A new sequential block partial update normalized least mean M-estimate algorithm and its convergence performance analysis

Abstract

This paper proposes a new sequential block partial update normalized least mean M-estimate (SB-NLMM) algorithm for adaptive filtering in impulsive noise environment. It utilizes the sequential partial update concept as in the sequential block partial update normalized least mean square (SB-NLMS) algorithm to reduce the computational complexity, while minimizing the M-estimate function for improved robustness to impulsive outliers. The mean and mean square convergence behavior of the SB-NLMM algorithm under Contaminated Gaussian (CG) noise is also analyzed by extending the approach of Bershad [8] and using an extension of Price's theorem to evaluate the expectation of the various quantities involved. New analytical expressions describing the convergence behavior are derived. The robustness of the proposed algorithm and accuracy of the performance analysis are verified by computer simulations. ©2007 IEEE.