Haykin examines both the mathematical theory behind various linear adaptive filters with finite-duration impulse response (FIR) and the elements of supervised neural networks. This edition has been updated and refined to keep current with the field and develop concepts in as unified and accessible a manner as possible. It: introduces a completely new chapter on Frequency-Domain Adaptive Filters; adds a chapter on Tracking Time-Varying Systems; adds two chapters on Neural Networks; enhances material on RLS algorithms; strengthens linkages to Kalman filter theory to gain a more unified treatment of the standard, square-root and order-recursive forms; and includes new computer experiments using MATLAB software that illustrate the underlying theory and applications of the LMS and RLS algorithms.
CONTENTS
Preface
Acknowledgments
Background and Preview
- Chapter 1 Stochastic Processes and Models
- Chapter 2 Wiener Filters
- Chapter 3 Linear Prediction
- Chapter 4 Method of Steepest Descent
- Chapter 5 Least-Mean-Square Adaptive Filters
- Chapter 6 Normalized Least-Mean-Square Adaptive Filters
- Chapter 7 Frequency-Domain and Subband Adaptive Filters
- Chapter 8 Method of Least Squares
- Chapter 9 Recursive Least-Square Adaptive Filters
- Chapter 10 Kalman Filters
- Chapter 11 Square-Root Adaptive Filters
- Chapter 12 Order-Recursive Adaptive Filters
- Chapter 13 Finite-Precision Effects
- Chapter 14 Tracking of Time-Varying Systems
- Chapter 15 Adaptive Filters Using Infinite-Duration Impulse Response Structures
- Chapter 16 Blind Deconvolution
- Chapter 17 Back-Propagation Learning
Epilogue
- Appendix A Complex Variables
- Appendix B Differentiation with Respect to a Vector
- Appendix C Method of Lagrange Multipliers
- Appendix D Estimation Theory
- Appendix E Eigenanalysis
- Appendix F Rotations and Reflections
- Appendix G Complex Wishart Distribution
- Glossary
- Bibliography
- Index