期刊文献+
共找到1篇文章
< 1 >
每页显示 20 50 100
Convergence analysis of filtered-X LMS algorithm with secondary path modeling error 被引量:1
1
作者 SUN Xu CHEN Duanshi( State Key Lab. of Vibration, Shock and Noise, Shanghai Jiaotong University Shanghai 200030) ( Institute of Power Plant and Automation, Shanghai Jiaotong University Shanghai 200030) 《Chinese Journal of Acoustics》 2003年第2期146-153,共8页
A more relaxed sufficient condition for the convergence of filtered-X LMS (FXLMS) algorithm is presented. It is pointed out that if some positive real condition for secondary path transfer function and its estimates i... A more relaxed sufficient condition for the convergence of filtered-X LMS (FXLMS) algorithm is presented. It is pointed out that if some positive real condition for secondary path transfer function and its estimates is satisfied within all the frequency bands, FXLMS algorithm converges whatever the reference signal is like. But if the above positive real condition is satisfied only within some frequency bands, the convergence of FXLMS algorithm is dependent on the distribution of power spectral density of the reference signal, and the convergence step size is determined by the distribution of some specific correlation matrix eigenvalues.Applying the conclusion above to the Delayed LMS (DLMS) algorithm, it is shown that DLMS algorithm with some error of time delay estimation converges in certain discrete frequency bands, and the width of which are determined only by the 'time-delay estimation error frequency' which is equal to one fourth of the inverse of estimated error of the time delay. 展开更多
关键词 LMS IT IS of convergence analysis of filtered-X LMS algorithm with secondary path modeling error with
原文传递
上一页 1 下一页 到第
使用帮助 返回顶部