• Advances in Heuristic Signal Processing and Applications
  • Preface
  • Contents
  • Contributors
• Chapter 1: Nonconvex Optimization via Joint Norm Relaxed SQP and Filled Function Method with Application to Minimax Two-Channel Linear Phase FIR QMF Bank Design
  • 1.1 Introduction
  • 1.2 Problem Formulation
  • 1.3 Joint Norm Relaxed Sequential Quadratic Programming and Filled Function Method
    • 1.3.1 Filled Function Method
    • 1.3.2 Norm Relaxed Sequential Quadratic Programming
  • 1.4 Computer Numerical Simulation Results
  • 1.5 Conclusions
  • References
• Chapter 2: Robust Reduced-Rank Adaptive LCMV Beamforming Algorithms Based on Joint Iterative Optimization of Parameters
  • 2.1 Introduction
  • 2.2 System Model
  • 2.3 Problem Statement and Design of Adaptive Beamformers
    • 2.3.1 Adaptive LCMV Beamformers
    • 2.3.2 Robust Adaptive LCMV Beamformers
  • 2.4 Robust Reduced-Rank Beamforming Based on Joint Iterative Optimization of Parameters
  • 2.5 Adaptive Algorithms
    • 2.5.1 Stochastic Gradient Algorithm
    • 2.5.2 Recursive Least-Squares Algorithms
    • 2.5.3 Complexity of RJIO Algorithms
    • 2.5.4 Rank Adaptation
  • 2.6 Simulations
  • 2.7 Conclusions
  • References
• Chapter 3: Designing OFDM Radar Waveform for Target Detection Using Multi-objective Optimization
  • 3.1 Introduction
  • 3.2 Problem Description and Modeling
    • 3.2.1 OFDM Signal Model
    • 3.2.2 Measurement Model
    • 3.2.3 Statistical Assumptions
  • 3.3 Sparse-Estimation Approach
    • 3.3.1 Sparse Model
    • 3.3.2 Sparse Recovery
  • 3.4 Adaptive Waveform Design
    • 3.4.1 Maximizing the Mahalanobis Distance
    • 3.4.2 Minimizing the Weighted Trace of CRB Matrix
    • 3.4.3 Minimizing the Upper Bound on Sparse Error
    • 3.4.4 Multi-objective Optimization
  • 3.5 Numerical Results
    • 3.5.1 Results of the MOO Problem
    • 3.5.2 Improvement in Detection and Estimation Performance
    • 3.5.3 Redistributions of Signal and Target Energies
  • 3.6 Conclusions
  • References
• Chapter 4: Multi-object Tracking Using Particle Swarm Optimization on Target Interactions
  • 4.1 Introduction
  • 4.2 Particle Swarm Optimization
  • 4.3 PSO-Based Object Tracking
    • 4.3.1 Multi-patch Based Object Tracking Using Region Covariance
    • 4.3.2 Foreground Prior
    • 4.3.3 Foreground Segmentation
    • 4.3.4 Re-diversification of the Swarm
  • 4.4 Multiple Object Tracking
    • 4.4.1 Multiple Object Tracking by Multiple Particle Swarms
    • 4.4.2 Refinement of the Tracklets by Particle Swarm Optimization
  • 4.5 Experiments
  • 4.6 Conclusions
  • References
• Chapter 5: A Comparative Study of Modified BBO Variants and Other Metaheuristics for Optimal Power Allocation in Wireless Sensor Networks
  • 5.1 Introduction
  • 5.2 Problem Statement
    • 5.2.1 Fusion Problem Formulation
    • 5.2.2 Node Decision Rules
    • 5.2.3 Transmission of Local Decisions
    • 5.2.4 The Decision Fusion Problem
  • 5.3 Optimal Power Allocation
    • 5.3.1 Independent Observations
    • 5.3.2 Correlated Observations
  • 5.4 Constrained BBO for Optimal Power Allocation
    • 5.4.1 Standard Unconstrained Biogeography-Based Optimization (BBO)
    • 5.4.2 Constrained Optimization
    • 5.4.3 Description of the Proposed Algorithms
      • 5.4.3.1 Solution Representation
      • 5.4.3.2 Objective Function
      • 5.4.3.3 Constraint Handling Approach
      • 5.4.3.4 Individual Initialization
      • 5.4.3.5 Individual Stochastic Ranking
      • 5.4.3.6 Population Update Strategy
      • 5.4.3.7 Selection
      • 5.4.3.8 Elitism
  • 5.5 Experimental Results and Analysis
    • 5.5.1 Parameter Configuration
    • 5.5.2 Numerical Results
  • 5.6 Conclusion
  • References
• Chapter 6: Joint Optimization of Detection and Tracking in Adaptive Radar Systems
  • 6.1 Introduction
    • 6.1.1 Related Work
    • 6.1.2 Chapter Outline
  • 6.2 Offline Performance Evaluation of Tracking Algorithms
    • 6.2.1 Target and Measurement Models
    • 6.2.2 NSPP Techniques for the PDAF
    • 6.2.3 The Modified Riccati Equation (MRE)
    • 6.2.4 The Hybrid Conditional Averaging (HYCA) Algorithm
  • 6.3 NSPP-Based Detector Threshold Optimization
    • 6.3.1 Static Threshold Optimization (STOP)
    • 6.3.2 Dynamic Threshold Optimization (DTOP)
      • 6.3.2.1 MRE-Based Dynamic Threshold Optimization
      • 6.3.2.2 HYCA-Based Dynamic Threshold Optimization
  • 6.4 Simulations
    • 6.4.1 Static Threshold Optimization Based on MRE and HYCA Algorithms
    • 6.4.2 Experiment 1: Comparison with Heuristic Approaches
    • 6.4.3 Experiment 2: Comparison of MRE-Based DTOP Schemes
  • 6.5 Conclusion and Future Directions
  • References
• Chapter 7: Iterative Design of FIR Filters
  • 7.1 Introduction
  • 7.2 Particle Swarm Optimization with Quantum Infusion
  • 7.3 Digital FIR Filter
  • 7.4 FIR Filter Design Using PSO-QI
  • 7.5 Studies and Results
  • 7.6 Conclusion
  • References
• Chapter 8: A Metaheuristic Approach to Two Dimensional Recursive Digital Filter Design
  • 8.1 Introduction
  • 8.2 The Design Problem Formulation
  • 8.3 An Outline of IWO Algorithm
    • 8.3.1 Generation of an Initial Population
    • 8.3.2 Reproduction
    • 8.3.3 Dispersal of Seeds Through the Search Space
    • 8.3.4 Competitive Exclusion
  • 8.4 Differential Q-Learning
  • 8.5 IIWO: The Proposed Approach
    • 8.5.1 Initialization
    • 8.5.2 Adaptive Selection of Memes
    • 8.5.3 Invasive Weed Optimization
    • 8.5.4 State Assignment
    • 8.5.5 Updating the Q-table
  • 8.6 The Filter Design Experiments and Results
  • 8.7 Conclusions
  • References
• Chapter 9: A Survey of Kurtosis Optimization Schemes for MISO Source Separation and Equalization
  • 9.1 Introduction
    • 9.1.1 Channel Equalization and Source Separation
    • 9.1.2 Why Kurtosis?
    • 9.1.3 Historical Overview
    • 9.1.4 Chapter Outline
    • 9.1.5 Mathematical Notations
  • 9.2 Blind Source Separation: Model and Assumptions
    • 9.2.1 Convolutive Mixtures
    • 9.2.2 Instantaneous Mixtures
  • 9.3 Deflationary Source Separation
    • 9.3.1 Source Extraction with MISO Contrast Functions
    • 9.3.2 Deflation Procedure
  • 9.4 Optimization Methods
    • 9.4.1 Gradient Search Algorithms
    • 9.4.2 Projected Gradient Search
    • 9.4.3 Gradient Algorithm with Filter Parametrization
    • 9.4.4 Approximate Newton Search: The FastICA Algorithm
      • 9.4.4.1 Derivation of the Algorithm
      • 9.4.4.2 FastICA as a Constant Step-Size Gradient Algorithm
    • 9.4.5 Algorithms with Optimal Step-Size Selection
      • 9.4.5.1 Step-Size Optimization
      • 9.4.5.2 Algebraic Exact Line Search: The RobustICA Algorithm
    • 9.4.6 Algorithms Based on Reference Signals
      • 9.4.6.1 Quadratic Contrast Functions
      • 9.4.6.2 Monotonically Convergent Algorithms Based on Quadratic Contrasts
  • 9.5 Illustrative Results
    • 9.5.1 Source Separation and Equalization in Digital Communications
    • 9.5.2 Artifact Rejection in Biomedical Recordings
  • 9.6 Conclusions
  • References
• Chapter 10: Swarm Intelligence Techniques Applied to Nonlinear Systems State Estimation
  • 10.1 Introduction
  • 10.2 Estimation Problem Formulation
  • 10.3 Generic Particle Filter and Limitations
    • 10.3.1 Initialization
    • 10.3.2 Propagation of Particles' Location
    • 10.3.3 Weight Computation and Normalization
    • 10.3.4 Resampling
    • 10.3.5 State Estimation
    • 10.3.6 Limitations
      • 10.3.6.1 Particle Impoverishment
      • 10.3.6.2 Sample Size Dependency
  • 10.4 Swarm Filters
    • 10.4.1 Particle Swarm Optimized Particle Filter (PSOPF)
      • 10.4.1.1 Computation of Cost Function
      • 10.4.1.2 Local Best and Global Best Update
      • 10.4.1.3 Velocity and Position Update Rules
      • 10.4.1.4 Weight Computation and Normalization
      • 10.4.1.5 Stopping Condition
    • 10.4.2 Ant Colony Optimization Assisted Particle Filter (PFACO)
      • 10.4.2.1 Computation of Probability Function
      • 10.4.2.2 Movement of the Ants
      • 10.4.2.3 Update Pheromone Distribution
      • 10.4.2.4 Stopping Condition
    • 10.4.3 Particle Filter with Ant Colony for Continuous Domains
      • 10.4.3.1 Initialization
      • 10.4.3.2 Computation of Cost Function
      • 10.4.3.3 External Archive Update
      • 10.4.3.4 Pheromone Update
      • 10.4.3.5 Movement of the Ants
      • 10.4.3.6 Computation of Weights
      • 10.4.3.7 State Estimation
    • 10.4.4 Continuous Ant Colony Filter (CACF)
      • 10.4.4.1 Initialization
      • 10.4.4.2 Computation of Cost Function
      • 10.4.4.3 Updating Pheromone Distribution
      • 10.4.4.4 Movement of Ants
      • 10.4.4.5 State Estimation
  • 10.5 Conclusion
  • References
• Chapter 11: Heuristic Optimal Design of Multiplier-less Digital Filter
  • 11.1 Introduction
  • 11.2 The Accumulated Hybrid Code (AHC)
    • 11.2.1 The Traditional Hybrid Code Method
    • 11.2.2 New Hybrid Code Method
  • 11.3 CSD-Based Genetic Algorithm
    • 11.3.1 Definition of the Fitness Function
    • 11.3.2 CSD-Based Crossover
    • 11.3.3 CSD-Based Mutation
  • 11.4 CSD-Based Design of FIR Filter
    • 11.4.1 Overview of Finite-Impulse Response (FIR) Filter
    • 11.4.2 Estimation of the Order
    • 11.4.3 Design Example of an FIR Filter [4]
  • 11.5 CSD-Based Design of IIR Filter
    • 11.5.1 Overview of Infinite Impulse Response (IIR) Filters
    • 11.5.2 Stability Criterion for an IIR Filter
    • 11.5.3 Design of an IIR Filter
    • 11.5.4 Design Example of an IIR Filter [5]
  • References
• Chapter 12: Hybrid Correlation-Neural Network Synergy for Gait Signal Classification
  • 12.1 Introduction
  • 12.2 The Acquisition of Gait Signals
  • 12.3 Cross-Correlation Based Feature Extraction Methodology
    • 12.3.1 Time Domain Features
    • 12.3.2 Frequency Domain Features
  • 12.4 Elman's Recurrent Neural Network Based Classification
  • 12.5 Time Domain Cross-Correlation Based Scheme for Gait Signal Classification
    • 12.5.1 Performance Evaluation
  • 12.6 Frequency Domain Cross-Correlation Based Scheme for Gait Signal Classification
    • 12.6.1 Performance Evaluation
  • 12.7 Conclusions
  • References
• Chapter 13: Image Denoising Using Wavelets: Application in Medical Imaging
  • 13.1 Introduction to Multiresolution Analysis
    • 13.1.1 Discovery and Contributions of Wavelets
    • 13.1.2 Continuous Wavelet Transforms
    • 13.1.3 Discrete Wavelet Transforms
    • 13.1.4 The Concept of MRA
    • 13.1.5 Implementation of MRA: Mallat Algorithm
  • 13.2 Redundant Multiresolution Analysis
    • 13.2.1 Undecimated Discrete Wavelet Transform
    • 13.2.2 Wavelet Packets
    • 13.2.3 Contourlet Transform
  • 13.3 Denoising or Noise Reduction
    • 13.3.1 Additive Gaussian White Noise Model
    • 13.3.2 Sigmoidal Wavelet Shrinkage
    • 13.3.3 Parametric Denoising: Wiener Filtering
    • 13.3.4 Suppression of Correlated Noise
  • 13.4 Applications in Medical Imaging
    • 13.4.1 Medical Imaging Methods and Techniques
    • 13.4.2 Wavelet-Based Denoising in fMRI, MRI, and Echography
      • 13.4.2.1 MRI Illustration
      • 13.4.2.2 Echography Illustration
  • 13.5 Conclusion
  • References
• Chapter 14: Signal Separation with A Priori Knowledge Using Sparse Representation
  • 14.1 Introduction
  • 14.2 Signal Separation Using Sparse Representation
    • 14.2.1 Dictionary Construction
    • 14.2.2 Pursuit Algorithms
      • 14.2.2.1 Greedy Algorithms
      • 14.2.2.2 Algorithms Based on Norm Minimization
  • 14.3 MRS Spectra Separation with A Priori Knowledge Using Sparse Representation
    • 14.3.1 Signal Models
    • 14.3.2 Dictionary Construction Based on the A Priori Knowledge
    • 14.3.3 Resonance Estimation with FOCUSS Algorithm
    • 14.3.4 Experiments and Results
      • 14.3.4.1 Simulated Experiments
      • 14.3.4.2 Quantitation of Human Brain 1H MRS Data
      • 14.3.4.3 Quantitation of Prostate 1H MRS Data
  • 14.4 Summary
  • References
• Chapter 15: Definition of a Discrete Color Monogenic Wavelet Transform
  • 15.1 Introduction
    • Notations
  • 15.2 Analytical Signal and 2D Generalization
    • 15.2.1 Analytic Signal (1D)
    • 15.2.2 Monogenic Signal (2D)
    • 15.2.3 Monogenic Multiresolution
      • 15.2.3.1 Primary Transform
      • 15.2.3.2 The Monogenic Transform
  • 15.3 First Extension of Color Monogenic Wavelet
    • 15.3.1 The Color Monogenic Signal
    • 15.3.2 The Color Monogenic Wavelet Transform
  • 15.4 Second Approach for Color Monogenic Signal: A Tensor Approach
    • 15.4.1 Link Between Riesz and Gradient
    • 15.4.2 Color Riesz Analysis
    • 15.4.3 Tensor Based Color Monogenic Wavelet Transform
    • 15.4.4 Algorithm Discussion
  • 15.5 The Radon Domain for Numerical Monogenic Transform
    • 15.5.1 The Radon Transform
    • 15.5.2 Discrete Radon
    • 15.5.3 Discrete Radon Based Riesz Transform
    • 15.5.4 Discrete Radon Based Monogenic Wavelet Transform
  • 15.6 Conclusion
  • References
• Chapter 16: On Image Matching and Feature Tracking for Embedded Systems: A State-of-the-Art
  • 16.1 Introduction
  • 16.2 Matching: Concepts, Algorithms, and Architectures
    • 16.2.1 Matching: Basic Concepts
    • 16.2.2 Characteristics for Matching
    • 16.2.3 Popular Matching Methods
      • 16.2.3.1 Raw Data/Area Based Methods
      • 16.2.3.2 Feature Based Approaches
    • 16.2.4 Hardware Systems for Image Matching
  • 16.3 Feature Tracking
    • 16.3.1 Correlation-Based Feature Tracking
    • 16.3.2 Bayesian Approaches
  • 16.4 Final Comments and Potential Future Developments
  • References
• Index

...see more