SSP'05 -- Proceedings

	IEEE/SP 13th workshop on Statistical Signal Processing July, 17-20, 2005 - Bordeaux - France

Information regarding the paper

Title

A New Proportionate Fast LMS/Newton Algorithm for Adaptive Filtering

Author(s)

Yi Zhou	The University of Hong Kong
S. C. Chan	The University of Hong Kong
K. L. Ho	The University of Hong Kong

Get the paper in PDF format

To obtain Acrobat Reader (version 5 minimum required) necessary to his read.

Abstract

Abstract—This paper proposes a new proportionate adaptive filtering algorithm which exploits the advantageous features of the generalized proportionate NLMS (GP-NLMS) algorithm and the fast LMS/Newton algorithm. By means of an efficient switching mechanism, the new algorithm works alternately between the GP-NLMS and the fast LMS/Newton algorithms in order to combine their respective advantages. The overall converging speed and steady state performance for both sparse and dispersive channels as well as tracking performance are thus significantly improved. Computer simulations on an echo cancellation problem verify the superior performance of the new algorithm over both the GP-NLMS algorithm and the conventional fast LMS/Newton algorithm.