Title :
Estimating the optimal adaptation gain for the LMS algorithm
Author :
Reason, Johnathan ; Wei Ren
Author_Institution :
Dept. of Electr. Eng. & Comput. Sci., California Univ., Berkeley, CA, USA
Abstract :
This paper evaluates two methods to see whether they are capable of estimating the optimal adaptation gain for the LMS algorithm in a stationary time-varying environment. The results show that the method of Kwong and Johnston (1992) runs counter to the goal of optimal trade-off between noise-rejection and tracking, while the method of Harris et al. (1986) (the VS method) comes closer to estimating the optimal adaptation gain, but there is much room for improvement-a preliminary improvement is suggested in the paper
Keywords :
adaptive filters; identification; least squares approximations; noise; time-varying systems; tracking; LMS algorithm; noise-rejection; optimal adaptation gain; stationary time-varying environment; tracking; variable step method; Adaptive filters; Algorithm design and analysis; Counting circuits; Equations; Gain; Least squares approximation; Parameter estimation; System identification; Variable structure systems; Yield estimation;
Conference_Titel :
Decision and Control, 1993., Proceedings of the 32nd IEEE Conference on
Conference_Location :
San Antonio, TX
Print_ISBN :
0-7803-1298-8
DOI :
10.1109/CDC.1993.325445
