Research on partitioned block frequency domain adaptive filter approach for acoustic echo cancellation systems

Abstract

This work deals with the development or cancellations of echo in complex scenario. Essentially. the new methods and techniques make changes in the conventional sequential pnrtuion sigunl improvements for adaptive filter operations. Furthermore, this type of time segmentation, IIl10ws step by step coefficient updating stages relating to different methods.The adaptive filter acoustic echo reduction algorithms (AERA) is proposed in this work. The MAT LAB simulation results indicate good tracking capability, high value of echo return loss enhancement (ERLE) value. reasonable convergence time, rate and speed, small misadjustment. Meanwhile. the acoustic echo cancellation based on frequency domain methods has fast processing and convergence rime compared to time domain Normalized least means square fNLMS) and Variable step size normalized least means square (VSS-NLMS)methods.