A noise constrained VS-LMS algorithm

Author

Bilcu, Radu Ciprian ; Kuosmanen, Pauli ; Rus, Corneliu

Author_Institution

Signal Process. Lab., Tampere Univ. of Technol., Finland

fYear

2000

fDate

2000

Firstpage

29

Lastpage

33

Abstract

In many practical applications from the field of system identification, an accurate estimation of the noise variance is available. Therefore, we introduce a new variable step-size LMS algorithm that exploits this information in order to provide a faster convergence speed. The new algorithm is called NCVS-LMS. The convergence speed and steady-state error performance of the new algorithm are analyzed. The experimental results show that the new algorithm outperforms the LMS and other variable step algorithms

Keywords

adaptive filters; convergence of numerical methods; filtering theory; identification; least mean squares methods; noise; NCVS-LMS; adaptive filtering; adaptive systems; convergence speed; experimental results; noise constrained VS-LMS algorithm; noise variance estimation; steady-state error performance; system identification; variable step algorithms; variable step-size LMS algorithm; Adaptive filters; Convergence; Digital signal processing; Laboratories; Least squares approximation; Performance analysis; Signal processing; Signal processing algorithms; Steady-state; System identification;

fLanguage

English

Publisher

ieee

Conference_Titel

EUROCOMM 2000. Information Systems for Enhanced Public Safety and Security. IEEE/AFCEA

Conference_Location

Munich

Print_ISBN

0-7803-6323-X

Type

conf

DOI

10.1109/EURCOM.2000.874766

Filename

874766