This instructional and reference text represents the proceedings of the Arab school on science and technology held in Zabadani (Syria) in 1983 and was edited by Thomas Kailath from the Stanford University. It contains 11 chapters written by authorities in the area of digital signal processing. Here are the chapters: (1) Fundamentals of digital signal processing (J. S. Lim; pp. 1-57); (2) Linear estimations for stationary and near-stationary processes (T. Kailath; pp. 59-128); (3) Spectral estimation with applications (S. L. Marple jun.; pp. 129-168); (4) Advanced digital filters (P. Dewilde; pp. 169-209); (5) Speech coding and processing (J. Makhoul; pp. 211-247); (6) Signal processing in the Arab world (M. Mrayati; pp. 249-260); (7) Signal processing in communications (J. G. Proakis; pp. 261-307); (8) Radar/sonar signal processing (H. J. Whitehouse; pp. 309-333); (9) Digital image processing: problems and methods (A. K. Jain; pp. 335-368); (10) Signal processing technology (H. J. Whitehouse; pp. 369-392); (11) VLSI array processor for signal processing (S. Y. Kung; pp. 393-440).
Chapter 1 gives a concise but comprehensive account of basic methods used in digital signal processing (DSP) such as: Z-transform, discrete Fourier transform, fast Fourier transform (FFT), finite impulse response filters (FIRs), infinite impulse response (IIR) digital filters.
Chapter 2 gives an extensive study of innovations for scalar and vector stationary stochastic processes, and Yule-Walker equations are derived. The theoretical basis of the Levinson-Durbin algorithm and of other ones related to it, is given in connection with their implementation. Schur algorithm is also presented. For nonstationary processes some extensions are given by introducing a measure of quasi-stationarity. The text represents an excellent account of the discussed problems and a rich and comprehensive bibliography is included at the end.
Chapter 3 represents a tutorial summary of modern spectral estimation methods. Here are given: classical methods, parametric methods (Autoregressive-moving average (ARMA) method, Prony method), and nonparametric methods (maximum likehood spectral estimation, Pisarenko method, MUSIC method). Various methods are discussed and compared with respect to the domain of application (Radar, Speech, Microwave arrays).
In chapter 4 two new advanced IIR classes of digital filters are presented. The first class contains digital orthogonal filters (DOF) which orthogonally map the input into the output, the second class are wave digital filters (WDF), which realize algebraically minimal mappings whose transfer is constrained in magnitude. It is shown that these filters give positive answers to two important problems of modern digital filters: a low sensitivity with respect to the variation of parameters and the absence of limit cycles. The theory, the synthesis and the implementation are extensively treated and Schur method is frequently used. The orthogonal embedding and the factorization of the transfer function are carefully discussed.
Chapter 5 gives an extensive presentation of various methods and techniques for coding Speech signals. The basic principles and performances with respect to the signal-to-noise ratio are discussed. Basic schemes for the implementation of these methods are given. The text represents a basic instance for the speech-coding domain.
Chapter 6 gives a brief account of the language dependent signal processing methods such as: synthesis, analysis and recognition of speech applied in the Arab world.
In chapter 7, methods and techniques applied in the modulation and demodulation of digital signals are presented. After a comprehensive classification of digital modulation techniques, the features of telephone channels and radio multipath channels are discussed in connection to the signal-to-noise ratio. Special attention is paid to adaptive techniques of demodulation.
In chapter 8, selected aspects of digital signal processing related to the real-time implementation for Radar/Sonar are discussed. Emphasis is on the large amount of computing needed for the ambiguity function and for the beam-forming. It results that the real-time operation could be obtained only by the use of new computation structures such as systolic arrays and matrix processors. In chapter 9, a review of digital image processing methods and techniques are presented. For each method, the theory, the advantages and the limitations are discussed. The operations considered here are the following: image representation, image enhancement, image restoration, image data compression, image data reconstruction from projections. Photographs illustrate the effects of the discussed operations.
Chapter 10 presents selected aspects of signal processing technology related to the real-time hardware implementation. The archetype of the FIR time invariant filter is taken to be the transverse filter which is implemented as: surface acoustic waves (SAW) devices or charge coupled devices (CCDs). It is pointed out that by the introduction of the chirp Z-Transform a parallel processing for FFT was achieved. It is argued that the new requirements of the real-time operation lead to a new architecture of digital processors in which many common set of matrix operations are to be performed in a very short time. This is realized by the use of systolic and wavefront arrays. In the last chapter, the architecture of very large scale integration (VLSI) array processors for signal processing is discussed. The systolic and wavefront arrays are analyzed.
The book gives an excellent account a) on the theory of digital signal processing (chapters 1, 2, 3, 4); b) on specifical methods and techniques used in DSP (chapters 5, 7, 8, 9); c) on perspectives of the large scale integration and real-time operation (chaters 10, 11) and at the same time establishes close relationships between the advanced theory, the algorithms and the implementation. Therefore we think that this excellent text will be very useful for the applied mathematician as well as for the research engineer working in the area of DSP.