Search

Proportionate-type normalized subband adaptive filter (PNSAF-type) algorithms are very attractive choices for echo cancellation. To further obtain both fast convergence rate and low steady-state error, in this paper, a variable step size (VSS) version of the presented improved PNSAF (IPNSAF) algorithm is proposed by minimizing the square of the noise-free a posterior subband error signals. A noniterative shrinkage method is used to recover the noise-free a priori subband error signals from the noisy subband error signals. Significantly, the proposed VSS strategy can be applied to any other PNSAF-type algorithm, since it is independent of the proportionate principles. Simulation results in the context of acoustic echo cancellation have demonstrated the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

Abstract: The bin-normalized frequency-domain block LMS (FBLMS) algorithm has low computational burden and potential fast convergence; however, it suffers from a biased steady-state solution when the reference signal lags behind the desired signal or the adaptive filter is of insufficient length. This paper proposes a unified framework for the FBLMS algorithm, which can be used to comprehensively analyze its steady-state behavior. Furthermore, a modified FBLMS algorithm with guaranteed optimal steady-state performance is proposed based on the framework. Simulations are carried out to demonstrate the benefit of the proposed algorithm. [Copyright &y& Elsevier]

• A novel robust mixed-norm algorithm with affine projection (APRMNA) is proposed. • The APRMNA based on the generalized maximum correntropy criterion (APRMNA-GMC) is proposed. • The simplified APRMNA-GMC (S-APRMNA-GMC) is proposed. In this paper, a novel adaptive filtering algorithm combining both affine projection (AP) method and robust mixed-norm algorithm (RMNA) is proposed, which is called APRMNA. The AP method has the feature of fast convergence speed under colored inputs and the RMNA exhibits stable performance against noise interference. The proposed APRMNA algorithm not only combines the advantages of both AP and RMNA but also utilizes the ℓ 2 -norm constraint on the weight vector to avoid matrix inversion. Then, applying the generalized maximum correntropy (GMC) criterion to APRMNA, we also develop the APRMNA-GMC. Finally, a simplified version of the proposed APRMNA-GMC (S-APRMNA-GMC) is derived to reduce the computation complexity. Numerical simulations for system identification show that the proposed algorithms outperform other AP-type algorithms. [ABSTRACT FROM AUTHOR]