A new modified robust variable step size LMS algorithm

Author

Kun, Zou ; Xiubing, Zhao

Author_Institution

Sch. of Telecommun. Eng., Air Force Eng. Univ., Xi´´An

fYear

2009

fDate

25-27 May 2009

Firstpage

2699

Lastpage

2703

Abstract

The variable step size (VSS) adaptive algorithm can be used to avoid the dilemma between convergence rate and mean square error of fix step size (FSS) least mean square (LMS) algorithm. Satisfying the basic requirement of the VSS adaptive algorithm, with the aim to the noising time-vary system, using the error square and error correlation information, a modified robust VSS algorithm is proposed, and its performance is analyzed. The theory analysis and computer simulation validate that the MRVSS algorithm can achieve the faster convergence rate and lower mean square error, and the anti-noise and tracking ability is better than the algorithms proposed in references, and the computer simulation result is correspondent with theory analysis.

Keywords

adaptive filters; least mean squares methods; adaptive filtering method; fix step size least mean square algorithm; robust variable step size LMS algorithm; Adaptive algorithm; Algorithm design and analysis; Computer errors; Computer simulation; Convergence; Frequency selective surfaces; Least squares approximation; Mean square error methods; Robustness; Variable structure systems; anti-noise ability; least mean square algorithm; tracking ability; variable steps size;

fLanguage

English

Publisher

ieee

Conference_Titel

Industrial Electronics and Applications, 2009. ICIEA 2009. 4th IEEE Conference on

Conference_Location

Xi´an

Print_ISBN

978-1-4244-2799-4

Electronic_ISBN

978-1-4244-2800-7

Type

conf

DOI

10.1109/ICIEA.2009.5138654

Filename

5138654