大多数介绍信息论和编码学的书，不是太过学术化，就是太过简单。本书避免了上述缺点，既考虑到 了数学的严谨性，又充分考虑了易读性。
　　本书的主要特点 详细介绍了网格编码调制(TCM)．并重点介绍了加性高斯白噪声(AWGN)和衰退倍道。 以大量示例仔细描述了信源编码。 深入讨论了TurboCode。
　　简要介绍了线性代数。 通过精心挑选的115道例题及115道练习题，清晰透彻地讲解了深奥的理论知识和定义。

　　通过大量篇幅论述了密码学，包括基本原理。私钥和公钥密码，当今通用的加密标准及最新发展趋势。 “叙述流畅，通过大量示例及精确的解说透彻地论述了各个主题。”
　　　　　　　　　　　　　　　　　　　　　　　　——审稿人

Ranjan Bose is an Associate Professor in the department of Electrical Engineering at the Indian
Institute of Technology (IIT), Delhi. He did his B. Tech. in Electrical Engineering from UT,
Kanpur and his M.S. and Ph. D. in Electrical Engineering from the University of Pennsylvania,
Philadelphia. He then worked at Alliance Semiconductors Inc. as a Senior Design Engineer.
Since November 1997, he has been with the Indian Institute of Technology (IIT), Delhi, as a
faculty member. Dr. Bose frequently lectures on coding and cryptography. Dr. Bose was
awarded the URSI Young Scientist award in 1999 and the Humboldt Fellowship inJuly 2000.

Information theory, error control coding and cryptography are the three load-bearing pillars of modern digital communication systems. All the three topics are vast, and there are many good books that deal with these topics individually. In this book, an attempt has been made to incorporate all the important concepts of information theory, error control coding and cryptography in-between the two covers, without making the covers too far apart. This is intended as a simple and lively book on the subject.
　 This book results from my teaching of different topics on information theory and coding at
Indian Institute of Technology, Delhi. While writing thisbook, I had to take a decision regarding
how mathematical the book should be. Quoting Richard W. Hamming: "Mathematits is. an
interesting intellectual sport but it should not be allowed to stand in the way of obtaining sensible informa-tion about physicalprocesses". Too mathematical a book has the potential danger of scaring away students who lack a strong background in mathematics. On the other hand, the use of mathe-matics cannot be reduced beyond a limit, if the concepts in information theory and error control coding have to be studied with a certain amount of rigor. But then, life is all about striking a balance. I have tried to traverse the path of golden mean in this book. Mathematics has been used wherever necessary, and only to the extent that it is essential. Intuitive explanations have been provided wherever possible. I also believe that teaching by example is a very effective method of instruction. Therefore, as soon as a new concept is introduced, I have tried to provide at least one numerical example.
　　HOW TO READ THIS BOOK
　　This book has been written to be both a lively introduction as well as a fairly detailed reference to the fascinating world of information theory, coding and cryptography. The entire book has been divided into three logical parts:
Part I--Information Theory and Source Coding,
Part II--Error Control Coding (Channel Coding), and Part III--Coding for Secure Communications.
Part I contains two chapters---Chapter 1 deals with the concept of information and its efficient
representation. Efficient representation of information leads to data compression. The chapter
also introduces the concept of run length coding, the rate distortion function and the design of
an optimal quantizer. The chapter concludes by giving a brief introduction to image compression.
　　Chapter 2 deals with the concepts of a communication channel and the channel capacity. This
chapter tries to answer the question: How many bits per second can be sent over a channel of a
given bandwidth and for a given signal to noise ratio It also brings out the need for error
control coding.
　Part II contains five chapters, all on error control coding--Chapter 3 introduces the reader to
the class of linear block codes. Linear block codes are useful, instructive and simple. Encoding
and decoding strategies are discussed for this class of codes. The notions of perfect codes,
optimal linear codes and maximum distance separable (MDS) codes are also introduced.
　Chapter 4 deals with cyclic codes, a sub-class of linear block codes. Cyclic codes are particularly useful for burst error correction. Fire codes, Golay codes and Cyclic Redundancy Check (CRC) codes are discussed as specific examples of cyclic codes. The chapter concludes with a section on the circuit implementation of cyclic codes.
　 Chapter 5 takes the reader to the world of Bose-Chaudhuri Hocquenghem (BCH) codes, a very
powerful class of multiple error correcting codes. The Reed-Solomon codes, a sub-class of BCH
codes, is also discussed in this chapter.
　 Chapter 6 takes a look at convolutional codes, which are essentially codes with memory. The
concept of Trellis codes is introduced to the reader and the Viterbi decoding technique is
discussed in detail. Some known good convolutional codes are also studied. Finally, the reader
is given a flavor of the not-so-old Turbo codes.
　Chapter 7 on Trellis Coded Modulation (TCM) talks about the combined coding and
modulation schemes. TCM encoding and decoding are discussed. The reader also learns how
to design TCM schemes for additive white Gaussian noise channels as well as fading channels.
Part III contains a chapter on Cryptography--Chapter 8 looks at yet another application of
coding, i.e. secure communications. The chapter discusses both the secret key and public key
encryption techniques using specific examples. Other techniques such as one-way hashing and
encryption using chaos functions are also discussed. The chapter concludes with a note on the
politics of cryptography.
　 I have tried to include numerical examples as and when required. Each chapter ends with a
concluding remark, which contains brief historical notes describing the origins of the important
results and contributions. Also, there is a telegraphic summary at the end of each chapter, which
can be used as a ready reference, or a quick search mechanism for a particular formula or
definition, or simply a confidence builder prior to an exam. The end of the chapter problems
should help the enthusiastic reader crystallize the concepts discussed in the text. I have also
added computer-based exercises at the end of each chapter. It is recommended that these
computer problems should be made a part of learning this subject.
　 I have tried my best to free the book from all errors. Unfortunately there does not exist a
foolproof error control technique for that! I have tried to include all the important, practical and interesting concepts related to this area. Comments from the readers regarding errors, omissions and other constructive suggestions are welcome at rbose@ee.iitd.ac.in Finally, I would like to quote Blaise Pascal, who said, "The last thing one knows when writing a book is what to put firsf .
　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　RANJAN Bose
　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　New Delhi

(美)Ranjan Bose：Ranjan Bose: 印度理工学院（IIT）电力工程系的副教授。他在IIT（Kanpur分校）的电力工程系获得工学学士学位，美国宾夕法尼亚大学电力工程系获得硕士和博士学位。之后在Alliance半导体公司任高级设计工程师。自1997年11月，他成为印度技术学院的教员。Bose博士经常做编码和密码学方面的演讲。他在1999年获得URSI青年科学家奖，在2000年7月获得Humboldt研究金。

Preface　　　　　　　　　　　　　　　　　
Acknowledgements　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　
Part I
Information Theory and Source Coding
1.　Source Coding　　　　　　　　　　　　　3
1.1　 Introduction to Information Theory　3
1.2　 Uncertainty And Information　4
1.3　 Average Mutual Information And Entropy　11
1.4　 Information Measures For Continuous Random Variables　14
1.5　 Source Coding Theorem　15
1.6　HuffmanCoding　21
1.7　 The Lempel-Ziv Algorithm　28
1.8　 Run Length Encoding and the PCX Format　30
1.9　 Rate Distortion Function　33
1.10　 Optimum Quantizer Design　36
1.11　 Introduction to Image Compression　37
1.19　 TheJpeg Standard for Lossless Compression　38
1.13　 TheJpeg Standard for Lossy Compression　39
1.14　 Concluding Remarks　41
Summary　42
Problems 44
Computer Problems 46
2. Channel Capacity and Coding　　　　　　　　　　　　　　　　　　　　 47
2.1　 Introduction　47
2.2　 Channel Models　48
2.3　Channel Capacity　50
2.4　 Channel Coding　52
2.5　 Information Capacity Theorem　56
2.6　The Shannon Limit　59
2.7　 Random Selection of Codes　61
2.8　 Concluding Remarks　67
Summary　68
Problems　69
Computer Problems　71
Part II
Error Control Coding
(Channel Coding)
3.　Linear Block Codes for Error Correction　　　　　　　　　　　　　　　　　　75
3.1　 Introduction to Error Correcting Codes　75
3.2　 Basic Definitions　77
3.3　 Matrix Description of Linear Block Codes　81
3.4　 Equivalent Codes　82
3.5　 Parity Check Matrix　85
3.6　 Decoding of a Linear Block Code　87
3.7　 Syndrome Decoding　94
3.8　 Error Probability after Coding (Probability of Error Correction)　95
3.9　 Perfect Codes　97
3.10　 Hamming Codes　100
3.11　 Optimal Linear Codes　102
3.12　 Maximum Distance Separable (MDS) Codes　102
3.13　 Concluding Remarks　102
Summary　103
Problems　105
Computer Problems　106
4.　Cyclic Codes　　　　　　　　　　　　　　　　　　　　　　　 108
4.1　 Introduction to Cyclic Codes　108
4.2　 Polynomials　109
4.3　 The Division Algorithm for Polynomials　110
4.4　 A Method for Generating Cyclic Codes　115
4.5　 Matrix Description of Cyclic Codes　119
4.6　 Burst Error Correction　121
4.7　 Fire Codes　123
4.8　 Golay Codes　124
4.9　 Cyclic Redundancy Check (CRC) Codes　125
4.10　 Circuit Implementation of Cyclic Codes　128
4.11　 Concluding Remarks　132
Summary　132
Problems　134
Computer Problems　135
5.　Bose-Chaudhuri Hocquenghem (BCH) Codes
5.1　 Introduction to BCH Codes　136
5.2　 Primitive Elements　137
5.3　 Minimal Polynomials　133
5.4　 Generator Polynomials in Terms of Minimal Polynomials141
5.5　 Some Examples of BCH Codes　143
5.6　 Decoding of BCH Codes　147
5.7　 Reed-Solomon Codes　150
5.8　 Implementation of Reed-Solomon Encoders and Decoders 153
5.9　Nested Codes　153
5.10　 Concluding Remarks　155
Summary　156
Problems　157
Computer Problems　158
6.　Convolutional Codes
6.1　 Introduction to Convolutional Codes　159
6.2　 Tree Codes and Trellis Codes　160
6.3　 Polynomial Description of Convolutional Codes
(Analytical Representation)　165
6.4　 Distance Notions for Convolutional Codes　170
6.5　 The Generating Function　173
6.6　 Matrix Description of Convolutional Codes　776
6.7　 Viterbi Decoding of Convolutional Codes　178
6.8　 Distance Bounds for Convolutional Codes　185
6.9　 Performance Bounds　187
6.10　 Known Good Convolutional Codes　188
6.11　 Turbo Codes　190
6.12　 Turbo Decoding　192
6.13　 Concluding Remarks　198
Summary　199
Problems 201
Computer Problems 203
7. Trellis Coded Modulation　　　　　　　　　　　　　　　　　　　　　　　　 206
7.1　 Introduction to TCM　206
7.2　 The Concept of Coded Modulation　207
7.3　Mapping by Set Partitioning 212
7.4　 Ungerboeck's TCM Design Rules　216
7.5　 Tcm Decoder　220
7.6　 Performance Evaluation for Awgn Channel　221
7.7　 Computation of dfree　227
7.8　 Tcm for Fading Channels　228
7.9　 Concluding Remarks　232
Summary　233
Problems 234
Computer Problems 238
Part III
Coding for Secure Communications
8.　Cryptography　　　　　　　　　　　　　　241
8.1　Introduction to Cryptography　241
8.2　 An Overview of Encryption Techniques　242
8.3　 Operations Used By Encryption Algorithms　245
8.4　Symmetric (Secret Key) Cryptography　246
8.5　Data Encryption Standard (DES)　248
8.6　International Data Encryption Algorithm (IDEA)　252
8.7　 RC Ciphers　253
8.8　 Asymmetric (Public-Key) Algorithms　254
8.9　The RSA Algorithm　254
8.10　Pretty Good Privacy (PGP)　256
8.11　 One-Way Hashing　253
8.12　 Other Techniques　260
8.13　 Secure Communication Using Chaos Functions　261
8.14　Cryptanalysis 262
8.15　Politics of Cryptography　264
8.16　 Concluding Remarks　265
Summary　268
Problems 269
Computer Problems 271
Index　　　　　　　　　　　　　　　 273