EM-based identification of sparse FIR systems having quantized data

Rodrigo Carvajal; Juan C. Agüero; Boris I. Godoy; Graham C. Goodwin; Juan I. Yuz

doi:10.3182/20120711-3-BE-2027.00253

EM-based identification of sparse FIR systems having quantized data

Rodrigo Carvajal, Juan C. Agüero, Boris I. Godoy, Graham C. Goodwin, Juan I. Yuz

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

2 Scopus citations

Abstract

In this paper, we explore the identification of sparse FIR systems having quantized output data. Our approach is based on the use of regularization. We explore several aspects concerning the implementation of the Expectation-Maximization (EM) algorithm, including: i) a general framework, based on mean-variance Gaussian mixtures, for incorporating a regularization term that forces sparsity, ii) utilization of Markov Chain Monte Carlo techniques (namely a Gibbs sampler) and scenarios to implement the EM algorithm for multiple input multiple output systems. We show that for single input single output systems, it is possible to obtain closed form expressions for solving the EM algorithm.

Original language	English
Title of host publication	SYSID 2012 - 16th IFAC Symposium on System Identification, Final Program
Publisher	IFAC Secretariat
Pages	553-558
Number of pages	6
Edition	PART 1
ISBN (Print)	9783902823069
DOIs	https://doi.org/10.3182/20120711-3-BE-2027.00253
State	Published - 2012
Externally published	Yes
Event	Universite Libre de Bruxelles - Bruxelles, Belgium Duration: 11 Jul 2012 → 13 Jul 2012

Publication series

Name	IFAC Proceedings Volumes (IFAC-PapersOnline)
Number	PART 1
Volume	16
ISSN (Print)	1474-6670

Conference

Conference	Universite Libre de Bruxelles
Country/Territory	Belgium
City	Bruxelles
Period	11/07/12 → 13/07/12

Keywords

EM algorithm
Quantized estimation
Regularization
Sparse optimization

Access to Document

10.3182/20120711-3-BE-2027.00253

Cite this

Carvajal, R., Agüero, J. C., Godoy, B. I., Goodwin, G. C., & Yuz, J. I. (2012). EM-based identification of sparse FIR systems having quantized data. In SYSID 2012 - 16th IFAC Symposium on System Identification, Final Program (PART 1 ed., pp. 553-558). (IFAC Proceedings Volumes (IFAC-PapersOnline); Vol. 16, No. PART 1). IFAC Secretariat. https://doi.org/10.3182/20120711-3-BE-2027.00253

@inproceedings{84bf6a72fde64730aa7373ea0d675334,

title = "EM-based identification of sparse FIR systems having quantized data",

abstract = "In this paper, we explore the identification of sparse FIR systems having quantized output data. Our approach is based on the use of regularization. We explore several aspects concerning the implementation of the Expectation-Maximization (EM) algorithm, including: i) a general framework, based on mean-variance Gaussian mixtures, for incorporating a regularization term that forces sparsity, ii) utilization of Markov Chain Monte Carlo techniques (namely a Gibbs sampler) and scenarios to implement the EM algorithm for multiple input multiple output systems. We show that for single input single output systems, it is possible to obtain closed form expressions for solving the EM algorithm.",

keywords = "EM algorithm, Quantized estimation, Regularization, Sparse optimization",

author = "Rodrigo Carvajal and Ag{\"u}ero, {Juan C.} and Godoy, {Boris I.} and Goodwin, {Graham C.} and Yuz, {Juan I.}",

note = "Funding Information: 1 Boris I. Godoy acknowledges support given by CONICYT - Chile through its Postdoctoral Fellowship Program 2011.; Universite Libre de Bruxelles ; Conference date: 11-07-2012 Through 13-07-2012",

year = "2012",

doi = "10.3182/20120711-3-BE-2027.00253",

language = "English",

isbn = "9783902823069",

series = "IFAC Proceedings Volumes (IFAC-PapersOnline)",

publisher = "IFAC Secretariat",

number = "PART 1",

pages = "553--558",

booktitle = "SYSID 2012 - 16th IFAC Symposium on System Identification, Final Program",

edition = "PART 1",

}

Carvajal, R, Agüero, JC, Godoy, BI, Goodwin, GC & Yuz, JI 2012, EM-based identification of sparse FIR systems having quantized data. in SYSID 2012 - 16th IFAC Symposium on System Identification, Final Program. PART 1 edn, IFAC Proceedings Volumes (IFAC-PapersOnline), no. PART 1, vol. 16, IFAC Secretariat, pp. 553-558, Universite Libre de Bruxelles, Bruxelles, Belgium, 11/07/12. https://doi.org/10.3182/20120711-3-BE-2027.00253

EM-based identification of sparse FIR systems having quantized data. / Carvajal, Rodrigo; Agüero, Juan C.; Godoy, Boris I. et al.
SYSID 2012 - 16th IFAC Symposium on System Identification, Final Program. PART 1. ed. IFAC Secretariat, 2012. p. 553-558 (IFAC Proceedings Volumes (IFAC-PapersOnline); Vol. 16, No. PART 1).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - EM-based identification of sparse FIR systems having quantized data

AU - Carvajal, Rodrigo

AU - Agüero, Juan C.

AU - Godoy, Boris I.

AU - Goodwin, Graham C.

AU - Yuz, Juan I.

N1 - Funding Information: 1 Boris I. Godoy acknowledges support given by CONICYT - Chile through its Postdoctoral Fellowship Program 2011.

PY - 2012

Y1 - 2012

N2 - In this paper, we explore the identification of sparse FIR systems having quantized output data. Our approach is based on the use of regularization. We explore several aspects concerning the implementation of the Expectation-Maximization (EM) algorithm, including: i) a general framework, based on mean-variance Gaussian mixtures, for incorporating a regularization term that forces sparsity, ii) utilization of Markov Chain Monte Carlo techniques (namely a Gibbs sampler) and scenarios to implement the EM algorithm for multiple input multiple output systems. We show that for single input single output systems, it is possible to obtain closed form expressions for solving the EM algorithm.

AB - In this paper, we explore the identification of sparse FIR systems having quantized output data. Our approach is based on the use of regularization. We explore several aspects concerning the implementation of the Expectation-Maximization (EM) algorithm, including: i) a general framework, based on mean-variance Gaussian mixtures, for incorporating a regularization term that forces sparsity, ii) utilization of Markov Chain Monte Carlo techniques (namely a Gibbs sampler) and scenarios to implement the EM algorithm for multiple input multiple output systems. We show that for single input single output systems, it is possible to obtain closed form expressions for solving the EM algorithm.

KW - EM algorithm

KW - Quantized estimation

KW - Regularization

KW - Sparse optimization

UR - http://www.scopus.com/inward/record.url?scp=84867056138&partnerID=8YFLogxK

U2 - 10.3182/20120711-3-BE-2027.00253

DO - 10.3182/20120711-3-BE-2027.00253

M3 - Conference contribution

AN - SCOPUS:84867056138

SN - 9783902823069

T3 - IFAC Proceedings Volumes (IFAC-PapersOnline)

SP - 553

EP - 558

BT - SYSID 2012 - 16th IFAC Symposium on System Identification, Final Program

PB - IFAC Secretariat

T2 - Universite Libre de Bruxelles

Y2 - 11 July 2012 through 13 July 2012

ER -

EM-based identification of sparse FIR systems having quantized data

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this