Signal Processing

and
Linear Systems

002110242

E.T.S.I.I. e I.T.lI.T.I.G.E.T.
BIBLIOTECA UN / EHU UBURUTEGIA
N2 Registro .. :~~.Zl. ... fecho .~;;.\;;Q:L...
N2 Entrada ... ~."!lJ.& .... Dplo ..~.'"t.f.d ..

') I

~r;

"

'...!

L Ii T

S19 .... .1:2.'<0.[: ..... ;':':.::; ; .. .... .. ..

Signal Processing
and
Linear Systems

B. P. Lathi

ij[]
Berkeley Cambridge Press
Carmichael, California

Preface
SIGNAL PROCESSING AND LINEAR SYSTEMS
Copyright @1998 by Berkeley-Cambridge Press.
All rights reserved. Printed in the United States of America. No part of this
book may be used or reproduced in any form or by any means without written
permission, except in the case of brief quotations embodied in critical articles and
reviews. For information address:
Berkeley-Cambridge Press
P.O. Box 947, Carmichael, CA. 95609-0947

10

This book presents a comprehensive treatment of signals and linear systems

suitable for juniors and seniors in electrical engineering. The book contains most of
the material from my earlier popular book Linear Systems and Signals (1992) with
added chapters on analog and digital filters and digital signal processing. There
are also additional applications to communications and controls. The sequence of
topics in this book is somewhat different from the earlier book. Here, the Laplace
transform follows Fourier, whereas in the 1992 book, the sequence was the exact
opposite. Moreover, the continuous-time and the discrete-time are treated sequentially, whereas in the 1992 book, both approaches were interwoven. The book
contains enough material in discrete-time systems so that it can be used not only
for a traditional course in Signals and Systems, but also for an introductory course
in Digital Signal processing.
A perceptive author has said: "The function of a teacher is not so much to
cover the topics of study as to uncover them for the students." The same can be
said of a textbook. This book, as all my previous books, emphasizes the physical
appreciation of concepts rather than mere mathematical manipulation of symbols.
There is a temptation to treat an engineering subject, such as this, as a branch of
applied mathematics. This view ignores the physical meaning behind various results
and derivations, which deprives a student of intuitive understanding of the subject.
I have used mathematics not so much to prove an axiomatic theory as to enhance
the physical and intuitive understanding. Wherever possible, theoretical results
are interpreted heuristicallyt and are supported by carefully chosen examples and
analogies.j:
Notable Features

Library of congress cataloging in pUblication data

Lathi, B. P. (Bhagawandas Pannalal)
Signal Processing and Linear Systems
Includes bibliographical references and index.
1. Signal processing--Mathematics. 2. System analysis
3. Linear time-invariant systems. 4. Digital filters (Mathematics)
I. Title.

TK5102.9.L38 1998
ISBN 0-941413-35-7

621.382'2-dc21

Cover and graphic design by Jeff Clinton.

Printed in the United States of America
iv

98-10609
CIP

The notable features of the book include the following:

1. Emphasis on intuitive and heuristic understanding of the concepts and physical meaning of mathematical results leading to deeper appreciation and easier
comprehension of the concepts. As one reviewer put it, "One thing I found very
appealing about this book is great balance of mathematical and intuitive explanation." Most reviewers of the book have noted the reader friendly character
of the book with unusual clarity of presentation.
2. The book provides extensive applications in the areas of communication, controls, and filtering.
3. For those who like to get students involved with computers, computer solutions of several examples are provided using MATLAJ3, which is becoming a
tHeuristic [Greek heuriskein, to invent, discover]: a method of education in which the pupil is
trained to find out things for himself. The word 'Eureka' (I have found it) is the 1st pers. perf.
indic. act., of heuriskein.

j:If these lines appear familiar to you, there is a good reason. 1 have used them in the preface of
some of my earlier books, including Signals, Systems, and Communication (Wiley, 1965). What
is mOre interesting, many other authors also have borrowed them for their preface.
v

az
Preface
standard software package in an electrical engineering curriculum.
4. Many students are handicapped by an inadequate background in basic material
such as complex numbers, sinusoids, sketching signals, Cramer's rule, partial
fraction expansion, and matrix algebra. I have added a chapter that addresses
these basic and pervasive topics in electrical engineering. Response by student
has been unanimously enthusiastic.
5. There are over 200 worked examples along with exercises (with answers) for students to test their understanding. There are also about 400 selected problems
of varying difficulty at the end of the chapters. Many problems are provided
with hints to steer a student in the proper direction.
6. The discrete-time and continuous-time systems are covered sequentially, with
flexibility to teach them concurrently if so desired.
7. The summary at the end of each chapter proves helpful to students in summing
up essential developments in the chapter, and is an effective tool in the study
for tests. Answers to selected problems are helpful in providing feedback to
students trying to assess their knowledge.
8. There are several historical notes to enhance student's interest in the subject.
These facts introduce students to historical background that influenced the
development of electrical engineering.
Organization

The book opens with a chapter titled Background, which deals with the mathematical background material that a student taking this course is expected to have
already mastered. It includes topics such as complex numbers, sinusoids, sketching
signals, Cramer's rule, partial fraction expansion, matrix algebra. The next 7 chapters deal with continuous-time signals and systems followed by 5 chapters treating
discrete-time signals and systems. The last chapter deals with state-space analysis.
There are MATLAB examples dispersed throught the book. The book can be readily tailored for a variety of courses of 30 to 90 lecture hours. It can also be used as
a text for a first undergraduate course in Digital signal Processing (DSP).
The organization of the book permits a great deal of flexibility in teaching the
continuous-time and discrete-time concepts. The natural sequence of chapters is
meant for a sequential approach in which all the continuous-time analysis is covered first, followed by discrete-time analysis. It is also possible to integrate (interweave) continuous-time and discrete-time analysis by using a appropriate sequence
of chapters.
Credits

The photographs of Gauss (p. 3), Laplace (p. 380), Heaviside (p. 380), Fourier
(p. 188), Michelson (p. 206) have been reprinted courtesy of the Smithsonian Institution. The photographs of Cardano (p. 3) and Gibbs (p. 206) have been reprinted
courtesy of the Library of Congress. Most of the MATLAB examples were prepared
by Dr. O. P. Mandhana of IBM, Austin, TX.
Acknowledgments

Several individuals have helped me in the preparation of this book. I am grateful for the helpful suggestions of the reviewers Professors. Dwight Day (Kansas State

vii

Preface

University), Prof. Mark Herro (University of Notre Dame), Hua Lee (University of
California, Santa Barbara), Tina Tracy (University of Missouri, Columbia), J.K.
Thgnait (Auburn University), R.L. Thmmala (Michigan State University). lowe
Dr. O.P. Mandhana a debt of gratitude for his helpful suggestions and his painstaking solutions to most of the MATLAB problems. Special thanks go to Prof. James
Simes for generous help with computer solution of several problems. I am much
obliged to Ing Ming Chang for his enthusiastic and crucial help in solving MATLAB problems and using computer to prepare the manuscript. Finally I would like
to mention the enormous but invisible sacrifices of my wife Rajani in this endeavor.
B. P. Lathi

MATLAB
Throughout this book, examples have been provided to familiarize the reader
with computer tools for systems design and analysis using the powerful and versatile
software package MATLAB. Much of the time and cost associated with the analysis
and design of systems can be reduced by using computer software packages for
simulation. Many corporations will no longer support the development systems
without prior computer simulation and numerical results which suggest a design
will work. The examples and problems in this book will assist the reader in learning
the value of computer packages for systems design and simulation.
MATLAB is the software package used throughout this book. MATLAB is a
powerful package developed to perform matrix manipulations for system designers.
MATLAB is easily expandable and uses its own high level language. These factors
make developing sophisticated systems easier. In addition, MATLAB has been
carefully written to yield numerically stable results to produce reliable simulations.
All the computer examples in this book are verified to be compatible with the
student edition of the MATLAB when used according to the instructions given in
its manual. The reader should make sure that \MATLAB\BIN is added in the DOS
search path. MATLAB can be invoked by executing the command MATLAB. The
MATLAB banner will appear after a moment with the prompt ''. MATLAB has
a useful on-line help. To get help on a specific command, type HELP COMMAND
NAME and then press the ENTER key. DIARY FILE is a command to record all
the important keyboard inputs to a file and the resulting output of your MATLAB
session to be written on the named file. MATLAB can be used interactively, or
by writing functions (subroutines) often called M files because of the .M extension
used for these files. Once familiar with the basics of MATLAB, the reader can easily
learn how to write functions and to use MATLAB's existing functions.
The MATLAB M-files have been created to supplement this text. This includes
all the examples solved by MATLAB in the text. These M-files may be retrieved
from the Mathworks anonymous FTP site at
ftp:/ / ftp.mathworks.com/pub/books/lathi/.

O. P. Mandhana

1 Introduction to Signals and Systems

1.7 Classification of Systems

Fig. 1.29 An example of a linear time-varying system.

1.7-3

lnstantaneous and Dynamic Systems

As observed earlier, a system's output a t any instant t generally depends upon

the entire past input. However, in a special class of systems, the output at any
instant t depends only on its input at that instant. In resistive networks, for example, any output of the network at some instant t depends only on the input at the
instant t. In these systems, past history is irrelevant in determining the response.
Such systems are said to be instantaneous or memoryless systems. More precisely, a system is said to be instantaneous (or memoryless) if its output at any
instant t depends, a t most, on the strength of its input(s) at the same instant but
not on any past or future values of the input(s). Otherwise, the system is said to be
dynamic (or a system with memory). A system whose response a t t is completely
determined by the input signals over the past T seconds [interval from (t - T) to
t] is a finite-memory system with a memory of T seconds. Networks containing inductive and capacitive elements generally have infinite memory because the
response of such networks a t any instant t is determined by their inputs over the
entire past (-m, t ) . This is true for the R C circuit of Fig. 1.26.
In this book we will generally examine dynamic systems. Instantaneous systems
are a special case of dynamic systems.

1.7-4

Causal and Noncausal Systems

A causal (also known as a physical or non-anticipative) system is one for

which the output at any instant to depends only on the value of the input f (t) for
t 5 to. In other words, the value of the output at the present instant depends only
on the past and present values of the input f (t), not on its future values. To put
it simply, in a causal system the output cannot start before the input is applied. If
the response starts before the input, it means that the system knows the input in
the future and acts on this knowledge before the input is applied. A system that
violates the condition of causality is called a noncausal (or anticipative) system.
Any practical system that operates in real timet must necessarily be causal.
We do not yet know how to build a system that can respond to future inputs (inputs
not yet applied). A noncausal system is a prophetic system that knows the future
input and acts on it in the present. Thus, if we apply an input starting at t = O
to a noncausal system, the output would begin even before t = O. As an example,
consider the system specified by

tIn real-time operations, the response to an input is essentially simultaneous (contemporaneous) with the input itself.

Fig. 1.30 A noncausal system and its realization by a delayed causal system.

For the input f ( t ) illustrated in Fig. 1.30a, the output y(t), as computed from
Eq. (1.46) (shown in Fig. 1.30b), starts even before the input is applied. Equation
(1.46) shows that y(t), the output at t, is given by the sum of the input values two
seconds before and two seconds after t (at t - 2 and t + 2 respectively). But if we
are operating the system in real time at t , we do not know what the value of the
input will be two seconds later. Thus it is impossible to implement this system in
real time. For this reason, noncausal systems are unrealizable in real time.
Why Study Noncausal Systems?

From the above discussion it may seem that noncausal systems have no practical
purpose. This is not the case; they are valuable in the study of systems for several
reasons. First, noncausal systems are realizable when the independent variable is
other than "time" (e.g., space). Consider, for example, an electric charge of density
q(x) placed along the x-axis for x 2 O. This charge density produces an electric field
E(x) that is present a t every point on the x-axis from x = -m to m. In this case the
input [i.e., the charge density q(x)] starts at x = 0, but its output [the electric field
E(x)] begins before x = O. Clearly, this space charge system is noncausal. This
discussion shows that only temporal systems (systems with time as independent
variable) must be causal in order to be realizable. The terms "before" and "after"
have a special connection to causality only when the independent variable is time.
This connection is lost for variables other than time. Nontemporal systems, such
as those occurring in optics, can be noncausal and still realizable.
Moreover, even for temporal systems, such as those used for signal processing,
the study of noncausal systems is important. In such systems we may have al1 input
data prerecorded. (This often happens with speech, geophysical, and meteorological
signals, and with space probes.) In such cases, the input's future values are available
to us. For example, suppose we had a set of input signal records available for the
system described by Eq. (1.46). We can then compute y(t) since, for any t , we need
only refer to the records to find the input's value two seconds before and two seconds
after t. Thus, noncausal systems can be realized, although not in real time. We

86

1 Introduction to Signals and Systems

1.7 Classification of Systems

1.7-5

Noncausal systems are realizable with time delay!

rnay therefore b e able to realize a noncausal system, provided that we are willing
to accept a time delay in the output. Consider a system whose output $(t) is the
same as y(t) in Eq. (1.46) delayed by two seconds (Fig 1.30c), so that

Here the value of the output $ at any instant t is the sum of the values of the input
j at t and at t h e instant four seconds earlier [at ( t - 4)]. In this case, the output
at any instant t does not depend on future values of the input, and the system is
causal. The output of this system, which is $ ( t ) , is identical to that in Eq. (1.46)
or Fig. 1.30b except for a delay of two seconds. Thus, a noncausal system rnay be
realized or satisfactorily approximated in real time by using a causal system with a
delay.
A third reason for studying noncausal systems is that they provide an upper
bound on the performance of causal systems. For example, if we wish to design
a filter for separating a signal from noise, then the optimum filter is invariably a
noncausal systern. Although unrealizable, this noncausal system's performance acts
as the upper limit on what can be achieved and gives us a standard for evaluating
the performance of causal filters.
At first glance, noncausal systems rnay seem inscrutable. Actually, there is
nothing mysterious about these systems and their approximate realization through
using physical systems with delay. If we want to know what will happen one year
from now, we have two choices: go to a prophet (an unrealizable person) who can
give the answers immediately, or go to a wise man and allow him a delay of one
p a r to give us the answer! If the wise man is truly wise, he rnay even be able to
shrewdly guess the future very closely with a delay of less than a year by studying
trends. Such is the case with noncausal systems-nothing more and nothing less.
A Exercise E1.15

Lumped-Parameter and Distributed-Parameter Systems

In the study of electrical systems, we make use of voltage-current relationships

for various components (Ohm's law, for example). In doing so, we implicitly assume'
that the current in any system component (resistor, inductor, etc.) is the same at
every point throughout that component. Thus, we assume that electrical signals are
propagated instantaneously throughout the system. In reality, however, electrical
signals are electromagnetic space waves requiring some finite propagation time.
An electric current, for example, propagates through a component with a finite
velocity and therefore rnay exhibit different values at different locations in the same
component. Thus, an electric current is a function not only of time but also of
space. However, if the physical dimensions of a component are small compared to
the wavelength of the signal propagated, we rnay assume that the current is constant
throughout the component. This is the assumption made in l u m p e d - p a r a m e t e r
systems, where each component is regarded as being lumped at one point in space.
Such an assumption is justified a t lower frequencies (higher wavelength). Therefore,
in lumped-parameter models, signals can be assumed to be functions of time alone.
For such systems, the system equations require only one independent variable (time)
and therefore are ordinary differential equations.
In contrast, for d i s t r i b u t e d - p a r a m e t e r s y s t e m s such as transmission lines,
waveguides, antennas, and microwave tubes, the system dimensions cannot be assumed to be small compared to the wavelengths of the signals; thus the lumpedparameter assumption breaks down. The signals here are functions of space as
well as of time, leading to mathematical models consisting of partial differential
equations.3 The discussion in this book will be restricted to lumped-parameter systems only.

1.7-6

Continuous-Time and Discrete-Time Systems

Distinction between discrete-time and continuous-time signals is discussed in

Sec. 1.2-1. Systems whose inputs and outputs are continuous-time signals are
continuous-time systems. On the other hand, systems whose inputs and outputs are discrete-time signals are discrete-time systems. If a continuous-time
signal is sampled, the resulting signal is a discrete-time signal. We can process a
continuous-time signal by processing its samples with a discrete-time system.

1.7-7

Analog and Digital Systems

Analog and digital signals are discussed in Sec. 1.2-2. A system whose input
and output signals are analog is an analog s y s t e m ; a system whose input and
output signals are digital is a digital system. A digital computer is an example
of a digital (binary) system. Observe that a digital computer is an example of a
system that is digital as well as discrete-time.

Show that a system described by the equation below is noncausal:

Additional Classification of Systems

Chow

that this system can be realized physically if

we

accept a delay of 5 seconds in the output.

There are additional classes of systems, such as invertible and noninvertible

systems. A system S performs certain operation(s) on input signal(s). If we can
obtain the input f ( t ) back from the output y ( t ) by some operation, the system