DSP环境下新的变步长LMS自适应滤波算法

An improved variable step-size LMS adaptive filtering algorithm on DSP

基于解决固定步长的LMS算法收敛速度快与均方误差小两个性能指标无法兼顾的目的,提出一种改进的变步长LMS自适应滤波算法,采用步长因子调整变步长的方法,建立u因子与超量误差信号之间的非线性函数关系,从而实现自适应滤波。通过TMS320C5509A平台进行实测数据试验可以得出,本文算法能在低信噪比的条件下对信号进行滤波保持较高性能,同时具有收敛速度较快和稳态误差较小的优点,在雷达信号处理行业中应用广泛。

Based on the traditional fixed step LMS algorithm cannot bothfast convergence speed and excess mean square error is small the two performance indicators,an improved variable step size LMS adaptive filtering algorithm is proposed to solve this contradiction.The methord is adopted that the step factor is used to realize variable step-size LMS The improved algorithm＇s step size establish a nonlinear function between the step size and the error signal. It has the advantages of fast convergence speed, small steady-state error, and in low SNR environment remains a good performance.It has more applications in Radar signal processing.