改进的滤波-e LMS算法

针对有源噪声控制中次级通路为非最小相位系统的情况,提出一种改进的滤波-e算法.该算法采用次级通路传递函数的逆来构建误差滤波器,依据Shannon-Bode方法将误差滤波器分解为白化滤波器和非因果滤波器,并对其进行改进从而构建一个物理可实现的滤波器,再将获得的误差信号通过该误差滤波器后对控制器的权系数进行更新.仿真实验表明,改进后的算法不仅可以提高系统的稳定性和降噪量还可以减少系统的超量均方误差,较传统的滤波-e算法在性能上有明显改进.

An Improved Filtered-e LMS Algorithm

Because the secondary-path is a non-minimum phase system in the active noise control, an improved filtered-e LMS algorithm is presented in which the inverse of the secondary-path transfer function is adopted to construct an error filter. According to the Shannon-Bode theory, the error filter is decomposed into a whitening filter and a non-causal filter which can be modified to build a physical realized filter. The acquired error signal filtered by the error filter is used to update the weight parameters of the active controller. The simulation results show that the modified algorithm can not only improve the stability and noise reduction but also reduce the excess squared residual noise. It is indicated that there is obvious improvement compared with the traditional fihered-e algorithm.