目錄:LIST OF SYMBOLS<br>part ⅠGENERAL INTRODUCTION Objectives of Part Ⅰ<br>1 ADAPTIVE SYSTEMS<br>Definition. and Characteristics<br>Areas of Application<br>General Properties<br>Open- and Closed-Loop Adaptation<br>Applications of Closed-Loop Adaptation<br>Example of an Adaptive System<br>The Chapters Ahead<br><br>2 THE ADAPTIVE LINEAR COMBINER<br>General Description<br>Input Signal and Weight Vectors<br>Desired Response and Error<br>The Performance Function<br>Gradient and Minimum Mean-Square Error<br>Example of a Performance Surface<br>Alternative Expression of the GradiEnt<br>Decorrelation of Error and Input Components<br>Exercises<br><br>part Ⅱ THEORY OF ADAPTATION WITH STATIONARY SIGNALS<br>Objectives of Part Ⅱ<br>3 PROPERTIES OF THE QUADRATIC PERFORMANCE SURFACE<br>Normal Form of the Input Correlation Matrix<br>Eigenvalues and Eigenvectors of the Input Correlation Matrix<br>An Example with Two Weights<br>Geometrical Significance of Eigenvectors and Eigenvalues<br>A Second Example<br>Exercises<br><br>4 SEARCHING THE PERFORMANCE SURFACE<br>Methods of Searching the Performance Surface<br>Basic Ideas of Gradient Search Methods<br>A Simple Gradient Search Algorithm and Its Solution<br>Stability and Rate of Convergence<br>The Learning Curve<br>Gradient Search by Newton's Method<br>Newton's Method in Multidimensional Space<br>Gradient Search by the Method of Steepest Descent<br>Comparison of Learning Curves<br>Exercises<br><br>5 GRADIENT ESTIMATION AND ITS EFFECTS ON ADAPTATION<br>Gradient Component Estimation by Derivative Measurement<br>part Ⅲ ADAPTIVE AL GORITHMS AND STRUCTURES<br>part Ⅳ APPLICATIONS<br>APPENDIX A A Portable Random Number Generator<br>INDEX