Principles of adaptive filters and self-learning systems /

Adaptive systems or self-learning systems have been studied by many disciplines, resulting in many different names: intelligent systems, machine learning systems, neural networks, simulated annealing, etc. The main characteristic of these systems is that the system behavior is not fixed but can be self-modified according to the present input stimulus. Zaknich (Murdoch Univ.; Univ. of Western Australia) comprehensively presents classical and adaptive systems from the perspective of control and signal processing applications. Coverage includes a solid introduction of system theory and system modeling. Classical filtering and spectral analysis are presented, followed by a robust treatment of adaptive filter theory and adaptive control systems. Nonclassical approaches are next discussed, including neural networks, fuzzy logic, and genetic algorithms. Instead of technical depth in a particular system area as in many other works, this book provides a system perspective to the various approaches and offers a comparison to the classical and nonclassical system approaches. An excellent tutorial for graduate students and a comprehensive introduction for researchers working in adaptive systems. ^BSumming Up: Highly recommended. Upper-division undergraduates through professionals. J. Y. Cheung emeritus, Compsys Consulting