Minimum mean cubic absolute value echo elimination method of convex combination

Abstract

The invention discloses a minimum mean cubic absolute value echo elimination method of convex combination. The method comprises the following steps: A, performing far-end signal filtering to obtain alarge step length filtering value y1(n) and a small step length filtering value y2(n); B, performing convex combination: performing convex combination on the large step length filtering value y1(n) and the small step length filtering value y2(n) to obtain a combined filtering value y(n); C, performing echo cancellation; D, performing filter tap weight coefficient update: respectively performing tap weight coefficient update of each of a large step length filter and a small step length filter based on the minimum mean cubic absolute value algorithm; E, performing weight update of the filter; F,performing weight limitation of the filter; G, enabling n=n+1, repeating the steps A, B, C, D, E and F until the call is ended. The cost function for the tap weight coefficient update of each of thesmall and large filters is the mean cubic absolute value of the total residual signal e(n), the error is amplified, the operational precision is high; through the fast convergence of the large step length filter and the small steady state error of the small step length filter, and the steady state error is small, the convergence speed is fast, and the echo cancellation effect is good.

Description

technical field [0001] The invention belongs to the technical field of adaptive echo cancellation for telephone communication. Background technique [0002] With the advancement of technology in recent years, communication technology has been developing day by day, and various new communications emerge in an endless stream, but the most important communication method is still voice communication, and users are more and more concerned about the quality of voice communication. Acoustic echo is the most important factor affecting the quality of voice calls. Acoustic echo in the communication process means that the user hears his own voice repeatedly during the communication process, which is divided into direct echo and indirect echo. Direct echo means that the sound from the speaker directly enters the microphone without any reflection. The delay of this echo is the shortest, which is related to the voice energy of the far-end speaker, the distance and angle between the speak...

AI Technical Summary

Problems solved by technology

However, because it uses a fixed step size, the initial step size is too short to converge faster, and when it converges, the step size is too large, making it difficult to accurately identify the system, resulting in steady-state imbalance

Its convergence speed and steady-state performance need to be improved

Images