{"product_id":"adaptive-blind-signal-and-image-processing-learning-algorithms-and-applications-hardback-9780471607915","title":"Adaptive Blind Signal and Image Processing; Learning Algorithms and Applications (Hardback) 9780471607915","description":"\u003cfont face=\"Georgia\"\u003e\r\n\u003cp\u003e\u003cfont size=\"6\"\u003eAdaptive Blind Signal and Image Processing\u003c\/font\u003e\u003cbr\u003e\r\n\u003cfont size=\"5\"\u003eLearning Algorithms and Applications\u003c\/font\u003e\u003c\/p\u003e\r\n\r\n\r\n\r\n\r\n\u003cp\u003e\u003cfont size=\"4\"\u003eAndrzej Cichocki (Author), Shun-ichi Amari (Author)\u003c\/font\u003e\u003c\/p\u003e\r\n\r\n\u003cp\u003e\u003cfont size=\"3\"\u003e9780471607915, Wiley\u003c\/font\u003e\u003c\/p\u003e\r\n\r\n\u003cp\u003e\u003cfont size=\"3\"\u003eHardback, published 26 April 2002\u003c\/font\u003e\u003c\/p\u003e\r\n\r\n\u003cp\u003e\u003cfont size=\"3\"\u003e528 pages\u003cbr\u003e25.2 x 17.6 x 3.7 cm, 1.157 kg\u003c\/font\u003e\u003c\/p\u003e\r\n\r\n\r\n\r\n\r\n\r\n\u003cp align=\"justify\"\u003e\u003cstrong\u003e\u003cfont size=\"3\"\u003eIm Mittelpunkt dieses modernen und spezialisierten Bandes stehen adaptive Strukturen und unüberwachte Lernalgorithmen, besonders im Hinblick auf effektive Computersimulationsprogramme. Anschauliche Illustrationen und viele Beispiele sowie eine interaktive CD-ROM ergänzen den Text.\u003c\/font\u003e\u003c\/strong\u003e\u003c\/p\u003e\r\n\r\n\u003cp\u003e\u003cfont size=\"3\"\u003e\u003cp\u003ePreface xxix\u003c\/p\u003e \u003cp\u003e\u003cb\u003e1 Introduction to Blind Signal Processing: Problems and Applications 1\u003c\/b\u003e\u003c\/p\u003e \u003cp\u003e1.1 Problem Formulations – An Overview 2\u003c\/p\u003e \u003cp\u003e1.1.1 Generalized Blind Signal Processing Problem 2\u003c\/p\u003e \u003cp\u003e1.1.2 Instantaneous Blind Source Separation and Independent Component Analysis 5\u003c\/p\u003e \u003cp\u003e1.1.3 Independent Component Analysis for Noisy Data 11\u003c\/p\u003e \u003cp\u003e1.1.4 Multichannel Blind Deconvolution and Separation 15\u003c\/p\u003e \u003cp\u003e1.1.5 Blind Extraction of Signals 19\u003c\/p\u003e \u003cp\u003e1.1.6 Generalized Multichannel Blind Deconvolution – State Space Models 20\u003c\/p\u003e \u003cp\u003e1.1.7 Nonlinear State Space Models – Semi-Blind Signal Processing 22\u003c\/p\u003e \u003cp\u003e1.1.8 Why State Space Demixing Models? 23\u003c\/p\u003e \u003cp\u003e1.2 Potential Applications of Blind and Semi-Blind Signal Processing 24\u003c\/p\u003e \u003cp\u003e1.2.1 Biomedical Signal Processing 25\u003c\/p\u003e \u003cp\u003e1.2.2 Blind Separation of Electrocardiographic Signals of Fetus and Mother 26\u003c\/p\u003e \u003cp\u003e1.2.3 Enhancement and Decomposition of EMG Signals 28\u003c\/p\u003e \u003cp\u003e1.2.4 EEG and MEG Data Processing 28\u003c\/p\u003e \u003cp\u003e1.2.5 Application of ICA\/BSS for Noise and Interference Cancellation in Multi-sensory Biomedical Signals 30\u003c\/p\u003e \u003cp\u003e1.2.6 Cocktail Party Problem 35\u003c\/p\u003e \u003cp\u003e1.2.7 Digital Communication Systems 36\u003c\/p\u003e \u003cp\u003e1.2.8 Image Restoration and Understanding 38\u003c\/p\u003e \u003cp\u003e\u003cb\u003e2 Solving a System of Algebraic Equations and Related Problems 43\u003c\/b\u003e\u003c\/p\u003e \u003cp\u003e2.1 Formulation of the Problem for Systems of Linear Equations 44\u003c\/p\u003e \u003cp\u003e2.2 Least-Squares Problems 45\u003c\/p\u003e \u003cp\u003e2.2.1 Basic Features of the Least-Squares Solution 45\u003c\/p\u003e \u003cp\u003e2.2.2 Weighted Least-Squares and Best Linear Unbiased Estimation 47\u003c\/p\u003e \u003cp\u003e2.2.3 Basic Network Structure-Least-Squares Criteria 48\u003c\/p\u003e \u003cp\u003e2.2.4 Iterative Parallel Algorithms for Large and Sparse Systems 49\u003c\/p\u003e \u003cp\u003e2.2.5 Iterative Algorithms with Non-negativity Constraints 51\u003c\/p\u003e \u003cp\u003e2.2.6 Robust Criteria and Iteratively Reweighted Least-Squares Algorithm 53\u003c\/p\u003e \u003cp\u003e2.2.7 Tikhonov Regularization and SVD 57\u003c\/p\u003e \u003cp\u003e2.3 Least Absolute Deviation (1-norm) Solution of Systems of Linear Equations 61\u003c\/p\u003e \u003cp\u003e2.3.1 Neural Network Architectures Using a Smooth Approximation and Regularization 62\u003c\/p\u003e \u003cp\u003e2.3.2 Neural Network Model for LAD Problem Exploiting Inhibition Principles 64\u003c\/p\u003e \u003cp\u003e2.4 Total Least-Squares and Data Least-Squares Problems 68\u003c\/p\u003e \u003cp\u003e2.4.1 Problems Formulation 68\u003c\/p\u003e \u003cp\u003e2.4.2 Total Least-Squares Estimation 70\u003c\/p\u003e \u003cp\u003e2.4.3 Adaptive Generalized Total Least-Squares 74\u003c\/p\u003e \u003cp\u003e2.4.4 Extended TLS for Correlated Noise Statistics 76\u003c\/p\u003e \u003cp\u003e2.4.5 An Illustrative Example - Fitting a Straight Line to a Set of Points 78\u003c\/p\u003e \u003cp\u003e2.5 Sparse Signal Representation and Minimum 1-norm Solution 80\u003c\/p\u003e \u003cp\u003e2.5.1 Approximate Solution of Minimum p-norm Problem Using Iterative LS Approach 81\u003c\/p\u003e \u003cp\u003e2.5.2 Uniqueness and Optimal Solution for Sparse Representation 84\u003c\/p\u003e \u003cp\u003e2.5.3 FOCUSS Algorithms 84\u003c\/p\u003e \u003cp\u003e\u003cb\u003e3 Principal\/Minor Component Analysis and Related Problems 87\u003c\/b\u003e\u003c\/p\u003e \u003cp\u003e3.1 Introduction 87\u003c\/p\u003e \u003cp\u003e3.2 Basic Properties of PCA 88\u003c\/p\u003e \u003cp\u003e3.2.1 Eigenvalue Decomposition 88\u003c\/p\u003e \u003cp\u003e3.2.2 Estimation of Sample Covariance Matrices 90\u003c\/p\u003e \u003cp\u003e3.2.3 Signal and Noise Subspaces - Automatic Choice of Dimensionality for PCA 91\u003c\/p\u003e \u003cp\u003e3.2.4 Basic Properties of PCA 94\u003c\/p\u003e \u003cp\u003e3.3 Extraction of Principal Components 95\u003c\/p\u003e \u003cp\u003e3.4 Basic Cost Functions and Adaptive Algorithms for PCA 99\u003c\/p\u003e \u003cp\u003e3.4.1 The Rayleigh Quotient – Basic Properties 99\u003c\/p\u003e \u003cp\u003e3.4.2 Basic Cost Functions for Computing Principal and Minor Components 100\u003c\/p\u003e \u003cp\u003e3.4.3 Fast PCA Algorithm Based on the Power Method 102\u003c\/p\u003e \u003cp\u003e3.4.4 Inverse Power Iteration Method 105\u003c\/p\u003e \u003cp\u003e3.5 Robust PCA 105\u003c\/p\u003e \u003cp\u003e3.6 Adaptive Learning Algorithms for MCA 108\u003c\/p\u003e \u003cp\u003e3.7 Unified Parallel Algorithms for PCA\/MCA and PSA\/MSA 111\u003c\/p\u003e \u003cp\u003e3.7.1 Cost Function for Parallel Processing 112\u003c\/p\u003e \u003cp\u003e3.7.2 Gradient of J(W) 113\u003c\/p\u003e \u003cp\u003e3.7.3 Stability Analysis 114\u003c\/p\u003e \u003cp\u003e3.7.4 Unified Stable Algorithms 117\u003c\/p\u003e \u003cp\u003e3.8 SVD in Relation to PCA and Matrix Subspaces 118\u003c\/p\u003e \u003cp\u003e3.9 Multistage PCA for BSS 120\u003c\/p\u003e \u003cp\u003e\u003cb\u003e4 Blind Decorrelation and SOS for Robust Blind Identification 129\u003c\/b\u003e\u003c\/p\u003e \u003cp\u003e4.1 Spatial Decorrelation - Whitening Transforms 130\u003c\/p\u003e \u003cp\u003e4.1.1 Batch Approach 130\u003c\/p\u003e \u003cp\u003e4.1.2 Optimization Criteria for Adaptive Blind Spatial Decorrelation 132\u003c\/p\u003e \u003cp\u003e4.1.3 Derivation of Equivariant Adaptive Algorithms for Blind Spatial Decorrelation 133\u003c\/p\u003e \u003cp\u003e4.1.4 Simple Local Learning Rule 136\u003c\/p\u003e \u003cp\u003e4.1.5 Gram-Schmidt Orthogonalization 138\u003c\/p\u003e \u003cp\u003e4.1.6 Blind Separation of Decorrelated Sources Versus Spatial Decorrelation 139\u003c\/p\u003e \u003cp\u003e4.1.7 Bias Removal for Noisy Data 139\u003c\/p\u003e \u003cp\u003e4.1.8 Robust Prewhitening - Batch Algorithm 140\u003c\/p\u003e \u003cp\u003e4.2 SOS Blind Identification Based on EVD 141\u003c\/p\u003e \u003cp\u003e4.2.1 Mixing Model 141\u003c\/p\u003e \u003cp\u003e4.2.2 Basic Principles: SD and EVD 143\u003c\/p\u003e \u003cp\u003e4.3 Improved Blind Identification Algorithms Based on EVD\/SVD 148\u003c\/p\u003e \u003cp\u003e4.3.1 Robust Orthogonalization of Mixing Matrices for Colored Sources 148\u003c\/p\u003e \u003cp\u003e4.3.2 An Improved Algorithm Based on GEVD 153\u003c\/p\u003e \u003cp\u003e4.3.3 An Improved Two-stage Symmetric EVD\/SVD Algorithm 155\u003c\/p\u003e \u003cp\u003e4.3.4 BSS and Identification Using a Bandpass Filters 156\u003c\/p\u003e \u003cp\u003e4.4 Joint Diagonalization - Robust SOBI Algorithms 157\u003c\/p\u003e \u003cp\u003e4.4.1 The Modified SOBI Algorithm for Nonstationary Sources: SONS Algorithm 160\u003c\/p\u003e \u003cp\u003e4.4.2 Computer Simulation Experiments 161\u003c\/p\u003e \u003cp\u003e4.4.3 Extensions of Joint Approximate Diagonalization Technique 162\u003c\/p\u003e \u003cp\u003e4.4.4 Comparison of the JAD and Symmetric EVD 163\u003c\/p\u003e \u003cp\u003e4.5 Cancellation of Correlation 164\u003c\/p\u003e \u003cp\u003e4.5.1 Standard Estimation of Mixing Matrix and Noise Covariance Matrix 164\u003c\/p\u003e \u003cp\u003e4.5.2 Blind Identification of Mixing Matrix Using the Concept of Cancellation of Correlation 165\u003c\/p\u003e \u003cp\u003e\u003cb\u003e5 Statistical Signal Processing Approach to Blind Signal Extraction 177\u003c\/b\u003e\u003c\/p\u003e \u003cp\u003e5.1 Introduction and Problem Formulation 178\u003c\/p\u003e \u003cp\u003e5.2 Learning Algorithms Using Kurtosis as a Cost Function 180\u003c\/p\u003e \u003cp\u003e5.2.1 A Cascade Neural Network for Blind Extraction of Non-Gaussian Sources with Learning Rule Based on Normalized Kurtosis 181\u003c\/p\u003e \u003cp\u003e5.2.2 Algorithms Based on Optimization of Generalized Kurtosis 184\u003c\/p\u003e \u003cp\u003e5.2.3 KuicNet Learning Algorithm 186\u003c\/p\u003e \u003cp\u003e5.2.4 Fixed-point Algorithms 187\u003c\/p\u003e \u003cp\u003e5.2.5 Sequential Extraction and Deflation Procedure 191\u003c\/p\u003e \u003cp\u003e5.3 On-Line Algorithms for Blind Signal Extraction of Temporally Correlated Sources 193\u003c\/p\u003e \u003cp\u003e5.3.1 On-Line Algorithms for Blind Extraction Using a Linear Predictor 195\u003c\/p\u003e \u003cp\u003e5.3.2 Neural Network for Multi-unit Blind Extraction 197\u003c\/p\u003e \u003cp\u003e5.4 Batch Algorithms for Blind Extraction of Temporally Correlated Sources 199\u003c\/p\u003e \u003cp\u003e5.4.1 Blind Extraction Using a First Order Linear Predictor 201\u003c\/p\u003e \u003cp\u003e5.4.2 Blind Extraction of Sources Using Bank of Adaptive Bandpass Filters 202\u003c\/p\u003e \u003cp\u003e5.4.3 Blind Extraction of Desired Sources Correlated with Reference Signals 205\u003c\/p\u003e \u003cp\u003e5.5 A Statistical Approach to Sequential Extraction of Independent Sources 206\u003c\/p\u003e \u003cp\u003e5.5.1 Log Likelihood and Cost Function 206\u003c\/p\u003e \u003cp\u003e5.5.2 Learning Dynamics 208\u003c\/p\u003e \u003cp\u003e5.5.3 Equilibrium of Dynamics 209\u003c\/p\u003e \u003cp\u003e5.5.4 Stability of Learning Dynamics and Newton’s Method 211\u003c\/p\u003e \u003cp\u003e5.6 A Statistical Approach to Temporally Correlated Sources 212\u003c\/p\u003e \u003cp\u003e5.7 On-line Sequential Extraction of Convolved and Mixed Sources 214\u003c\/p\u003e \u003cp\u003e5.7.1 Formulation of the Problem 214\u003c\/p\u003e \u003cp\u003e5.7.2 Extraction of Single i.i.d. Source Signal 215\u003c\/p\u003e \u003cp\u003e5.7.3 Extraction of Multiple i.i.d. Sources 217\u003c\/p\u003e \u003cp\u003e5.7.4 Extraction of Colored Sources from Convolutive Mixture 218\u003c\/p\u003e \u003cp\u003e5.8 Computer Simulations: Illustrative Examples 219\u003c\/p\u003e \u003cp\u003e5.8.1 Extraction of Colored Gaussian Signals 220\u003c\/p\u003e \u003cp\u003e5.8.2 Extraction of Natural Speech Signals from Colored Gaussian Signals 222\u003c\/p\u003e \u003cp\u003e5.8.3 Extraction of Colored and White Sources 222\u003c\/p\u003e \u003cp\u003e5.8.4 Extraction of Natural Image Signal from Interferences 224\u003c\/p\u003e \u003cp\u003e5.9 Concluding Remarks 224\u003c\/p\u003e \u003cp\u003e\u003cb\u003e6 Natural Gradient Approach to Independent Component Analysis 231\u003c\/b\u003e\u003c\/p\u003e \u003cp\u003e6.1 Basic Natural Gradient Algorithms 232\u003c\/p\u003e \u003cp\u003e6.1.1 Kullback–Leibler Divergence - Relative Entropy as a Measure of Stochastic Independence 232\u003c\/p\u003e \u003cp\u003e6.1.2 Derivation of Natural Gradient Basic Learning Rules 235\u003c\/p\u003e \u003cp\u003e6.2 Generalizations of the Basic Natural Gradient Algorithm 237\u003c\/p\u003e \u003cp\u003e6.2.1 Nonholonomic Learning Rules 237\u003c\/p\u003e \u003cp\u003e6.2.2 Natural Riemannian Gradient in Orthogonality Constraint 239\u003c\/p\u003e \u003cp\u003e6.3 NG Algorithms for Blind Extraction 242\u003c\/p\u003e \u003cp\u003e6.3.1 Stiefel and Grassmann-Stiefel Manifolds Approaches 242\u003c\/p\u003e \u003cp\u003e6.4 Generalized Gaussian Distribution Model 244\u003c\/p\u003e \u003cp\u003e6.4.1 Moments of the Generalized Gaussian Distribution 248\u003c\/p\u003e \u003cp\u003e6.4.2 Kurtosis and Gaussian Exponent 250\u003c\/p\u003e \u003cp\u003e6.4.3 The Flexible ICA Algorithm 250\u003c\/p\u003e \u003cp\u003e6.4.4 Pearson System 254\u003c\/p\u003e \u003cp\u003e6.5 Natural Gradient Algorithms for Non-stationary Sources 255\u003c\/p\u003e \u003cp\u003e6.5.1 Model Assumptions 255\u003c\/p\u003e \u003cp\u003e6.5.2 Second Order Statistics Cost Function 256\u003c\/p\u003e \u003cp\u003e6.5.3 Derivation of Natural Gradient Learning Algorithms 256\u003c\/p\u003e \u003cp\u003e\u003cb\u003e7 Locally Adaptive Algorithms for ICA and their Implementations 273\u003c\/b\u003e\u003c\/p\u003e \u003cp\u003e7.1 Modified Jutten-H´erault Algorithms for Blind Separation of Sources 274\u003c\/p\u003e \u003cp\u003e7.1.1 Recurrent Neural Network 274\u003c\/p\u003e \u003cp\u003e7.1.2 Statistical Independence 274\u003c\/p\u003e \u003cp\u003e7.1.3 Self-normalization 277\u003c\/p\u003e \u003cp\u003e7.1.4 Feed-forward Neural Network and Associated Learning Algorithms 278\u003c\/p\u003e \u003cp\u003e7.1.5 Multilayer Neural Networks 281\u003c\/p\u003e \u003cp\u003e7.2 Iterative Matrix Inversion Approach to the Derivation of a Family of Robust ICA Algorithms 284\u003c\/p\u003e \u003cp\u003e7.2.1 Derivation of Robust ICA Algorithm Using Generalized Natural Gradient Approach 287\u003c\/p\u003e \u003cp\u003e7.2.2 Practical Implementation of the Algorithms 288\u003c\/p\u003e \u003cp\u003e7.2.3 Special Forms of the Flexible Robust Algorithm 290\u003c\/p\u003e \u003cp\u003e7.2.4 Decorrelation Algorithm 290\u003c\/p\u003e \u003cp\u003e7.2.5 Natural Gradient Algorithms 290\u003c\/p\u003e \u003cp\u003e7.2.6 Generalized EASI Algorithm 290\u003c\/p\u003e \u003cp\u003e7.2.7 Non-linear PCA Algorithm 291\u003c\/p\u003e \u003cp\u003e7.2.8 Flexible ICA Algorithm for Unknown Number of Sources and their Statistics 292\u003c\/p\u003e \u003cp\u003e7.3 Blind Source Separation with Non-negativity Constraints 293\u003c\/p\u003e \u003cp\u003e7.4 Computer Simulations 294\u003c\/p\u003e \u003cp\u003e\u003cb\u003e8 Robust Techniques for BSS and ICA with Noisy Data 305\u003c\/b\u003e\u003c\/p\u003e \u003cp\u003e8.1 Introduction 305\u003c\/p\u003e \u003cp\u003e8.2 Bias Removal Techniques for Prewhitening and ICA Algorithms 306\u003c\/p\u003e \u003cp\u003e8.2.1 Bias Removal for Whitening Algorithms 306\u003c\/p\u003e \u003cp\u003e8.2.2 Bias Removal for Adaptive ICA Algorithms 307\u003c\/p\u003e \u003cp\u003e8.3 Blind Separation of Signals Buried in Additive Convolutive Reference Noise 310\u003c\/p\u003e \u003cp\u003e8.3.1 Learning Algorithms for Noise Cancellation 311\u003c\/p\u003e \u003cp\u003e8.4 Cumulant-Based Adaptive ICA Algorithms 314\u003c\/p\u003e \u003cp\u003e8.4.1 Cumulant-Based Cost Functions 314\u003c\/p\u003e \u003cp\u003e8.4.2 Family of Equivariant Algorithms Employing Higher Order Cumulants 315\u003c\/p\u003e \u003cp\u003e8.4.3 Possible Extensions 317\u003c\/p\u003e \u003cp\u003e8.4.4 Cumulants for Complex Valued Signals 318\u003c\/p\u003e \u003cp\u003e8.4.5 Blind Separation with More Sensors than Sources 318\u003c\/p\u003e \u003cp\u003e8.5 Robust Extraction of Arbitrary a Group of Source Signals 320\u003c\/p\u003e \u003cp\u003e8.5.1 Blind Extraction of Sparse Sources with Largest Positive Kurtosis Using Prewhitening and Semi-Orthogonality Constraint 320\u003c\/p\u003e \u003cp\u003e8.5.2 Blind Extraction of an Arbitrary Group of Sources without Prewhitening 323\u003c\/p\u003e \u003cp\u003e8.6 Recurrent Neural Network Approach for Noise Cancellation 325\u003c\/p\u003e \u003cp\u003e8.6.1 Basic Concept and Algorithm Derivation 325\u003c\/p\u003e \u003cp\u003e8.6.2 Simultaneous Estimation of a Mixing Matrix and Noise Reduction 328\u003c\/p\u003e \u003cp\u003e8.6.2.1 Regularization 329\u003c\/p\u003e \u003cp\u003e8.6.3 Robust Prewhitening and Principal Component Analysis (PCA) 331\u003c\/p\u003e \u003cp\u003e8.6.4 Computer Simulation Experiments for the Amari-Hopfield Network 331\u003c\/p\u003e \u003cp\u003e\u003cb\u003e9 Multichannel Blind Deconvolution: Natural Gradient Approach 335\u003c\/b\u003e\u003c\/p\u003e \u003cp\u003e9.1 SIMO Convolutive Models and Learning Algorithms for Estimation of a Source Signal 336\u003c\/p\u003e \u003cp\u003e9.1.1 Equalization Criteria for SIMO Systems 338\u003c\/p\u003e \u003cp\u003e9.1.2 SIMO Blind Identification and Equalization via Robust ICA\/BSS 340\u003c\/p\u003e \u003cp\u003e9.1.3 Feed-forward Deconvolution Model and Natural Gradient Learning Algorithm 342\u003c\/p\u003e \u003cp\u003e9.1.4 Recurrent Neural Network Model and Hebbian Learning Algorithm 343\u003c\/p\u003e \u003cp\u003e9.2 Multichannel Blind Deconvolution with Constraints Imposed on FIR Filters 346\u003c\/p\u003e \u003cp\u003e9.3 General Models for Multiple-Input Multiple-Output Blind Deconvolution 349\u003c\/p\u003e \u003cp\u003e9.3.1 Fundamental Models and Assumptions 349\u003c\/p\u003e \u003cp\u003e9.3.2 Separation-Deconvolution Criteria 351\u003c\/p\u003e \u003cp\u003e9.4 Relationships Between BSS\/ICA and MBD 354\u003c\/p\u003e \u003cp\u003e9.4.1 Multichannel Blind Deconvolution in the Frequency Domain 354\u003c\/p\u003e \u003cp\u003e9.4.2 Algebraic Equivalence of Various Approaches 355\u003c\/p\u003e \u003cp\u003e9.4.3 Convolution as a Multiplicative Operator 357\u003c\/p\u003e \u003cp\u003e9.4.4 Natural Gradient Learning Rules for Multichannel Blind Deconvolution (MBD) 358\u003c\/p\u003e \u003cp\u003e9.4.5 NG Algorithms for Double Infinite Filters 359\u003c\/p\u003e \u003cp\u003e9.4.6 Implementation of Algorithms for a Minimum Phase Non-causal System 360\u003c\/p\u003e \u003cp\u003e9.5 Natural Gradient Algorithms with Nonholonomic Constraints 362\u003c\/p\u003e \u003cp\u003e9.5.1 Equivariant Learning Algorithm for Causal FIR Filters in the Lie Group Sense 363\u003c\/p\u003e \u003cp\u003e9.5.2 Natural Gradient Algorithm for a Fully Recurrent Network 367\u003c\/p\u003e \u003cp\u003e9.6 MBD of Non-minimum Phase System Using Filter Decomposition Approach 368\u003c\/p\u003e \u003cp\u003e9.6.1 Information Back-propagation 370\u003c\/p\u003e \u003cp\u003e9.6.2 Batch Natural Gradient Learning Algorithm 371\u003c\/p\u003e \u003cp\u003e9.7 Computer Simulation Experiments 373\u003c\/p\u003e \u003cp\u003e9.7.1 The Natural Gradient Algorithm vs. the Ordinary Gradient Algorithm 373\u003c\/p\u003e \u003cp\u003e9.7.2 Information Back-propagation Example 375\u003c\/p\u003e \u003cp\u003e\u003cb\u003e10 Estimating Functions and Superefficiency for ICA and Deconvolution 383\u003c\/b\u003e\u003c\/p\u003e \u003cp\u003e10.1 Estimating Functions for Standard ICA 384\u003c\/p\u003e \u003cp\u003e10.1.1 What is an Estimating Function? 384\u003c\/p\u003e \u003cp\u003e10.1.2 Semiparametric Statistical Model 385\u003c\/p\u003e \u003cp\u003e10.1.3 Admissible Class of Estimating Functions 386\u003c\/p\u003e \u003cp\u003e10.1.4 Stability of Estimating Functions 389\u003c\/p\u003e \u003cp\u003e10.1.5 Standardized Estimating Function and Adaptive Newton Method 392\u003c\/p\u003e \u003cp\u003e10.1.6 Analysis of Estimation Error and Superefficiency 393\u003c\/p\u003e \u003cp\u003e10.1.7 Adaptive Choice of ' Function 395\u003c\/p\u003e \u003cp\u003e10.2 Estimating Functions in Noisy Cases 396\u003c\/p\u003e \u003cp\u003e10.3 Estimating Functions for Temporally Correlated Source Signals 397\u003c\/p\u003e \u003cp\u003e10.3.1 Source Model 397\u003c\/p\u003e \u003cp\u003e10.3.2 Likelihood and Score Functions 399\u003c\/p\u003e \u003cp\u003e10.3.3 Estimating Functions 400\u003c\/p\u003e \u003cp\u003e10.3.4 Simultaneous and Joint Diagonalization of Covariance Matrices and Estimating Functions 401\u003c\/p\u003e \u003cp\u003e10.3.5 Standardized Estimating Function and Newton Method 404\u003c\/p\u003e \u003cp\u003e10.3.6 Asymptotic Errors 407\u003c\/p\u003e \u003cp\u003e10.4 Semiparametric Models for Multichannel Blind Deconvolution 407\u003c\/p\u003e \u003cp\u003e10.4.1 Notation and Problem Statement 408\u003c\/p\u003e \u003cp\u003e10.4.2 Geometrical Structures on FIR Manifold 409\u003c\/p\u003e \u003cp\u003e10.4.3 Lie Group 410\u003c\/p\u003e \u003cp\u003e10.4.4 Natural Gradient Approach for Multichannel Blind Deconvolution 410\u003c\/p\u003e \u003cp\u003e10.4.5 Efficient Score Matrix Function and its Representation 413\u003c\/p\u003e \u003cp\u003e10.5 Estimating Functions for MBD 415\u003c\/p\u003e \u003cp\u003e10.5.1 Superefficiency of Batch Estimator 418\u003c\/p\u003e \u003cp\u003e\u003cb\u003e11 Blind Filtering and Separation Using a State-Space Approach 423\u003c\/b\u003e\u003c\/p\u003e \u003cp\u003e11.1 Problem Formulation and Basic Models 424\u003c\/p\u003e \u003cp\u003e11.1.1 Invertibility by State Space Model 426\u003c\/p\u003e \u003cp\u003e11.1.2 Controller Canonical Form 428\u003c\/p\u003e \u003cp\u003e11.2 Derivation of Basic Learning Algorithms 428\u003c\/p\u003e \u003cp\u003e11.2.1 Gradient Descent Algorithms for Estimation of Output Matrices W= [C;D] 429\u003c\/p\u003e \u003cp\u003e11.2.2 Special Case - Multichannel Blind Deconvolution with Causal FIR Filters 432\u003c\/p\u003e \u003cp\u003e11.2.3 Derivation of the Natural Gradient Algorithm for the State Space Model 432\u003c\/p\u003e \u003cp\u003e11.3 Estimation of Matrices [A;B] by Information Back–propagation 434\u003c\/p\u003e \u003cp\u003e11.4 State Estimator – The Kalman Filter 437\u003c\/p\u003e \u003cp\u003e11.4.1 Kalman Filter 437\u003c\/p\u003e \u003cp\u003e11.5 Two–stage Separation Algorithm 439\u003c\/p\u003e \u003cp\u003e\u003cb\u003e12 Nonlinear State Space Models – Semi-Blind Signal Processing 443\u003c\/b\u003e\u003c\/p\u003e \u003cp\u003e12.1 General Formulation of The Problem 443\u003c\/p\u003e \u003cp\u003e12.1.1 Invertibility by State Space Model 447\u003c\/p\u003e \u003cp\u003e12.1.2 Internal Representation 447\u003c\/p\u003e \u003cp\u003e12.2 Supervised-Unsupervised Learning Approach 448\u003c\/p\u003e \u003cp\u003e12.2.1 Nonlinear Autoregressive Moving Average Model 448\u003c\/p\u003e \u003cp\u003e12.2.2 Hyper Radial Basis Function Neural Network Model (HRBFN) 449\u003c\/p\u003e \u003cp\u003e12.2.3 Estimation of Parameters of HRBF Networks Using Gradient Approach 451\u003c\/p\u003e \u003cp\u003eReferences 453\u003c\/p\u003e \u003cp\u003e\u003cb\u003e13 Appendix – Mathematical Preliminaries 535\u003c\/b\u003e\u003c\/p\u003e \u003cp\u003e13.1 Matrix Analysis 535\u003c\/p\u003e \u003cp\u003e13.1.1 Matrix inverse update rules 535\u003c\/p\u003e \u003cp\u003e13.1.2 Some properties of determinant 536\u003c\/p\u003e \u003cp\u003e13.1.3 Some properties of the Moore-Penrose pseudo-inverse 536\u003c\/p\u003e \u003cp\u003e13.1.4 Matrix Expectations 537\u003c\/p\u003e \u003cp\u003e13.1.5 Differentiation of a scalar function with respect to a vector 538\u003c\/p\u003e \u003cp\u003e13.1.6 Matrix differentiation 539\u003c\/p\u003e \u003cp\u003e13.1.7 Trace 540\u003c\/p\u003e \u003cp\u003e13.1.8 Matrix differentiation of trace of matrices 541\u003c\/p\u003e \u003cp\u003e13.1.9 Important Inequalities 542\u003c\/p\u003e \u003cp\u003e13.1.10Inequalities in Information Theory 543\u003c\/p\u003e \u003cp\u003e13.2 Distance measures 544\u003c\/p\u003e \u003cp\u003e13.2.1 Geometric distance measures 544\u003c\/p\u003e \u003cp\u003e13.2.2 Distances between sets 544\u003c\/p\u003e \u003cp\u003e13.2.3 Discrimination measures 545\u003c\/p\u003e \u003cp\u003e\u003cb\u003e14 Glossary of Symbols and Abbreviations 547\u003c\/b\u003e\u003c\/p\u003e \u003cp\u003eIndex 552\u003c\/p\u003e\u003c\/font\u003e\u003c\/p\u003e\r\n\r\n\u003cp\u003e\u003cfont size=\"3\"\u003eSubject Areas: Electronics \u0026amp; communications engineering [\u003ca title=\"See our other books on Electronics \u0026amp; communications engineering\" href=\"https:\/\/freshlyprintedbooks.co.uk\/search?q=%22Electronics%20\u0026amp;%20communications%20engineering%20%5BTJ%5D%22\"\u003eTJ\u003c\/a\u003e]\u003c\/font\u003e\u003c\/p\u003e\r\n\r\n\r\n\u003c\/font\u003e","brand":"Wiley","offers":[{"title":"Brand New","offer_id":52298007347480,"sku":"9780471607915","price":106.99,"currency_code":"GBP","in_stock":true}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0730\/2037\/5320\/files\/9780471607915.jpg?v=1781730674","url":"https:\/\/freshlyprintedbooks.co.uk\/products\/adaptive-blind-signal-and-image-processing-learning-algorithms-and-applications-hardback-9780471607915","provider":"Freshly Printed Books","version":"1.0","type":"link"}