Research And Implementation Of Acoustic Echo Cancellation Based On Webrtc

Adaptive Echo Canceller is mainly used for Acoustic Echo Cancellation. And its core component is an adaptive filter. The basic principle of AEC is to simulate echo channel and simulate the echo signal, then the microphone input signal subtracts the estimation signal, so that the input voice signal is no longer containing echo. At last acoustic echo is cancelled. How to eliminate echo on the phone call has been a very important research direction in the multimedia communication, and many research results have been widely used in many business areas. The topic of this paper is also directly derived from practical multimedia communications engineering research projects.This paper firstly analyzes the echo generation process, as well as the principle of acoustic echo canceller. Then the paper still analyzes the main cause of telephone echo and points out that the research focus of echo cancellation has shifted to "acoustic echo" from "line echo". In accordance with the characteristics of the acoustic echo, this paper analysis and studies adaptive echo canceller’s structural features and give its design ideas. On this basis, the thesis starts its work. This paper’s work and new results are as follows:1. Putting forward a NLMS algorithm(normalized LMS algorithm), compared with some other algorithms in convergence and stability, the structure of NLMS algorithm is simple and easy to implement.2. Proposing a new adaptive scheme to to adjust the adaptive coefficient based on Google WebRTC AECM’s adaptive algorithm.3. Acoustic Echo Canceller’s software design and implementation, including the design and implementation of the various modules and specific algorithms. The results of echo canceller test through Audacity shows that the AEC system has a more efficient adaptive acoustic echo cancellation function.The result also proves that WebRTC AECM is better on processing echo cancellation than other software.