1 Constant Directivity Beamforming
1.1 Introduction
1.2 Problem Formulation
1.3 Theoretical Solution
1.3.1 Continuous sensor
1.3.2 Beam-shaping function
1.4 Practical Implementation
1.4.1 Dimension-reducing parameterization
1.4.2 Reference beam-shaping filter
1.4.3 Sensor placelllent
1.4.4 SUllllllary of illlplelllentation
1.5 Examples
1.6 Conclusions
2 Superdirective Microphone Arrays
2.1 Introduction
2.2 Evaluation of Beamformers
2.2.1 Array-Gain
2.2.2 Beampattern
2.2.3 Directivity
2.2.4 Front-to-Back Ratio
2.2.5 White Noise Gain
2.3 Design of Superdirective Beamformers
2.3.1 Delay-and-Sum Beamformer
2.3.2 Design for spherical isotropic noise
2.3.3 Design for Cylindrical Isotropic Noise
2.3.4 Design for an Optimal Front-to-Back Ratio
2.3.5 Design for Measured Noise Fields
2.4 Extensions and Details
2.4.1 Alternative Form
2.4.2 Comparison with Gradient Microphones
2.5 Conclusion
3 Post-Filtering Techniques
3.1 Introduction
3.2 Multi-channel Wiener Filtering in Subbands
3.2.1 Derivation of the Optimum Solution
3.2.2 Factorization of the Wiener Solution
3.2.3 Interpretation
3.3 Algorithms for Post-Filter Estimation
3.3.1 Analysis of Post-Filter Algorithms
3.3.2 Properties of Post-Filter Algorithms
3.3.3 A New Post-Filter Algorithm
3.4 Performance Evaluation
3.4.1 Simulation System
3.4.2 Objective Measures
3.4.3 Simulation Results
3.5 Conclusion
4 Spatial Coherence Functions for Differential Microphones in Isotropic NoiseFields
4.1 Introduction
4.2 Adaptive Noise Cancellation
4.3 Spherically Isotropic Coherence
4.4 Cylindrically Isotropic Fields
4.5 Conclusions
5 Robust Adaptive Beamforming
5.1 Introduction
5.2 Adaptive Beamformers
5.3 Robustness Problem in the GJBF
5.4 Robust Adaptive Microphone Arrays - Solutionsto Steering-Vector Errors
5.4.1 LAF-LAF Structure
5.4.2 CCAF-LAF Structure
5.4.3 CCAF-NCAF Structure
5.4.4 CCAF-NCAF Structure with an AMC
5.5 Software Evaluation of a Robust AdaptiveMicrophone Array
5.5.1 Simulated Anechoic Environment
5.5.2 Reverberant Environment
5.6 Hardware Evaluation of a Robust AdaptiveMicrophone Array
5.6.1 Implementation
5.6.2 Evaluation in a Real Environment
5.7 Conclusion
6 GSVD-Based Optimal Filtering forMulti-Microphone Speech Enhancement
6.1 Introduction
6.2 GSVD-Based Optimal Filtering Technique
6.2.1 Optimal Filter Theory
6.2.2 General Class of Estimators
6.2.3 Symmetry Properties for Time-Series Filtering
6.3 Performance of GSVD-Based Optimal Filtering
6.3.1 Simulation Environment
6.3.2 Spatial Directivity Pattern
6.3.3 Noise Reduction Performance
6.3.4 Robustness Issues
6.4 Complexity Reduction
6.4.1 Linear Algebra Techniques for Computing GSVD
6.4.2 Recursive and Approximate GSVD-Updating Algorithms
6.4.3 Downsampling Techniques
6.4.4 Simulations
6.4.5 Computational Complexity
6.5 Combination with ANC Postprocessing Stage
6.5.1 Creation of Speech and Noise References
6.5.2 Noise Reduction Performance of ANC PostprocessingStage
6.5.3 Comparison with Standard Beamforming Techniques
6.6 Conclusion
7 Explicit Speech Modeling for MicrophoneArray Applications
7.1 Introduction
7.2 Model-Based Strategies
7.2.1 Example 1: A Frequency-Domain Model-Based Algorithm
7.2.2 Example 2: A Time-Domain Model-Based Algorithm