Maximum Likelihood identification for Linear Dynamic Systems with finite Gaussian mixture noise distribution

Gustavo Bittner, Rafael Orellana, Rodrigo Carvajal, Juan C. Aguero

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

8 Scopus citations

Abstract

This paper considers the identification of a linear dynamic system driven by a non-Gaussian noise distribution. The noise is approximated by a finite Gaussian mixture, whilst the parameters of the system and the parameters that approximate the noise distribution are simultaneously estimated using the principle of Maximum Likelihood. To this end, a global optimization algorithm is utilized to solve the resulting non-convex optimization problem. It is shown that our approach improves the accuracy of the estimates, when compared with classic estimation techniques such as the prediction error method (PEM), in terms of covariance of the estimation error, while also obtaining an approximation of the noise distribution. The benefits of the proposed technique are illustrated by numerical simulations.

Original languageEnglish
Title of host publicationIEEE CHILEAN Conference on Electrical, Electronics Engineering, Information and Communication Technologies, CHILECON 2019
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781728131856
DOIs
StatePublished - Nov 2019
Externally publishedYes
Event2019 IEEE CHILEAN Conference on Electrical, Electronics Engineering, Information and Communication Technologies, CHILECON 2019 - Valparaiso, Chile
Duration: 13 Nov 201927 Nov 2019

Publication series

NameIEEE CHILEAN Conference on Electrical, Electronics Engineering, Information and Communication Technologies, CHILECON 2019

Conference

Conference2019 IEEE CHILEAN Conference on Electrical, Electronics Engineering, Information and Communication Technologies, CHILECON 2019
Country/TerritoryChile
CityValparaiso
Period13/11/1927/11/19

Keywords

  • Gaussian Mixture Model
  • Linear Dynamical Systems
  • Maximum Likelihood
  • Non-Gaussian Noise Distribution

Fingerprint

Dive into the research topics of 'Maximum Likelihood identification for Linear Dynamic Systems with finite Gaussian mixture noise distribution'. Together they form a unique fingerprint.

Cite this