⊕ (b ⊕ c) = (a ⊕ b) ⊕ c.
Lattices and Boolean Algebra 233

Theorem 5.3.6 [Commutative law] The operations of meet and join on


hL, ≤i satisfy commutative property. That is, for any a, b ∈ L, we have the
following:
(i) a ? b = b ? a
(ii) a ⊕ b = b ⊕ a.

Proof.
Given: a, b ∈ L. Both a ? b and b ? a are GLB of a and b. By the uniqueness
of GLB of a and b, we have a ? b = b ? a. Similarly, a ⊕ b = b ⊕ a holds good.
Theorem 5.3.7 [Absorption law] For any a, b ∈ L, we have the following:
(i) a ? (a ⊕ b) = a
(ii) a ⊕ (a ? b) = a.
Proof.
Let a, b ∈ L. Then, a ≤ a and a ≤ a ⊕ b. So, a ≤ a ? (a ⊕ b). On the other
hand, a ? (a ⊕ b) ≤ a. By antisymmetric property of ≤, we have a = a ? (a ⊕ b).
Similarly, we have a ⊕ (a ? b) = a, for all a, b ∈ L.

5.3.2 Theorems on Lattices


Theorem 5.3.8 Let hL, ≤i be a lattice in which ? and ⊕ denote the operations
of meet and join respectively. For any a, b ∈ L,
a ≤ b ⇐⇒ a ? b = a ⇐⇒ a ⊕ b = b.

Proof.
First, let us prove that a ≤ b ⇐⇒ a ? b = a ⇐⇒ a ⊕ b = b.
Let us assume that a ≤ b, and also, we know that a ≤ a.

∴ a ≤ a ? b. (5.5)

But, from the definition of a ? b, we have

a ? b ≤ a. (5.6)

Hence, a ≤ b =⇒ a ? b = a [using (5.5) and (5.6)].


Next, assume that a ? b = a. But it is only possible if a ≤ b.
That is, a ? b = a =⇒ a ≤ b.
Combining these two results, we get
a ≤ b ⇐⇒ a ? b = a.
Similarly, we can prove that a ≤ b ⇐⇒ a ⊕ b = b.
From a ? b = a, we have
b ⊕ (a ? b) = b ⊕ a = a ⊕ b.
But b ⊕ (a ? b) = b.
Hence, a ⊕ b = b follows that a ? b = a.
234 Discrete Mathematical Structures

Theorem 5.3.9 Let hL, ≤i be a lattice. For any a, b ∈ L, the following are
equivalent:
(i) a ≤ b
(ii) a ? b = a
(iii) a ⊕ b = b.
Proof.
First, consider (i) ⇐⇒ (ii).
We have a ≤ a. Assume a ≤ b. Therefore, a ≤ a ? b. By the definition of
GLB, we have
a ? b ≤ a.
Hence, by antisymmetric property, a ? b = a.
Assume that a ? b = a, but it is only possible if
a ≤ b =⇒ a ? b = a =⇒ a ≤ b.
Combining these two results, we have a ≤ b ⇐⇒ a ? b = a.
Similarly, a ≤ b ⇐⇒ a ⊕ b = b.
Now, consider (ii) ⇐⇒ (iii).
Assume a ? b = a, we have b ⊕ (a ? b) = b ⊕ a = a ⊕ b, but by absorption,
b ⊕ (a ? b) = b.
Hence, a ⊕ b = b.
By similar arguments, we can show that a ? b = a follows from a ⊕ b = b.
(ii) ⇐⇒ (iii)
Hence, the theorem is proved.
Theorem 5.3.10 Let hL, ≤i be a lattice. For any a, b ∈ L, the following
inequalities hold:
(1) Distributive Inequalities
(i) a ⊕ (b ? c) ≤ (a ⊕ b) ? (a ⊕ c)
(ii) a ? (b ⊕ c) ≥ (a ? b) ⊕ (a ? c).
(2) Modular Inequalities
(i) a ≤ c ⇐⇒ a ⊕ (b ? c) ≤ (a ⊕ b) ? c
(ii) a ≥ c ⇐⇒ a ? (b ⊕ c) ≥ (a ? b) ⊕ c.
Proof.
Since (ii) in (1) and (ii) in (2) are duals of (i) in (1) and (i) in (2) respectively,
it is enough to prove (i) in (1) and (i) in (2) only.
Consider (i) in (1).
Let a, b, c ∈ L. Since a ≤ a ⊕ b and a ≤ a ⊕ c, we have
a ≤ [(a ⊕ b) ? (a ⊕ c)].
Since b ? c ≤ b ≤ a ⊕ b and b ? c ≤ c ≤ a ⊕ c, we have
(b ? c) ≤ (a ⊕ b) ? (a ⊕ c).
Therefore, (a ⊕ b) ? (a ⊕ c) is an upper bound for a and b ? c, and hence
a ⊕ (b ? c) ≤ (a ⊕ b) ? (a ⊕ c).
Thus, (i) in (1) is proved.
Lattices and Boolean Algebra 235

The inequality (i) in (2) is a special case of (i) in (1).


If a ≤ c, then a ⊕ c = c, and from (i) in (1), we obtain
a ⊕ (b ? c) ≤ (a ⊕ b) ? (a ⊕ c) = (a ⊕ b) ? c, which is inequality (i) in (2).
Hence, the theorem is proved.
Theorem 5.3.11 In a lattice hL, ≤i, for all a, b, c ∈ L, we have the following:
(i) (a ? b) ⊕ (c ? d) ≤ (a ⊕ c) ? (b ⊕ d)
(ii) (a ? b) ⊕ (b ? c) ⊕ (c ? a) ≤ (a ⊕ b) ? (b ⊕ c) ? (c ⊕ a).
Proof.
Let a, b, c ∈ L. Then
a ? b ≤ a (or) b ≤ a ⊕ b. (5.7)
a ? b ≤ a ≤ c ⊕ a. (5.8)
a ? b ≤ b ≤ b ⊕ c. (5.9)
Using (5.7), (5.8), and (5.9), we get
a ? b ≤ (a ⊕ b) ? (b ⊕ c) ? (c ⊕ a).
Similarly, b ? c ≤ (a ⊕ b) ? (b ⊕ c) ? (c ⊕ a),
c ? a ≤ (a ⊕ b) ? (b ⊕ c) ? (c ⊕ a).
This proves (ii).
We have a ≤ a ⊕ c and b ≤ b ⊕ d.
We know that
c ≤ a ⊕ c. (5.10)
d ≤ b ⊕ d. (5.11)
Therefore, c ? d ≤ (a ⊕ c) ? (b ⊕ d).
By (5.10) and (5.11), we have
(a ? b) ⊕ (c ? d) ≤ (a ⊕ c) ? (b ⊕ d).
This proves (i).
Theorem 5.3.12 In a lattice hL, ≤i, prove that for a, b, c ∈ L,
(i) (a ? b) ⊕ (a ? c) ≤ a ? (b ⊕ (a ? c))
(ii) (a ⊕ b) ? (a ⊕ c) ≥ a ⊕ (b ? (a ⊕ c)).
Proof.
We know that a ? b ≤ a, a ? c ≤ a.

Therefore, (a ? b) ⊕ (a ? c) ≤ a ⊕ a = a. (5.12)

Also, a ? b ≤ b, a ? c ≤ a ? c

=⇒ (a ? b) ⊕ (a ? c) ≤ b ⊕ (a ? c). (5.13)

From (5.12) and (5.13), we have


(a ? b) ⊕ (a ? c) ≤ a ? (b ⊕ (a ? c)).
236 Discrete Mathematical Structures

This proves (i).


We know that a ≤ a ⊕ b; a ≤ a ⊕ c

=⇒ a = a ? a ≤ (a ⊕ b) ? (a ⊕ c). (5.14)

Further, b ≤ a ⊕ b; a ⊕ c ≤ a ⊕ c

=⇒ b ? (a ⊕ c) ≤ (a ⊕ b) ? (a ⊕ c). (5.15)

Using (5.14) and (5.15), we have


a ⊕ (b ? (a ⊕ c)) ≤ (a ⊕ b) ? (a ⊕ c).
This proves (ii).
Theorem 5.3.13 In a lattice if a ≤ b ≤ c, show that
(i) a ⊕ b = b ? c
(ii) (a ? b) ⊕ (b ? c) = (a ⊕ b) ? (a ⊕ c) = b.
Proof.
Let a ≤ b ≤ c.
a ≤ b =⇒ a ⊕ b = b, a ? b = a.
b ≤ c =⇒ b ⊕ d = c, b ? c = b.
a ≤ c =⇒ a ⊕ c = c, a ? c = a.
Therefore, a ⊕ b = b = b ? c, which is (i).
Now, (a ? b) ⊕ (b ? c) = a ⊕ b = b
(a ⊕ b) ? (a ⊕ c) = b ? c = b, which is (ii).

5.3.3 Solved Problems


1. Determine whether the posets represented by each of the Hasse
diagrams are lattices.

Solution.

f h

e e
d e g
f

c d
c d
b b c b

a a

(i) (ii) (iii)


Hasse diagrams of the given posets
Lattices and Boolean Algebra 237

The posets represented by the Hasse diagrams in (i) and (iii) are
both lattices because in each poset, every pair of elements has both
a least upper bound and a greatest lower bound. On the other hand,
the poset with the Hasse diagram shown in (ii) is not a lattice, since
the elements b and c have no least upper bound.
It is to be noted that each of the elements d, e, and f is an upper
bound, but none of these three elements precede the other two with
respect to the ordering of this poset.
2. Is the poset hZ+ , |i a lattice?
Solution.
Let a and b be two positive integers. The least upper bound
and greatest lower bound of these two integers are the least
common multiple and the greatest common divisor of these integers,
respectively. Hence, it follows that this poset is a lattice.
3. Explain why the partially ordered sets of Figures 5.4 and 5.5 are
not lattices.

(i) (ii) (iii)

FIGURE 5.4
Hasse diagrams of the given posets

e e f

d e
d d

b c
c b c

a b a
a

(i) (ii) (iii)

FIGURE 5.5
Hasse diagrams of the given posets
238 Discrete Mathematical Structures

Solution. Given:
(i) does not represent a lattice since e ⊕ f does not exist.
(ii) does not represent a lattice since b ⊕ c does not exist.
(iii) does not represent a lattice because neither d ⊕ c nor b ? c
exists.
4. Let the sets S0 , S1 , S2 , . . . , S7 be given by

S0 = {a, b, c, d, e, f }, S1 = {a, b, c, d, e},


S2 = {a, b, c, d, f }, S3 = {a, b, c, e},
S4 = {a, b, c}, S5 = {a, b},
S6 = {a, c}, S7 = {a}.

Draw the diagram of hL, ⊆i where L = {S0 , S1 , S2 , . . . , S7 }.


Solution.
The diagram is shown below.

S0

S1 S2

S3

S4

S5 S6

S7
Hasse diagram of hL, ⊆i

5. Show that every non-empty subset of a lattice has a least upper


bound and a greatest lower bound.
6. Show that every totally ordered set is a lattice.
7. Let A = {1, 2, 5, 10} with the relation “divides”. Draw the Hasse
diagram.

Definition 5.3.14 Sublattice: Let hL, ?, ⊕i be a lattice, and let S ⊆ L be a


subset of L. The algebra hS, ?, ⊕i is a sublattice of hL, ?, ⊕i if and only if S is
closed under both operations ? and ⊕.

Example 5.3.15 Let hL, ≤i be a lattice in which L = {a1 , a2 , . . . , a8 } and


S1 , S2 , and S3 be the sublattices of L given by S1 = {a1 , a2 , a4 , a6 },
S2 = {a3 , a5 , a7 , a8 }, and S3 = {a1 , a2 , a4 , a8 }. The diagram of hL, ≤i is below.
Lattices and Boolean Algebra 239
a1

a2 a3
a4

a6
a5 a7

a8
Hasse diagram of hL, ⊆i

Note that hS1 , ≤i and hS2 , ≤i are sublattices of hL, ≤i, but hS3 , ≤i is not a
sublattice since a2 , a4 ∈ S3 but a2 ? a4 = a6 ∈ S3 . Also, note that hS3 , ≤i is a
lattice.
Definition 5.3.16 Direct Product of Lattices: Let hL, ?, ⊕i and hS, ∧, ∨i
be two lattices. The algebraic system hL×S, ·, +i in which the binary operations
“+” and “·” on L × S are such that for any (a1 , b1 ) and (a2 , b2 ) in L × S
(a1 , b1 ) · (a2 , b2 ) = (a1 ? a2 , b1 ∧ b2 )
(a1 , b1 ) + (a2 , b2 ) = (a1 ⊕ a2 , b1 ∨ b2 )
is called the direct product of the lattices hL, ?, ⊕i and hS, ∧, ∨i.

Definition 5.3.17 Lattice Homomorphism: Let hL, ?, ⊕i and hS, ∧, ∨i be


two lattices. A mapping g : L −→ S is called a lattice homomorphism from
the lattice hL, ?, ⊕i to hS, ∧, ∨i if for any a, b ∈ L,
g(a ? b) = g(a) ∧ g(b) and g(a ⊕ b) = g(a) ∨ g(b).

Remark 5.3.18 Observe that both the operations of meet and join are
preserved. These may be mappings which preserve only one of the two
operations. Such mappings are not lattice homomorphisms.

Definition 5.3.19 Lattice Isomorphism: If a homomorphism g : L −→ S


of two lattices hL, ?, ⊕i and hS, ∧, ∨i is bijective, that is, one-to-one and onto,
then g is called an isomorphism. If there exists an isomorphism between two
lattices, then the lattices are called isomorphic.

Definition 5.3.20 Lattice Endomorphism: Let hL, ?, ⊕i be a lattice. A


homomorphism g : L −→ L is called an endomorphism.

Definition 5.3.21 Lattice Automorphism: Let hL, ?, ⊕i be a lattice. If g :


L −→ L is an isomorphism, then g is called an automorphism.

Remark 5.3.22 Let hL, ?, ⊕i be a lattice. If g : L −→ L is an endomorphism,


then the image set of g is a sublattice of L.
240 Discrete Mathematical Structures

Definition 5.3.23 Order-Preserving Mapping: Let hP, ≤i and hQ, ≤0 i be


two partially ordered sets. A mapping f : P −→ Q is said to be order-preserving
relative to the ordering ≤ in P and ≤0 in Q if and only if for any a, b ∈ P
such that a ≤ b, f (a) ≤0 f (b) in Q.
Remark 5.3.24 If hP, ≤i and hQ, ≤0 i are lattices and g : P −→ Q is a lattice
homomorphism, then g is order-preserving.
Definition 5.3.25 Order-isomorphic Partially Ordered Sets: Two
partially ordered sets hP, ≤i and hQ, ≤0 i are called order-isomorphic if there
exists a mapping f : P −→ Q which is bijective and if both f and f −1 are
order-preserving.

5.3.4 Problem for Practice


1. Let hL, ?, ⊕i and hS, ∧, ∨i be any two lattices with the partial
orderings ≤ and ≤0 respectively. If g is a lattice homomorphism,
then g preserves the partial ordering.

5.4 Special Lattices


Let hL, ?, ⊕i be a lattice and S ⊆ L be a finite subset of L where
S = {a1 , a2 , . . . , an }. The greatest lower bound and the least upper bound
of S can be expressed as
GLB S = ?ni=1 ai and LU B S = ⊕ni=1 ai
where ?2i=1 ai = a1 ? a2 and ?ki=1 ai = ?k−1
i=1 (ai ? ak ), k = 3, 4, . . .
A similar representation can be given for ⊕ni=1 . In lieu of the associative
property of the operations ? and ⊕, we can write
?ni=1 ai = a1 ? a2 ? · · · ? an
and ⊕ni=1 ai = a1 ⊕ a2 ⊕ · · · ⊕ an .
Definition 5.4.1 Complete Lattice: A lattice is called complete if each of
its non-empty subsets has a least upper bound and a greatest lower bound.
Definition 5.4.2 Complement Element: In a bounded lattice hL, ?, ⊕, 0, 1i,
an element b ∈ L is called a complement of an element a ∈ L if
a?b=0 and a ⊕ b = 1.
Definition 5.4.3 Complemented Lattice: A lattice hL, ?, ⊕, 0, 1i is said to
be a complemented lattice if every element of L has at least one complement.
Lattices and Boolean Algebra 241

Definition 5.4.4 Distributive Lattice: A lattice hL, ?, ⊕i is called a


distributive lattice if for any a, b, c ∈ L,
a ? (b ⊕ c) = (a ? b) ⊕ (a ? c)
a ⊕ (b ? c) = (a ⊕ b) ? (a ⊕ c).
In other words, in a distributive lattice, the operations ? and ⊕ are distributed
over each other.
Definition 5.4.5 Modular Lattice: A lattice hL, ∧, ∨i is called modular if
for all x, y, z ∈ L,
x ≤ z =⇒ x ∨ (y ∧ z) = (x ∨ y) ∧ z (modular equations).
Remark 5.4.6 We have (by modular inequality) if x ≤ z =⇒ x ∨ (y ∧ z) =
(x ∨ y) ∧ z holds in any lattice. Therefore, to show that a lattice L is modular,
it is enough to show if
x ≤ z =⇒ x ∨ (y ∧ z) ≥ (x ∨ y) ∧ z holds in L.
Theorem 5.4.7 Every chain is a distributive lattice.
Proof.
Let hL, ≤i be a chain. Let a, b, c ∈ L. Consider the following possible cases:
(i) a ≤ b or a ≤ c
(ii) a ≥ b and a ≥ c.
We shall now show the distributive law
a ? (b ⊕ c) = (a ? b) ⊕ (a ? c).
In case (i), if a ≤ b or a ≤ c, then we have
a ? b = a, a ⊕ a = a, a ? c = c and
=⇒ a ≤ b ⊕ c. (5.16)
Hence, a ? (b ⊕ c) = a
and (a ? b) ⊕ (a ? c) = a ⊕ a = a. (5.17)
From (5.16) and (5.17), we get
a ? (b ⊕ c) = (a ? b) ⊕ (a ? c).
In case (ii), if a ≥ b and a ≥ c, then we have a ? b = b, a ? c = c and b ⊕ c ≤ a,
so that
a ? (b ⊕ c) = b ⊕ c (5.18)
and (a ? b) ⊕ (a ? c) = b ⊕ c. (5.19)
From (5.18) and (5.19), we get
a ? (b ⊕ c) = (a ? b) ⊕ (a ? c).
242 Discrete Mathematical Structures

Theorem 5.4.8 Let hL, ?, ⊕i be a distributive lattice. For any a, b, c ∈ L,


(a ? b = a ? c) ∧ (a ⊕ b = a ⊕ c) =⇒ b = c.
Proof.
(a ? b) ⊕ c = (a ? c) ⊕ c = c. (5.20)

(a ? b) ⊕ c = (a ⊕ c) ? (b ⊕ c)
= (a ⊕ b) ? (b ⊕ c)
= b ⊕ (a ? c)
= b ⊕ (a ? b)
= b. (5.21)
From (5.20) and (5.21), we have
b = c.
Theorem 5.4.9 Every distributive lattice is modular.
Proof.
Let hL, ≤i be a distributive lattice.
For all a, b, c ∈ L, we have
a ⊕ (b ? c) = (a ⊕ b) ? (a ⊕ c).
Thus, if a ≤ c, then a ⊕ c = c and
a ⊕ (b ? c) = (a ⊕ b) ? c.
Hence, if a ≤ c, the modular equation is satisfied, and L is modular.

5.4.1 Solved Problems


1. Show that a chain of three or more elements is not complemented.
Solution.
In a chain, we have that any two elements are comparable.
Let 0, x, 1 be any three elements in a chain hL, ≤i with least
element 0 and greatest element 1.
We have 0 ≤ x ≤ 1.
Now, 0 ∧ x = 0 and 0 ∨ x = x.
Similarly, x ∧ 1 = x and x ∨ 1 = 1.
Therefore, x does not have any complement.
Hence, any chain with three or more elements is not
complemented.
2. Find all sublattices of hD30 , |i where | is the divisor relation.
Solution.
The Hasse diagram of hD30 , |i is shown in Figure 5.6.
Lattices and Boolean Algebra 243
30

10 6
15

2 5 3

FIGURE 5.6
Hasse diagram of hD30 , |i

Therefore, the sublattices are

D6 = {1, 2, 3, 6}
D10 = {1, 2, 5, 10}
D15 = {1, 3, 5, 15}
S1 = {5, 10, 15, 30}
S2 = {3, 5, 15, 30}, etc. are lattices.

In general, if m|n, then Dm is a sublattice of Dn , and Dkm is also


a sublattice of Dn .
3. Show that the lattices given by the diagrams are not distributive.

1 1

a2
b1 b3
b2
a1

a3

0 0
(i) (ii)
Hasse diagrams of given lattices

Solution.
In lattice (i),
a3 ? (a1 ⊕ a2 ) = a3 ? 1 = a3 = (a3 ? a1 ) ⊕ (a3 ? a2 )
a1 ? (a2 ⊕ a3 ) = 0 = (a1 ? a2 ) ⊕ (a1 ? a3 )
but a2 ? (a1 ⊕ a3 ) = a2 ? 1 = a2
(a2 ? a1 ) ⊕ (a2 ? a3 ) = 0 ⊕ a3 = a3 .
Hence, the lattice (i) is not distributive.
In lattice (ii),
244 Discrete Mathematical Structures

b1 ? (b2 ⊕ b3 ) = b1 while (b1 ? b2 ) ⊕ (b1 ? b3 ) = 0 which shows


that the lattice is not distributive.
4. If Dn denotes the lattice of all the divisors of the integer n, draw
the Hasse diagrams of D10 , D15 , D32 , and D45 .
Solution.
The Hasse diagrams are shown below.

45
32
10 15
16 9 15

2 5 3 5 8

4 3 5
1 1 2
1
1

D10 D15 D32 D45


Hasse diagrams of given lattices

5. Prove that in a distributive lattice, the complement of an element


is unique.
Solution.
Let a be an element with two distinct complements b and c. Then
a ? b = 0 and a ? c = 0
=⇒ a ? b = a ? c.
Also, a ⊕ b = 1 and a ⊕ c = 1
=⇒ a ⊕ b = a ⊕ c.
By a theorem, we have b = c.
6. Let L be a complemented, distributive lattice. Then for a, b ∈ L,
show the following are equivalent:
(i) a ≤ b
(ii) a ? b0 = 0
(iii) a0 ⊕ b = 1
(iv) b0 ≤ a0
where 0 denotes corresponding complement.
or
Show that the following hold in a distributive and complemented
lattice L:
a ≤ b ⇐⇒ a ? b0 = 0 ⇐⇒ a0 ⊕ b = 1 ⇐⇒ b0 ≤ a0 for a, b ∈ L.
Lattices and Boolean Algebra 245

Solution.

a ≤ b =⇒ a ⊕ b = b
=⇒ (a ⊕ b) ? b0 = 0 since b ? b0 = 0
=⇒ (a ? b0 ) ⊕ (b ? b0 ) = 0
=⇒ a ? b0 = 0 since b ? b0 = 0.

Hence (i) =⇒ (ii).

a ? b0 = 0 =⇒ (a ? b)0 = 1
=⇒ a0 ⊕ (b0 )0 = 1
=⇒ a0 ⊕ b = 1.

Hence (ii) =⇒ (iii).

a0 ⊕ b = 1 =⇒ (a0 ⊕ b) ? b0 = b0
=⇒ (a0 ? b0 ) ⊕ (b ? b0 ) = b0 (using distributive law)
=⇒ a0 ? b0 = b0 since b ? b0 = 0
=⇒ b0 ≤ a0 .

Hence (iii) =⇒ (iv).


7. Let hL, ∧, ∨i be a distributive lattice and a, b, c ∈ L. If a ∧ b = a ∧ c
and a ∨ b = a ∨ c, then b = c.
or
Show that the cancellation laws are valid in a distributive lattice.
Solution.
Let hL, ∧, ∨i be a distributive lattice and a, b, c ∈ L, such that
a ∧ b = a ∧ c and a ∨ b = a ∨ c. Now,

(a ∧ b) ∨ c = (a ∨ c) ∧ (b ∨ c) (since L is distributive)
= (a ∨ b) ∧ (b ∨ c)
= (b ∨ a) ∧ (b ∨ c)
= b ∨ (a ∧ c)
= b ∨ (a ∧ b)
=b

and (a ∧ b) ∨ c = (a ∧ c) ∨ c = c.
Thus, b = (a ∧ b) ∨ c = c, so that
a ∧ b = a ∧ c and a ∨ b = a ∨ c =⇒ b = c.
That is, the cancellation law is valid in a distributive lattice.
8. Show that the direct product of any two distributive lattices is a
distributive lattice.
246 Discrete Mathematical Structures

Solution.
Let L1 and L2 be two distributive lattices. Let x, y, z ∈ L1 × L2 ,
the direct product (lattice) of L1 and L2 . Then, x = (a1 , a2 ),
y = (b1 , b2 ), and z = (c1 , c2 ) for some a1 , b1 , c1 ∈ L1 and
a2 , b2 , c2 ∈ L2 . Now,

x ∨ (y ∧ z)
= (a1 , a2 ) ∨ ((b1 , b2 ) ∧ (c1 , c2 ))
= (a1 , a2 ) ∨ (b1 ∧ c1 , b2 ∧ c2 )
= (a1 ∨ (b1 ∧ c1 ), a2 ∨ (b2 ∧ c2 ))
= ((a1 ∨ b1 ) ∧ (a1 ∨ c1 ), (a2 ∨ b2 ) ∧ (a2 ∨ c2 ))
(since L1 and L2 are distributive lattices)
= ((a1 ∨ b1 ), (a2 ∨ b2 )) ∧ ((a1 ∨ c1 ), (a2 ∨ c2 ))
= ((a1 , a2 ) ∨ (b1 , b2 )) ∧ ((a1 , a2 ) ∨ (c1 , c2 ))
= (x ∨ y) ∧ (x ∨ z).

Hence, for all x, y, z ∈ L1 × L2 , x ∨ (y ∧ z) = (x ∨ z) ∧ (x ∨ z).


Therefore, if L1 and L2 are distributive lattices, then the direct
product L1 × L2 is also a distributive lattice.
9. Prove that the lattice is modular.
1

a b c

0
Hasse diagram of given lattice
Solution.
The elements a, b, and c are symmetric in the lattice. It is enough
to prove for any one of a, b, c.
We have the cases a < 1 and 0 < a.
Case (i): Let a < 1.
Let x1 = a and x3 = 1. Then
x1 ∨ (x2 ∧ x3 ) = a ∨ (x1 ∧ 1) = a ∨ x2
and (x1 ∨ x2 ) ∧ x3 = (a ∨ x2 ) ∧ 1 = a ∨ x2 .
Hence, x1 ∨ (x2 ∧ x3 ) = (x1 ∨ x2 ) ∧ x3 .
Case (ii): Let 0 < a.
Let x1 = 0 and x3 = a. Then
x1 ∨ (x2 ∧ x3 ) = 0 ∨ (x2 ∧ a) = x2 ∧ a
and (x1 ∨ x2 ) ∧ x3 = (0 ∨ x2 ) ∧ a = x2 ∧ a.
Hence, x1 ∨ (x2 ∧ x3 ) = (x1 ∨ x2 ) ∧ x3 .
Therefore, the above lattice is modular.
Lattices and Boolean Algebra 247

5.4.2 Problems for Practice


1. Find the complements, if they exist, of the elements a, b, c of the
lattice, whose Hasse diagram is given below. Can the lattice be
complemented?
1

d e

a b

2. Give an example for a distributive and complemented lattice.


3. Examine whether the lattice given in the following Hasse diagram
is distributive or not.
1

a b c

4. In a distributive complemented lattice, show that the following are


equivalent:
(i) a≤b
(ii) a ∧ b̄ = 0
(iii) ā ∨ b = 1
(iv) b̄ ≤ ā.
5. Let hL, ≤, ∨, ∧i be a distributive lattice and a, b ∈ L if a ∧ b = a ∧ c
and a ∨ b = a ∨ c. Then, show that b = c.
6. Define a lattice. Give a suitable example.
7. In a complemented and distributive lattice, prove that the
complement of each element is unique.
8. State modular inequality of lattices.
9. Show that cancellation laws are valid in a distributive lattice.
248 Discrete Mathematical Structures

5.5 Boolean Algebra


Definition 5.5.1 Boolean Algebra: A Boolean algebra is a complemented
distributive lattice.
A Boolean algebra will generally be denoted by hB, ?, ⊕, 0 , 0, 1i, and it satisfies
the following properties in which a, b, and c denote any element of the set B.
1. hB, ?, ⊕, 0 , 0, 1i is a lattice and satisfies the following:
(i) a?a=a
(ii) a?b=b?a
(iii) (a ? b) ? c = a ? (b ? c)
(iv) a ? (a ⊕ b) = a
(v) a⊕a=a
(vi) a⊕b=b⊕a
(vii) (a ⊕ b) ⊕ c = a ⊕ (b ⊕ c)
(viii) a ⊕ (a ? b) = a.
2. hB, ?, ⊕i is a distributive lattice and satisfies the following:
(i) a ? (b ⊕ c) = (a ? b) ⊕ (a ? c)
(ii) a ⊕ (b ? c) = (a ⊕ b) ? (a ⊕ c)
(iii) (a ? b) ⊕ (b ? c) ⊕ (c ? a) = (a ⊕ b) ? (b ⊕ c) ? (c ⊕ a)
(iv) (a ? b) = (a ? c) and (a ⊕ b) = (a ⊕ c) =⇒ b = c.
3. hB, ?, ⊕, 0 , 0, 1i is a bounded lattice and satisfies the following:
(i) 0≤a≤1
(ii) a?0=0
(iii) a?1=a
(iv) a⊕0=a
(v) a ⊕ 1 = 1.
4. hB, ?, ⊕, 0 , 0, 1i is a complemented lattice in which the complement
of any element a ∈ B is denoted by a0 ∈ B and satisfies the
following:
(i) a ? a0 = 0
(ii) a ⊕ a0 = 1
(iii) 00 = 1
(iv) 10 = 0
(v) (a ? b)0 = a0 ⊕ b0
(vi) a ⊕ b0 = a0 ? b0 .
Lattices and Boolean Algebra 249

5. There exists a partial ordering ≤ on B such that


(i) a ? b = GLB{a, b}
(ii) a ⊕ b = LU B{a, b}
(iii) a ≤ b ⇐⇒ a ? b = a ⇐⇒ a ⊕ b = b
(iv) a ≤ b ⇐⇒ a ? b0 = 0 ⇐⇒ b0 ≤ a0 ⇐⇒ a0 ⊕ b = 1.

Example 5.5.2 Let A = {a, b, c} and consider the lattice hP (A), ∩, ∪i as


shown below.

{a,b,c}

{a,b} {a,c} {b,c}

{a} {b} {c}

Hasse diagram of hP (A), ∩, ∪i

Clearly, hP (A), ∩, ∪i is a Boolean algebra.

Example 5.5.3 Let B = {0, 1} be a set. The operations ?, ⊕,0 on B are


defined in the table below.
Tables showing Operations of ?, ⊕,0 on B
? 0 1 ⊕ 0 1 x x0
0 0 0 0 0 1 0 1
1 0 1 1 1 1 1 0

Clearly, hB, ?, ⊕,0 , 0, 1i is a Boolean algebra.

Definition 5.5.4 Sub-Boolean Algebra: Let hB, ?, ⊕,0 , 0, 1i be a Boolean


algebra and S ⊆ B. If S contains the elements 0 and 1 and is closed under
the operations ?, ⊕, 0 , then hS, ?, ⊕,0 , 0, 1i is called a sub-Boolean algebra.

Remark 5.5.5 A sub-Boolean algebra of a Boolean algebra is itself a Boolean


algebra.

Remark 5.5.6 A subset of a Boolean algebra can be a Boolean algebra.


However, it may not be a sub-Boolean algebra because it may not close with
respect to the operations in the Boolean algebra.
250 Discrete Mathematical Structures

Definition 5.5.7 Direct Product of Boolean Algebra: Let


hB1 , ?1 , ⊕1 ,0 , 01 , 11 i and hB2 , ?2 , ⊕2 ,00 , 02 , 12 i be any two Boolean algebras.
The direct product of the two Boolean algebras is defined to be a Boolean
algebra that is given by hB1 × B2 , ?3 , ⊕3 ,000 , 03 , 13 i in which the following
operations are defined for any (a1 , b1 ), (a2 , b2 ) ∈ B1 × B2 as

(a1 , b1 ) ?3 (a2 , b2 ) = ((a1 ?1 a2 ), (b1 ?2 b2 ))


(a1 , b1 ) ⊕3 (a2 , b2 ) = ((a1 ⊕1 a2 ), (b1 ⊕2 b2 ))
(a1 , b1 )000 = (a01 , b001 )
03 = (01 , 02 ) and 13 = (11 , 12 ).

Definition 5.5.8 Join-irreducible: Let hL, ?, ⊕i be a lattice. An element


a ∈ L is called join-irreducible if it cannot be expressed as the join of two
distinct elements of L.
In other words, a ∈ L is join-irreducible, if for any a1 , a2 ∈ L,
a = a1 ⊕ a2 =⇒ (a = a1 ) ? (a = a2 ).

Definition 5.5.9 Boolean Homomorphism: Let hB, ?, ⊕,0 , 0, 1i and


hP (A), ∪, ∩,c , α, βi be any two Boolean algebras, where A is a set. Then,
a mapping f : B −→ P (A) is called a Boolean homomorphism, if for any
a, b ∈ B,

f (a ? b) = f (a) ∩ f (b)
f (a ⊕ b) = f (a) ∪ f (b)
f (a0 ) = [f (a)]c
f (0) = α
f (1) = β.

Remark 5.5.10 The binary operations ? and ⊕ are preserved under Boolean
homomorphism.

Remark 5.5.11 Let hL, ?, ⊕, ≤i and hS, ∧, ∨, ≤0 i be two Boolean algebras.


Then, a mapping g : L =⇒ S is called an order homomorphism, then
a ≤ b =⇒ g(a) ≤0 g(b), for all a, b ∈ L.

Theorem 5.5.12 In a Boolean algebra, De Morgan’s laws hold.

Proof.
Let hL, ?, ⊕,− , 0, 1i be a Boolean algebra. Then, L is a complemented and
distributive lattice.
De Morgan’s laws are
a⊕¯ b = ā ? b̄, a ¯? b = ā ⊕ b̄, for all ā, a, b ∈ L.
Assume that a, b ∈ L. There exist elements ā, b̄ ∈ L such that
a ⊕ ā = 1, a ? ā = 0, b ⊕ b̄ = 1, b ? b̄ = 0.
Lattices and Boolean Algebra 251
¯ b = ā ? b̄.
(i) Claim: a ⊕
(a ⊕ b) ⊕ (ā ? b̄) = [(a ⊕ b) ⊕ ā] ? [(a ⊕ b) ⊕ b̄]
= [a ⊕ ? a ⊕ b] ? [a ⊕ b ⊕ b̄]
= [a ⊕ b] ? [a ⊕ a]
= 1 ? 1 = 1.
(a ⊕ b) ? (ā ? b̄) = [(a ⊕ b) ? ā] ? [(a ⊕ b) ? b̄]
= [(a ? ā) ⊕ (b ? ā)] ? [(a ? b̄) ⊕ (b ? b̄)]
= [0 ⊕ (b ? ā)] ? [(a ? b̄) ⊕ 0]
= (b ? ā) ? (a ? b̄)
= b ? (ā ? a) ? b = b̄ ? 0 ? b̄ = 0.
Hence, claim (i) is proved.
(ii) Claim: a ¯? b = ā ⊕ b̄.
(a ? b) ⊕ (ā ⊕ b̄) = [(a ? b) ⊕ ā] ⊕ [(a ? b) ⊕ b̄]
= [(a ⊕ ā) ? (b ⊕ ā)] ⊕ [(a ⊕ b̄) ? (b ⊕ b̄)]
= [1 ? (b ⊕ ā)] ⊕ [(a ⊕ b̄) ? 1]
= (b ⊕ ā) ⊕ (a ⊕ b̄)
= b ⊕ (ā ⊕ a) ⊕ b̄
= b ⊕ 1 ⊕ b̄ = b ⊕ b̄ = 1.
(a ? b) ? (ā ⊕ b̄) = [(a ? b) ? ā] ⊕ [(a ? b) ? b̄]
= (a ? ā ? b) ⊕ (a ? b ? b̄)
= (0 ? b) ⊕ (a ? 0)
= 0 ? 0 = 0.
Hence, claim (ii) is proved.
Therefore, De Morgan’s laws are proved.
Theorem 5.5.13 In a Boolean algebra hL, ?, ⊕i, the complement ā of any
element a ∈ L is unique.
Proof.
Let a ∈ L have two complements b, c ∈ L.
By definition, we have a ? b = 0, a ⊕ b = 1, a ? c = 0, a ⊕ c = 1.
Then, we have
b=b?1
= b ? (a ⊕ c)
= (b ? a) ⊕ (b ? c)
= 0 ⊕ (b ? c)
=b?c (5.22)
252 Discrete Mathematical Structures

and
c=c?1
= c ? (a ⊕ b)
= (c ? a) ⊕ (c ? b)
= 0 ⊕ (c ? b)
=c?b
= b ? c. (5.23)
From (5.22) and (5.23), we have b = c.
Therefore, every element of L has a unique complement.

5.5.1 Solved Problems


1. Show that hP (A), ∪, ∩, ⊆i is a Boolean algebra, where A is any set.
Solution.
We know that hP (A), ∪, ∩, ⊆i is a lattice.
For any X, Y, Z ∈ P (A),
X ∩ (Y ∪ Z) = (X ∩ Y ) ∪ (X ∩ Z)
X ∪ (Y ∩ Z) = (X ∪ Y ) ∩ (X ∪ Z).
Also, for all X ∈ P (A), there exists a subset X̄ of A such that
X ∪ X̄ = A, X ∩ X̄ = { } = φ.
Zero element of P (A) is { } = least element.
The greatest element of P (A) is A.
Therefore, hP (A), ∪, ∩, ⊆i is a Boolean algebra.
2. Show that in any Boolean algebra,
(a + b)(a0 + c) = ac + a0 b + bc.
Solution.
Let hB, +, ·, 0 i be a Boolean algebra. Let a, b, c ∈ B.
(a + b)(a0 + c) = (a + b)a0 + (a + b)c
= aa0 + ba0 + ac + bc
= 0 + a0 b + ac + bc
= ac + a0 b + bc.

3. In any Boolean algebra, show that a = b if and only if ab̄ + āb = 0.


Solution.
Let hB, +·, − , 0, 1i be any Boolean algebra. Let a, b ∈ B and a = b.
To show that: ab̄ + āb = 0.
a · b̄ + ā · b = aā + āa = 0 + 0 = 0.
Now, let ab̄ + āb = 0, for all a, b ∈ B. Then
ab̄ + āb = 0
Lattices and Boolean Algebra 253

=⇒ aā + ab̄ + āb + bb̄ = 0


=⇒ a(ā + b̄) + b(ā + b̄) = 0
=⇒ (a + b)(ā + b̄) = 0
=⇒ (a + b)ab¯ =0
=⇒ ab = a + b
=⇒ GLB{a, b} = LU B{a, b}
=⇒ a = b.

4. Simplify (i) (a ? b)0 ⊕ (a ⊕ b)0


(ii) (a0 ? b0 ? c) ⊕ (a ? b0 ? c) ⊕ (a ? b0 ? c0 ).
Solution.

(i) (a ? b)0 ⊕ (a ⊕ b)0 = (a0 ⊕ b0 ) ⊕ (a0 ? b0 )


= (a0 ⊕ b0 ⊕ a0 ) ? (a0 ⊕ b0 ⊕ b0 )
= (a0 ⊕ b0 ) ? (a0 ? b0 )
= a0 ? b0 .
(ii) (a0 ? b0 ? c) ⊕ (a ? b0 ? c) ⊕ (a ? b0 ? c0 )
= (a0 ⊕ a) ? (b0 ? c)
= 1 ? (b0 ? c) = b0 ? c.

5. Let a, b, c be any elements in a Boolean algebra B. Prove that

(i) a ? a = a (ii) a ⊕ a = a.

Solution.
(i) To prove: a ? a = a.

Let a=a?1 (by identity law)


= a ? (a ⊕ a0 ) (by complement law)
= a ? a ⊕ a ? a0 (by distributive law)
= (a ? a) ⊕ 0 (by complement law)
= a ? a. (by identity law)

(ii) To prove: a ⊕ a = a.

Let a=a⊕0 (by identity law)


= a ⊕ (a ? a0 ) (by complement law)
= (a ⊕ a) ? (a ⊕ a0 ) (by distributive law)
= (a ⊕ a) ? 1 (by complement law)
= a ⊕ a. (by identity law)
254 Discrete Mathematical Structures

6. Let a, b, c be any elements in a Boolean algebra B. Show that

(i) a ⊕ a = 1 (ii) a ? 0 = 0.

Solution.
(i) To prove: a ⊕ 1 = 1.

Let a ⊕ 1 = (a ⊕ 1) ? 1 (by identity law)


= (a ⊕ 1) ? (a ⊕ a0 ) (by complement law)
= a ⊕ (a ? a0 ) (by distributive law)
= a ⊕ (a0 ? 1) (by commutative law)
= a ⊕ a0 (by identity law)
= 1. (by complement law)

(ii) To prove: a ? 0 = 0.

Let a ? 0 = (a ? 0) ⊕ 0 (by identity law)


= (a ? 0) ⊕ (a ? a0 ) (by complement law)
= a ? (0 ⊕ a0 ) (by distributive law)
= a ? (a0 ⊕ 0) (by commutative law)
= a ? a0 (by identity law)
= 0. (by complement law)

7. Prove that a ⊕ (a0 ? b) = a ⊕ b.


Solution.
a ⊕ (a0 ? b) = (a ⊕ a0 ) ? (a ⊕ b) = 1 ? (a ⊕ b) = a ⊕ b.
8. Prove that a ? (a0 ⊕ b) = a ? b.
Solution.
a ? (a0 ⊕ b) = (a ? a0 ) ⊕ (a ? b) = 0 ⊕ (a ? b) = a ? b.
9. Prove that (a ? b) ⊕ (a ? b0 ) = a.
Solution.
(a ? b) ⊕ (a ? b0 ) = a ? (b ⊕ b0 ) = a ? 1 = a.
10. In any Boolean algebra, hB, ·, +, 0 , 0, 1i, show that
(a + b0 )(b + c0 )(c + a0 ) = (a0 + b)(b0 + c)(c0 + a).
Solution.

(a + b0 )(b + c0 )(c + a0 ) = (a + b0 + 0)(b + c0 + 0)(c + a0 + 0)


= (a + b0 + cc0 )(b + c0 + aa0 )(c + a0 + bb0 )
= (a + b0 + c)(a + b0 + c0 )(b + c0 + a)
(b + c0 + a0 )(c + a0 + b)(c + a0 + b0 )
= [(a0 + b + c)(a0 + b + c0 )]
Lattices and Boolean Algebra 255

[(b0 + c + a)(b0 + c + a0 )]
[(c0 + a + b)(c0 + a + b0 )]
= (a0 + b + cc0 )(b0 + c + aa0 )(c0 + a + bb0 )
= (a0 + b + 0)(b0 + c + 0)(c0 + a + 0)
= (a0 + b)(b0 + c)(c0 + a).

11. In any Boolean algebra, hB, ·, +, 0 , 0, 1i, show that


a = 0 ⇐⇒ ab0 + a0 b = b.
Solution.
If a = 0, then it directly follows that
ab0 + a0 b = 0 + 1b = 0 + b = b.
Suppose b = ab0 + a0 b. (5.24)
Therefore, 0 = b0 b = b0 (ab0 + a0 b) = ab0 + 0 = ab0 .
Using De Morgan’s law, from (5.24) we obtain b0 (a0 + b)(a + b0 ).
Therefore,
0 = ab0 = a(a0 + b)(a + b0 )
= (aa0 + ab)(a + b0 )
= (0 + ab)(a + b0 )
= ab(a + b0 )
= aba + abb0 = ab + 0 = ab.
Therefore, 0 = ab = ab0 .
Therefore, 0 = ab + ab0 = a(b + b0 ) = a1 = a.
Hence, a = 0.

5.5.2 Problems for Practice


1. What values of the Boolean variables x and y satisfy xy = x + y?
2. Show that De Morgan’s laws hold in a Boolean algebra. That is,
show that for all x and y, x ∨ y = x ∧ y and x ∧ y = x ∨ y.
3. Does a Boolean algebra contain six elements? Justify your answer.
4. If P (S) is the power set of a non-empty set S, prove that
hP (S), ∪, ∩,c , φ, Si is a Boolean algebra.
5. Prove that in a Boolean algebra, (a ∨ b)0 = a0 ∧ b0 .
6. Give an example of a two-element Boolean algebra.
7. Write the Boolean algebra whose Hasse diagram is a chain.
8. Is there a Boolean algebra with five elements? Justify your answer.
9. Show that a lattice homomorphism on a Boolean algebra which
preserves 0 and 1 is a Boolean homomorphism.
256 Discrete Mathematical Structures

10. Prove that a lattice with five elements is not a Boolean algebra.
11. Show that in a Boolean algebra, a ⊕ (a0 ? b) = a ⊕ b.
12. Show that in a Boolean algebra, a ? (a0 ⊕ b) = a ? b.
13. Show that in a Boolean algebra, (a ? b) ⊕ (a ? b0 ) = a.
14. Show that in a Boolean algebra, (a ? b ? c) ⊕ (a ? b) = a ? b.
15. Show that in a Boolean algebra, a ≤ b =⇒ a + bc = b(a + c).
16. Simplify the Boolean expression: (a ? c) ⊕ c ⊕ [(b ⊕ b0 ) ? a].
17. Simplify the Boolean expression: (1 ? a) ⊕ (0 ? a0 ).
Bibliography

G. Balaji, Discrete Mathematics, G. Balaji Publishers, Chennai, India, 2017.


R. K. Bisht and H. S. Dhami, Discrete Mathematics, Oxford University Press,
New Delhi, India, 2015.
Susanna S. Epp, Discrete Mathematics with Applications, Fourth Edition,
Brooks/Cole, Cengage Learning, U.S.A., 2011.
Rowan Garnier and John Taylor, Discrete Mathematics: Proofs, Structures
and Applications, Third Edition, CRC Press, Boca Raton, FL, 2010.
Seymour Lipschutz and Marc Lipson, Discrete Mathematics, Third Edition,
Schaum’s Outlines, Tata McGraw-Hill Company, New Delhi, India, 2007.
Kenneth H. Rosen, Discrete Mathematics and its Applications, Seventh
Edition, McGraw-Hill Education, New York, U.S.A., 2007.
J. P. Tremblay and R. Manohar, Discrete Mathematical Structures with
Applications to Computer Science, Tata McGraw-Hill Publishing
Company Limited, New Delhi, India, 2008.

257
Index

A Complemented lattice, 240


Abelian group, 174, 184 Complete bipartite graph, 144
Absorption law, 233 Complete graph, 142
Adjacency matrix, 149 Complete lattice, 240
Adjacent edges, 137 Component, 158
Adjacent vertices, 137 Conditional statement, 3
Algebraic structure, 174 Congruence relation, 175
Algebra, 173 Conjunction, 2
Algebraic system, 173 Connected graph, 158
Associative property, 173 Complement element, 240
Connectivity, 156
B Contingency, 6
Biconditional, 4 Contradiction, 6
Binary operation, 173 Cosets, 197
Bipartite graph, 143 Cycle, 157
Block, 158 Cycle graph, 142
Boolean algebra, 248 Cyclic group, 193
Boolean homomorphism, 250 Cyclic permutation, 209
Bounded variable, 23
D
C Degree of a vertex, 138
Cancellation property, 174 Directed graph, 138
Canonical form, 9 Direct product of Boolean
Cayley’s representation algebra, 250
theorem, 196 Direct product of lattices, 239
Chain, 231 Direct proof, 22
Characteristic equation, 91 Disconnected graph, 158
Chinese postman problem, 163 Disjoint cycles, 209
Circuit, 157 Disjunction, 3
Circular path, 157 Distributive lattice, 241
Closed walk, 157 Distributive properties, 174
Closure property, 173 Duality law, 7
Combination, 80 Dual lattices, 231
Combinatorics, 39
Commutative group, 184 E
Commutative ring, 219 Elementary cycle, 157
Commutativity, 173 Elementary path, 157

259
260 Index

Endomorphism, 196 Integral domain, 218


Eulerian circuit, 161 Isolated vertex, 137
Eulerian graph, 161 Isomorphic graphs, 151
Eulerian path, 161 Isomorphism, 196
Eulerian trail, 161
Even permutation, 210 J
Existence of identity, 173, 183 Join-irreducible, 250
Existence of inverse, 173, 183
Existential generalization, 28 K
Existential specification, 28 Kernel of a homormorphism, 195
Konisberg bridge, 168
F L
Factor group, 199
Field, 218 Lagrange’s theorem, 198
Free variable, 23 Lattice, 231
Fundamental theorem, 202 Lattice automorphism, 239
Lattice endomorphism, 239
Lattice homomorphism, 239
G Lattice isomorphism, 239
Generalized pigeonhole
Left coset, 197
principle, 58
Linearly ordered set, 231
Generating function, 103
Linear recurrence relation, 88
Graph colouring, 145
Lower bound, 231
Graph isomorphism, 149
Graphs, 135 M
Group, 183 Mathematical induction, 39
Method of contradiction, 19
H Method of contrapositive, 21
Hamiltonian circuit, 163 Mixed graph, 138
Hamiltonian path, 163 Modular lattice, 241
Handshaking theorem, 139 Monoid, 174
Hasse diagram, 224 Multigraph, 138
Homogenous recurrence
relation, 88 N
Homomorphism, 195 Negation, 2
Normal form, 9
I Normal subgroup, 199
Idempotent element, 174
Idempotent law, 232 O
Incidence matrix, 150 Odd permutation, 210
Inclusion-exclusion principle, 117 Open walk, 157
In-degree of a vertex, 139 Order isomorphic partially
Indirect method, 19 ordered set, 240
Inference theory, 13 Order of a group, 193
Inference theory for predicate Order-preserving mapping, 240
calculus, 28 Out-degree of a vertex, 139
Index 261

P Star graph, 144


Parallel edges, 136 Strong induction, 57
Partially ordered set, 223 Sub-Boolean algebra, 249
Partial order relation, 223 Subgraph, 142
Path, 157 Subgroup, 192
Permutation, 70 Sublattice, 238
Permutation function, 208 Subring, 218
Permutation group, 185 Substitution property, 175
Permutation with repetition, 71 Symmetric group, 185
Pigeonhole principle, 58
Poset, 223 T
Predicate calculus, 22 Tautology, 6
Principle of mathematical Terminal vertex, 157
induction, 39 Totally ordered set, 223
Proposition, 1 Transposition, 210
Pseudo graph, 138 Trivial proof, 21
Truth table, 2
Q
Quantifier, 23 U
Quotient group, 199 Undirected graph, 138
Unicursal graph, 164
R Universal generalization, 28
Recurrence relation, 87
Universal specification, 28
Regular graph, 142
Universe of discourse, 23
Right coset, 197
Upper bound, 231
Ring, 217
Ring homomorphism, 218
Rules of inference, 14
V
Vacuous proof, 22
S
Self loop, 136 W
Semigroup, 174 Walk, 156
Simple graph, 137 Well-ordering property, 57
Simple path, 157 Well-ordered set, 224
Special lattices, 240 Wheel graph, 143
Taylor & Francis eBooks
www.taylorfrancis.com

A single destination for eBooks from Taylor & Francis


with increased functionality and an improved user
experience to meet the needs of our customers.

90,000+ eBooks of award-winning academic content in


Humanities, Social Science, Science, Technology, Engineering,
and Medical written by a global network of editors and authors.

TAYLO R & FRA N CIS EBO OKS O FFERS :

Improved
A streamlined A single point search and
experience for of discovery discovery of
our library for all of our content at both
customers eBook content book and
chapter level

REQUEST A FREE TRIAL


support@taylorfrancis.com

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy