Research Of Adaptive Noise Reduction Algorithm For Microphone Array

With the development of science and technology, wireless communication technology advances every day. The microphone array is the basic application of voice communication, but the noise is always appeared in the phonetic environments inevitably, so the study of the microphone array noise reduction technology is very important.There are a lot of difficulties in the study of noise reduction system of microphone array, for example, the influence of the structure of the microphone for noise reduction (the study of robustness),the influence of source location and the movement of the source for the signal acquisition, and when the frequency of human voice is close to the main source as the noise, how to separate and extract the signal. In addition algorithm (the core of the noise reduction), the convergence speed of algorithm and iteration complexity are the factors that should be considered in the design process.In this paper, on the basis of the traditional adaptive algorithm, the shortcoming is optimized and the effect of noise reduction is analysis. Firstly, this paper introduces the model of microphone array and the microphone array noise reduction technology, such as the time delay estimation, the principle of adaptive filter. Secondly, comparing the two basic algorithms adaptive algorithm (LMS and RLS) and matlab simulation, and comparing their convergence speed and error function, so RLS algorithm is more suitable for study of this paper. Finally, aiming at shortcomings of the high computational complexity of RLS algorithm.We design a low complexity algorithm as QR-RLS algorithm and we use the simulation results to verify the feasibility of the algorithm.