Statistical estimation of multiple parameters via symbolic dynamic filtering

Chinmay Rao, Kushal Mukherjee, Soumik Sarkar, Asok Ray

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

This paper addresses statistical estimation of multiple parameters that may vary simultaneously but slowly relative to the process response in nonlinear dynamical systems. The estimation algorithm is sensor-data-driven and is built upon the concept of symbolic dynamic filtering for real-time execution on limited-memory platforms, such as local nodes in a sensor network. In this approach, the behavior patterns of the dynamical system are compactly generated as quasi-stationary probability vectors associated with the finite-state automata for symbolic dynamic representation. The estimation algorithm is validated on nonlinear electronic circuits that represent externally excited Duffing and unforced van der Pol systems. Confidence intervals are obtained for statistical estimation of two parameters in each of the systems.

Original languageEnglish (US)
Pages (from-to)981-988
Number of pages8
JournalSignal Processing
Volume89
Issue number6
DOIs
StatePublished - Jun 1 2009

Fingerprint

Nonlinear dynamical systems
Finite automata
Sensor networks
Dynamical systems
Data storage equipment
Networks (circuits)
Sensors

All Science Journal Classification (ASJC) codes

  • Control and Systems Engineering
  • Software
  • Signal Processing
  • Computer Vision and Pattern Recognition
  • Electrical and Electronic Engineering

Cite this

Rao, Chinmay ; Mukherjee, Kushal ; Sarkar, Soumik ; Ray, Asok. / Statistical estimation of multiple parameters via symbolic dynamic filtering. In: Signal Processing. 2009 ; Vol. 89, No. 6. pp. 981-988.
@article{8f993fb3c47545328534226c1a906392,
title = "Statistical estimation of multiple parameters via symbolic dynamic filtering",
abstract = "This paper addresses statistical estimation of multiple parameters that may vary simultaneously but slowly relative to the process response in nonlinear dynamical systems. The estimation algorithm is sensor-data-driven and is built upon the concept of symbolic dynamic filtering for real-time execution on limited-memory platforms, such as local nodes in a sensor network. In this approach, the behavior patterns of the dynamical system are compactly generated as quasi-stationary probability vectors associated with the finite-state automata for symbolic dynamic representation. The estimation algorithm is validated on nonlinear electronic circuits that represent externally excited Duffing and unforced van der Pol systems. Confidence intervals are obtained for statistical estimation of two parameters in each of the systems.",
author = "Chinmay Rao and Kushal Mukherjee and Soumik Sarkar and Asok Ray",
year = "2009",
month = "6",
day = "1",
doi = "10.1016/j.sigpro.2008.11.018",
language = "English (US)",
volume = "89",
pages = "981--988",
journal = "Signal Processing",
issn = "0165-1684",
publisher = "Elsevier",
number = "6",

}

Statistical estimation of multiple parameters via symbolic dynamic filtering. / Rao, Chinmay; Mukherjee, Kushal; Sarkar, Soumik; Ray, Asok.

In: Signal Processing, Vol. 89, No. 6, 01.06.2009, p. 981-988.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Statistical estimation of multiple parameters via symbolic dynamic filtering

AU - Rao, Chinmay

AU - Mukherjee, Kushal

AU - Sarkar, Soumik

AU - Ray, Asok

PY - 2009/6/1

Y1 - 2009/6/1

N2 - This paper addresses statistical estimation of multiple parameters that may vary simultaneously but slowly relative to the process response in nonlinear dynamical systems. The estimation algorithm is sensor-data-driven and is built upon the concept of symbolic dynamic filtering for real-time execution on limited-memory platforms, such as local nodes in a sensor network. In this approach, the behavior patterns of the dynamical system are compactly generated as quasi-stationary probability vectors associated with the finite-state automata for symbolic dynamic representation. The estimation algorithm is validated on nonlinear electronic circuits that represent externally excited Duffing and unforced van der Pol systems. Confidence intervals are obtained for statistical estimation of two parameters in each of the systems.

AB - This paper addresses statistical estimation of multiple parameters that may vary simultaneously but slowly relative to the process response in nonlinear dynamical systems. The estimation algorithm is sensor-data-driven and is built upon the concept of symbolic dynamic filtering for real-time execution on limited-memory platforms, such as local nodes in a sensor network. In this approach, the behavior patterns of the dynamical system are compactly generated as quasi-stationary probability vectors associated with the finite-state automata for symbolic dynamic representation. The estimation algorithm is validated on nonlinear electronic circuits that represent externally excited Duffing and unforced van der Pol systems. Confidence intervals are obtained for statistical estimation of two parameters in each of the systems.

UR - http://www.scopus.com/inward/record.url?scp=60749098242&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=60749098242&partnerID=8YFLogxK

U2 - 10.1016/j.sigpro.2008.11.018

DO - 10.1016/j.sigpro.2008.11.018

M3 - Article

AN - SCOPUS:60749098242

VL - 89

SP - 981

EP - 988

JO - Signal Processing

JF - Signal Processing

SN - 0165-1684

IS - 6

ER -