An improved order estimation of MSF for stereophonic acoustic echo cancellation

Abstract

In this paper the order estimation of each sub-filter, in Multiple Sub-Filters (MSF), parallel structure has been used for Stereophonic Acoustic Echo Cancellation (SAEC). The performance of the MSF with suitable sub-filter order has been studied for two types of algorithms namely; different error algorithm and common error algorithm. The performances of these filter structures and adaptive algorithms have been compared with conventional Single Long Filter (SLF) echo canceller via computer simulations. The order estimation algorithm designed to estimate the order of each sub-filter in MSF not only reduces the computational complexity but also perform better at relatively low SNR ranges. Simulation and results show that MSF with both adaptation algorithms provide better convergence speed as compared to the conventional SAEC. © 2012 Springer-Verlag GmbH Berlin Heidelberg.