改进的LMS算法在噪声对消中的应用

摘要

To address the issue of contradiction between convergence rate and steady⁃state error of the LMS algorithm, an improved LMS algorithm is proposed.The new algorithm makes three improvements based on the existing algorithms:using the autocorre⁃lation of error signals to suppress uncorrelated noise;setting the first step adaptive after a fixed one to improve the convergence rate and reduce the steady⁃state error;regulating the fixed parameters in the step factor and making them be adjusted as the step changes to fur⁃ther reduce the steady⁃state error. Both the existing algorithm and the improved algorithm are applied in the simulation experiment of adaptive noise,and the results show that the new algorithm has advantages of faster convergence and less steady⁃state error,as well as good noise cancellation capability.%针对最小均方误差（ Least Mean Squares， LMS）算法收敛速度和稳态失调之间的矛盾，在已有算法的基础上进行3个方面改进：利用误差信号自相关性对不相关噪声进行抑制、将步长设置为先固定后自适应，提高算法的收敛速度、减小稳态误差；调节步长因子中固定的参数，使其伴随步长变化进行调节，进一步减小稳态误差。将已有算法与改进算法同时用于Matlab仿真实验，仿真结果表明，新算法具有更快收敛速度、更小稳态误差和良好的消噪能力。