时变参数比例自适应滤波算法  

Time-Varying Parameter Proportionate Adaptive Filtering Algorithm

作　　者：倪锦根[1] 

机构地区：[1]苏州大学电子信息学院,江苏苏州215006

出　　处：《电子学报》2016年第5期1208-1212,共5页Acta Electronica Sinica

基　　金：国家自然科学基金(No.61471251;No.61101217);江苏省自然科学基金(No.BK20131164)

摘　　要：在免提电话和视频会议系统中,自适应滤波器估计的回声路径通常是稀疏的.改进的比例归一化最小均方(IPNLMS)算法能够加快自适应滤波器在估计稀疏系统时的收敛速度,但与归一化最小均方(NLMS)算法相比,其稳态失调的波动性较大.为了解决这一问题,本文提出了一种时变参数IPNLMS(TV-IPNLMS)算法.该算法根据系统的均方误差(MSE)与噪声功率的比值,使用一个sigmoid函数来调整时变参数的值.该时变参数能够降低IPNLMS算法在滤波器到达稳态时的比例增益.仿真结果表明,时变参数方法能够降低IPNLMS算法稳态失调的波动性.该算法可用于回声消除、主动噪声控制等领域.In hands-free telephones and teleconferencing systems,the echo path to be estimated by the adaptive filter is usually sparse. The improved proportionate normalized least-mean-square （IPNLMS） algorithm can increase the conver- gence rate of the adaptive filter when it is used to estimate sparse systems. However, the steady-state misalignment of the IPNLMS algorithm may suffer from much larger fluctuations than that of the normalized least-mean-square （ NLMS ） algo- rithm. To address this problem, a time-varying parameter IPNLMS （TV-IPNLMS） algorithm is proposed, which uses a sig- moid function to adjust the value of the time-varying parameter according to the ratio of the mean square error （MSE） to the power of the system noise. This time-varying parameter can reduce the proportionate gains of the IPNLMS algorithm when the adaptive filter arrives at steady state. Simulation results show that the time-varying parameter method can reduce the fluc- tuations of the steady-state misalignment of the IPNLMS algorithms. This algorithm can be used in the fields of echo cancel- lation, active noise control, and so on.

关 键 词：免提电话 视频会议 比例自适应 时变参数 稳态失调 

分 类 号：TN911.72[电子电信—通信与信息系统]