Collaborative Adaptive Learning using Hybrid Filters

Author

Mandic, D. ; Vayanos, P. ; Boukis, C. ; Jelfs, B. ; Goh, S.L. ; Gautama, T. ; Rutkowski, T.

Author_Institution

Imperial Coll. London

Volume

3

fYear

2007

fDate

15-20 April 2007

Abstract

A novel stable and robust algorithm for training of finite impulse response adaptive filters is proposed. This is achieved based on a convex combination of the least mean square (LMS) and a recently proposed generalised normalised gradient descent (GNGD) algorithm. In this way, the desirable fast convergence and stability of GNGD is combined with the robustness and small steady state misadjustment of LMS. Simulations on linear and nonlinear signals in the prediction setting support the analysis.

Keywords

FIR filters; adaptive filters; filtering theory; gradient methods; least mean squares methods; LMS; collaborative adaptive learning; finite impulse response adaptive filters; generalised normalised gradient descent algorithm; hybrid filters; least mean square; steady state misadjustment; Adaptive filters; Analytical models; Collaboration; Convergence; Finite impulse response filter; Least squares approximation; Predictive models; Robust stability; Robustness; Steady-state; Distributed; adaptive; collaborative SP;

fLanguage

English

Publisher

ieee

Conference_Titel

Acoustics, Speech and Signal Processing, 2007. ICASSP 2007. IEEE International Conference on

Conference_Location

Honolulu, HI

ISSN

1520-6149

Print_ISBN

1-4244-0727-3

Electronic_ISBN

1520-6149

Type

conf

DOI

10.1109/ICASSP.2007.366831

Filename

4217861