Title :
Performance of least mean absolute third (LMAT) adaptive algorithm in various noise environments
Author :
Lee, Young Hwan ; Mok, Jin Dam ; Kim, Sang Duck ; Cho, Sung Ho
Author_Institution :
Wireless Commun. Stand. Sect., ETRI, Taejon, South Korea
fDate :
2/5/1998 12:00:00 AM
Abstract :
Convergence properties of the stochastic gradient adaptive algorithm based on the least mean absolute third (LMAT) error criterion are presented. The performance of the algorithm is examined and compared for several different probability densities of the measurement noise in the system identification mode. It is observed that the LMAT algorithm outperforms the LMS algorithm for most noise probability densities, except in the case of exponentially distributed noise
Keywords :
adaptive signal processing; convergence of numerical methods; identification; least squares approximations; noise; probability; LMAT error criterion; convergence properties; exponentially distributed noise; least mean absolute third adaptive algorithm; noise environment; noise probability densities; stochastic gradient adaptive algorithm; system identification mode;
Journal_Title :
Electronics Letters
DOI :
10.1049/el:19980181
