Symbolization of dynamic data-driven systems for signal representation

Soumalya Sarkar, Pritthi Chattopdhyay, Asok Ray

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

The underlying theory of symbolic time series analysis (STSA) has led to the development of signal representation tools in the paradigm of dynamic data-driven application systems (DDDAS), where time series of sensor signals are partitioned to obtain symbol strings that, in turn, lead to the construction of probabilistic finite state automata (PFSA). Although various methods for construction of PFSA from symbol strings have been reported in literature, similar efforts have not been expended on identification of an appropriate alphabet size for partitioning of time series, so that the symbol strings can be optimally or suboptimally generated in a specified sense. The paper addresses this critical issue and proposes an information-theoretic procedure for partitioning of time series to extract low-dimensional features, where the key idea is suboptimal identification of boundary locations of the partitioning segments via maximization of the mutual information between the state probability vector of PFSA and the members of the pattern classes. Robustness of the symbolization process has also been addressed. The proposed alphabet size selection and time series partitioning algorithm have been validated by two examples. The first example addresses parameter identification in a simulated Duffing system with sinusoidal input excitation. The second example is built upon an ensemble of time series of chemiluminescence data to predict lean blowout (LBO) phenomena in a laboratory-scale swirl-stabilized combustor apparatus.

Original languageEnglish (US)
Pages (from-to)1535-1542
Number of pages8
JournalSignal, Image and Video Processing
Volume10
Issue number8
DOIs
StatePublished - Nov 1 2016

Fingerprint

Time series
Finite automata
Chemiluminescence
Time series analysis
Combustors
Identification (control systems)
Sensors

All Science Journal Classification (ASJC) codes

  • Signal Processing
  • Electrical and Electronic Engineering

Cite this

Sarkar, Soumalya ; Chattopdhyay, Pritthi ; Ray, Asok. / Symbolization of dynamic data-driven systems for signal representation. In: Signal, Image and Video Processing. 2016 ; Vol. 10, No. 8. pp. 1535-1542.
@article{bd24b1dd937c4bb59590ebe12e6add14,
title = "Symbolization of dynamic data-driven systems for signal representation",
abstract = "The underlying theory of symbolic time series analysis (STSA) has led to the development of signal representation tools in the paradigm of dynamic data-driven application systems (DDDAS), where time series of sensor signals are partitioned to obtain symbol strings that, in turn, lead to the construction of probabilistic finite state automata (PFSA). Although various methods for construction of PFSA from symbol strings have been reported in literature, similar efforts have not been expended on identification of an appropriate alphabet size for partitioning of time series, so that the symbol strings can be optimally or suboptimally generated in a specified sense. The paper addresses this critical issue and proposes an information-theoretic procedure for partitioning of time series to extract low-dimensional features, where the key idea is suboptimal identification of boundary locations of the partitioning segments via maximization of the mutual information between the state probability vector of PFSA and the members of the pattern classes. Robustness of the symbolization process has also been addressed. The proposed alphabet size selection and time series partitioning algorithm have been validated by two examples. The first example addresses parameter identification in a simulated Duffing system with sinusoidal input excitation. The second example is built upon an ensemble of time series of chemiluminescence data to predict lean blowout (LBO) phenomena in a laboratory-scale swirl-stabilized combustor apparatus.",
author = "Soumalya Sarkar and Pritthi Chattopdhyay and Asok Ray",
year = "2016",
month = "11",
day = "1",
doi = "10.1007/s11760-016-0967-5",
language = "English (US)",
volume = "10",
pages = "1535--1542",
journal = "Signal, Image and Video Processing",
issn = "1863-1703",
publisher = "Springer London",
number = "8",

}

Symbolization of dynamic data-driven systems for signal representation. / Sarkar, Soumalya; Chattopdhyay, Pritthi; Ray, Asok.

In: Signal, Image and Video Processing, Vol. 10, No. 8, 01.11.2016, p. 1535-1542.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Symbolization of dynamic data-driven systems for signal representation

AU - Sarkar, Soumalya

AU - Chattopdhyay, Pritthi

AU - Ray, Asok

PY - 2016/11/1

Y1 - 2016/11/1

N2 - The underlying theory of symbolic time series analysis (STSA) has led to the development of signal representation tools in the paradigm of dynamic data-driven application systems (DDDAS), where time series of sensor signals are partitioned to obtain symbol strings that, in turn, lead to the construction of probabilistic finite state automata (PFSA). Although various methods for construction of PFSA from symbol strings have been reported in literature, similar efforts have not been expended on identification of an appropriate alphabet size for partitioning of time series, so that the symbol strings can be optimally or suboptimally generated in a specified sense. The paper addresses this critical issue and proposes an information-theoretic procedure for partitioning of time series to extract low-dimensional features, where the key idea is suboptimal identification of boundary locations of the partitioning segments via maximization of the mutual information between the state probability vector of PFSA and the members of the pattern classes. Robustness of the symbolization process has also been addressed. The proposed alphabet size selection and time series partitioning algorithm have been validated by two examples. The first example addresses parameter identification in a simulated Duffing system with sinusoidal input excitation. The second example is built upon an ensemble of time series of chemiluminescence data to predict lean blowout (LBO) phenomena in a laboratory-scale swirl-stabilized combustor apparatus.

AB - The underlying theory of symbolic time series analysis (STSA) has led to the development of signal representation tools in the paradigm of dynamic data-driven application systems (DDDAS), where time series of sensor signals are partitioned to obtain symbol strings that, in turn, lead to the construction of probabilistic finite state automata (PFSA). Although various methods for construction of PFSA from symbol strings have been reported in literature, similar efforts have not been expended on identification of an appropriate alphabet size for partitioning of time series, so that the symbol strings can be optimally or suboptimally generated in a specified sense. The paper addresses this critical issue and proposes an information-theoretic procedure for partitioning of time series to extract low-dimensional features, where the key idea is suboptimal identification of boundary locations of the partitioning segments via maximization of the mutual information between the state probability vector of PFSA and the members of the pattern classes. Robustness of the symbolization process has also been addressed. The proposed alphabet size selection and time series partitioning algorithm have been validated by two examples. The first example addresses parameter identification in a simulated Duffing system with sinusoidal input excitation. The second example is built upon an ensemble of time series of chemiluminescence data to predict lean blowout (LBO) phenomena in a laboratory-scale swirl-stabilized combustor apparatus.

UR - http://www.scopus.com/inward/record.url?scp=84982291990&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84982291990&partnerID=8YFLogxK

U2 - 10.1007/s11760-016-0967-5

DO - 10.1007/s11760-016-0967-5

M3 - Article

AN - SCOPUS:84982291990

VL - 10

SP - 1535

EP - 1542

JO - Signal, Image and Video Processing

JF - Signal, Image and Video Processing

SN - 1863-1703

IS - 8

ER -