| Preface | p. xiii |
| Introduction to Adaptive Filtering | p. 1 |
| Introduction | p. 1 |
| Adaptive Signal Processing | p. 3 |
| Introduction to Adaptive Algorithms | p. 5 |
| Applications | p. 8 |
| Fundamentals of Adaptive Filtering | p. 15 |
| Introduction | p. 15 |
| Signal Representation | p. 16 |
| Deterministic Signals | p. 16 |
| Random Signals | p. 17 |
| Ergodicity | p. 25 |
| The Correlation Matrix | p. 27 |
| Wiener Filter | p. 38 |
| Linearly-Constrained Wiener Filter | p. 42 |
| The Generalized Sidelobe Canceller | p. 45 |
| Mean-Square Error Surface | p. 47 |
| Bias and Consistency | p. 49 |
| Newton Algorithm | p. 50 |
| Steepest-Descent Algorithm | p. 52 |
| Applications Revisited | p. 58 |
| System Identification | p. 58 |
| Signal Enhancement | p. 60 |
| Signal Prediction | p. 61 |
| Channel Equalization | p. 62 |
| Digital Communication System | p. 66 |
| Concluding Remarks | p. 68 |
| The Least-Mean-Square (LMS) Algorithm | p. 79 |
| Introduction | p. 79 |
| The LMS Algorithm | p. 80 |
| Some Properties of the LMS Algorithm | p. 82 |
| Gradient Behavior | p. 83 |
| Convergence Behavior of the Coefficient Vector | p. 83 |
| Coefficient-Error-Vector Covariance Matrix | p. 86 |
| Behavior of the Error Signal | p. 89 |
| Minimum Mean-Square Error | p. 90 |
| Excess Mean-Square Error and Misadjustment | p. 91 |
| Transient Behavior | p. 94 |
| LMS Algorithm Behavior in Nonstationary Environments | p. 96 |
| Examples | p. 101 |
| Analytical Examples | p. 101 |
| System Identification Simulations | p. 111 |
| Channel Equalization Simulations | p. 120 |
| Fast Adaptation Simulations | p. 122 |
| The Linearly-Constrained LMS Algorithm | p. 126 |
| Concluding Remarks | p. 128 |
| LMS-Based Algorithms | p. 139 |
| Introduction | p. 139 |
| Quantized-Error Algorithms | p. 140 |
| Sign-Error Algorithm | p. 141 |
| Dual-Sign Algorithm | p. 150 |
| Power-of-Two Error Algorithm | p. 151 |
| Sign-Data Algorithm | p. 152 |
| The LMS-Newton Algorithm | p. 153 |
| The Normalized LMS Algorithm | p. 157 |
| The Transform-Domain LMS Algorithm | p. 159 |
| The Affine Projection Algorithm | p. 169 |
| Simulation Examples | p. 174 |
| Signal Enhancement Simulation | p. 177 |
| Signal Prediction Simulation | p. 182 |
| Concluding Remarks | p. 183 |
| Conventional RLS Adaptive Filter | p. 195 |
| Introduction | p. 195 |
| The Recursive Least-Squares Algorithm | p. 196 |
| Properties of the Least-Squares Solution | p. 201 |
| Orthogonality Principle | p. 201 |
| Relation Between Least-Squares and Wiener Solutions | p. 204 |
| Influence of the Deterministic Autocorrelation Initialization | p. 204 |
| Steady-State Behavior of the Coefficient Vector | p. 205 |
| Coefficient-Error-Vector Covariance Matrix | p. 208 |
| Behavior of the Error Signal | p. 209 |
| Excess Mean-Square Error and Misadjustment | p. 213 |
| Behavior in Nonstationary Environments | p. 218 |
| Simulation Examples | p. 222 |
| Concluding Remarks | p. 226 |
| Adaptive Lattice-Based RLS Algorithms | p. 235 |
| Introduction | p. 235 |
| Recursive Least-Squares Prediction | p. 236 |
| Forward Prediction Problem | p. 236 |
| Backward Prediction Problem | p. 240 |
| Order-Updating Equations | p. 242 |
| A New Parameter [delta](k, i) | p. 243 |
| Order Updating of [xi superscript d subscript b subscript min](k, i) and w[subscript b](k, i) | p. 245 |
| Order Updating of [xi superscript d subscript f subscript min](k, i) and w[subscript f](k, i) | p. 246 |
| Order Updating of Prediction Errors | p. 247 |
| Time-Updating Equations | p. 249 |
| Time Updating for Prediction Coefficients | p. 249 |
| Time Updating for [delta](k, i) | p. 251 |
| Order Updating for [gamma](k, i) | p. 254 |
| Joint-Process Estimation | p. 257 |
| Time Recursions of the Least-Squares Error | p. 260 |
| Normalized Lattice RLS Algorithm | p. 265 |
| Basic Order Recursions | p. 265 |
| Feedforward Filtering | p. 267 |
| Error-Feedback Lattice RLS Algorithm | p. 269 |
| Recursive Formulas for the Reflection Coefficients | p. 272 |
| Lattice RLS Algorithm Based on A Priori Errors | p. 273 |
| Quantization Effects | p. 275 |
| Concluding Remarks | p. 280 |
| Fast Transversal RLS Algorithms | p. 287 |
| Introduction | p. 287 |
| Recursive Least-Squares Prediction | p. 288 |
| Forward Prediction Relations | p. 289 |
| Backward Prediction Relations | p. 290 |
| Joint-Process Estimation | p. 292 |
| Stabilized Fast Transversal RLS Algorithm | p. 295 |
| Concluding Remarks | p. 303 |
| QR-Decomposition-Based RLS Filters | p. 309 |
| Introduction | p. 309 |
| Triangularization Using QR-Decomposition | p. 310 |
| Initialization Process | p. 311 |
| Input data matrix triangularization | p. 312 |
| QR-Decomposition RLS Algorithm | p. 320 |
| Systolic Array Implementation | p. 325 |
| Some Implementation Issues | p. 334 |
| Fast QR-RLS Algorithm | p. 335 |
| Backward Prediction Problem | p. 339 |
| Forward Prediction Problem | p. 341 |
| Conclusions and Further Reading | p. 350 |
| Adaptive IIR Filters | p. 361 |
| Introduction | p. 361 |
| Output-Error IIR Filters | p. 362 |
| General Derivative Implementation | p. 369 |
| Adaptive Algorithms | p. 371 |
| Recursive least-squares algorithm | p. 371 |
| The Gauss-Newton algorithm | p. 372 |
| Gradient-based algorithm | p. 375 |
| Alternative Adaptive Filter Structures | p. 375 |
| Cascade Form | p. 376 |
| Lattice Structure | p. 378 |
| Parallel Form | p. 382 |
| Frequency-Domain Parallel Structure | p. 383 |
| Mean-Square Error Surface | p. 394 |
| Influence of the Filter Structure on MSE Surface | p. 402 |
| Alternative Error Formulations | p. 405 |
| Equation Error Formulation | p. 405 |
| The Steiglitz-McBride Method | p. 409 |
| Conclusion | p. 413 |
| Nonlinear Adaptive Filtering | p. 423 |
| Introduction | p. 423 |
| The Volterra Series Algorithm | p. 424 |
| LMS Volterra Filter | p. 427 |
| RLS Volterra Filter | p. 431 |
| Adaptive Bilinear Filters | p. 439 |
| Multilayer Perceptron Algorithm | p. 445 |
| Radial Basis Function Algorithm | p. 451 |
| Conclusion | p. 460 |
| Subband Adaptive Filters | p. 467 |
| Introduction | p. 467 |
| Multirate Systems | p. 468 |
| Decimation and Interpolation | p. 468 |
| Filter Banks | p. 471 |
| Two-Band Perfect Reconstruction Filter Banks | p. 477 |
| Analysis of Two-Band Filter Banks | p. 478 |
| Analysis of M-Band Filter Banks | p. 478 |
| Hierarchical M-Band Filter Banks | p. 479 |
| Cosine-Modulated Filter Banks | p. 479 |
| Block Representation | p. 481 |
| Subband Adaptive Filters | p. 481 |
| Subband Identification | p. 486 |
| Two-Band Identification | p. 487 |
| Closed-Loop Structure | p. 488 |
| Cross-Filters Elimination | p. 494 |
| Fractional Delays | p. 497 |
| Delayless Subband Adaptive Filtering | p. 501 |
| Computational Complexity | p. 508 |
| Frequency-Domain Adaptive Filtering | p. 510 |
| Conclusion | p. 519 |
| Quantization Effects in the LMS and RLS Algorithms | p. 527 |
| Quantization Effects in the LMS Algorithm | p. 527 |
| Error Description | p. 527 |
| Error Models for Fixed-Point Arithmetic | p. 529 |
| Coefficient-Error-Vector Covariance Matrix | p. 530 |
| Algorithm Stop | p. 533 |
| Mean-Square Error | p. 533 |
| Floating-Point Arithmetic Implementation | p. 534 |
| Floating-Point Quantization Errors in LMS Algorithm | p. 537 |
| Quantization Effects in the RLS Algorithm | p. 540 |
| Error Description | p. 540 |
| Error Models for Fixed-Point Arithmetic | p. 543 |
| Coefficient-Error-Vector Covariance Matrix | p. 544 |
| Algorithm Stop | p. 548 |
| Mean-Square Error | p. 549 |
| Fixed-Point Implementation Issues | p. 549 |
| Floating-Point Arithmetic Implementation | p. 550 |
| Floating-Point Quantization errors in RLS Algorithm | p. 554 |
| Table of Contents provided by Syndetics. All Rights Reserved. |