| On a Class of Exponentiated Adaptive Algorithms for the Identification of Sparse Impulse Responses | p. 1 |
| Introduction | p. 1 |
| Derivation of the Different Algorithms | p. 2 |
| Link Between the LMS and EG Algorithms and Normalized Versions | p. 5 |
| The RLS and ERLS Algorithms | p. 9 |
| Link Between the PNLMS and EG[plus and minus] Algorithms | p. 10 |
| Application of the EG[plus and minus] Algorithm for Blind Channel Identification | p. 12 |
| Simulations | p. 15 |
| Conclusions | p. 20 |
| References | p. 21 |
| Adaptive Feedback Cancellation in Hearing Aids | p. 23 |
| Introduction | p. 23 |
| Steady-State Analysis | p. 26 |
| The Feedback Path | p. 29 |
| Real-World Processing Concerns | p. 30 |
| Feedback Cancellation System | p. 32 |
| Initialization | p. 32 |
| Running Adaptation | p. 35 |
| Performance Metric | p. 36 |
| Constrained Adaptation | p. 37 |
| Adaptation with Clamp | p. 37 |
| Adaptation with Cost Function | p. 40 |
| Simulation Results | p. 41 |
| Filtered-X System | p. 44 |
| Room Reverberation Effects | p. 47 |
| Test Configuration | p. 47 |
| Initialization and Measurement Procedure | p. 49 |
| Measured Feedback Path | p. 49 |
| Maximum Stable Gain | p. 52 |
| Conclusions | p. 53 |
| References | p. 55 |
| Single-Channel Acoustic Echo Cancellation | p. 59 |
| Introduction | p. 59 |
| Settings | p. 61 |
| Loudspeaker-Enclosure-Microphone Systems | p. 61 |
| Electronic Replica of LEM Systems | p. 63 |
| Speech Signals | p. 65 |
| Background Noise | p. 67 |
| Regulations | p. 68 |
| Methods for Acoustic Echo Control | p. 69 |
| Loss Control | p. 70 |
| Echo Cancellation | p. 70 |
| Echo Suppression | p. 71 |
| Adaptive Algorithms | p. 72 |
| NLMS Algorithm | p. 73 |
| AP Algorithm | p. 74 |
| RLS Algorithm | p. 76 |
| Adaptation Control | p. 77 |
| Optimal Step Size for the NLMS Algorithm | p. 78 |
| A Method for Estimating the Optimal Step Size | p. 80 |
| Suppression of Residual Echoes | p. 82 |
| Processing Structures | p. 87 |
| Fullband Processing | p. 87 |
| Block Processing | p. 88 |
| Subband Processing | p. 88 |
| Conclusions | p. 89 |
| References | p. 91 |
| Multichannel Frequency-Domain Adaptive Filtering with Application to Multichannel Acoustic Echo Cancellation | p. 95 |
| Introduction | p. 95 |
| General Derivation of Multichannel Frequency-Domain Algorithms | p. 98 |
| Optimization Criterion | p. 98 |
| Normal Equation | p. 102 |
| Adaptation Algorithm | p. 103 |
| Convergence Analysis | p. 105 |
| Analysis Model | p. 106 |
| Convergence in Mean | p. 107 |
| Convergence in Mean Square | p. 107 |
| Generalized Frequency-Domain Adaptive MIMO Filtering | p. 110 |
| Approximation and Special Cases | p. 112 |
| Approximation of the Frequency-Domain Kalman Gain | p. 113 |
| Special Cases | p. 114 |
| A Dynamical Regularization Strategy | p. 116 |
| Efficient Multichannel Realization | p. 117 |
| Efficient Calculation of the Frequency-Domain Kalman Gain | p. 117 |
| Dynamical Regularization for Proposed Kalman Gain Approach | p. 119 |
| Efficient DFT Calculation of Overlapping Data Blocks | p. 120 |
| Simulations and Real-World Applications | p. 122 |
| Multichannel Acoustic Echo Cancellation | p. 123 |
| Adaptive MIMO Filtering for Hands-Free Speech Communication | p. 125 |
| Conclusions | p. 126 |
| References | p. 127 |
| Filtering Techniques for Noise Reduction and Speech Enhancement | p. 129 |
| Introduction | p. 129 |
| Noise Reduction with an Array | p. 131 |
| Adaptive Noise Cancellation | p. 137 |
| Spectral Modification with a Single Microphone | p. 144 |
| Parametric Spectral Subtraction | p. 145 |
| Estimation of the Noise Spectrum | p. 146 |
| Parametric Wiener Filtering | p. 148 |
| Estimation of the Wiener Gain Filter | p. 150 |
| Conclusions | p. 151 |
| References | p. 153 |
| Adaptive Beamforming for Audio Signal Acquisition | p. 155 |
| Introduction | p. 155 |
| Signal Model, Sensor Arrays, and Concepts | p. 157 |
| Sensor Array, Sensor Signals, and Beamformer Setup | p. 157 |
| Interference-Independent Beamformer Performance Measures | p. 159 |
| Interference-Dependent Beamformer Performance Measures | p. 161 |
| Data-Independent Beamformer Design | p. 162 |
| Optimum Data-Dependent Beamformer Designs | p. 165 |
| Least-Squares Error (LSE) Design | p. 166 |
| Least-Squares Formulation of Linearly Constrained Minimum Variance (LCMV) Beamforming: LCMV-LS Design | p. 169 |
| Eigenvector Beamformers | p. 175 |
| Suppression of Correlated Interference | p. 177 |
| Adaptation of LCMV-LS Beamformers | p. 178 |
| A Practical Audio Acquisition System Using a Robust GSC | p. 180 |
| Spatio-Temporal Constraints | p. 181 |
| Robust GSC After [1] as an LCMV-LS Beamformer with Spatio-Temporal Constraints | p. 182 |
| Realization in the DFT-Domain | p. 185 |
| Experimental Evaluation | p. 186 |
| Conclusions | p. 187 |
| References | p. 188 |
| Blind Source Separation of Convolutive Mixtures of Speech | p. 195 |
| Introduction | p. 195 |
| What Is BSS? | p. 196 |
| Mixed Signal Model for Speech Signals in a Room | p. 197 |
| Unmixed Signal Model | p. 197 |
| Task of Blind Source Separation of Speech Signals | p. 198 |
| Instantaneous Mixtures vs. Convolutive Mixtures | p. 198 |
| Time-Domain Approach vs. Frequency-Domain Approach | p. 199 |
| Time-Domain Approach for Convolutive Mixtures | p. 199 |
| Frequency-Domain Approach for Convolutive Mixtures | p. 200 |
| Scaling and Permutation Problems | p. 200 |
| What Is ICA? | p. 201 |
| What Is Independence? | p. 201 |
| Minimization of Mutual Information | p. 202 |
| Maximization of Nongaussianity | p. 202 |
| Maximization of Likelihood | p. 202 |
| Three ICA Theories Are Identical | p. 203 |
| Learning Rules | p. 204 |
| How Speech Signals Can Be Separated? | p. 204 |
| Second Order Statistics vs. Higher Order Statistics | p. 205 |
| Second Order Statistics (SOS) Approach | p. 206 |
| Higher Order Statistics (HOS) Approach | p. 207 |
| Physical Interpretation of BSS | p. 207 |
| Frequency-Domain Adaptive Beamformer (ABF) | p. 208 |
| ABF for Target S[subscript 1] and Jammer S[subscript 2] | p. 208 |
| ABF for Target S[subscript 2] and Jammer S[subscript 1] | p. 210 |
| Two Sets of ABFs | p. 210 |
| Equivalence Between BSS and Adaptive Beamformers | p. 210 |
| When S[subscript 1] [not equal] 0 and S[subscript 2] [not equal] 0 | p. 211 |
| When S[subscript 1] [not equal] 0 and S[subscript 2] = 0 | p. 212 |
| When S[subscript 1] = 0 and S[subscript 2] [not equal] 0 | p. 213 |
| Separation Mechanism of BSS | p. 213 |
| Fundamental Limitation of BSS | p. 214 |
| When Sources Are Near the Microphones | p. 214 |
| When Sources Are Not "Independent" | p. 215 |
| BSS Is Upper Bounded by ABF | p. 216 |
| BSS Is an Intelligent Version of ABF | p. 216 |
| Sound Quality | p. 217 |
| Directivity Patterns of NBF, BSS, and ABF | p. 219 |
| Experimental Conditions | p. 220 |
| Mixing Systems | p. 220 |
| Source Signals | p. 220 |
| Evaluation Measure | p. 221 |
| Scaling and Permutation | p. 222 |
| Conclusions | p. 222 |
| References | p. 223 |
| Adaptive Multichannel Time Delay Estimation Based on Blind System Identification for Acoustic Source Localization | p. 227 |
| Introduction | p. 227 |
| Problem Formulation | p. 229 |
| Notation | p. 229 |
| Signal Model | p. 230 |
| Channel Properties and Assumptions | p. 231 |
| Generalized Multichannel Time Delay Estimation | p. 232 |
| The Principle | p. 232 |
| Time-Domain Multichannel LMS Approach | p. 233 |
| Frequency-Domain Adaptive Algorithms | p. 234 |
| Algorithm Implementation | p. 237 |
| Simulations | p. 238 |
| Experimental Setup | p. 238 |
| Performance Measure | p. 242 |
| Experimental Results | p. 242 |
| Conclusions | p. 243 |
| References | p. 246 |
| Algorithms for Adaptive Equalization in Wireless Applications | p. 249 |
| Introduction | p. 249 |
| Criteria for Equalization | p. 251 |
| Channel Equalization | p. 253 |
| Infinite Filter Length Solutions for Single Channels | p. 254 |
| Finite and Infinite Filter Length Solutions for Multiple Channels | p. 255 |
| Finite Filter Length Solutions for Single Channels | p. 258 |
| Decision Feedback Equalizers | p. 261 |
| Adaptive Algorithms for Channel Equalization | p. 265 |
| Adaptively Minimizing ZF | p. 265 |
| Adaptively Minimizing MMSE | p. 267 |
| Training and Tracking | p. 267 |
| Channel Estimation | p. 269 |
| Channel Estimation in MIMO Systems | p. 270 |
| Estimation of Wireless Channels | p. 271 |
| Channel Estimation by Basis Functions | p. 271 |
| Channel Estimation by Predictive Methods | p. 272 |
| Maximum Likelihood Equalization | p. 273 |
| Viterbi Algorithm | p. 274 |
| Blind Algorithms | p. 275 |
| Conclusions | p. 277 |
| References | p. 277 |
| Adaptive Space-Time Processing for Wireless CDMA | p. 283 |
| Introduction | p. 283 |
| Data Model | p. 285 |
| The Blind 2D RAKE Receiver | p. 287 |
| The Blind 2D STAR | p. 288 |
| Decision-Feedback Identification (DFI) | p. 288 |
| Parallel and Soft DFI | p. 289 |
| Parallel and Hard DFI | p. 291 |
| Common and Soft DFI | p. 293 |
| Common and Hard DFI | p. 295 |
| Performance Gains of the DFI Versions | p. 300 |
| The Blind 1D-ST STAR | p. 301 |
| 1D-ST Structured Data Model | p. 302 |
| 2D STAR with Common DFI Reinterpreted | p. 302 |
| 1D-ST Structured DFI | p. 303 |
| Performance Gains of 1D-ST STAR over 2D STAR | p. 305 |
| The Pilot-Assisted 1D-ST STAR | p. 311 |
| Data Model with Pilot Signals | p. 311 |
| 1D-ST STAR with Conventional Pilot-Channel Use | p. 313 |
| 1D-ST STAR with Enhanced Pilot-Channel Use | p. 314 |
| 1D-ST STAR with Conventional Pilot-Symbol Use | p. 315 |
| 1D-ST STAR with Enhanced Pilot-Symbol Use | p. 316 |
| Performance Gains with Enhanced Pilot Use | p. 316 |
| Conclusions | p. 318 |
| References | p. 319 |
| The IEEE 802.11 System with Multiple Receive Antennas | p. 323 |
| Introduction | p. 323 |
| System Model with Multiple Receive Antennas | p. 326 |
| Estimating the Receive Antenna Weights | p. 328 |
| The IEEE 802.11 Distributed Coordination Function with Multiple Packet Reception | p. 328 |
| Throughput Analysis | p. 330 |
| All Successful Packets Acknowledged | p. 332 |
| Only a Single Successful Packet Acknowledged | p. 334 |
| Performance Results | p. 335 |
| Conclusions | p. 338 |
| References | p. 339 |
| Adaptive Estimation of Clock Skew and Different Types of Delay in the Internet Network | p. 341 |
| Introduction | p. 341 |
| Terminology and Problem Formulation | p. 342 |
| Delay Jitter Model | p. 344 |
| The Least-Squares (LS) Estimator | p. 344 |
| The Maximum Likelihood (ML) Estimator and Linear Programming | p. 346 |
| An Unbiased RLS Algorithm | p. 347 |
| Simulations | p. 348 |
| Conclusions | p. 349 |
| References | p. 351 |
| Index | p. 353 |
| Table of Contents provided by Ingram. All Rights Reserved. |