THIS BOOK IS an excellent first level textbook aimed principally at graduates and post-graduates studying random and statistical signal processing. Among other things it is meant to provide a comprehensive and detailed survey of the field that students can cover in two semesters. This survey is well done: it is efficient, lucid and engaging. For teaching purposes the book can be faulted little. Well-organized and progressive exercises are included for students' use. For computer users, a floppy diskette (3.5") containing data for computer-aided analysis is supplied, a useful and thoughtful inclusion. The reader is also recommended to use MATLAB computer language with these exercises.
The book is divided into an introduction and three parts. The introduction covers the motivation behind random and statistical analysis and is very instructional in its explanation of the problems considered. Notwithstanding, the book assumes that the reader has completed a course in discrete-time signals and systems and has a working knowledge of probability theory and statistics.
Part I deals with random processes. Part II concerns itself with classical statistical signal processing and part III deals with modern statistical signal processing.
The first part describes in systematic and easy-to-handle fashion random vectors' processes and their second moment analysis and linear transformation. Chapters 2 and 3 mainly portray a correlation analysis of random vectors, properties of Markov's processes, martingales, periodic and almost periodic random processes. Chapter 4 gives frequency-and transform-domain descriptions of random processes, the discrete Karhunen-Loeve Transform, white noise and higher-order moments of random processes. Chapter 5 deals with Linear Shift-Invariant Transforms and their spectral