Approximate maximum-likelihood identification of linear systems from quantized measurements

Riccardo Sven Risuleo, Giulio Bottegal, Håkan Hjalmarsson

Research output: Contribution to journalConference articleAcademicpeer-review

Abstract

We analyze likelihood-based identification of systems that are linear in the parameters from quantized output data; in particular, we propose a method to find approximate maximum-likelihood and maximum-a-posteriori solutions. The method consists of appropriate least-squares projections of the middle point of the active quantization intervals. We show that this approximation maximizes a variational approximation of the likelihood and we provide an upper bound for the approximation error. In a simulation study, we compare the proposed method with the true maximum-likelihood estimate of a finite impulse response model.

LanguageEnglish
Pages724-729
Number of pages6
JournalIFAC-PapersOnLine
Volume51
Issue number15
DOIs
StatePublished - 1 Jan 2018
Event18th IFAC Symposium on System Identification (SYSID 2018) - Stockholm, Sweden
Duration: 9 Jul 201811 Jul 2018

Fingerprint

Maximum likelihood
Linear systems
Identification (control systems)
Impulse response

Keywords

  • Least-squares approximation
  • Maximum-likelihood estimators
  • quantized signals

Cite this

Risuleo, Riccardo Sven ; Bottegal, Giulio ; Hjalmarsson, Håkan. / Approximate maximum-likelihood identification of linear systems from quantized measurements. In: IFAC-PapersOnLine. 2018 ; Vol. 51, No. 15. pp. 724-729
@article{b151d1c7a8a54a469795338d4be8c600,
title = "Approximate maximum-likelihood identification of linear systems from quantized measurements⁎",
abstract = "We analyze likelihood-based identification of systems that are linear in the parameters from quantized output data; in particular, we propose a method to find approximate maximum-likelihood and maximum-a-posteriori solutions. The method consists of appropriate least-squares projections of the middle point of the active quantization intervals. We show that this approximation maximizes a variational approximation of the likelihood and we provide an upper bound for the approximation error. In a simulation study, we compare the proposed method with the true maximum-likelihood estimate of a finite impulse response model.",
keywords = "Least-squares approximation, Maximum-likelihood estimators, quantized signals",
author = "Risuleo, {Riccardo Sven} and Giulio Bottegal and H{\aa}kan Hjalmarsson",
year = "2018",
month = "1",
day = "1",
doi = "10.1016/j.ifacol.2018.09.169",
language = "English",
volume = "51",
pages = "724--729",
journal = "IFAC-PapersOnLine",
issn = "2405-8963",
publisher = "Elsevier",
number = "15",

}

Approximate maximum-likelihood identification of linear systems from quantized measurements. / Risuleo, Riccardo Sven; Bottegal, Giulio; Hjalmarsson, Håkan.

In: IFAC-PapersOnLine, Vol. 51, No. 15, 01.01.2018, p. 724-729.

Research output: Contribution to journalConference articleAcademicpeer-review

TY - JOUR

T1 - Approximate maximum-likelihood identification of linear systems from quantized measurements⁎

AU - Risuleo,Riccardo Sven

AU - Bottegal,Giulio

AU - Hjalmarsson,Håkan

PY - 2018/1/1

Y1 - 2018/1/1

N2 - We analyze likelihood-based identification of systems that are linear in the parameters from quantized output data; in particular, we propose a method to find approximate maximum-likelihood and maximum-a-posteriori solutions. The method consists of appropriate least-squares projections of the middle point of the active quantization intervals. We show that this approximation maximizes a variational approximation of the likelihood and we provide an upper bound for the approximation error. In a simulation study, we compare the proposed method with the true maximum-likelihood estimate of a finite impulse response model.

AB - We analyze likelihood-based identification of systems that are linear in the parameters from quantized output data; in particular, we propose a method to find approximate maximum-likelihood and maximum-a-posteriori solutions. The method consists of appropriate least-squares projections of the middle point of the active quantization intervals. We show that this approximation maximizes a variational approximation of the likelihood and we provide an upper bound for the approximation error. In a simulation study, we compare the proposed method with the true maximum-likelihood estimate of a finite impulse response model.

KW - Least-squares approximation

KW - Maximum-likelihood estimators

KW - quantized signals

UR - http://www.scopus.com/inward/record.url?scp=85054371213&partnerID=8YFLogxK

U2 - 10.1016/j.ifacol.2018.09.169

DO - 10.1016/j.ifacol.2018.09.169

M3 - Conference article

VL - 51

SP - 724

EP - 729

JO - IFAC-PapersOnLine

T2 - IFAC-PapersOnLine

JF - IFAC-PapersOnLine

SN - 2405-8963

IS - 15

ER -