Proposing SVS-PNLMS algorithm for sparse echo cancellation

Abstract

In this paper segment variable-step-size proportionate normalized least mean square (SVS-PNLMS) algorithm is proposed for acoustic echo cancellation (AEC) application which is introduced as an important issue in services like video conferencing. The analysis reveals that it performs a faster convergence rate compared to that of the recently introduced SPNLMS (segment proportionate normalized least mean square), PNLMS (proportionate normalized least mean square) algorithms. Compared with its proportionate counterparts e.g. PNLMS and SPNLMS, the proposed SVS-PNLMS algorithm not only results in a faster convergence rate for both white and colored noise inputs, but also preserves this initial fast convergence rate until it reaches to steady state. It also presents a higher tracking behavior for quasi-stationary inputs such as speech signal in addition to better performance in terms of computational complexity and resulting ERLE (echo return loss enhancement).