State splitting and merging in probabilistic finite state automata for signal representation and analysis

Kushal Mukherjee, Asok Ray

Research output: Contribution to journalArticle

58 Citations (Scopus)

Abstract

Probabilistic finite state automata (PFSA) are often constructed from symbol strings that, in turn, are generated by partitioning time series of sensor signals. This paper focuses on a special class of PFSA, which captures finite history of the symbol strings; these PFSA, called D-Markov machines, have a simple algebraic structure and are computationally efficient to construct and implement. The procedure of PFSA construction is based on (i) state splitting that generates symbol blocks of different lengths based on their information contents; and (ii) state merging that assimilates histories by combining two or more symbol blocks without any significant loss of the embedded information. A metric on the probability distribution of symbol blocks is introduced for trade-off between loss of information (e.g., entropy rate) and the number of PFSA states. The underlying algorithms have been validated with three test examples. While the first and second examples elucidate the key concepts and the pertinent numerical steps, the third example presents the results of analysis of time series data, generated from laboratory experimentation, for detection of fatigue crack damage in a polycrystalline alloy.

Original languageEnglish (US)
Pages (from-to)105-119
Number of pages15
JournalSignal Processing
Volume104
DOIs
StatePublished - Jan 1 2014

Fingerprint

Finite automata
Merging
Time series
Probability distributions
Entropy
History
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

@article{0468ffa9569b40eb8807f8de0d5240de,
title = "State splitting and merging in probabilistic finite state automata for signal representation and analysis",
abstract = "Probabilistic finite state automata (PFSA) are often constructed from symbol strings that, in turn, are generated by partitioning time series of sensor signals. This paper focuses on a special class of PFSA, which captures finite history of the symbol strings; these PFSA, called D-Markov machines, have a simple algebraic structure and are computationally efficient to construct and implement. The procedure of PFSA construction is based on (i) state splitting that generates symbol blocks of different lengths based on their information contents; and (ii) state merging that assimilates histories by combining two or more symbol blocks without any significant loss of the embedded information. A metric on the probability distribution of symbol blocks is introduced for trade-off between loss of information (e.g., entropy rate) and the number of PFSA states. The underlying algorithms have been validated with three test examples. While the first and second examples elucidate the key concepts and the pertinent numerical steps, the third example presents the results of analysis of time series data, generated from laboratory experimentation, for detection of fatigue crack damage in a polycrystalline alloy.",
author = "Kushal Mukherjee and Asok Ray",
year = "2014",
month = "1",
day = "1",
doi = "10.1016/j.sigpro.2014.03.045",
language = "English (US)",
volume = "104",
pages = "105--119",
journal = "Signal Processing",
issn = "0165-1684",
publisher = "Elsevier",

}

State splitting and merging in probabilistic finite state automata for signal representation and analysis. / Mukherjee, Kushal; Ray, Asok.

In: Signal Processing, Vol. 104, 01.01.2014, p. 105-119.

Research output: Contribution to journalArticle

TY - JOUR

T1 - State splitting and merging in probabilistic finite state automata for signal representation and analysis

AU - Mukherjee, Kushal

AU - Ray, Asok

PY - 2014/1/1

Y1 - 2014/1/1

N2 - Probabilistic finite state automata (PFSA) are often constructed from symbol strings that, in turn, are generated by partitioning time series of sensor signals. This paper focuses on a special class of PFSA, which captures finite history of the symbol strings; these PFSA, called D-Markov machines, have a simple algebraic structure and are computationally efficient to construct and implement. The procedure of PFSA construction is based on (i) state splitting that generates symbol blocks of different lengths based on their information contents; and (ii) state merging that assimilates histories by combining two or more symbol blocks without any significant loss of the embedded information. A metric on the probability distribution of symbol blocks is introduced for trade-off between loss of information (e.g., entropy rate) and the number of PFSA states. The underlying algorithms have been validated with three test examples. While the first and second examples elucidate the key concepts and the pertinent numerical steps, the third example presents the results of analysis of time series data, generated from laboratory experimentation, for detection of fatigue crack damage in a polycrystalline alloy.

AB - Probabilistic finite state automata (PFSA) are often constructed from symbol strings that, in turn, are generated by partitioning time series of sensor signals. This paper focuses on a special class of PFSA, which captures finite history of the symbol strings; these PFSA, called D-Markov machines, have a simple algebraic structure and are computationally efficient to construct and implement. The procedure of PFSA construction is based on (i) state splitting that generates symbol blocks of different lengths based on their information contents; and (ii) state merging that assimilates histories by combining two or more symbol blocks without any significant loss of the embedded information. A metric on the probability distribution of symbol blocks is introduced for trade-off between loss of information (e.g., entropy rate) and the number of PFSA states. The underlying algorithms have been validated with three test examples. While the first and second examples elucidate the key concepts and the pertinent numerical steps, the third example presents the results of analysis of time series data, generated from laboratory experimentation, for detection of fatigue crack damage in a polycrystalline alloy.

UR - http://www.scopus.com/inward/record.url?scp=84899660625&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84899660625&partnerID=8YFLogxK

U2 - 10.1016/j.sigpro.2014.03.045

DO - 10.1016/j.sigpro.2014.03.045

M3 - Article

AN - SCOPUS:84899660625

VL - 104

SP - 105

EP - 119

JO - Signal Processing

JF - Signal Processing

SN - 0165-1684

ER -