Title :
Maximum likelihood forgetting stochastic gradient estimation algorithm for Hammerstein CARARMA systems
Author :
Li, Junhong ; Gu, Juping ; Ma, Weiguo ; Ding, Rui
Author_Institution :
Sch. of Electr. Eng., Nantong Univ., Nantong, China
Abstract :
This paper considers the identification problem of Hammerstein CARARMA systems, and derives a maximum likelihood stochastic gradient algorithm (ML-SG) by using the maximum likelihood principle and the negative gradient search. Furthermore, a forgetting factor is introduced to improve the convergence rate of the ML-SG algorithm. The simulation results indicate that the proposed algorithm are effective.
Keywords :
gradient methods; identification; maximum likelihood estimation; stochastic processes; Hammerstein CARARMA systems; ML-SG algorithm; convergence rate; identification problem; maximum likelihood forgetting stochastic gradient estimation algorithm; maximum likelihood principle; negative gradient search; Educational institutions; Maximum likelihood estimation; Nonlinear systems; Parameter estimation; Signal processing algorithms; Stochastic processes; Vectors; Hammerstein models; Maximum likelihood; Parameter estimation; Stochastic gradient;
Conference_Titel :
Control and Decision Conference (CCDC), 2012 24th Chinese
Conference_Location :
Taiyuan
Print_ISBN :
978-1-4577-2073-4
DOI :
10.1109/CCDC.2012.6244405
