Maximum likelihood based multi-innovation stochastic gradient estimation for controlled autoregressive ARMA systems using the data filtering technique

Author

Feiyan Chen ; Feng Ding

Author_Institution

Key Lab. of Adv. Process Control for Light Ind. (Minist. of Educ.), Jiangnan Univ., Wuxi, China

fYear

2014

Firstpage

5993

Lastpage

5998

Abstract

This paper considers parameter estimation problems of a controlled autoregressive ARMA system. We decompose this system into two subsystems, use the data filtering technique to derive a maximum likelihood multi-innovation stochastic gradient algorithm. The simulation results show that the proposed algorithm has a higher computational efficiency than the maximum likelihood gradient algorithm and the filtering-based maximum likelihood stochastic gradient algorithm.

Keywords

autoregressive moving average processes; estimation theory; filtering theory; gradient methods; maximum likelihood sequence estimation; parameter estimation; computational efficiency; controlled autoregressive ARMA system; data filtering technique; filtering-based maximum likelihood stochastic gradient algorithm; maximum likelihood based multiinnovation stochastic gradient estimation; maximum likelihood gradient algorithm; maximum likelihood multiinnovation stochastic gradient algorithm; parameter estimation; Computational modeling; Mathematical model; Maximum likelihood estimation; Parameter estimation; Signal processing algorithms; Stochastic processes; Vectors; Filtering; Maximum likelihood; Parameter estimation; Stochastic gradient;

fLanguage

English

Publisher

ieee

Conference_Titel

Intelligent Control and Automation (WCICA), 2014 11th World Congress on

Type

conf

DOI

10.1109/WCICA.2014.7053747

Filename

7053747