The quantification of large SNR for MLE of ARARMAX models

Author

Zou, Yiqun ; Heath, William P.

Author_Institution

Control Syst. Centre, Univ. of Manchester, Manchester, UK

fYear

2009

fDate

15-18 Dec. 2009

Firstpage

5108

Lastpage

5113

Abstract

Maximum likelihood estimation(MLE) is widely applied in system identification because it is consistent and has excellent convergence properties. However gradient based optimization of likelihood function might end up in local convergence. It is known that for ARMAX and ARARX models, providing a large enough Signal-to-Noise-Ratio(SNR) will avoid the potential local convergence. We show the same condition can be extended to ARARMAX models in this paper. To ease the application of this condition, the exact value of such SNR needs to be quantified. Here we realize the quantification by constrained optimization.

Keywords

autoregressive moving average processes; convergence; gradient methods; maximum likelihood estimation; optimisation; ARARMAX Models; ARARX models; ARMAX models; convergence properties; gradient based optimization; large SNR quantification; likelihood function; maximum likelihood estimation; system identification; Constraint optimization; Control system synthesis; Convergence; Frequency domain analysis; Heat engines; Maximum likelihood estimation; Open loop systems; Resistance heating; System identification; Temperature control; ARARMAX; Constrained Optimization; MLE; Quantification; SNR;

fLanguage

English

Publisher

ieee

Conference_Titel

Decision and Control, 2009 held jointly with the 2009 28th Chinese Control Conference. CDC/CCC 2009. Proceedings of the 48th IEEE Conference on

Conference_Location

Shanghai

ISSN

0191-2216

Print_ISBN

978-1-4244-3871-6

Electronic_ISBN

0191-2216

Type

conf

DOI

10.1109/CDC.2009.5399593

Filename

5399593