Statistical models of reconstructed phase spaces for signal classification

Richard J. Povinelli; Michael T. Johnson; Andrew C. Lindgren; Felice M. Roberts; Jinjin Ye

doi:10.1109/TSP.2006.873479

Statistical models of reconstructed phase spaces for signal classification

Richard J. Povinelli, Michael T. Johnson, Andrew C. Lindgren, Felice M. Roberts, Jinjin Ye

Research output: Contribution to journal › Article › peer-review

68 Scopus citations

Abstract

This paper introduces a novel approach to the analysis and classification of time series signals using statistical models of reconstructed phase spaces. With sufficient dimension, such reconstructed phase spaces are, with probability one, guaranteed to be topologically equivalent to the state dynamics of the generating system, and, therefore, may contain information that is absent in analysis and classification methods rooted in linear assumptions. Parametric and nonparametric distributions are introduced as statistical representations over the multidimensional reconstructed phase space, with classification accomplished through methods such as Bayes maximum likelihood and artificial neural networks (ANNs). The technique is demonstrated on heart arrhythmia classification and speech recognition. This new approach is shown to be a viable and effective alternative to traditional signal classification approaches, particularly for signals with strong nonlinear characteristics.

Original language	English
Pages (from-to)	2178-2186
Number of pages	9
Journal	IEEE Transactions on Signal Processing
Volume	54
Issue number	6 I
DOIs	https://doi.org/10.1109/TSP.2006.873479
State	Published - Jun 2006

Bibliographical note

Funding Information:
Manuscript received July 22, 2004; revised July 11, 2005. This work was supported in part by the National Science Foundation by Grant IIS-0113508, by the Department of Education GAANN Fellowship, by an American Heart Association Fellowship, and by the Milwaukee Foundation Frank Rogers Bacon Research Assistantship. The associate editor coordinating the review of this manuscript and approving it for publication was Prof. Tulay Adali.

Keywords

Reconstructed phase spaces (RPSs)
Signal classification
Statistical models

ASJC Scopus subject areas

Signal Processing
Electrical and Electronic Engineering

Access to Document

10.1109/TSP.2006.873479

Cite this

@article{95645007c23c427095282db2c198b74d,

title = "Statistical models of reconstructed phase spaces for signal classification",

abstract = "This paper introduces a novel approach to the analysis and classification of time series signals using statistical models of reconstructed phase spaces. With sufficient dimension, such reconstructed phase spaces are, with probability one, guaranteed to be topologically equivalent to the state dynamics of the generating system, and, therefore, may contain information that is absent in analysis and classification methods rooted in linear assumptions. Parametric and nonparametric distributions are introduced as statistical representations over the multidimensional reconstructed phase space, with classification accomplished through methods such as Bayes maximum likelihood and artificial neural networks (ANNs). The technique is demonstrated on heart arrhythmia classification and speech recognition. This new approach is shown to be a viable and effective alternative to traditional signal classification approaches, particularly for signals with strong nonlinear characteristics.",

keywords = "Reconstructed phase spaces (RPSs), Signal classification, Statistical models",

author = "Povinelli, {Richard J.} and Johnson, {Michael T.} and Lindgren, {Andrew C.} and Roberts, {Felice M.} and Jinjin Ye",

note = "Funding Information: Manuscript received July 22, 2004; revised July 11, 2005. This work was supported in part by the National Science Foundation by Grant IIS-0113508, by the Department of Education GAANN Fellowship, by an American Heart Association Fellowship, and by the Milwaukee Foundation Frank Rogers Bacon Research Assistantship. The associate editor coordinating the review of this manuscript and approving it for publication was Prof. Tulay Adali.",

year = "2006",

month = jun,

doi = "10.1109/TSP.2006.873479",

language = "English",

volume = "54",

pages = "2178--2186",

number = "6 I",

}

TY - JOUR

T1 - Statistical models of reconstructed phase spaces for signal classification

AU - Povinelli, Richard J.

AU - Johnson, Michael T.

AU - Lindgren, Andrew C.

AU - Roberts, Felice M.

AU - Ye, Jinjin

N1 - Funding Information: Manuscript received July 22, 2004; revised July 11, 2005. This work was supported in part by the National Science Foundation by Grant IIS-0113508, by the Department of Education GAANN Fellowship, by an American Heart Association Fellowship, and by the Milwaukee Foundation Frank Rogers Bacon Research Assistantship. The associate editor coordinating the review of this manuscript and approving it for publication was Prof. Tulay Adali.

PY - 2006/6

Y1 - 2006/6

N2 - This paper introduces a novel approach to the analysis and classification of time series signals using statistical models of reconstructed phase spaces. With sufficient dimension, such reconstructed phase spaces are, with probability one, guaranteed to be topologically equivalent to the state dynamics of the generating system, and, therefore, may contain information that is absent in analysis and classification methods rooted in linear assumptions. Parametric and nonparametric distributions are introduced as statistical representations over the multidimensional reconstructed phase space, with classification accomplished through methods such as Bayes maximum likelihood and artificial neural networks (ANNs). The technique is demonstrated on heart arrhythmia classification and speech recognition. This new approach is shown to be a viable and effective alternative to traditional signal classification approaches, particularly for signals with strong nonlinear characteristics.

AB - This paper introduces a novel approach to the analysis and classification of time series signals using statistical models of reconstructed phase spaces. With sufficient dimension, such reconstructed phase spaces are, with probability one, guaranteed to be topologically equivalent to the state dynamics of the generating system, and, therefore, may contain information that is absent in analysis and classification methods rooted in linear assumptions. Parametric and nonparametric distributions are introduced as statistical representations over the multidimensional reconstructed phase space, with classification accomplished through methods such as Bayes maximum likelihood and artificial neural networks (ANNs). The technique is demonstrated on heart arrhythmia classification and speech recognition. This new approach is shown to be a viable and effective alternative to traditional signal classification approaches, particularly for signals with strong nonlinear characteristics.

KW - Reconstructed phase spaces (RPSs)

KW - Signal classification

KW - Statistical models

UR - http://www.scopus.com/inward/record.url?scp=33744520854&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33744520854&partnerID=8YFLogxK

U2 - 10.1109/TSP.2006.873479

DO - 10.1109/TSP.2006.873479

M3 - Article

AN - SCOPUS:33744520854

SN - 1053-587X

VL - 54

SP - 2178

EP - 2186

JO - IEEE Transactions on Signal Processing

JF - IEEE Transactions on Signal Processing

IS - 6 I

ER -

Statistical models of reconstructed phase spaces for signal classification

Abstract

Bibliographical note

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this