归一化子带自适应滤波器步长控制

Step-Size Control for the Normalized Subband Adaptive Filter

定步长子带自适应滤波器必须在快的收敛速度和低的稳态失调之间进行折中。根据自适应滤波器系数向量均方偏差与步长之间的函数关系,该文采用使自适应滤波器系数向量均方偏差在每次迭代更新时最速下降的方法,提出一种步长控制算法来解决上述问题。该算法可以兼得快的收敛速度和低的稳态失调。实验结果验证了该方法的有效性。

Fixed step-size Subband Adaptive Filters（SAFs） must carry out a trade-off between fast convergence rate and low steady-state misadjustment.According to the functional relationship between the Mean-Square Deviation（MSD） of the coefficient vector of the adaptive filter and step-size,this paper proposes a step-size control algorithm to address the problem above,which is derived by the largest decrease method of the MSD of the coefficient vector of the adaptive filter for each iterative update.This algorithm can obtain both fast convergence rate and low steady-state misadjustment.Experimental results verify the validity of the proposed method.