Search

This paper analyzes the statistical behavior of the bias-compensated sign subband adaptive filtering algorithm with individual-weighting-factors (BC-IWF-SSAF) under errors-in-variables (EIV) model. The theoretical expressions for transient and steady-state behaviors of BC-IWF-SSAF in the mean and mean-square senses are derived through utilizing some reasonable assumptions and Price's theorem. Furthermore, the stability ranges of algorithm are provided. To accomplish performance enhancement, the variable step-size version (VSS) of BC-IWF-SSAF, called VSS-BC-IWF-SSAF algorithm, is developed based on the transient model via minimizing the mean-square deviation (MSD) at every iteration to tackle the contradictory requirements of quick convergence rate and high filtering accuracy caused by fixed step-size (FSS). Finally, Monte-Carlo simulations under system identification and acoustic echo cancellation applications certificate the effectiveness of statistical analysis and variable step-size strategy. [ABSTRACT FROM AUTHOR]

Although the least mean p th power (LMP) and normalized LMP (NLMP) algorithms of adaptive Volterra filters outperform the conventional least mean square (LMS) algorithm in the presence of α -stable noise, they still exhibit slow convergence and high steady-state kernel error in nonlinear system identification. To overcome these limitations, an enhanced recursive least mean p th power algorithm with logarithmic transformation (RLogLMP) is proposed in this paper. The proposed algorithm is adjusted to minimize the new cost function with the p -norm logarithmic transformation of the error signal. The logarithmic transformation, which can diminish the significance of outliers under α -stable noise environment, increases the robustness of the proposed algorithm and reduces the steady-state kernel error. Moreover, the proposed method improves the convergence rate by the enhanced recursive scheme. Finally, simulation results demonstrate that the proposed algorithm is superior to the LMP, NLMP, normalized least mean absolute deviation (NLMAD), recursive least squares (RLS) and nonlinear iteratively reweighted least squares (NIRLS) algorithms in terms of convergence rate and steady-state kernel error. [ABSTRACT FROM AUTHOR]