本文介绍了自适应滤波算法的原理及性能,并介绍了自适应滤波在降噪方面的应用。在MATLAB环境下编写LMS算法的程序,然后运用MATLAB软件分析语音信号的时域波形图和频域波形图,验证最小均方(LMS)自适应滤波算法的降噪效果。LMS算法的降噪效果比较明显,但是这种算法依旧存在收敛速度慢的缺点。本文对LMS算法进行改进并利用MATLAB进行仿真与分析,改进的算法比LMS算法的计算复杂度低,计算时间缩短,降噪效果更好。 This paper introduces the principle and performance of adaptive filtering algorithm, and introduces the application of adaptive filtering in noise reduction. In the MATLAB environment, the program of LMS algorithm was compiled, and then the MATLAB software was used to analyze the time-domain waveform and frequency-domain waveform of the speech signal to verify the noise reduction effect of the LMS adaptive filtering algorithm. The noise reduction effect of LMS algorithm is obvious, but this algorithm still has the disadvantage of slow convergence speed. In this paper, the LMS algorithm is improved and simulated and analyzed by using MATLAB. The improved algorithm has less computational complexity than LMS algorithm, and the calculation time is shortened, and the noise reduction effect is better.