The quantification of large SNR for MLE of ARARMAX models

Y. Zou; W.P. Heath

doi:10.1109/CDC.2009.5399593

The quantification of large SNR for MLE of ARARMAX models

Y. Zou
, W.P. Heath

University of Manchester

Allbwn ymchwil: Cyfraniad at gynhadledd › Papur › adolygiad gan gymheiriaid

Crynodeb

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.

Iaith wreiddiol	Anadnabyddus
Tudalennau	5108-5113
Nifer y tudalennau	6
Dynodwyr Gwrthrych Digidol (DOIs)	https://doi.org/10.1109/CDC.2009.5399593
Statws	Cyhoeddwyd - 29 Ion 2010
Cyhoeddwyd yn allanol	Ie
Digwyddiad	Proceedings of the 48h IEEE Conference on Decision and Control (CDC) held jointly with 2009 28th Chinese Control Conference - Shanghai, Tsieina Hyd: 15 Rhag 2009 → 18 Rhag 2009

Cynhadledd

Cynhadledd	Proceedings of the 48h IEEE Conference on Decision and Control (CDC) held jointly with 2009 28th Chinese Control Conference
Gwlad/Tiriogaeth	Tsieina
Dinas	Shanghai
Cyfnod	15/12/09 → 18/12/09

Mynediad at Ddogfen

10.1109/CDC.2009.5399593

Dyfynnu hyn

@conference{d6f19b807b164af6ad54d0afb83435bc,

title = "The quantification of large SNR for MLE of ARARMAX models",

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.",

author = "Y. Zou and W.P. Heath",

year = "2010",

month = jan,

day = "29",

doi = "10.1109/CDC.2009.5399593",

language = "Anadnabyddus",

pages = "5108--5113",

note = "Proceedings of the 48h IEEE Conference on Decision and Control (CDC) held jointly with 2009 28th Chinese Control Conference ; Conference date: 15-12-2009 Through 18-12-2009",

}

TY - CONF

T1 - The quantification of large SNR for MLE of ARARMAX models

AU - Zou, Y.

AU - Heath, W.P.

PY - 2010/1/29

Y1 - 2010/1/29

N2 - 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.

AB - 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.

U2 - 10.1109/CDC.2009.5399593

DO - 10.1109/CDC.2009.5399593

M3 - Papur

SP - 5108

EP - 5113

T2 - Proceedings of the 48h IEEE Conference on Decision and Control (CDC) held jointly with 2009 28th Chinese Control Conference

Y2 - 15 December 2009 through 18 December 2009

ER -