不相关高斯输入情况下LMS滤波器的收敛性分析  

Convergence Analysis of Adaptive Filter with Uncorrelated Gaussian Input

作　　者：李琳[1] 路军[1] 张尔扬[1] 

机构地区：[1]国防科技大学电子科学与工程学院

出　　处：《信号处理》2004年第4期336-341,共6页Journal of Signal Processing

基　　金：JS63空间微波技术国防科技重点室基金试点项目(项目编号:2000JS63.3.1.KG0111)

摘　　要：在分析最小均方自适应滤波器(LMS AF)均方误差(MSE)的收敛性时,文献常用统计自相关矩阵代替瞬时自相关矩阵以简化分析,由此得出的收敛条件比较粗糙。本文指出:不相关高斯输入情况下,无需如上简化,可依据高斯阶矩因式分解定理得到确切的MSE收敛条件,相应的失调表式能更准确地预报失调。When analyzing the convergence behavior of the mean squared error (MSE) of the least mean square (LMS) adaptive filter, many references often replace the instantaneous correlation matrix of the tap inputs with statistical correlation matrix in order to simplify analysis. It is clear that the condition derived from this method is coarse. In fact, for uncorrelated Gaussian input, the accurate necessary and sufficient conditions for the convergence of the MSE can be obtained by Gaussian moment factoring theorem, and the accurate misadjustment can be predicted by the corresponding misadjustment expression.

关 键 词：LMS滤波器 收敛性分析 均方误差 高斯阶矩因式分解定理 

分 类 号：TN713[电子电信—电路与系统]