A new class of gradient adaptive step-size LMS algorithms
Authors:
- B. Farhang-BoroujenyBFB. Farhang-Boroujeny
- 1 April 2001
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Signal Processing
- Vol. 49 (4) , 805-810
Abstract
The gradient adaptive step-size least-mean-square (LMS) algorithms [an important family of variable step-size LMS (VSLMS) algorithms] are revisited. We propose a simplification to a class of the studied algorithms and show that this leads to a new class of VSLMS algorithms with reduced complexity but with no observable loss in performance.Keywords
This publication has 7 references indexed in Scilit:
- Performance of joint data and channel estimation using tap variable step-size (TVSS) LMS for multipath fast fading channel2002
- Analysis of adaptive step-size SA algorithms for parameter trackingIEEE Transactions on Automatic Control, 1995
- Variable-step-size LMS algorithm: new developments and experimentsIEE Proceedings - Vision, Image, and Signal Processing, 1994
- A stochastic gradient adaptive filter with gradient adaptive step sizeIEEE Transactions on Signal Processing, 1993
- Exponentially weighted stepsize NLMS adaptive filter based on the statistics of a room impulse responseIEEE Transactions on Speech and Audio Processing, 1993
- Adaptive Algorithms and Stochastic Approximations1990
- Measures of tracking performance for the LMS algorithmIEEE Transactions on Acoustics, Speech, and Signal Processing, 1990