PREFACEThis second edition of Digital Communications: Fundamentals and Applications
represents an update of the original publication. The key features that have been updated are:
- The error-correction coding chapters have been expanded, particularly in the areas of Reed-Solomon codes, turbo codes, and trellis-coded modulation.
- A new chapter on fading channels and how to mitigate the degrading effects of fading has been introduced.
- Explanations and descriptions of essential digital communication concepts have been amplified.
- End-of-chapter problem sets have been expanded. Also, end-of-chapter question sets (and where to find the answers), as well as end-of-chapter CD exercises have been added.
- A compact disc (CD) containing an educational version of the design software SystemView by ELANIX accompanies the textbook. The CD contains a workbook with over 200 exercises, as well as a concise tutorial on digital signal processing (DSP). CD exercises in the workbook reinforce material in the textbook; concepts can be explored by viewing waveforms with a windows-based PC and by changing parameters to see the effects on the overall system. Some of the exercises provide basic training in using SystemView; others provide additional training in DSP techniques.
The teaching of a one-semester university course proceeds in a very different manner compared with that of a short-course in the same subject. At the university, one has the luxury of timetime to develop the needed skills and mathematical tools, time to practice the ideas with homework exercises. In a short-course, the treatment is almost backwards compared with the university. Because of the time factor, a short-course teacher must "jump in" early with essential concepts and applications. One of the vehicles that I found useful in structuring a short course was to start by handing out a check list. This was not merely an outline of the curriculum. It represented a collection of concepts and nomenclature that are not clearly documented, and are often misunderstood. The short-course students were thus initiated into the course by being challenged. I promised them that once they felt comfortable describing each issue, or answering each question on the list, they would be well on their way toward becoming knowledgeable in the field of digital communications. I have learned that this list of essential concepts is just as valuable for teaching full-semester courses as it is for short courses. Here then is my "check list" for digital communications.
- What mathematical dilemma is the cause for there being several definitions of bandwidth? (See Section 1.7.2.)
- Why is the ratio of bit energy-to-noise power spectral density, Eb/N0, a natural figure-to-merit for digital communication systems? (See Section 3.1.5.)
- When representing timed events, what dilemma can easily result in confusing the most-significant bit (MSB) and the least-significant bit (LSB)? (See Section 220.127.116.11.)
- The error performance of digital signaling suffers primarily from two degradation types. a) loss in signal-to-noise ratio, b) distortion resulting in an irreducible bit-error probability. How do they differ? (See Section 3.3.2.)
- Often times, providing more Eb/N0 will not mitigate the degradation due to intersymbol interference (ISI). Explain why. (See Section 3.3.2.)
- At what location in the system is Eb/N0 defined? (See Section 4.3.2.)
- Digital modulation schemes fall into one of two classes with opposite behavior characteristics. a) orthogonal signaling, b) phase/amplitude signaling. Describe the behavior of each class. (See Section 4.8.2 and 9.7.)
- Why do binary phase shift keying (BPSK) and quaternary phase shift keying (QPSK) manifest the same bit-error-probability relationship? Does the same hold true for M-ary pulse amplitude modulation (M-PAM) and M2-ary quadrature amplitude modulation (M2-QAM) bit-error probability? (See Sections 4.8.4 and 18.104.22.168.)
- In orthogonal signaling, why does error-performance improve with higher dimensional signaling? (See Section 4.8.5.)
- Why is free-space loss a function of wavelength? (See Section 5.3.3.)
- What is the relationship between received signal to noise (S/N) ratio and carrier to noise (C/N) ratio? (See Section 5.4.)
- Describe four types of trade-offs that can be accomplished by using an error-correcting code. (See Section 6.3.4.)
- Why do traditional error-correcting codes yield error-performance degradation at low values of Eb/N0? (See Section 6.3.4.)
- Of what use is the standard array in understanding a block code, and in evaluating its capability? (See Section 6.6.5.)
- Why is the Shannon limit of -1.6 dB not a useful goal in the design of real systems? (See Section 22.214.171.124.)
- 16. What are the consequences of the fact that the Viterbi decoding algorithm does not yield a posteriori probabilities? What is a more descriptive name for the Viterbi algorithm? (See Section 8.4.6.)
- 17. Why do binary and 4-ary orthogonal frequency shift keying (FSK) manifest the same bandwidth-efficiency relationship? (See Section 9.5.1.)
- 18. Describe the subtle energy and rate transformations of received signals: from data-bits to channel-bits to symbols to chips. (See Section 9.7.7.)
- 19. Define the following terms: Baud, State, Communications Resource, Chip, Robust Signal. (See Sections 1.1.3 and 7.2.2, Chapter 11, and Sections 12.3.2 and 12.4.2.)
- 20. In a fading channel, why is signal dispersion independent of fading rapidity? (See Section 126.96.36.199.)
I hope you find it useful to be challenged in this way. Now, let us describe the purpose of the book in a more methodical way. This second edition is intended to provide a comprehensive coverage of digital communication systems for senior level undergraduates, first year graduate students, and practicing engineers. Though the emphasis is on digital communications, necessary analog fundamentals are included since analog waveforms are used for the radio transmission of digital signals. The key feature of a digital communication system is that it deals with a finite set of discrete messages, in contrast to an analog communication system in which messages are defined on a continuum. The objective at the receiver of the digital system is not to reproduce a waveform with precision; it is instead to determine from a noise-perturbed signal, which of the finite set of waveforms had been sent by the transmitter. In fulfillment of this objective, there has arisen an impressive assortment of signal processing techniques.
The book develops these techniques in the context of a unified structure. The structure, in block diagram form, appears at the beginning of each chapter; blocks in the diagram are emphasized, when appropriate, to correspond to the subject of that chapter. Major purposes of the book are to add organization and structure to a field that has grown and continues to grow rapidly, and to insure awareness of the "big picture" even while delving into the details. Signals and key processing steps are traced from the information source through the transmitter, channel, receiver, and ultimately to the information sink. Signal transformations are organized according to nine functional classes: Formatting and source coding, Baseband signaling, Bandpass signaling, Equalization, Channel coding, Muliplexing and multiple access, Spreading, Encryption, and Synchronization. Throughout the book, emphasis is placed on system goals and the need to trade off basic system parameters such as signal-to-noise ratio, probability of error, and bandwidth expenditure.
ORGANIZATION OF THE BOOK
Chapter 1 introduces the overall digital communication system and the basic signal transformations that are highlighted in subsequent chapters. Some basic ideas of random variables and the additive white Gaussian noise (AWGN) model are reviewed. Also, the relationship between power spectral density and autocorrelation, and the basics of signal transmission through linear systems are established. Chapter 2 covers the signal processing step, known as formatting, in order to render an information signal compatible with a digital system. Chapter 3 emphasizes baseband signaling, the detection of signals in Gaussian noise, and receiver optimization. Chapter 4 deals with bandpass signaling and its associated modulation and demodulation/detection techniques. Chapter 5 deals with link analysis, an important subject for providing overall system insight; it considers some subtleties that are often missed. Chapters 6, 7, and 8 deal with channel codinga cost-effective way of providing a variety of system performance trade-offs. Chapter 6 emphasizes linear block codes, Chapter 7 deals with convolutional codes, and Chapter 8 deals with Reed-Solomon codes and concatenated codes such as turbo codes.
Chapter 9 considers various modulation/coding system trade-offs dealing with probability of bit-error performance, bandwidth efficiency, and signal-to-noise ratio. It also treats the important area of coded modulation, particularly trellis-coded modulation. Chapter 10 deals with synchronization for digital systems. It covers phase-locked loop implementation for achieving carrier synchronization. It covers bit synchronization, frame synchronization, and network synchronization, and it introduces some ways of performing synchronization using digital methods.
Chapter 11 treats multiplexing and multiple access. It explores techniques that are available for utilizing the communication resource efficiently. Chapter 12 introduces spread spectrum techniques and their application in such areas as multiple access, ranging, and interference rejection. This technology is important for both military and commercial applications. Chapter 13 deals with source coding which is a special class of data formatting. Both formatting and source coding involve digitization of data; the main difference between them is that source coding additionally involves data redundancy reduction. Rather than considering source coding immediately after formatting, it is purposely treated in a later chapter so as not to interrupt the presentation flow of the basic processing steps. Chapter 14 covers basic encryption/decryption ideas. It includes some classical concepts, as well as a class of systems called public key cryptosystems, and the widely used E-mail encryption software known as Pretty Good Privacy (PGP). Chapter 15 deals with fading channels. Here, we deal with applications, such as mobile radios, where characterization of the channel is much more involved than that of a nonfading one. The design of a communication system that will withstand the degradation effects of fading can be much more challenging than the design of its nonfading counterpart. In this chapter, we describe a variety of techniques that can mitigate the effects of fading, and we show some successful designs that have been implemented.
It is assumed that the reader is familiar with Fourier methods and convolution. Appendix A reviews these techniques, emphasizing those properties that are particularly useful in the study of communication theory. It also assumed that the reader has a knowledge of basic probability and has some familiarity with random variables. Appendix B builds on these disciplines for a short treatment on statistical decision theory with emphasis on hypothesis testingso important in the understanding of detection theory. A new section, Appendix E, has been added to serve as a short tutorial on s-domain, z-domain, and digital filtering. A concise DSP tutorial also appears on the CD that accompanies the book.
If the book is used for a two-term course, a simple partitioning is suggested; the first seven chapters can be taught in the first term, and the last eight chapters in the second term. If the book is used for a one-term introductory course, it is suggested that the course material be selected from the following chapters: 1, 2, 3, 4, 5, 6, 7, 9, 10, and 12.
- The clear, easy-to-understand introduction to digital communications
- Completely updated coverage of today's most critical technologies
- Step-by-step implementation coverage
- Trellis-coded modulation, fading channels, Reed-Solomon codes, encryption, and more
- Exclusive coverage of maximizing performance with advanced "turbo codes"
"This is a remarkably comprehensive treatment of the field, covering in considerable detail modulation, coding (both source and channel), encryption, multiple access and spread spectrum. It can serve both as an excellent introduction for the graduate student with some background in probability theory or as a valuable reference for the practicing ommunication system engineer. For both communities, the treatment is clear and well presented." Andrew Viterbi, The Viterbi Group
Master every key digital communications technology, concept, and technique.
Digital Communications, Second Edition is a thoroughly revised and updated edition of the field's classic, best-selling introduction. With remarkable clarity, Dr. Bernard Sklar introduces every digital communication technology at the heart of today's wireless and Internet revolutions, providing a unified structure and context for understanding them -- all without sacrificing mathematical precision.
Sklar begins by introducing the fundamentals of signals, spectra, formatting, and baseband transmission. Next, he presents practical coverage of virtually every contemporary modulation, coding, and signal processing technique, with numeric examples and step-by-step implementation guidance. Coverage includes:
- Signals and processing steps: from information source through transmitter, channel, receiver, and information sink
- Key tradeoffs: signal-to-noise ratios, probability of error, and bandwidth expenditure
- Trellis-coded modulation and Reed-Solomon codes: what's behind the math
- Synchronization and spread spectrum solutions
- Fading channels: causes, effects, and techniques for withstanding fading
- The first complete how-to guide to turbo codes: squeezing maximum performance out of digital connections
- Implementing encryption with PGP, the de facto industry standard
Whether you're building wireless systems, xDSL, fiber or coax-based services, satellite networks, or Internet infrastructure, Sklar presents the theory and the practical implementation details you need. With nearly 500 illustrations and 300 problems and exercises, there's never been a faster way to master advanced digital communications. CD-ROM INCLUDED
The CD-ROM contains a complete educational version of Elanix' SystemView DSP design software, as well as detailed notes for getting started, a comprehensive DSP tutorial, and over 50 additional communications exercises.