Performance analysis of an improved variable tap-length LMS algorithm

Author

Wang, Junfeng ; Luo, Junwei

Author_Institution

Coll. of Comput. Sci.&Technol., Henan Polytech. Univ., Jiaozuo

fYear

2009

fDate

11-12 April 2009

Firstpage

377

Lastpage

380

Abstract

An improved variable tap-length least mean square(LMS) algorithm is proposed in this paper. As compared with the original fractional tap-length LMS (FT-LMS) algorithm, the proposed algorithm can obtain both a fast convergence rate and a small steady-state error of the tap-length. Algorithm analysis is given which also provides a guideline for choosing the parameter of the proposed algorithm. Simulations are performed under low noise and high noise conditions with time varying impulse responses. All simulation results show the advantages of the proposed algorithm as compared with the FT-LMS algorithm, and confirm the algorithm analysis.

Keywords

adaptive filters; convergence of numerical methods; least squares approximations; adaptive filter; convergence; fractional tap-length LMS algorithm; least mean square method; performance analysis; steady-state error; time varying impulse response; variable tap-length LMS algorithm; 1f noise; Adaptive filters; Algorithm design and analysis; Analytical models; Computational complexity; Convergence; Flyback transformers; Least squares approximation; Performance analysis; Steady-state; Adaptive filter; FT-LMS; Variable tap-length;

fLanguage

English

Publisher

ieee

Conference_Titel

Image Analysis and Signal Processing, 2009. IASP 2009. International Conference on

Conference_Location

Taizhou

Print_ISBN

978-1-4244-3987-4

Type

conf

DOI

10.1109/IASP.2009.5054621

Filename

5054621