中北大学学报（自然科学版）2017，Vol.38Issue(2)：140-144,5.DOI:10.3969/j.issn.1673-3193.2017.02.008
一种新的变步长LMS自适应滤波算法及其应用
A New Variable Step Size LMS Adaptive Filtering Algorithm and Its Application
摘要
Abstract
For the problemthat variable step size LMS(Least Mean Square) adaptive filtering algorithm from a lot of literature could not meet the higher convergence rate and lower steady-state error,a new variable step size LMS adaptive filtering algorithm was proposed based on the feedback theory.A new nonlinear function model on step and error was created by introducing a feedback control function in the original algorithm model,making the current step value associated with the square of the current error and the previous error.The impact of new function model on the filtering performance was analyzed by MATLAB,and reasonable key parameters were determined.Simulation results show that:compared to the original algorithm,the new improved algorithm greatly improves the convergence rate,and the steady-state error is also reduced.The new algorithm has good performance it is applied to the filtering process of ultra-wideband radio fuze echo signal,the error suppression ability is improved by 4 times,and the filtering effect is better.关键词
变步长LMS/自适应滤波/收敛速度/稳态误差/超宽带无线电引信Key words
variable step size LMS/adaptive filtering/convergence rate/steady-state error/UWB radio fuze分类
信息技术与安全科学引用本文复制引用
韩广超,王锋,赵河明,彭志凌,柏迅..一种新的变步长LMS自适应滤波算法及其应用[J].中北大学学报（自然科学版）,2017,38(2):140-144,5.基金项目
国家青年自然科学基金资助项目(51305409) （51305409）
