改进子带自适应滤波算法及其在回波抵消中的应用  

Improved Subband Adaptive Filter and Its Application in Echo Cancellation

作　　者：刘小蒙[1] 邵高平[1] 戚晓慧 

机构地区：[1]解放军信息工程大学,河南郑州450001

出　　处：《信号处理》2016年第8期973-981,共9页Journal of Signal Processing

摘　　要：本文提出了两种基于多带结构的仿射投影符号子带自适应滤波器(Affine Projection Sign Subband Adaptive Filter,AP-SSAF)的改进方法。针对稀疏系统的系统识别,设计了两种子带自适应滤波器。首先给出了AP-SSAF的变正则化参数更新方程,文中采用随机梯度下降法来更新正则化参数,来使系统的均方偏差最小化,该方法能同时兼顾快速收敛及低稳态失调。其次将权重分布矩阵引入AP-SSAF得到系数比例AP-SSAF,该方法能够利用系统的稀疏性提高AP-SSAF的收敛性能。仿真中将本文所提算法用于一般系统识别以及回波抵消,实验结果验证了本文的算法对脉冲噪声具有稳健性,具有较好的跟踪性能,并具有较快的收敛速度及低稳态失调。This paper proposes two improved affine projection sign subband adaptive filters （AP-SSAF） based on the multi- band structure, which called variable regularization parameter AP-SSAF （VRP-AP-SSAF） and proportionate AP-SSAF （ P- AP-SSAF）. For sparse system identification, two AP-SSAFs are designed. On the one hand, VRP-AP-SSAF is presented to get a trade-off between fast convergence rate and low steady-state misalignment by minimizing the mean-square deviation （MSD） of the system weight vector. In addition, the update of regularization parameter is achieved by the normalized sto- chastic gradient of MSD. On the other hand, P-AP-SSAF incorporates a gain distribution matrix into the AP-SSAF to pro- portionately adapt the tap-weight vector of adaptive filter, which can both utilize the sparsity of system and raise conver- gence performance for sparse system identification. Simulation results show that the proposed VRP-AP-SSAF and P-AP- SSAF used in general system identification and echo cancellation, not only maintain the robustness against impulsive noise and have a good result for tracking capacity, but also have an improved performance in convergence rate and steady-state misalignment in their proper context.

关 键 词：正则化参数 符号子带自适应滤波 仿射投影 系数比例自适应滤波 回波抵消 

分 类 号：TN911[电子电信—通信与信息系统]