A fast robust LMS algorithm utilizing the dynamics of a damped pendulum
โ Scribed by T.F. Haddad; M.A. Khasawneh
- Book ID
- 104134751
- Publisher
- Elsevier Science
- Year
- 1998
- Tongue
- English
- Weight
- 853 KB
- Volume
- 335
- Category
- Article
- ISSN
- 0016-0032
No coin nor oath required. For personal study only.
โฆ Synopsis
In this paper, a ne\c,fa.st LMS-based udaptive algorithm is proposed. It is derived by incorporatin~g a damping ,force in the LMS update recursion similar to the ,force ucting upon a damped planar pendulum. An expression ,for evolution and the steady-state behavior. for the mean weight rector-is developed. This expression provides u mathematical bound which constrains the parameter that controls the ma.xitnum contribution of'the introduced dampiyforce.
Furthermore.
an expression for the mean square error is developed to esumine algorithm behavior in thr mror~ .rquurc sense. Simulution results .show an improved robust performance ,for the new algorithm as compured with the conventional LMS and Momentum LMS (MLMS) ulgorithms in .smoothi~~ rrcuking the optimul solution in stationary, correlated and nonstationary noisy environments. Mowowr, simulation examples include system identljication with varying power environment.s. P , 1991
๐ SIMILAR VOLUMES
In this paper, we propose a new gradient-based adaptive algorithm, in which the conventional LMS update recursion is modi"ed by introducing a damping factor, which emulates the force acting on a free-moving object in a gravitational "eld. The contribution of the damping factor being introduced is co