Partial-Update Adaptive Signal Processing
Design Analysis and ImplementationBy
- Kutluyil Dogancay, University of South Australia
Partial-update adaptive signal processing algorithms not only permit significant complexity reduction in adaptive filter implementations, but can also improve adaptive filter performance in telecommunications applications. This book gives state-of-the-art methods for the design and development of partial-update adaptive signal processing algorithms for use in systems development.Partial-Update Adaptive Signal Processing provides a comprehensive coverage of key partial updating schemes, giving detailed information on the theory and applications of acoustic and network echo cancellation, channel equalization and multiuser detection. It also examines convergence and stability issues for partial update algorithms, providing detailed complexity analysis and a unifying treatment of partial-update techniques.Features: Advanced analysis and design tools Application examples illustrating the use of partial-update adaptive signal processing MATLAB codes for developed algorithms This unique reference will be of interest to signal processing and communications engineers, researchers, R&D engineers and graduate students."This is a very systematic and methodical treatment of an adaptive signal processing topic, of particular significance in power limited applications such as in wireless communication systems and smart ad hoc sensor networks. I am very happy to have this book on my shelf, not to gather dust, but to be consulted and used in my own research and teaching activities" Professor A. G. Constantinides, Imperial College, LondonAbout the author:Kutluyil Dogançay is an associate professor of Electrical Engineering at the University of South Australia. His research interests span statistical and adaptive signal processing and he serves as a consultant to defence and private industry. He was the Signal Processing and Communications Program Chair of IDC Conference 2007, and is currently chair of the IEEE South Australia Communications and Signal Processing Chapter.
Signal processing systems engineers,R&D engineers,university applied researchers and postgraduate students.
Hardbound, 296 Pages
Imprint: Academic Press
"This is a very systematic and methodical treatment of an adaptive signal processing topic, of particular significance in power limited applications such as in wireless communication systems and smart ad hoc sensor networks. I am very happy to have this book on my shelf, not to gather dust, but to be consulted and used in my own research and teaching activities" Professor A. G. Constantinides, Imperial College, London
- TABLE OF CONTENTS 1. Introduction 1.1 Adaptive Signal Processing 1.2 Examples of Adaptive Filtering 1.2.1 Adaptive System Identification 1.2.2 Adaptive Inverse System Identification 1.3 Raison D'etre for Partial Coefficient Updates 1.3.1 Resource Constraints 1.3.2 Convergence Performance 1.3.3 System Identification with White Input Signal 1.3.4 System Identification with Correlated Input Signal2. Approaches to Partial Coefficient Updates 2.1 Introduction 2.2 Periodic Partial Updates 2.2.1 Example 1: Convergence Performance 2.2.2 Example 2: Convergence Difficulties 2.3 Sequential Partial Updates 2.3.1 Example 1: Convergence Performance 2.3.2 Example 2: Cyclostationary Inputs 2.3.3 Example 3: Instability 2.4 Stochastic Partial Updates 2.4.1 System Identification Example 2.5 M-Max Updates 2.5.1 Example 1: Eigenvalue Spread of R_M 2.5.2 Example 2: Convergence Performance 2.5.3 Example 3: Convergence Rate and Eigenvalues of R_M 2.5.4 Example 4: Convergence Difficulties 2.5.5 Example 5: Instability 2.6 Selective Partial Updates 2.6.1 Constrained Optimization 2.6.2 Instantaneous Approximation of Newton's Method 2.6.3 q-Norm Constrained Optimization 2.6.4 Averaged System 2.6.5 Example 1: Eigenanalysis 2.6.6 Example 2: Convergence Performance 2.6.7 Example 3: Instability 2.7 Set Membership Partial Updates 2.7.1 Example 1: Convergence Performance 2.7.2 Example 2: Instability 2.8 Block Partial Updates 2.9 Complexity Considerations3. Convergence and Stability Analysis 3.1 Introduction 3.2 Convergence Performance 3.3 Steady-State Analysis 3.3.1 Partial-Update LMS Algorithms Periodic partial updates Sequential partial updates Stochastic partial updates M-max partial updates 3.3.2 Partial-Update NLMS Algorithms 3.3.3 Simulation Examples for Steady-State Analysis 3.4 Convergence Analysis 3.4.1 Partial-Update LMS Algorithms Vectorization Steady-State Analysis Mean-Square Stability Mean Stability 3.4.2 Partial-Update NLMS Algorithms 3.4.3 Simulation Examples for Convergence Analysis4. Partial-Update Adaptive Filters 4.1 Introduction 4.2 Least-Mean-Square Algorithm 4.3 Partial-Update LMS Algorithms 4.3.1 Periodic-Partial-Update LMS Algorithm 4.3.2 Sequential-Partial-Update LMS Algorithm 4.3.3 Stochastic-Partial-Update LMS Algorithm 4.3.4 M-max LMS Algorithm 4.3.5 Computational Complexity 4.4 Normalized Least-Mean-Square Algorithm 4.5 Partial-Update NLMS Algorithms 4.5.1 Periodic-Partial-Update NLMS Algorithm 4.5.2 Sequential-Partial-Update NLMS Algorithm 4.5.3 Stochastic-Partial-Update NLMS Algorithm 4.5.4 M-max NLMS Algorithm 4.5.5 Selective-Partial-Update NLMS Algorithm 4.5.6 Set-Membership Partial-Update NLMS Algorithm 4.5.7 Computational Complexity 4.6 Affine Projection Algorithm 4.7 Partial-Update Affine Projection Algorithms 4.7.1 Periodic-Partial-Update APA 4.7.2 Sequential-Partial-Update APA 4.7.3 Stochastic-Partial-Update APA 4.7.4 M-max APA 4.7.5 Selective-Partial-Update APA 4.7.6 Set-Membership Partial-Update APA 4.7.7 Selective-Regressor APA 4.7.8 Computational Complexity 4.8 Recursive Least Squares Algorithm 4.9 Partial-Update RLS Algorithms 4.9.1 Periodic-Partial-Update RLS Algorithm 4.9.2 Sequential-Partial-Update RLS Algorithm 4.9.3 Stochastic-Partial-Update RLS Algorithm 4.9.4 Selective-Partial-Update RLS Algorithm 4.9.5 Set-Membership Partial-Update RLS Algorithm 4.9.6 Partial-Update RLS Simulations 4.9.7 Computational Complexity 4.10 Transform-Domain Least-Mean-Square Algorithm 4.10.1 Power Normalization 4.10.2 Comparison of Power Normalization Algorithms 4.11 Partial-Update Transform-Domain LMS Algorithms 4.11.1 Periodic-Partial-Update Transform-Domain LMS Algorithm 4.11.2 Sequential-Partial-Update Transform-Domain LMS Algorithm 4.11.3 Stochastic-Partial-Update Transform-Domain LMS Algorithm 4.11.4 M-max Transform-Domain LMS Algorithm 4.11.5 Computational Complexity 4.12 Generalized-Subband-Decomposition Least-Mean-Square Algorithm 4.12.1 Relationship Between GSD-LMS Coefficients and Equivalent Time Domain Response 4.12.2 Eigenvalue Spread of GSD Input Correlation Matrix 4.13 Partial-Update GSD-LMS Algorithms 4.13.1 Periodic-Partial-Update GSD-LMS Algorithm 4.13.2 Sequential-Partial-Update GSD-LMS Algorithm 4.13.3 Stochastic-Partial-Update GSD-LMS Algorithm 4.13.4 M-max GSD-LMS Algorithm 4.13.5 Computational Complexity 4.14 Simulation Examples: Channel Equalization5. Selected Applications 5.1 Introduction 5.2 Acoustic Echo Cancellation 5.3 Network Echo Cancellation 5.3.1 PNLMS and mu-Law PNLMS with Selective Partial Updates PNLMS mu-Law PNLMS Selective Partial Updates Simulation Examples 5.4 Blind Channel Equalization 5.4.1 Normalized CMA 5.4.2 Selective-Partial-Update NCMA 5.4.3 Simulation Examples 5.5 Blind Adaptive Linear Multiuser Detection 5.5.1 MUD in Synchronous DS-CDMA 5.5.2 Blind Multiuser NLMS Algorithm 5.5.3 Selective-Partial-Update NLMS for Blind Multiuser Detection 5.5.4 Simulation ExamplesA. Overview of Fast Sorting Algorithms A.1 Introduction A.2 Running Min/Max and Sorting Algorithms A.2.1 Divide-and-Conquer Approaches A.2.2 Maxline Algorithm A.2.3 The Gil-Werman Algorithm A.2.4 Sortline Algorithm A.3 Heapsort Algorithm