Title :
Modified Extrapolated LMS Algorithm
Author :
Shehata, Tarik S. ; Aboulnasr, T. ; Tanany, M. El
Author_Institution :
Dept. of Syst. & Comp. Eng., Carleton Univ., Ottawa, ON
Abstract :
This paper proposes a modification to extrapolated LMS algorithm proposed in the work of Sau-Gee Chen et al. (1996) to guarantee consistent improved convergence speed. Unlike the original extrapolated LMS algorithm that exhibits improved performance on specific runs but is not guaranteed to perform well at any given run, the proposed algorithm is based on the repeated application of the extrapolation process in a way that allows the algorithm to retain the ELMS advantage for any arbitrary run with limited additional complexity.
Keywords :
adaptive filters; convergence of numerical methods; extrapolation; least mean squares methods; adaptive filter; convergence speed; modified extrapolated LMS algorithm; Adaptive algorithm; Adaptive filters; Convergence; Extrapolation; Finite impulse response filter; Instrumentation and measurement; Iterative algorithms; Least squares approximation; Real time systems; Steady-state; Adaptive algorithms; LMS;
Conference_Titel :
Instrumentation and Measurement Technology Conference Proceedings, 2008. IMTC 2008. IEEE
Conference_Location :
Victoria, BC
Print_ISBN :
978-1-4244-1540-3
Electronic_ISBN :
1091-5281
DOI :
10.1109/IMTC.2008.4547081
