摘要
基于箕舍线的变步长最小均方(Variable step least mean square,VSLMS)算法是一种经典的LMS算法,但其步长更新公式易受噪声干扰的影响,导致自适应滤波器权值在最优权值附近波动。为解决箕舌线变步长LMS算法步长更新公式易受噪声干扰的问题,根据高斯白噪声相关性比较差的特性,对箕舌线变步长LMS算法进行改进,提出基于相关特性的改进箕舌线变步长LMS算法,使算法的抗噪声干扰能力明显增强。理论分析和仿真结果表明:若两算法选取相同参数,则基于相关特性的改进箕舌线变步长LMS算法相对于箕舌线变步长LMS算法具有小的稳态误差;在保证算法收敛的条件下,基于相关特性的改进箕舌线变步长LMS算法相对箕舌线变步长LMS算法具有较快的收敛速度。
The tonguelike curve variable step least mean square(VSLMS)algorithm is a classical LMS algorithm.The disadvantage of the algorithm is that its step-formula can be disturbed easily by noise jamming,thus causing adaptive filter′s weights fluctuate around the optimal weights.To solve the problem,the tonguelike curve VSLMS algorithm is improved according to the characteristics that the correlation of white Gaussian noise is weak.The improved tonguelike curve VSLMS algorithm based on the correlation characteristic is presented.Tonguelike curve VSLMS algorithm′s ability of anti-noise jamming is improved evidently.If two algorithms choose the same parameters,the improved tonguelike curve VSLMS algorithm based on the correlation characteristic has the less steady-state error than the tonguelike curve VSLMS algorithm.Under the condition that the two algorithms are both convergent,the convergence rate of the improved tonguelike curve VSLMS algorithm is faster than that of the tonguelike curve VSLMS algorithm.The above conclusions are testified through theoretical analysis and simulation.
出处
《数据采集与处理》
CSCD
北大核心
2015年第4期896-901,共6页
Journal of Data Acquisition and Processing
关键词
自适应滤波器
噪声抑制
稳态误差
收敛速度
adaptive filter
noise suppression
steady-state error
convergence rate