Browse Books

Reviewer: Klaus Galensa

Information theory provides a quantitative measure of the information contained in message signals, and allows us to determine the maximum capability of a system to transfer information from the source to the destination. In classical information theory, the system is simply a one-way point-to-point communication channel. Network information theory extends the classical theory from a simple channel to general networks with various sources and destinations and shared resources. Some of the techniques developed in network information theory, like network coding, multiple description coding, and successive cancellation decoding, can be found in real network implementations. This book is divided into 24 chapters. After a very brief introduction, chapters 2 and 3 present background on the information theory used throughout the book, particularly differential entropy and mutual information. The next 11 chapters discuss networks with single-round one-way communication, where each node is either a sender or a receiver. This section starts with a discussion of independent messages over noisy channels in many-to-one and one-to-many communication; continues with a review of multiple, interfering one-to-one communication and multiple-input multiple-output (MIMO) systems; and concludes with a description of correlated sources. The next six chapters deal with multi-hop networks, again beginning with independent messages and continuing to correlated sources. The final four chapters present some extensions of the theory. Of particular practical interest is the treatment of wireless fading channels in chapter 23, because wireless channels are time varying, not time invariant, due to multiple signal paths and user mobility. The wireless fading channel model captures these effects very well. Each chapter concludes with a summary, which is basically a table of contents, bibliographic notes, several exercises (solutions aren't provided), and occasionally an appendix with proofs of lemmas and propositions. As might be expected, the mathematics is sometimes heavy, but at least not obscure. The illustrations are very helpful. The authors cover an impressively large part of information theory; furthermore, they make it accessible. I do not know of any other book on the subject that contains so many topics of practical relevance to communication engineers. I therefore recommend the book to all students, researchers, and practitioners working in the field of communication. Online Computing Reviews Service