%0 Journal Article
%T Performance Analysis of Least Mean Square Algorithm for Time-Varying Systems<br>时变系统最小均方算法的性能分析(英文)
%A DING Feng
%A YANG Jia-ben
%A DING Tao
%A <br>丁　锋
%A 杨家本
%A 丁　韬
%J 控制理论与应用
%D 2001
%I 
%X By means of stochastic process theory, the bounded convergence of least mean square algorithm (LMS) is studied without data stationary assumption and ergodicity condition. The upper bound of the estimation error is given, and the way of choosing the convergence factor or stepsize is stated so that the upper bound of the parameter estimation error is minimized. The convergence analyses indicate that i) for deterministic time invariant systems, LMS algorithm is convergent exponentially, ii) for deterministic time varying systems, the estimation error upper bound is minimal as the stepsize goes to unity, and iii) for time varying or time invariant stochastic systems, the estimation error is uniformly bounded.
%K time
%K varying system
%K identification
%K parameter estimation
%K least mean square algorithm<br>时变系统
%K 最小均方算法
%K 性能分析
%K 参数估计
%U http://www.alljournals.cn/get_abstract_url.aspx?pcid=5B3AB970F71A803DEACDC0559115BFCF0A068CD97DD29835&cid=8240383F08CE46C8B05036380D75B607&jid=970898A57DFC021F93AB51667BAED7F7&aid=A1085D3A7C90B39E&yid=14E7EF987E4155E6&vid=13553B2D12F347E8&iid=38B194292C032A66&sid=DDEED1BDDBFAA8A7&eid=B40AD8FE6FA88DE9&journal_id=1000-8152&journal_name=控制理论与应用&referenced_num=4&reference_num=6