Abstract:Aiming to accelerate convergence speed and reduce steady-state misalignment of adaptive filter, a new data-reusing algorithm is proposed. Different from conventional affine projection algorithm (APA) based on minimum disturbance principle (MDP), the proposed algorithm obtains its cost function based on a convex combination of the minimum norm principle (MNP) and the MDP. Weight factor (named as mixing parameter) used in the combination determines the approach for updating. The proposed algorithm has a performance similar to the APA when the mixing parameter closes to 0 and yields results to the data-reusing algorithm based on the MNP if the mixing parameter closes to 1. By minimizing mean-square deviation, the variable mixing parameter has a large value at initial stage and gradually decreases as iteration number increases. At the initial stage, the proposed algorithm achieves fast convergence since the MNP plays a leading role. At the steady-state, the proposed algorithm obtains a low misalignment since the MDP gives a major impact and the updated results have minimum disturbance from the previous ones. At the transition stage, the proposed algorithm is a combination of the APA and the MNP-based data-reusing algorithm. Simulation results show that the proposed algorithm exhibits faster convergence rate and lower steady-state misalignment than some other derivatives of the APA without significantly increasing computational complexity.

Affine projection mixed-norm algorithms for robust filtering

Combined Step-size Affine Projection Mix-norm Algorithm with Generalized Maximum Correntropy for Robust Filtering

Combined Step Sizes Affine Projection Mixed-Norm Algorithm for Robust Filtering

Generalized Maximum Correntropy Algorithm with Affine Projection for Robust Filtering under Impulsive-Noise Environments.

Robust hybrid affine projection filtering algorithm under α -stable environment

Affine Projection Algorithm with Outlier Detection for Robust Filtering

Combined-Statistical-Distribution Generalized Maximum Correntropy Algorithm with Affine Projection for Robust Adaptive Filtering

Generalized mixed-norm maximum correntropy for robust adaptive filtering

An Improved Affine Projection Sign Algorithm Based on Individual-Weighting Factors

Affine Projection Algorithm by Employing Maximum Correntropy Criterion for System Identification of Mixed Noise

A New Data-Reusing Algorithm Based on Minimum Norm and Minimum Disturbance Principles

Affine Projection Algorithm Based on Least Mean Fourth Algorithm for System Identification

A Data Selective Segment Proportionate Affine Projection Algorithm for Echo Cancellation

Variable step-size affine projection normalized subband adaptive filter design

Robust Normalized Least Mean Absolute Third Algorithms.

An Improved Affine Projection Algorithm for Active Noise Cancellation

Variable-Step-Size Efficient Proportionate Affine Projection Sign Algorithms

Kernel robust mixed-norm adaptive filtering

P-Norm Based Subband Adaptive Filtering Algorithm

Evolving Order Based Affine Projection Sign Algorithm For Enhanced Adaptive Filtering

P-Norm Based Subband Adaptive Filtering Algorithm: Performance Analysis and Improvements