基于粒子群的LMS算法在信号滤波降噪中的应用

摘要

在自适应滤波算法中,LMS算法是最常用的算法之一,因为具备结构简单,易于实现,性能稳定,计算复杂度低等特点.然而,LMS算法也存在缺点,比如,收敛速度较慢,收敛精度低的问题,这就影响LMS算法在收敛性要求较高的领域中的应用.使用粒子群算法对LMS算法进行改进,可以将LMS滤波设计变成对LMS滤波参数优化的问题,利用粒子群算法的优化能力,使得滤波参数得到全局最优解.以此可以提高LMS滤波算法的收敛性能,从而提高滤波性能.%In the adaptive filter algorithms,LMS algorithm is one of the most common algorithms because of its simple structure,facile realization and stability,low computational complexity.But,LMS algorithm has also some defects,for example,low rate of convergence and convergence precision.Those problems have impacted on the application in high convergence fields.PSO algorithm can be used to improve the LMS algorithm.The designation of the LMS filter can be transferred to optimization problem for the parameters of LMS filter.The optimization of PSO algorithm is utilized to get the global optimal solution of filter parameters.In this way,convergence of LMS algorithm can be improved,and the performance of filter also can be improved.