Digital Signal Processing

Dr. Shekha Rai

Department of Electrical Engineering

National Institute of Technology Rourkela
India

1 / 23
Syllabus

An introduction to signals and systems.

Representation of signals in time domain.
Linear, time-invariant systems.
Impulse response and convolution sum.
Sampling and Reconstruction of continuous time signals.
Characterization and properties of discrete time signals and systems.
Computation of the discrete time Fourier transform and its properties.
Fast Fourier transform algorithms.
The Z-transform and its properties.
The inverse z-transform.
System function and system stability.
Transform analysis of linear time invariant systems.
Implementation of structures for discrete time systems.
Digital filter design techniques.
Design of FIR and IIR filters.
Applications of DSP.

2 / 23
Books recommended

John G.Proakis, Dimitris G. Manolakis, and D Sharma, Digital

Signal Processing: Principles, Algorithms and, Pearson
Education India , 3rd edition
A. V. Oppenheim, R. W. Schafer, Digital Signal Processing,
Pearson Education , 2004
Vinay K. Ingle, John G. Proakis, Digital Signal Processing using
MATLAB, Brooks/Cole - Thomson Learning
S. K. Mitra, Digital Signal Processing: A computer based
approach„ TMH , 2001
T. K. Rawat, Digital Signal Processing, Oxford University Press,
2015

3 / 23
An introduction to signals and systems

Signal:
Any time varying physical phenomenon that is intended to
convey information. It varies with time, space or any other
independent variable.
Examples:
Speech- telephony, radio
Biomedical signals- electrocardiogram, electroencephalogram
Images
Radar signals

4 / 23
An introduction to signals and systems

Figure: Speech signal

x-axis → time (t)

y-axis → f(t)
5 / 23
An introduction to signals and systems

Figure: Electrocardiogram signal

6 / 23
An introduction to signals and systems

Figure: 2-D signal

y → row index
x → column index
s[x,y] → pixel
7 / 23
An introduction to signals and systems

System:
It can modify the signal or extract additional information from
them.
It processes a set of input signals to obtain another set of desired
output signals
It may be
made up of physical components such as electrical, mechanical, or
hydraulic systems
algorithm that computes an output from an input

8 / 23
An introduction to signals and systems

Figure:

Input → excitation
Output → response

9 / 23
An introduction to signals and systems

Signal processing:
Analyse, modify or extract useful information from the signal.
Mathematical representation of the signal and the algorithmic
operation carried out on it to extract the information present
It may be
Analog Signal Processing (ASP)- analog signal processor (like
filter or frequency analysers)
Digital Signal Processing (DSP)- digital representation of signals
and use of digital processors
Mixed Signal Processing- Combination of both ASP and DSP

10 / 23
An introduction to signals and systems

Figure: Digital signal processing system

11 / 23
An introduction to signals and systems

Advantages of DSP over ASP:

Guaranteed accuracy
Flexibility
Superior performance
Adaptability
Drift
Perfect reproducibility
Multiplexing
Data logging
Low-frequency capability

12 / 23
An introduction to signals and systems

Disadvantages of DSP over ASP:

Increased system complexity
Speed and cost
Finite-word length effects

13 / 23
Classification of signals

Nature of independent variable, value of function defining the

signal
Continuous- time signal or analog signal → defined for
continuous values of independent variable time t
Discrete- time signal → defined only at discrete time k, and
independent variable k has only discrete set of values. x[k] is not
defined for non-integer values of k.
Digital signal→Discrete- time signal with set of discrete values

14 / 23
Classification of signals

Continuous-time and Discrete-time signals

CT sine wave
10

5
x(t)

−5

−10
0 5 10 15 20 25 30 35 40
t
DT sine wave
10
x[k]

−10
0 10 20 30 40
k
15 / 23
Classification of signals

Digital signals

Figure:

16 / 23
Classification of signals

Multi-channel signals
Multi-channel signal → signals generated from multiple sources
or sensors
if xk (t) → electrical signal from kth sensor as a function of time,
where k = 1, 2, 3 then
 
x1 (t)
x(t) = x2 (t)
x3 (t)

17 / 23
Classification of signals

Multidimensional signals
One-dimensional signal → function of one independent variable
Multi-dimensional signal → function of M independent variable,
M dimensional signal

18 / 23
Classification of signals

I(x,y) → 2-D signal, e.g picture

I(x,y,t) → 3-D signal,e.g black and white TV picture

I(x,y,t) → 3-D signal, 3-channel, e.g color TV picture

 
Ir (x, y, t)
I(x, y, t) = Ig (x, y, t)
Ib (x, y, t)

19 / 23
Classification of signals

Deterministic signal
Physical description is completely known, mathematically or
graphically
Nature and amplitude at any time is predicted

Random Signal
Values cannot be predicted precisely
Known in terms of probabilistic description, like mean,
mean-square value

20 / 23
Multiple choice questions

1. Monthly sales of a corporation is

(a) continuous time signal

(b) digital time signal

21 / 23
Multiple choice questions

1. Monthly sales of a corporation is

(a) continuous time signal

(b) digital time signal

Answer (b)

22 / 23
Thank You

23 / 23