john g. proakis masoud salehi 2nd ed. - u-cursos · john g. proakis masoud salehi 2nd ed. ......

Proakis-50210 proa-fm August 9, 2001 14:2

COMMUNICATION SYSTEMSENGINEERING

John G. Proakis

Masoud Salehi

2nd Ed.

Upper Saddle River, New Jersey 07458

i


To Felia, George, and Elena.John G. Proakis

To Fariba, Omid, Sina, and my parents.Masoud Salehi

Library of Congress Cataloging-in-Publication Data

CIP data available on file.

Vice President and Editorial Director, ECS: Marcia J. HortonPublisher: Tom RobbinsEditorial Assistant: Jody McDonnellVice President and Director of Production and Manufacturing, ESM: David W. RiccardiExecutive Managing Editor: Vince OBrienManaging Editor: David A. GeorgeProduction Editor: Chanda WakefieldDirector of Creative Services: Paul BelfantiCreative Director: Carole AnsonArt Director: Jayne ConteArt Editor: Greg DullesManufacturing Manager: Trudy PisciottiManufacturing Buyer: Lynda CastilloMarketing Manager: Holly StarkMarketing Assistant: Karen Moon

c 2002 by Prentice-Hall, Inc.Upper Saddle River, New Jersey

All rights reserved. No part of this book may be reproduced, in any formor by any means without permission in writing from the publisher.

Printed in the United States of America10 9 8 7 6 5 4 3 2 1

ISBN 0-13-061793-8

Pearson Education Ltd., LondonPearson Education Australia Pty. Ltd., SydneyPearson Education Singapore, Pte. Ltd.Pearson Education North Asia Ltd., Hong KongPearson Education Canada, Inc., TorontoPearson Educacon de Mexico, S.A. de C.V.Pearson EducationJapan, TokyoPearson Education Malaysia, Pte. Ltd.

ii


Contents

PREFACE xi

1 INTRODUCTION 1

1.1 Historical Review 1

1.2 Elements of an Electrical Communication System 41.2.1 Digital Communication System, 71.2.2 Early Work in Digital Communications, 10

1.3 Communication Channels and Their Characteristics 12

1.4 Mathematical Models for Communication Channels 19

1.5 Organization of the Book 22

1.6 Further Reading 23

2 FREQUENCY DOMAIN ANALYSIS OF SIGNALSAND SYSTEMS 24

2.1 Fourier Series 242.1.1 Fourier Series for Real Signals: the Trigonometric Fourier Series, 29

2.2 Fourier Transforms 312.2.1 Fourier Transform of Real, Even, and Odd Signals, 35

iii


iv Contents

2.2.2 Basic Properties of the Fourier Transform, 362.2.3 Fourier Transform for Periodic Signals, 39

2.3 Power and Energy 402.3.1 Energy-Type Signals, 412.3.2 Power-Type Signals, 42

2.4 Sampling of Bandlimited Signals 45

2.5 Bandpass Signals 49


Problems 57

3 ANALOG SIGNAL TRANSMISSION AND RECEPTION 70

3.1 Introduction to Modulation 70

3.2 Amplitude Modulation (AM) 713.2.1 Double-Sideband Suppressed Carrier AM, 713.2.2 Conventional Amplitude Modulation, 783.2.3 Single-Sideband AM, 813.2.4 Vestigial-Sideband AM, 853.2.5 Implementation of AM Modulators and Demodulators, 883.2.6 Signal Multiplexing, 94

3.3 Angle Modulation 963.3.1 Representation of FM and PM Signals, 973.3.2 Spectral Characteristics of Angle-Modulated Signals, 1013.3.3 Implementation of Angle Modulators and Demodulators, 107

3.4 Radio and Television Broadcasting 1153.4.1 AM Radio Broadcasting, 1153.4.2 FM Radio Broadcasting, 1163.4.3 Television Broadcasting, 120

3.5 Mobile Radio Systems 128


Problems 131

4 RANDOM PROCESSES 144

4.1 Probability and Random Variables 144

4.2 Random Processes: Basic Concepts 1594.2.1 Description of Random Processes, 1624.2.2 Statistical Averages, 1644.2.3 Stationary Processes, 1664.2.4 Random Processes and Linear Systems, 174


Contents v

4.3 Random Processes in the Frequency Domain 1774.3.1 Power Spectrum of Stochastic Processes, 1774.3.2 Transmission over LTI Systems, 183

4.4 Gaussian and White Processes 1864.4.1 Gaussian Processes, 1864.4.2 White Processes, 188

4.5 Bandlimited Processes and Sampling 192

4.6 Bandpass Processes 194


Problems 202

5 EFFECT OF NOISE ON ANALOG COMMUNICATIONSYSTEMS 217

5.1 Effect of Noise on Linear-Modulation Systems 2175.1.1 Effect of Noise on a Baseband System, 2185.1.2 Effect of Noise on DSB-SC AM, 2185.1.3 Effect of Noise on SSB AM, 2205.1.4 Effect of Noise on Conventional AM, 221

5.2 Carrier-Phase Estimation with a Phase-Locked Loop (PLL) 2255.2.1 The Phase-Locked Loop (PLL), 2265.2.2 Effect of Additive Noise on Phase Estimation, 229

5.3 Effect of Noise on Angle Modulation 2345.3.1 Threshold Effect in Angle Modulation, 2445.3.2 Pre-emphasis and De-emphasis Filtering, 248

5.4 Comparison of Analog-Modulation Systems 251

5.5 Effects of Transmission Losses and Noise in AnalogCommunication Systems 2525.5.1 Characterization of Thermal Noise Sources, 2535.5.2 Effective Noise Temperature and Noise Figure, 2545.5.3 Transmission Losses, 2575.5.4 Repeaters for Signal Transmission, 258


Problems 261

6 INFORMATION SOURCES AND SOURCE CODING 267

6.1 Modeling of Information Sources 2686.1.1 Measure of Information, 2696.1.2 Joint and Conditional Entropy, 271


vi Contents

6.2 Source-Coding Theorem 273

6.3 Source-Coding Algorithms 2766.3.1 The Huffman Source-Coding Algorithm, 2766.3.2 The Lempel-Ziv Source-Coding Algorithm, 280

6.4 Rate-Distortion Theory 2826.4.1 Mutual Information, 2836.4.2 Differential Entropy, 2846.4.3 Rate-Distortion Function, 285

6.5 Quantization 2906.5.1 Scalar Quantization, 2916.5.2 Vector Quantization, 300

6.6 Waveform Coding 3026.6.1 Pulse-Code Modulation (PCM), 3026.6.2 Differential Pulse-Code Modulation (DPCM), 3076.6.3 Delta Modulation (M), 310

6.7 Analysis-Synthesis Techniques 312

6.8 Digital Audio Transmission and Digital Audio Recording 3166.8.1 Digital Audio in Telephone Transmission Systems, 3176.8.2 Digital Audio Recording, 319

6.9 The JPEG Image-Coding Standard 323


Problems 327

7 DIGITAL TRANSMISSION THROUGH THE ADDITIVE WHITEGAUSSIAN NOISE CHANNEL 340

7.1 Geometric Representation of Signal Waveforms 341

7.2 Pulse Amplitude Modulation 345

7.3 Two-dimensional Signal Waveforms 3507.3.1 Baseband Signals, 3507.3.2 Two-dimensional Bandpass SignalsCarrier-Phase Modulation, 3547.3.3 Two-dimensional Bandpass SignalsQuadrature Amplitude

Modulation, 357

7.4 Multidimensional Signal Waveforms 3607.4.1 Orthogonal Signal Waveforms, 3607.4.2 Biorthogonal Signal Waveforms, 3657.4.3 Simplex Signal Waveforms, 3667.4.4 Binary-Coded Signal Waveforms, 367


Contents vii

7.5 Optimum Receiver for Digitally Modulated Signals in Additive WhiteGaussian Noise 3707.5.1 Correlation-Type Demodulator, 3707.5.2 Matched-Filter-Type Demodulator, 3757.5.3 The Optimum Detector, 3817.5.4 Demodulation and Detection of Carrier-Amplitude Modulated

Signals, 3867.5.5 Demodulation and Detection of Carrier-Phase Modulated Signals, 3887.5.6 Demodulation and Detection of Quadrature Amplitude Modulated

Signals, 3967.5.7 Demodulation and Detection of Frequency-Modulated Signals, 398

7.6 Probability of Error for Signal Detection in Additive WhiteGaussian Noise 4057.6.1 Probability of Error for Binary Modulation, 4057.6.2 Probability of Error for M-ary PAM, 4087.6.3 Probability of Error for Phase-Coherent PSK Modulation, 4137.6.4 Probability of Error for DPSK, 4177.6.5 Probability of Error for QAM, 4187.6.6 Probability of Error for M-ary Orthogonal Signals, 4237.6.7 Probability of Error for M-ary Biorthogonal Signals, 4287.6.8 Probability of Error for M-ary Simplex Signals, 4297.6.9 Probability of Error for Noncoherent Detection of FSK, 4307.6.10 Comparison of Modulation Methods, 432

7.7 Performance Analysis for Wireline and Radio CommunicationChannels 4367.7.1 Regenerative Repeaters, 4377.7.2 Link Budget Analysis for Radio Channels, 438

7.8 Symbol Synchronization 4427.8.1 EarlyLate Gate Synchronizers, 4437.8.2 Minimum Mean-Square-Error Method, 4457.8.3 Maximum-Likelihood Methods, 4487.8.4 Spectral Line Methods, 4497.8.5 Symbol Synchronization for Carrier-Modulated Signals, 451


Problems 453

8 DIGITAL TRANSMISSION THROUGH BANDLIMITEDAWGN CHANNELS 474

8.1 Digital Transmission through Bandlimited Channels 4748.1.1 Digital PAM Transmission through Bandlimited Baseband

Channels, 4788.1.2 Digital Transmission through Bandlimited Bandpass Channels, 480


viii Contents

8.2 The Power Spectrum of Digitally Modulated Signals 4828.2.1 The Power Spectrum of the Baseband Signal, 4838.2.2 The Power Spectrum of a Carrier-Modulated Signal, 488

8.3 Signal Design for Bandlimited Channels 4908.3.1 Design of Bandlimited Signals for Zero ISIThe Nyquist Criterion, 4928.3.2 Design of Bandlimited Signals with Controlled ISIPartial Response

Signals, 497

8.4 Probability of Error in Detection of Digital PAM 4998.4.1 Probability of Error for Detection of Digital PAM with Zero ISI, 5008.4.2 Symbol-by-Symbol Detection of Data with Controlled ISI, 5018.4.3 Probability of Error for Detection of Partial Response Signals, 504

8.5 Digitally Modulated Signals with Memory 5078.5.1 Modulation Codes and Modulation Signals with Memory, 5088.5.2 The Maximum-Likelihood Sequence Detector, 5218.5.3 Maximum-Likelihood Sequence Detection of Partial Response

Signals, 5258.5.4 The Power Spectrum of Digital Signals with Memory, 530

8.6 System Design in the Presence of Channel Distortion 5348.6.1 Design of Transmitting and Receiving Filters for a Known Channel, 5358.6.2 Channel Equalization, 538

8.7 Multicarrier Modulation and OFDM 5568.7.1 An OFDM System Implemented via the FFT Algorithm, 557


Problems 561

9 CHANNEL CAPACITY AND CODING 576

9.1 Modeling of Communication Channels 576

9.2 Channel Capacity 5799.2.1 Gaussian Channel Capacity, 583

9.3 Bounds on Communication 5869.3.1 Transmission of Analog Sources by PCM, 590

9.4 Coding for Reliable Communication 5919.4.1 A Tight Bound on Error Probability of Orthogonal Signals, 5929.4.2 The Promise of Coding, 595

9.5 Linear Block Codes 6019.5.1 Decoding and Performance of Linear Block Codes, 6069.5.2 Burst-Error-Correcting-Codes, 614

9.6 Cyclic Codes 6159.6.1 The Structure of Cyclic Codes, 615


Contents ix

9.7 Convolutional Codes 6239.7.1 Basic Properties of Convolutional Codes, 6249.7.2 Optimum Decoding of Convolutional CodesThe Viterbi

Algorithm, 6299.7.3 Other Decoding Algorithms for Convolutional Codes, 6349.7.4 Bounds on Error Probability of Convolutional Codes, 634

9.8 Complex Codes Based on Combination of Simple Codes 6389.8.1 Product Codes, 6399.8.2 Concatenated Codes, 6409.8.3 Turbo Codes, 6409.8.4 The BCJR Algorithm, 6429.8.5 Performance of Turbo Codes, 644

9.9 Coding for Bandwidth-Constrained Channels 6469.9.1 Combined Coding and Modulation, 6479.9.2 Trellis-Coded Modulation, 649

9.10 Practical Applications of Coding 6559.10.1 Coding for Deep-Space Communications, 6569.10.2 Coding for Telephone-Line Modems, 6579.10.3 Coding for Compact Discs, 658


Problems 661

10 WIRELESS COMMUNICATIONS 674

10.1 Digital Transmission on Fading Multipath Channels 67410.1.1 Channel Models for Time-Variant Multipath Channels, 67610.1.2 Signal Design for Fading Multipath Channels, 68410.1.3 Performance of Binary Modulation in Frequency Nonselective Rayleigh

Fading Channels, 68610.1.4 Performance Improvement Through Signal Diversity, 68910.1.5 Modulation and Demodulation on Frequency Selective Channels

The RAKE Demodulator, 69410.1.6 Multiple Antenna Systems and Space-Time Codes, 697

10.2 Continuous Carrier-Phase Modulation 70210.2.1 Continuous-Phase FSK (CPFSK), 70210.2.2 Continuous-Phase Modulation (CPM), 71110.2.3 Spectral Characteristics of CPFSK and CPM Signals, 71510.2.4 Demodulation and Detection of CPM Signals, 72010.2.5 Performance of CPM in AWGN and Rayleigh Fading Channels, 726

10.3 Spread-Spectrum Communication Systems 72910.3.1 Model of a Spread-Spectrum Digital Communication System, 73010.3.2 Direct-Sequence Spread-Spectrum Systems, 73110.3.3 Some Applications of DS Spread-Spectrum Signals, 742


x Contents

10.3.4 Effect of Pulsed Interference and Fading, 74610.3.5 Generation of PN Sequences, 74810.3.6 Frequency-Hopped Spread Spectrum, 75210.3.7 Synchronization of Spread-Spectrum Systems, 758

10.4 Digital Cellular Communication Systems 76610.4.1 The GSM System, 76610.4.2 CDMA System Based on IS-95, 768


Problems 775

APPENDIX A: THE PROBABILITY OF ERROR FORMULTICHANNEL RECEPTION OF BINARY SIGNALS 782

REFERENCES 785

INDEX 794


Preface

The objective of this book is to provide an introduction to the basic principles in theanalysis and design of communication systems. It is primarily intended for use as a textfor a first course in communications, either at a senior level or at a first-year graduatelevel.

BROAD TOPICAL COVERAGE

Although we have placed a very strong emphasis on digital communications, we haveprovided a review of important mathematical foundational topics and a solid introduc-tion to analog communications. The major topics covered are:

A review of frequency domain analysis of signals and systems, and the charac-terization of random processes (Chapters 2 and 4)

An introduction to analog signal transmission and reception (Chapters 3 and 5) An introduction to digital communications (Chapters 610)

EMPHASIS ON DIGITAL COMMUNICATIONS

Our motivation for emphasizing digital communications is due to the technologicaldevelopments that have occurred during the past five decades. Today, digital communi-cation systems are in common use and generally carry the bulk of our daily informationtransmission through a variety of communications media, such as wireline telephonechannels, microwave radio, fiber optic channels, and satellite channels. We are currentlywitnessing an explosive growth in the development of personal communication systems

xi


xii Preface

and ultrahigh speed communication networks, which are based on digital transmissionof the information, whether it is voice, still images, or video. We anticipate that, in thenear future, we will witness a replacement of the current analog AM and FM radio andtelevision broadcast by digital transmission systems.

The development of sophisticated, high-speed digital communication systemshas been accelerated by concurrent developments in inexpensive high speed integratedcircuits (IC) and programmable digital signal processing chips. The developments inMicroelectronic IC fabrication have made possible the implementation of high-speed,high precision A/D converters, of powerful error-correcting coders/decoders, and ofcomplex digital modulation techniques. All of these technological developments pointto a continuation in the trend toward increased use of digital communications as ameans for transmitting information.

OVERVIEW OF THE TEXT

It is assumed that students using this book have a basic understanding of linear systemtheory, both continuous and discrete, including a working knowledge of Fourier seriesand Fourier transform techniques. Chapter 2 provides a review of basic material on sig-nals and systems and establishes the necessary notation used in subsequent chapters.It is also assumed that students have had a first course in probability. Such courses arecurrently required in many undergraduate electrical engineering and computer engi-neering programs. Chapter 4 provides a review of probability and random processes tothe extent that is necessary for a first course in communications.

Chapter 3 treats modulation and demodulation of analog signals. This treatmentincludes amplitude modulation (AM), frequency modulation (FM), and phase modu-lation (PM). Radio and television broadcasting and mobile radio cellular systems arediscussed as examples of analog communication systems. Chapter 5 continues the treat-ment of analog communication systems by analyzing the effect of additive noise in thedemodulation of AM, FM, and PM signals. The phase-locked loop, which is used forestimating the phase of a sinusoidal carrier in both analog and digital communicationsystems is also described in Chapter 5. The chapter concludes with a treatment of the ef-fect of transmission losses and the characterization of noise sources in communicationsystems.

A logical beginning in the introduction of digital communication systems analysisand design is the characterization of information sources and source encoding. Chapter 6is devoted to this topic. In this chapter we introduce the reader to the modeling ofinformation sources, both discrete and continuous (analog), and the basic mathematicalconcepts of entropy and mutual information. Our discussion of source encoding fordiscrete sources includes the Huffman coding algorithm and the Lempel-Ziv algorithm.For the case of analog sources, we treat both scalar and vector quantization and describethe common waveform-coding techniques, namely, PCM, DPCM, and DM. We alsodescribe the LPC-based source modeling method. As practical examples of the source-coding methods described in this chapter we cite the digital speech transmission systems


Preface xiii

in the telephone plant, the digital audio recording systems as embodied in the compactdisc (CD) player and the JPEG image-coding standard.

Digital modulation and demodulation techniques are described in Chapter 7. Bi-nary and nonbinary modulation methods are described based on a geometric representa-tion of signals, and their error-rate performance is evaluated and compared. This chapteralso describes symbol synchronization methods for digital communication systems.

Chapter 8 treats digital transmission through bandlimited AWGN channels. In thischapter we derive the power-spectral density of linearly modulated baseband signalsand consider the problem of signal design for a bandlimited channel. We show that theeffect of channel distortion is to introduce intersymbol interference (ISI), which canbe eliminated or minimized by proper signal design. The use of linear and nonlinearadaptive equalizers for reducing the effect of ISI is also described.

Chapter 9 treats the topic of channel coding and decoding. The capacity of acommunication channel is first defined, and the capacity of the Gaussian channel isdetermined. Linear block codes and convolutional codes are introduced and appropriatedecoding algorithms are described. The benefits of coding for bandwidth constrainedchannels are also described. The final section of this chapter presents three practicalapplications of coding.

The last chapter of this book treats topics in wireless communications. First, weconsider the characterization of fading multipath channels and describe the effects ofsuch channels on wireless digital communication systems. The design of signals thatare effective in mitigating this type of channel distortion is also considered. Second, wedescribe the class of continuous-phase modulated signals, which are especially suitablefor digital communication in wireless channels. Finally, we treat the class of spread-spectrum signals, which are suitable for multi-user wireless communication systems.

EXAMPLES AND HOMEWORK PROBLEMS

We have included a large number of carefully chosen examples and homework prob-lems. The text contains over 180 worked-out examples and over 480 problems. Ex-amples and problems range from simple exercises to more challenging and thought-provoking problems. A Solutions Manual is available free to all adopting faculty, whichis provided in both typeset form and as a diskette formatted in LATEX. Solutions are notavailable for sale to students. This will enable instructors to print out solutions in anyconfiguration easily.

COURSE OPTIONS

This book can serve as a text in either a one- or two-semester course in communicationsystem. An important consideration in the design of the course is whether or not thestudents have had a prior course in probability and random processes. Another importantconsideration is whether or not analog modulation and demodulation techniques are tobe covered. Here, we outline three scenarios. Others are certainly possible.


xiv Preface

1. A one-term course in analog and digital communication: Selected review sectionsfrom Chapters 2 and 4, all of chapters 3, 5, 7, and 8, and selections from chapters 6,9, and 10.

2. A one-term course in digital communication: Selected review sections from Chap-ters 2 and 4, and Chapters 610.

3. A two-term course sequence on analog and digital communications:

(a) Chapters 26 for the first course.(b) Chapters 710 for the second course.

We wish to thank Gloria Doukakis for her assistance in the preparation of themanuscript.

John ProakisAdjunct Professor,

University of California at San Diegoand Professor Emeritus,

Masoud SalehiNortheastern University

Proakis-50210 book August 3, 2001 13:2

1

Introduction

Every day, in our work and in our leisure time, we come in contact with and use a varietyof modern communication systems and communication media, the most common beingthe telephone, radio, television, and the Internet. Through these media we are able tocommunicate (nearly) instantaneously with people on different continents, transact ourdaily business, and receive information about various developments and events of notethat occur all around the world. Electronic mail and facsimile transmission have madeit possible to rapidly communicate written messages across great distances.

Can you imagine a world without telephones, radio, and TV? Yet, when you thinkabout it, most of these modern-day communication systems were invented and devel-oped during the past century. Here, we present a brief historical review of major develop-ments within the last two hundred years that have had a major role in the development ofmodern communication systems.

1.1 HISTORICAL REVIEW

Telegraphy and Telephony. One of the earliest inventions of major signifi-cance to communications was the invention of the electric battery by Alessandro Voltain 1799. This invention made it possible for Samuel Morse to develop the electric tele-graph, which he demonstrated in 1837. The first telegraph line linked Washington withBaltimore and became operational in May 1844. Morse devised the variable-length bi-nary code given in Table 1.1, in which letters of the English alphabet were representedby a sequence of dots and dashes (code words). In this code, more frequently occurringletters are represented by short code words, while letters occurring less frequently arerepresented by longer code words.

1


2 Introduction Chapter 1

TABLE 1.1 MORSE CODE

A N B O C P D Q 1 E R 2 F S 3 G T 4 H U 5 I V 6 J W 7 K X 8 L Y 9 M Z 0

(a) Letters (b) Numbers

Period () Wait sign (AS) Comma (,) Double dash (break) Interrogation (?) Error sign Quotation Mark () Fraction bar (/) Colon (:) End of message (AR) Semicolon (;) End of transmission (SK) Parenthesis ( )

(c) Punctuation and Special Characters

The Morse code was the precursor to the variable-length source-coding methodsthat are described in Chapter 6. It is remarkable that the earliest form of electricalcommunications that was developed by Morse, namely telegraphy, was a binary digitalcommunication system in which the letters of the English alphabet were efficientlyencoded into corresponding variable-length code words having binary elements.

Nearly forty years later, in 1875, Emile Baudot developed a code for telegraphyin which each letter was encoded into fixed-length binary code words of length 5. Inthe Baudot code the binary code elements have equal length and are designated as markand space.

An important milestone in telegraphy was the installation of the first transatlanticcable in 1858 that linked the United States and Europe. This cable failed after about fourweeks of operation. A second cable was laid a few years later and became operationalin July 1866.

Telephony came into being with the invention of the telephone in the 1870s.Alexander Graham Bell patented his invention of the telephone in 1876, and in 1877 es-tablished the Bell Telephone Company. Early versions of telephone communication sys-tems were relatively simple and provided service over several hundred miles. Significantadvances in the quality and range of service during the first two decades of the twentiethcentury resulted from the invention of the carbon microphone and the induction coil.


Section 1.1 Historical Review 3

The invention of the triode amplifier by Lee De Forest in 1906 made it possible tointroduce signal amplification in telephone communication systems and, thus, to allowfor telephone signal transmission over great distances. For example, transcontinentaltelephone transmission became operational in 1915.

Two world wars and the Great Depression during the 1930s must have been adeterrent to the establishment of transatlantic telephone service. It was not until 1953,when the first transatlantic cable was laid, that telephone service became availablebetween the United States and Europe.

Automatic switching was another important advance in the development of tele-phony. The first automatic switch, developed by Strowger in 1897, was an electrome-chanical step-by-step switch. This type of switch was used for several decades. With theinvention of the transistor, electronic (digital) switching became economically feasible.After several years of development at the Bell Telephone Laboratories, a digital switchwas placed in service in Illinois in June 1960.

During the past thirty years there have been numerous significant advances in tele-phone communications. Fiber optic cables are rapidly replacing copper wire in the tele-phone plant and electronic switches have replaced the old electromechanical systems.

Wireless Communications. The development of wireless communicationsstems from the works of Oersted, Faraday, Gauss, Maxwell, and Hertz. In 1820, Oersteddemonstrated that an electric current produces a magnetic field. On August 29, 1831,Michael Faraday showed that an induced current is produced by moving a magnet in thevicinity of a conductor. Thus, he demonstrated that a changing magnetic field producesan electric field. With this early work as background, James C. Maxwell in 1864predicted the existence of electromagnetic radiation and formulated the basic theorythat has been in use for over a century. Maxwells theory was verified experimentallyby Hertz in 1887.

In 1894, a sensitive device that could detect radio signals, called the coherer,was used by its inventor Oliver Lodge to demonstrate wireless communication over adistance of 150 yards at Oxford, England. Guglielmo Marconi is credited with the devel-opment of wireless telegraphy. Marconi demonstrated the transmission of radio signalsat a distance of approximately 2 kilometers in 1895. Two years later, in 1897, he patenteda radio telegraph system and established the Wireless Telegraph and Signal Company.On December 12, 1901, Marconi received a radio signal at Signal Hill in Newfoundland,which was transmitted from Cornwall, England, a distance of about 1700 miles.

The invention of the vacuum tube was especially instrumental in the developmentof radio communication systems. The vacuum diode was invented by Fleming in 1904and the vacuum triode amplifier was invented by De Forest in 1906, as previously indi-cated. The invention of the triode made radio broadcast possible in the early part of thetwentieth century. Amplitude modulation (AM) broadcast was initiated in 1920 whenradio station KDKA, Pittsburgh, went on the air. From that date, AM radio broadcast-ing grew rapidly across the country and around the world. The superheterodyne AMradio receiver, as we know it today, was invented by Edwin Armstrong during WorldWar I. Another significant development in radio communications was the invention



of Frequency modulation (FM), also by Armstrong. In 1933, Armstrong built anddemonstrated the first FM communication system. However, the use of FM was slowto develop compared with AM broadcast. It was not until the end of World War II thatFM broadcast gained in popularity and developed commercially.

The first television system was built in the United States by V. K. Zworykin anddemonstrated in 1929. Commercial television broadcasting began in London in 1936by the British Broadcasting Corporation (BBC). Five years later the Federal Commu-nications Commission (FCC) authorized television broadcasting in the United States.

The Past Fifty Years. The growth in communications services over the pastfifty years has been phenomenal. The invention of the transistor in 1947 by WalterBrattain, John Bardeen, and William Shockley; the integrated circuit in 1958 by JackKilby and Robert Noyce; and the laser by Townes and Schawlow in 1958, have madepossible the development of small-size, low-power, low-weight, and high-speed elec-tronic circuits which are used in the construction of satellite communication systems,wideband microwave radio systems, and lightwave communication systems using fiberoptic cables. A satellite named Telstar I was launched in 1962 and used to relay TVsignals between Europe and the United States. Commercial satellite communicationservices began in 1965 with the launching of the Early Bird satellite.

Currently, most of the wireline communication systems are being replaced byfiber optic cables which provide extremely high bandwidth and make possible thetransmission of a wide variety of information sources, including voice, data, and video.Cellular radio has been developed to provide telephone service to people in automobiles,buses, and trains. High-speed communication networks link computers and a varietyof peripheral devices literally around the world.

Today we are witnessing a significant growth in the introduction and use of per-sonal communications services, including voice, data, and video transmission. Satelliteand fiber optic networks provide high-speed communication services around the world.Indeed, this is the dawn of the modern telecommunications era.

There are several historical treatments in the development of radio and telecom-munications covering the past century. We cite the books by McMahon, entitled TheMaking of a ProfessionA Century of Electrical Engineering in America (IEEE Press,1984); Ryder and Fink, entitled Engineers and Electronics (IEEE Press, 1984); andS. Millman, Ed., entitled A History of Engineering and Science in the Bell SystemCommunications Sciences (19251980) (AT & T Bell Laboratories, 1984).

1.2 ELEMENTS OF AN ELECTRICAL COMMUNICATION SYSTEM

Electrical communication systems are designed to send messages or information from asource that generates the messages to one or more destinations. In general, a communi-cation system can be represented by the functional block diagram shown in Figure 1.1.The information generated by the source may be of the form of voice (speech source),a picture (image source), or plain text in some particular language, such as English,Japanese, German, French, etc. An essential feature of any source that generates infor-


Section 1.2 Elements of an Electrical Communication System 5

Informationsource and

input transducerTransmitter

Channel

Outputtransducer

Outputsignal Receiver

Figure 1.1 Functional block diagram of a communication system.

mation is that its output is described in probabilistic terms; i.e., the output of a sourceis not deterministic. Otherwise, there would be no need to transmit the message.

A transducer is usually required to convert the output of a source into an elec-trical signal that is suitable for transmission. For example, a microphone serves as thetransducer that converts an acoustic speech signal into an electrical signal, and a videocamera converts an image into an electrical signal. At the destination, a similar trans-ducer is required to convert the electrical signals that are received into a form that issuitable for the user; e.g., acoustic signals, images, etc.

The heart of the communication system consists of three basic parts, namely,the transmitter, the channel, and the receiver. The functions performed by these threeelements are described next.

The Transmitter. The transmitter converts the electrical signal into a form thatis suitable for transmission through the physical channel or transmission medium. Forexample, in radio and TV broadcast, the Federal Communications Commission (FCC)specifies the frequency range for each transmitting station. Hence, the transmitter musttranslate the information signal to be transmitted into the appropriate frequency rangethat matches the frequency allocation assigned to the transmitter. Thus, signals trans-mitted by multiple radio stations do not interfere with one another. Similar functionsare performed in telephone communication systems where the electrical speech signalsfrom many users are transmitted over the same wire.

In general, the transmitter performs the matching of the message signal to thechannel by a process called modulation. Usually, modulation involves the use of theinformation signal to systematically vary either the amplitude, frequency, or phase ofa sinusoidal carrier. For example, in AM radio broadcast, the information signal that istransmitted is contained in the amplitude variations of the sinusoidal carrier, which isthe center frequency in the frequency band allocated to the radio transmitting station.This is an example of amplitude modulation. In FM radio broadcast, the informationsignal that is transmitted is contained in the frequency variations of the sinusoidalcarrier. This is an example of frequency modulation. Phase modulation (PM) is yet athird method for impressing the information signal on a sinusoidal carrier.



In general, carrier modulation such as AM, FM, and PM is performed at the trans-mitter, as indicated above, to convert the information signal to a form that matches thecharacteristics of the channel. Thus, through the process of modulation, the informationsignal is translated in frequency to match the allocation of the channel. The choice ofthe type of modulation is based on several factors, such as the amount of bandwidthallocated, the types of noise and interference that the signal encounters in transmissionover the channel, and the electronic devices that are available for signal amplificationprior to transmission. In any case, the modulation process makes it possible to accom-modate the transmission of multiple messages from many users over the same physicalchannel.

In addition to modulation, other functions that are usually performed at the trans-mitter are filtering of the information-bearing signal, amplification of the modulatedsignal, and in the case of wireless transmission, radiation of the signal by means of atransmitting antenna.

The Channel. The communications channel is the physical medium that isused to send the signal from the transmitter to the receiver. In wireless transmission, thechannel is usually the atmosphere (free space). On the other hand, telephone channelsusually employ a variety of physical media, including wirelines, optical fiber cables,and wireless (microwave radio). Whatever the physical medium for signal transmission,the essential feature is that the transmitted signal is corrupted in a random manner by avariety of possible mechanisms. The most common form of signal degradation comesin the form of additive noise, which is generated at the front end of the receiver, wheresignal amplification is performed. This noise is often called thermal noise. In wirelesstransmission, additional additive disturbances are man-made noise, and atmosphericnoise picked up by a receiving antenna. Automobile ignition noise is an example ofman-made noise, and electrical lightning discharges from thunderstorms is an exampleof atmospheric noise. Interference from other users of the channel is another form ofadditive noise that often arises in both wireless and wireline communication systems.

In some radio communication channels, such as the ionospheric channel that isused for long range, short-wave radio transmission, another form of signal degradationis multipath propagation. Such signal distortion is characterized as a nonadditive signaldisturbance which manifests itself as time variations in the signal amplitude, usuallycalled fading. This phenomenon is described in more detail in Section 1.3.

Both additive and nonadditive signal distortions are usually characterized as ran-dom phenomena and described in statistical terms. The effect of these signal distortionsmust be taken into account on the design of the communication system.

In the design of a communication system, the system designer works with mathe-matical models that statistically characterize the signal distortion encountered on phys-ical channels. Often, the statistical description that is used in a mathematical model isa result of actual empirical measurements obtained from experiments involving signaltransmission over such channels. In such cases, there is a physical justification for themathematical model used in the design of communication systems. On the other hand,in some communication system designs, the statistical characteristics of the channel



may vary significantly with time. In such cases, the system designer may design acommunication system that is robust to the variety of signal distortions. This can be ac-complished by having the system adapt some of its parameters to the channel distortionencountered.

The Receiver. The function of the receiver is to recover the message signalcontained in the received signal. If the message signal is transmitted by carrier modu-lation, the receiver performs carrier demodulation in order to extract the message fromthe sinusoidal carrier. Since the signal demodulation is performed in the presence ofadditive noise and possibly other signal distortion, the demodulated message signal isgenerally degraded to some extent by the presence of these distortions in the receivedsignal. As we shall see, the fidelity of the received message signal is a function of thetype of modulation, the strength of the additive noise, the type and strength of any otheradditive interference, and the type of any nonadditive interference.

Besides performing the primary function of signal demodulation, the receiveralso performs a number of peripheral functions, including signal filtering and noisesuppression.

1.2.1 Digital Communication System

Up to this point we have described an electrical communication system in rather broadterms based on the implicit assumption that the message signal is a continuous time-varying waveform. We refer to such continuous-time signal waveforms as analog sig-nals and to the corresponding information sources that produce such signals as analogsources. Analog signals can be transmitted directly via carrier modulation over thecommunication channel and demodulated accordingly at the receiver. We call such acommunication system an analog communication system.

Alternatively, an analog source output may be converted into a digital form and themessage can be transmitted via digital modulation and demodulated as a digital signalat the receiver. There are some potential advantages to transmitting an analog signal bymeans of digital modulation. The most important reason is that signal fidelity is bettercontrolled through digital transmission than analog transmission. In particular, digitaltransmission allows us to regenerate the digital signal in long-distance transmission,thus eliminating effects of noise at each regeneration point. In contrast, the noise addedin analog transmission is amplified along with the signal when amplifiers are usedperiodically to boost the signal level in long-distance transmission. Another reasonfor choosing digital transmission over analog is that the analog message signal maybe highly redundant. With digital processing, redundancy may be removed prior tomodulation, thus conserving channel bandwidth. Yet a third reason may be that digitalcommunication systems are often cheaper to implement.

In some applications, the information to be transmitted is inherently digital; e.g.,in the form of English text, computer data, etc. In such cases, the information sourcethat generates the data is called a discrete (digital) source.

In a digital communication system, the functional operations performed at thetransmitter and receiver must be expanded to include message signal discretization at



Channelencoder

Sourceencoder

Informationsource and

input transducer

Digitalmodulator

Channel

Channeldecoder

Sourcedecoder

Outputtransducer

Outputsignal

Digitaldemodulator

Figure 1.2 Basic elements of a digital communication system.

the transmitter and message signal synthesis or interpolation at the receiver. Additionalfunctions include redundancy removal, and channel coding and decoding.

Figure 1.2 illustrates the functional diagram and the basic elements of a digitalcommunication system. The source output may be either an analog signal, such as audioor video signal, or a digital signal, such as the output of a computer which is discrete intime and has a finite number of output characters. In a digital communication system,the messages produced by the source are usually converted into a sequence of binarydigits. Ideally, we would like to represent the source output (message) by as few binarydigits as possible. In other words, we seek an efficient representation of the sourceoutput that results in little or no redundancy. The process of efficiently converting theoutput of either an analog or a digital source into a sequence of binary digits is calledsource encoding or data compression. We shall describe source-encoding methods inChapter 6.

The sequence of binary digits from the source encoder, which we call the in-formation sequence is passed to the channel encoder. The purpose of the channelencoder is to introduce, in a controlled manner, some redundancy in the binary infor-mation sequence which can be used at the receiver to overcome the effects of noiseand interference encountered in the transmission of the signal through the channel.Thus, the added redundancy serves to increase the reliability of the received data andimproves the fidelity of the received signal. In effect, redundancy in the informationsequence aids the receiver in decoding the desired information sequence. For example,a (trivial) form of encoding of the binary information sequence is simply to repeateach binary digit m times, where m is some positive integer. More sophisticated (non-trivial) encoding involves taking k information bits at a time and mapping each k-bitsequence into a unique n-bit sequence, called a code word. The amount of redun-dancy introduced by encoding the data in this manner is measured by the ratio n/k.The reciprocal of this ratio, namely, k/n, is called the rate of the code or, simply, thecode rate.

The binary sequence at the output of the channel encoder is passed to the digitalmodulator, which serves as the interface to the communications channel. Since nearlyall of the communication channels encountered in practice are capable of transmittingelectrical signals (waveforms), the primary purpose of the digital modulator is to map



the binary information sequence into signal waveforms. To elaborate on the point, letus suppose that the coded information sequence is to be transmitted one bit at a time atsome uniform rate R bits/s. The digital modulator may simply map the binary digit 0into a waveform s0(t) and the binary digit 1 into a waveform s1(t). In this manner, eachbit from the channel encoder is transmitted separately. We call this binary modulation.Alternatively, the modulator may transmit b coded information bits at a time by usingM = 2b distinct waveforms si (t), i = 0, 1, . . . , M 1, one waveform for each of the 2bpossible b-bit sequences. We call this M-ary modulation (M > 2). Note that a new b-bitsequence enters the modulator every b/R seconds. Hence, when the channel bit rate Ris fixed, the amount of time available to transmit one of the M waveforms correspondingto a b-bit sequence is b times the time period in a system that uses binary modulation.

At the receiving end of a digital communications system, the digital demodulatorprocesses the channel-corrupted transmitted waveform and reduces each waveform to asingle number that represents an estimate of the transmitted data symbol (binary or M-ary). For example, when binary modulation is used, the demodulator may process thereceived waveform and decide on whether the transmitted bit is a 0 or a 1. In such a case,we say the demodulator has made a binary decision. As one alternative, the demodulatormay make a ternary decision; that is, it decides that the transmitted bit is either a 0 or1 or it makes no decision at all, depending on the apparent quality of the receivedsignal. When no decision is made on a particular bit, we say that the demodulator hasinserted an erasure in the demodulated data. Using the redundancy in the transmitteddata, the decoder attempts to fill in the positions where erasures occurred. Viewing thedecision process performed by the demodulator as a form of quantization, we observethat binary and ternary decisions are special cases of a demodulator that quantizes to Qlevels, where Q 2. In general, if the digital communications system employs M-arymodulation, where m = 0, 1, . . . , M 1 represent the M possible transmitted symbols,each corresponding to b = log2 M bits, the demodulator may make a Q-ary decision,where Q M . In the extreme case where no quantization is performed, Q = .

When there is no redundancy in the transmitted information, the demodulatormust decide which of the M waveforms was transmitted in any given time interval.Consequently Q = M , and since there is no redundancy in the transmitted information,no discrete channel decoder is used following the demodulator. On the other hand, whenthere is redundancy introduced by a discrete channel encoder at the transmitter, the Q-ary output from the demodulator occurring every b/R seconds is fed to the decoder,which attempts to reconstruct the original information sequence from knowledge of thecode used by the channel encoder and the redundancy contained in the received data.

A measure of how well the demodulator and encoder perform is the frequency withwhich errors occur in the decoded sequence. More precisely, the average probabilityof a bit-error at the output of the decoder is a measure of the performance of thedemodulator-decoder combination. In general, the probability of error is a function ofthe code characteristics, the types of waveforms used to transmit the information overthe channel, the transmitter power, the characteristics of the channel; i.e., the amount ofnoise, the nature of the interference, etc., and the method of demodulation and decoding.These items and their effect on performance will be discussed in detail in Chapters 79.



As a final step, when an analog output is desired, the source decoder acceptsthe output sequence from the channel decoder and, from knowledge of the source-encoding method used, attempts to reconstruct the original signal from the source. Dueto channel-decoding errors and possible distortion introduced by the source encoderand, perhaps, the source decoder, the signal at the output of the source decoder is anapproximation to the original source output. The difference or some function of thedifference between the original signal and the reconstructed signal is a measure of thedistortion introduced by the digital communications system.

1.2.2 Early Work in Digital Communications

Although Morse is responsible for the development of the first electrical digital commu-nication system (telegraphy), the beginnings of what we now regard as modern digitalcommunications stem from the work of Nyquist (1924), who investigated the problemof determining the maximum signaling rate that can be used over a telegraph channelof a given bandwidth without intersymbol interference. He formulated a model of atelegraph system in which a transmitted signal has the general form

s(t) =

n

ang(t nT )

where g(t) represents a basic pulse shape and {an} is the binary data sequence of{1} transmitted at a rate of 1/T bits/sec. Nyquist set out to determine the optimumpulse shape that was bandlimited to W Hz and maximized the bit rate 1/T underthe constraint that the pulse caused no intersymbol interference at the sampling timesk/T, k = 0, 1, 2, . . . . His studies led him to conclude that the maximum pulserate 1/T is 2W pulses/sec. This rate is now called the Nyquist rate. Moreover, thispulse rate can be achieved by using the pulses g(t) = (sin 2W t)/2W t . This pulseshape allows the recovery of the data without intersymbol interference at the samplinginstants. Nyquists result is equivalent to a version of the sampling theorem for band-limited signals, which was later stated precisely by Shannon (1948). The samplingtheorem states that a signal s(t) of bandwidth W can be reconstructed from samplestaken at the Nyquist rate of 2W samples/sec using the interpolation formula

s(t) =

n

s(

n

2W

)sin 2W (t n/2W )

2W (t n/2W )In light of Nyquists work, Hartley (1928) considered the issue of the amount

of data that can be transmitted reliably over a bandlimited channel when multipleamplitude levels are used. Due to the presence of noise and other interference, Hartleypostulated that the receiver can reliably estimate the received signal amplitude to someaccuracy, say A . This investigation led Hartley to conclude that there is a maximumdata rate that can be communicated reliably over a bandlimited channel, when themaximum signal amplitude is limited to Amax (fixed power constraint) and the amplituderesolution is A .



Another significant advance in the development of communications was the workof Wiener (1942) who considered the problem of estimating a desired signal waveforms(t) in the presence of additive noise n(t), based on observation of the received signalr(t) = s(t)+n(t). This problem arises in signal demodulation. Wiener determined thelinear filter whose output is the best mean-square approximation to the desired signals(t). The resulting filter is called the optimum linear (Wiener) filter.

Hartleys and Nyquists results on the maximum transmission rate of digitalinformation were precursors to the work of Shannon (1948a,b) who established themathematical foundations for information theory and derived the fundamental limitsfor digital communication systems. In his pioneering work, Shannon formulated thebasic problem of reliable transmission of information in statistical terms, using prob-abilistic models for information sources and communication channels. Based on sucha statistical formulation, he adopted a logarithmic measure for the information contentof a source. He also demonstrated that the effect of a transmitter power constraint, abandwidth constraint, and additive noise can be associated with the channel and incor-porated into a single parameter, called the channel capacity. For example, in the caseof an additive white (spectrally flat) Gaussian noise interference, an ideal bandlimitedchannel of bandwidth W has a capacity C given by

C = W log2(

1 + PW N0

)bits/s

where P is the average transmitted power and N0 is the power-spectral density of theadditive noise. The significance of the channel capacity is as follows: If the informationrate R from the source is less than C (R < C), then it is theoretically possible toachieve reliable transmission through the channel by appropriate coding. On the otherhand if R > C , reliable transmission is not possible regardless of the amount of signalprocessing performed at the transmitter and receiver. Thus, Shannon established basiclimits on communication of information and gave birth to a new field that is now calledinformation theory.

Initially, the fundamental work of Shannon had a relatively small impact on thedesign and development of new digital communications systems. In part, this was due tothe small demand for digital information transmission during the decade of the 1950s.Another reason was the relatively large complexity and, hence, the high cost of digitalhardware required to achieve the high efficiency and the high reliability predicted byShannons theory.

Another important contribution to the field of digital communications is the workof Kotelnikov (1947) which provided a coherent analysis of the various digital com-munication systems based on a geometrical approach. Kotelnikovs approach was laterexpanded by Wozencraft and Jacobs (1965).

The increase in the demand for data transmission during the last three decades,coupled with the development of more sophisticated integrated circuits, has led to thedevelopment of very efficient and more reliable digital communications systems. Inthe course of these developments, Shannons original results and the generalization



of his results on maximum transmission limits over a channel and on bounds on theperformance achieved, have served as benchmarks relative to which any given commu-nications system design is compared. The theoretical limits derived by Shannon andother researchers that contributed to the development of information theory serve asan ultimate goal in the continuing efforts to design and develop more efficient digitalcommunications systems.

Following Shannons publications came the classic work of Hamming (1950) onerror detecting and error-correcting codes to combat the detrimental effects of channelnoise. Hammings work stimulated many researchers in the years that followed and avariety of new and powerful codes were discovered, many of which are used today inthe implementation of modern communication systems.

1.3 COMMUNICATION CHANNELS AND THEIR CHARACTERISTICS

As indicated in our preceding discussion, the communication channel provides theconnection between the transmitter and the receiver. The physical channel may be apair of wires that carry the electrical signal, or an optical fiber that carries the informationon a modulated light beam, or an underwater ocean channel in which the informationis transmitted acoustically, or free space over which the information-bearing signal isradiated by use of an antenna. Other media that can be characterized as communicationchannels are data storage media, such as magnetic tape, magnetic disks, and opticaldisks.

One common problem in signal transmission through any channel is additivenoise. In general, additive noise is generated internally by components such as resistorsand solid-state devices used to implement the communication system. This is sometimescalled thermal noise. Other sources of noise and interference may arise externally tothe system, such as interference from other users of the channel. When such noiseand interference occupy the same frequency band as the desired signal, its effect canbe minimized by proper design of the transmitted signal and its demodulator at thereceiver. Other types of signal degradations that may be encountered in transmissionover the channel are signal attenuation, amplitude and phase distortion, and multipathdistortion.

The effects of noise may be minimized by increasing the power in the transmittedsignal. However, equipment and other practical constraints limit the power level inthe transmitted signal. Another basic limitation is the available channel bandwidth. Abandwidth constraint is usually due to the physical limitations of the medium and theelectronic components used to implement the transmitter and the receiver. These twolimitations result in constraining the amount of data that can be transmitted reliablyover any communications channel. Shannons basic results relate the channel capacityto the available transmitted power and channel bandwidth.

Next, we describe some of the important characteristics of several communicationchannels.

Wireline Channels. The telephone network makes extensive use of wire linesfor voice signal transmission, as well as data and video transmission. Twisted-pair wire-


Section 1.3 Communication Channels and Their Characteristics 13

lines and coaxial cable are basically guided electromagnetic channels which providerelatively modest bandwidths. Telephone wire generally used to connect a customer toa central office has a bandwidth of several hundred kilohertz (KHz). On the other hand,coaxial cable has a usable bandwidth of several megahertz (MHz). Figure 1.3 illustratesthe frequency range of guided electromagnetic channels which includes waveguidesand optical fibers.

Signals transmitted through such channels are distorted in both amplitude andphase and further corrupted by additive noise. Twisted-pair wireline channels are alsoprone to crosstalk interference from physically adjacent channels. Because wirelinechannels carry a large percentage of our daily communications around the country and

Visible light

Waveguide

Wirelinechannels

Coaxial cablechannels

Ultraviolet

Infrared106 m

1014 Hz

100 mm

1 cm

10 cm

1 m

10 m

Wav

elen

gth

Freq

uenc

y

100 m

1 km

10 km

100 km

1 kHz

10 kHz

100 kHz

1 MHz

10 MHz

100 MHz

1 GHz

10 GHz

100 GHz

1015 Hz

Figure 1.3 Frequency range for guidedwireline channels.



the world, much research has been performed on the characterization of their trans-mission properties and on methods for mitigating the amplitude and phase distortionencountered in signal transmission. In Chapter 8, we describe methods for designingoptimum transmitted signals and their demodulation, including the design of channelequalizers that compensate for amplitude and phase distortion.

Fiber Optic Channels. Optical fibers offer the communications system de-signer a channel bandwidth that is several orders of magnitude larger than coaxialcable channels. During the past decade optical fiber cables have been developed whichhave a relatively low signal attenuation and highly reliable photonic devices have beendeveloped for signal generation and signal detection. These technological advanceshave resulted in a rapid deployment of optical fiber channels both in domestic telecom-munication systems as well as for transatlantic and transpacific communications. Withthe large bandwidth available on fiber optic channels it is possible for the telephonecompanies to offer subscribers a wide array of telecommunication services, includingvoice, data, facsimile, and video.

The transmitter or modulator in a fiber optic communication system is a lightsource, either a light-emitting diode (LED) or a laser. Information is transmitted byvarying (modulating) the intensity of the light source with the message signal. The lightpropagates through the fiber as a light wave and is amplified periodically (in the case ofdigital transmission, it is detected and regenerated by repeaters) along the transmissionpath to compensate for signal attenuation. At the receiver, the light intensity is detectedby a photodiode, whose output is an electrical signal that varies in direct proportion tothe power of the light impinging on the photodiode.

It is envisioned that optical fiber channels will replace nearly all wireline channelsin the telephone network in the next few years.

Wireless Electromagnetic Channels. In radio communication systems,electromagnetic energy is coupled to the propagation medium by an antenna whichserves as the radiator. The physical size and the configuration of the antenna dependprimarily on the frequency of operation. To obtain efficient radiation of electromag-netic energy, the antenna must be longer than 1/10 of the wavelength. Consequently, aradio station transmitting in the AM frequency band, say at 1 MHz (corresponding toa wavelength of = c/ fc = 300 m) requires an antenna of at least 30 meters.

Figure 1.4 illustrates the various frequency bands of the electromagnetic spectrum.The mode of propagation of electromagnetic waves in the atmosphere and in free spacemay be subdivided into three categories, namely, ground-wave propagation, sky-wavepropagation, and line-of-sight (LOS) propagation. In the VLF and ELF frequency bands,where the wavelengths exceed 10 km, the earth and the ionosphere act as a waveguide forelectromagnetic wave propagation. In these frequency ranges, communication signalspractically propagate around the globe. For this reason, these frequency bands areprimarily used to provide navigational aids from shore to ships around the world. Thechannel bandwidths available in these frequency bands are relatively small (usually from110% of the center frequency), and hence, the information that is transmitted through



Visible light

Super High Frequency(SHF)

Ultra High Frequency(UHF)

Very High Frequency(VHF)

High Frequency(HF)

Medium Frequency(MF)

Low Frequency(LF)

Very Low Frequency(VLF)

Audioband

Millimeter waves

Ultraviolet

Frequency band

Infrared106 m

1 cm

10 cm

1 m

10 m

100 m

Wav

elen

gth

1 km

10 km

100 km

Experimental

ExperimentalNavigation

Satellite to satelliteMicrowave relayEarthsatellite

Radar

BusinessAmateur radio

International radioCitizens band

AeronauticalNavigation

Radio teletype

UHF TVMobile, aeronautical

VHF TV and FM Broadcast

Mobile radio

AM broadcast

Use

1014 Hz

Freq

uenc

y

1 kHz

10 kHz

100 kHz

1 MHz

10 MHz

100 MHz

1 GHz

10 GHz

100 GHz

1015 Hz

Microwaveradio

Longwaveradio

Shortwaveradio

Figure 1.4 Frequency range for wireless electromagnetic channels. (Adapted from Carlson,Sec. Ed.; c 1975 McGraw-Hill. Reprinted with permission of the publisher.)

these channels is relatively slow speed and, generally, confined to digital transmission.A dominant type of noise at these frequencies is generated from thunderstorm activityaround the globe, especially in tropical regions. Interference results from the manyusers of these frequency bands.

Ground-wave propagation, illustrated in Figure 1.5, is the dominant mode ofpropagation for frequencies in the MF band (0.33 MHz). This is the frequency band



EarthFigure 1.5 Illustration of ground-wavepropagation.

Ionosphere

EarthFigure 1.6 Illustration of sky-wavepropagation.

used for AM broadcasting and maritime radio broadcasting. In AM broadcast, the rangewith ground-wave propagation of even the more powerful radio stations is limited toabout 100 miles. Atmospheric noise, man-made noise, and thermal noise from electroniccomponents at the receiver are dominant disturbances for signal transmission of MF.

Sky-wave propagation, as illustrated in Figure 1.6, results from transmitted sig-nals being reflected (bent or refracted) from the ionosphere, which consists of severallayers of charged particles ranging in altitude from 30250 miles above the surface ofthe earth. During the daytime hours, the heating of the lower atmosphere by the suncauses the formation of the lower layers at altitudes below 75 miles. These lower layers,especially the D-layer serve to absorb frequencies below 2 MHz, thus, severely limitingsky-wave propagation of AM radio broadcast. However, during the night-time hours theelectron density in the lower layers of the ionosphere drops sharply and the frequencyabsorption that occurs during the day time is significantly reduced. As a consequence,powerful AM radio broadcast stations can propagate over large distances via sky-waveover the F-layer of the ionosphere, which ranges from 90250 miles above the surfaceof the earth.

A frequently occurring problem with electromagnetic wave propagation via sky-wave in the HF frequency range is signal multipath. Signal multipath occurs whenthe transmitted signal arrives at the receiver via multiple propagation paths at differ-ent delays. Signal multipath generally results in intersymbol interference in a digitalcommunication system. Moreover, the signal components arriving via different prop-agation paths may add destructively, resulting in a phenomenon called signal fading,which most people have experienced when listening to a distant radio station at night,when sky-wave is the dominant propagation mode. Additive noise at HF is a combina-tion of atmospheric noise and thermal voice.

Sky-wave ionospheric propagation ceases to exist at frequencies above approx-imately 30 MHz, which is the end of the HF band. However, it is possible to have



ionospheric scatter propagation at frequencies in the range of 3060 MHz, resultingfrom signal scattering from the lower ionosphere. It is also possible to communicateover distances of several hundred miles by use of tropospheric scattering at frequenciesin the range of 40300 MHz. Troposcatter results from signal scattering due to particlesin the atmosphere at altitudes of 10 miles or less. Generally, ionospheric scatter andtropospheric scatter involve large signal propagation losses and require a large amountof transmitter power and relatively large antennas.

Frequencies above 30 MHz propagate through the ionosphere with relativelylittle loss and make satellite and extraterrestrial communications possible. Hence, atfrequencies in the VHF band and higher, the dominant mode of electromagnetic propa-gation is line-of-sight (LOS) propagation. For terrestrial communication systems, thismeans that the transmitter and receiver antennas must be in direct LOS with relativelylittle or no obstruction. For this reason television stations transmitting in the VHF andUHF frequency bands mount their antennas on high towers in order to achieve a broadcoverage area.

In general, the coverage area for LOS propagation is limited by the curvature ofthe earth. If the transmitting antenna is mounted at a height h feet above the surfaceof the earth, the distance to the radio horizon, assuming no physical obstructions sucha mountains, is approximately d = 2h miles. For example, a TV antenna mountedon a tower of 1000 ft in height provides a coverage of approximately 50 miles. Asanother example, microwave radio relay systems used extensively for telephone andvideo transmission at frequencies above 1 GHz have antennas mounted on tall towersor on the top of tall buildings.

The dominant noise limiting the performance of communication systems in theVHF and UHF frequency ranges is thermal noise generated in the receiver front end andcosmic noise picked up by the antenna. At frequencies in the SHF band above 10 GHz,atmospheric conditions play a major role in signal propagation. Figure 1.7 illustratesthe signal attenuation in dB/mile due to precipitation for frequencies in the range of10100 GHz. We observe that heavy rain introduces extremely high propagation lossesthat can result in service outages (total breakdown in the communication system).

At frequencies above the EHF band, we have the infrared and visible light regionsof the electromagnetic spectrum which can be used to provide LOS optical commu-nication in free space. To date, these frequency bands have been used in experimentalcommunication systems, such as satellite-to-satellite links.

Underwater Acoustic Channels. Over the past few decades, ocean explo-ration activity has been steadily increasing. Coupled with this increase in ocean ex-ploration is the need to transmit data, collected by sensors placed underwater, to thesurface of the ocean. From there it is possible to relay the data via a satellite to a datacollection center.

Electromagnetic waves do not propagate over long distances underwater, except atextremely low frequencies. However, the transmission of signals at such low frequenciesis prohibitively expensive because of the large and powerful transmitters required. Theattenuation of electromagnetic waves in water can be expressed in terms of the skin



10.0

1.0

0.1

0.01

0.00110 30 100

Heavy rain(1 in. /h)

Medium rain(0.1 in. /h)

Fog(0.01 in. /h)

Frequency, GHz

Atte

nuat

ion,

dB

/mi

Light rain(0.05 in. /h)

Figure 1.7 Signal attenuation due to precipitation. (From Ziemer and Tranter(1990); c Houghton Mifflin Reprinted with permission of the publisher.)

depth, which is the distance a signal is attenuated by 1/e. For sea water, the skin depth = 250/ f , where f is expressed in Hz and is in meters. For example, at 10 kHz,the skin depth is 2.5 m. In contrast, acoustic signals propagate over distances of tensand even hundreds of kilometers.

A shallow water acoustic channel is characterized as a multipath channel dueto signal reflections from the surface and the bottom of the sea. Due to wave mo-tion, the signal multipath components undergo time-varying propagation delays whichresult in signal fading. In addition, there is frequency-dependent attenuation, which isapproximately proportional to the square of the signal frequency.

Ambient ocean acoustic noise is caused by shrimp, fish, and various mammals.Near harbors, there is also man-made acoustic noise in addition to the ambient noise.


Section 1.4 Mathematical Models for Communication Channels 19

In spite of this hostile environment, it is possible to design and implement efficient andhighly reliable underwater acoustic communication systems for transmitting digitalsignals over large distances.

Storage Channels. Information storage and retrieval systems constitute avery significant part of our data-handling activities on a daily basis. Magnetic tape,including digital audio tape and video tape, magnetic disks used for storing largeamounts of computer data, and optical disks used for computer data storage, music(compact disks), and video are examples of data storage systems that can be charac-terized as communication channels. The process of storing data on a magnetic tape ora magnetic or optical disk is equivalent to transmitting a signal over a telephone ora radio channel. The readback process and the signal processing involved in storagesystems to recover the stored information is equivalent to the functions performed bya receiver in a telephone or radio communication system to recover the transmittedinformation.

Additive noise generated by the electronic components and interference fromadjacent tracks is generally present in the readback signal of a storage system, just asis the case in a telephone or a radio communication system.

The amount of data that can be stored is generally limited by the size of the diskor tape and the density (number of bits stored per square inch) that can be achievedby the write/read electronic systems and heads. For example, a packing density of109 bits/sq. in. has been recently demonstrated in an experimental magnetic disk storagesystem. (Current commercial magnetic storage products achieve a much lower density.)The speed at which data can be written on a disk or tape and the speed at which it canbe read back is also limited by the associated mechanical and electrical subsystems thatconstitute an information storage system.

Channel coding and modulation are essential components of a well-designeddigital magnetic or optical storage system. In the readback process, the signal is de-modulated and the added redundancy introduced by the channel encoder is used tocorrect errors in the readback signal.

1.4 MATHEMATICAL MODELS FOR COMMUNICATION CHANNELS

In the design of communication systems for transmitting information through physicalchannels, we find it convenient to construct mathematical models that reflect the mostimportant characteristics of the transmission medium. Then, the mathematical modelfor the channel is used in the design of the channel encoder and modulator at thetransmitter and the demodulator and channel decoder at the receiver. Next, we providea brief description of the channel models that are frequently used to characterize manyof the physical channels that we encounter in practice.

The Additive Noise Channel. The simplest mathematical model for a com-munication channel is the additive noise channel, illustrated in Figure 1.8. In thismodel the transmitted signal s(t) is corrupted by an additive random noise process



Channel

n(t)

s(t)r (t) s(t) n(t)

Figure 1.8 The additive noise channel.

n(t). Physically, the additive noise process may arise from electronic components andamplifiers at the receiver of the communication system, or from interference encoun-tered in transmission, as in the case of radio signal transmission.

If the noise is introduced primarily by electronic components and amplifiers at thereceiver, it may be characterized as thermal noise. This type of noise is characterizedstatistically as a Gaussian noise process. Hence, the resulting mathematical modelfor the channel is usually called the additive Gaussian noise channel. Because thischannel model applies to a broad class of physical communication channels and becauseof its mathematical tractability, this is the predominant channel model used in ourcommunication system analysis and design. Channel attenuation is easily incorporatedinto the model. When the signal undergoes attenuation in transmission through thechannel, the received signal is

r(t) = as(t) + n(t) (1.4.1)

where a represents the attenuation factor.

The Linear Filter Channel. In some physical channels such as wireline tele-phone channels, filters are used to ensure that the transmitted signals do not exceedspecified bandwidth limitations and, thus, do not interfere with one another. Such chan-nels are generally characterized mathematically as linear filter channels with additivenoise, as illustrated in Figure 1.9. Hence, if the channel input is the signal s(t), the

Channel

Linearfilterh (t)

n (t)

s (t) r (t) s (t) h (t) n (t)

Figure 1.9 The linear filter channel with additive noise.


Section 1.4 Mathematical Models for Communication Channels 21

Channel

s(t)r (t)

n (t)

Linear

time-invariantfilter h( ; t)

Figure 1.10 Linear time-variant filterchannel with additive noise.

channel output is the signal

r(t) = s(t) h(t) + n(t)

= +

h( )s(t ) d + n(t) (1.4.2)

where h(t) is the impulse response of the linear filter and denotes convolution.

The Linear Time-Variant Filter Channel. Physical channels such as under-water acoustic channels and ionospheric radio channels which result in time-variantmultipath propagation of the transmitted signal may be characterized mathematically astime-variant linear filters. Such linear filters are characterized by time-variant channelimpulse response h( ; t) where h( ; t) is the response of the channel at time t , due toan impulse applied at time t . Thus, represents the age (elapsed time) variable.The linear time-variant filter channel with additive noise is illustrated Figure 1.10. Foran input signal s(t), the channel output signal is

r(t) = s(t) h( ; t) + n(t)

= +

h( ; t)s(t ) d + n(t) (1.4.3)

A good model for multipath signal propagation through physical channels, such asthe ionosphere (at frequencies below 30 MHz) and mobile cellular radio channels, is aspecial case of Equation (1.4.3) is which the time-variant impulse response has the form

h( ; t) =L

k=1ak(t)( k) (1.4.4)

where the {ak(t)} represent the possibly time-variant attenuation factors for the Lmultipath propagation paths. If Equation (1.4.4) is substituted into Equation (1.4.3),the received signal has the form

r(t) =L

k=1ak(t)s(t k) + n(t) (1.4.5)

Hence, the received signal consists of L multipath components, where each componentis attenuated by {ak} and delayed by {k}.



The three mathematical models described above adequately characterize a largemajority of physical channels encountered in practice. These three channel models areused in this text for the analysis and design of communication systems.

1.5 ORGANIZATION OF THE BOOK

Before we embark on the analysis and design of communication systems, we providea brief review of basic frequency-domain characteristics of signals and linear systemsin Chapter 2. Emphasis is placed on the Fourier series and the Fourier transform repre-sentation of signals and the use of transforms in linear systems analysis. The processof sampling a bandlimited analog signal is also considered.

In Chapter 3, we treat the modulation and demodulation of analog signals. Thischapter provides detailed description of amplitude modulation (AM), frequency mod-ulation (FM), and phase modulation (PM). As examples of analog signal transmissionand reception, we consider radio and television broadcasting, and mobile radio cellularcommunication systems.

In Chapter 4, we present a review of the basic definitions and concepts in prob-ability and random processes. These topics are particularly important in our study ofelectrical communications, because information sources produce random signals attheir output and communication channels generally corrupt the transmitted signals ina random manner, through the addition of noise and other channel distortion. Spe-cial emphasis is placed on Gaussian random processes, which provide mathematicallytractable models for additive noise disturbances. Both time domain and frequency do-main representations of random signals are presented.

Chapters 5 provides a treatment of the effects of additive noise in the demodulationof amplitude modulated (AM) and angle modulated (FM, PM) analog signals and acomparison of these analog signal modulations in terms of their signal-to-noise ratioperformance. Also discussed in this chapter is the problem of estimating the carrierphase using a phase-locked loop (PLL). Finally, we describe the characterization ofthermal noise and the effect of transmission losses in analog communication systems.

The remainder of the book is focused on digital communication systems.Chapter 6 is devoted to the modeling and characterization of information sources andsource coding. In this chapter, we introduce a measure for the information content of adiscrete source and describe two algorithms, the Huffman algorithm and the Lempel-Ziv algorithm, for efficient encoding of the source output. The major part of the chapteris devoted to the problem of encoding the outputs of analog sources. Several waveform-encoding methods are described, including pulse-code modulation (PCM), differentialPCM, and delta modulation (DM). We also describe a model-based approach to ana-log source encoding, based on linear prediction. As practical examples of the theorypresented in this chapter, we consider digital audio recording systems, digital audiotransmission in telephone systems, and the JPEG image-coding standard.

Chapter 7 treats modulation methods for digital signal transmission throughan additive white Gaussian noise channel. Various types of binary and nonbinary


Section 1.6 Further Reading 23

modulation methods are described based on a geometric representation of signals andtheir performance is evaluated in terms of the probability of error. A link budget anal-ysis for radio communication systems is also described. The final topic of this chapteris focused on signal synchronization methods for digital communication systems.

In Chapter 8, we consider the problem of digital communication through bandlim-ited, additive white Gaussian noise channels. In particular, we describe the design ofbandlimited signal waveforms for such channels and the spectral characteristics of dig-itally modulated signals. Digitally modulated signals with memory, their modulation,and their spectral characteristics are also described. The effect of channel distortion onthe transmitted signals is characterized in terms of intersymbol interference (ISI), andthe design of adaptive equalizers for suppressing ISI is described.

Channel coding and decoding is the topic of Chapter 9. In this chapter, we describethe concept of channel capacity, and derive the capacity of an additive white Gaussiannoise channel. Linear block codes and convolutional codes are considered for enhancingthe performance of a digital communication system in the presence of additive noise.Decoding algorithms for both block codes and convolutional codes are also described.The final topic of this chapter provides a treatment of trellis-coded modulation, whichis widely used in the implementation of high speed modems.

The last chapter of the book, Chapter 10, focuses on topics dealing with wirelesscommunications. We begin by characterizing channel fading and multipath effects inwireless communication systems and describe the design of signals that are effective inmitigating these channel distortions. Then, we describe the class of continuous-phasemodulated signals, which are especially suitable for digital communications in wirelesschannels due to their constant amplitude and narrow bandwidth characteristics. Finally,we treat the class of spread-spectrum signals, which are suitable for multi-user wirelesscommunication systems. As examples of practical wireless communication systems,we briefly describe two digital cellular communication systems, the pan European GSMsystem and the North-American IS-95 CDMA system.

In an introductory book of this level, we have not attempted to provide a largenumber of references to the technical literature. However, we have included in eachchapter several supplementary references for further reading, including textbooks andbasic or tutorial treatments of important topics. References are cited by giving theauthors name, with the year of publication in parentheses; e.g., Nyquist (1924).

1.6 FURTHER READING

We have already cited several historical treatments of radio and telecommunicationsduring the past century. These include the book by McMahon (1984), Ryder and Fink(1984), and Millman (1984). The classical works of Nyquist (1924), Hartley (1928),Kotelnikov (1947), Shannon (1948), and Hamming (1950) are particularly importantbecause they lay the groundwork of modern communication systems engineering.


2

Frequency Domain Analysisof Signals and Systems

In this chapter, we review the basics of signals and linear systems in the frequencydomain. The motivation for studying these fundamental concepts stems from the basicrole they play in modeling various types of communication systems. In particular signalsare used to transmit the information over a communication channel. Such signals areusually called information-bearing signals.

In the transmission of an information-bearing signal over a communication chan-nel, the shape of the signal is changed, or distorted, by the channel. In other words,the output of the communication channel, which is called the received signal, is not anexact replica of the channel input due to the channel distortion. The communicationchannel is an example of a system; i.e., an entity that produces an output signal whenexcited by an input signal. A large number of communication channels can be mod-eled closely by a subclass of systems called linear systems. Linear systems are a largesubclass of systems which arise naturally in many practical applications and are rathereasy to analyze. We have devoted this entire chapter to the study of the basics of signalsand linear systems in the frequency domain.

2.1 FOURIER SERIES

The analysis of signals and linear systems in the frequency domain is based on represen-tation of signals in terms of the frequency variable and this is done through employingFourier series and Fourier transforms. Fourier series is applied to periodic signalswhereas the Fourier transform can be applied to periodic and nonperiodic signals.

24


Section 2.1 Fourier Series 25

Theorem 2.1.1. [Fourier Series] Let the signal x(t) be a periodic signal withperiod T0. If the following conditions (known as the Dirichlet conditions) are satisfied

1. x(t) is absolutely integrable over its period; i.e., T0

0|x(t)| dt < ,

2. The number of maxima and minima of x(t) in each period is finite,3. The number of discontinuities of x(t) in each period is finite,

then x(t) can be expanded in terms of the complex exponential signals {e j2 nT0 t}+n= as

x(t) =+

n=xne

j2 nT0t (2.1.1)

where

xn = 1T0

+T0

x(t)e j2n

T0t dt (2.1.2)

for some arbitrary and

x(t) =

x(t) if x(t) is continuous at tx(t+) + x(t)

2if x(t) is discontinuous at t

(2.1.3)

Some observations concerning this theorem are in order.

The coefficients xn are called the Fourier series coefficients of the signal x(t).These are, in general, complex numbers.

The parameter in the limits of the integral is arbitrary. It can be chosen tosimplify computation of the integral. Usually = 0 or = T0/2 are goodchoices.

For all practical purposes, x(t) equals x(t). From now on, we will use x(t)instead of x(t) but will keep in mind that the Fourier series expansion, at pointsof discontinuity of x(t), gives the midpoint between right and left limits of thesignal.

The Dirichlet conditions are only sufficient conditions for the existence of theFourier series expansion. For some signals that do not satisfy these conditions,we can still find the Fourier series expansion.

john g. proakis masoud salehi 2nd ed. - u-cursos · john g. proakis masoud salehi 2nd ed. ......

Documents