A justification for the improved performance of the multi-split LMS algorithm

Author

Souza, R.D. ; Resende, L.S. ; Bellanger, M.G.

Author_Institution

Commun. Res. Group, Univ. Fed. de Santa Catarina, Florianopolis, Brazil

Volume

6

fYear

2003

fDate

6-10 April 2003

Abstract

This paper presents an analysis that justifies the improved performance of the multi-split LMS algorithm. It is shown that instead of reducing the eigenvalue ratio, the multi-split operation increases the diagonalization factor of the transformed input signal autocorrelation matrix, which assists the power normalized and time-varying step-size LMS algorithm used for updating the single parameters independently. Case studies and simulation results enable us to evaluate the improved performance of the multi-split LMS algorithm.

Keywords

adaptive signal processing; correlation methods; least mean squares methods; matrix algebra; time-varying systems; adaptive signal processing; diagonalization factor; multi-split LMS algorithm; parameter updating; performance; power normalization; time-varying step-size; transformed input signal autocorrelation matrix; Algorithm design and analysis; Autocorrelation; Computational complexity; Data preprocessing; Discrete cosine transforms; Eigenvalues and eigenfunctions; Finite impulse response filter; Least squares approximation; Performance evaluation; Robustness;

fLanguage

English

Publisher

ieee

Conference_Titel

Acoustics, Speech, and Signal Processing, 2003. Proceedings. (ICASSP '03). 2003 IEEE International Conference on

ISSN

1520-6149

Print_ISBN

0-7803-7663-3

Type

conf

DOI

10.1109/ICASSP.2003.1201625

Filename

1201625