Acoustic noise and echo cancellation microphone system for videoconferencing

Acoustic noise and echo cancellation using two adaptive filters and the least-mean-square (LMS) algorithm is the most effective technique of reducing acoustic noises and echoes in a hands-free telephone system. Unfortunately, high order filters are required in this traditional structure, which results in high cost, slow convergence, and high excess mean-square and numerical errors. This paper presents a new acoustic noise and echo cancellation microphone (ANECM) system, which consists of two closely-spaced directional microphones pointing in opposite directions and two low order adaptive filters. By using two highly correlated microphone signals, the ANECM system provides satisfied performance at much lower cost than the traditional techniques. To test and perfect the design, computer simulations were conducted based on real data collected from a typical conference room.