Segmentation and identification of drifting dynamical systems

J. Kohlmorgen, Klaus Muller, K. Pawelzik

Research output: Chapter in Book/Report/Conference proceedingConference contribution

10 Citations (Scopus)

Abstract

A method for the analysis of nonstationary time series with multiple operating modes is presented. In particular, it is possible to detect and to model a switching of the dynamics and also a less abrupt, time consuming drift from one mode to another. This is achieved by an unsupervised algorithm that segments the data according to inherent modes, and a subsequent search through the space of possible drifts. An application to physiological wake/sleep data demonstrates that analysis and modeling of real-world time series can be improved when the drift paradigm is taken into account. In the case of wake/sleep data, we hope to gain more insight into the physiological processes that are involved in the transition from wake to sleep.

Original languageEnglish
Title of host publicationNeural Networks for Signal Processing - Proceedings of the IEEE Workshop
Place of PublicationPiscataway, NJ, United States
PublisherIEEE
Pages326-335
Number of pages10
Publication statusPublished - 1997 Dec 1
Externally publishedYes
EventProceedings of the 1997 7th IEEE Workshop on Neural Networks for Signal Processing, NNSP'97 - Amelia Island, FL, USA
Duration: 1997 Sep 241997 Sep 26

Other

OtherProceedings of the 1997 7th IEEE Workshop on Neural Networks for Signal Processing, NNSP'97
CityAmelia Island, FL, USA
Period97/9/2497/9/26

Fingerprint

Dynamical systems
Time series
Sleep

ASJC Scopus subject areas

  • Signal Processing
  • Software
  • Electrical and Electronic Engineering

Cite this

Kohlmorgen, J., Muller, K., & Pawelzik, K. (1997). Segmentation and identification of drifting dynamical systems. In Neural Networks for Signal Processing - Proceedings of the IEEE Workshop (pp. 326-335). Piscataway, NJ, United States: IEEE.

Segmentation and identification of drifting dynamical systems. / Kohlmorgen, J.; Muller, Klaus; Pawelzik, K.

Neural Networks for Signal Processing - Proceedings of the IEEE Workshop. Piscataway, NJ, United States : IEEE, 1997. p. 326-335.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Kohlmorgen, J, Muller, K & Pawelzik, K 1997, Segmentation and identification of drifting dynamical systems. in Neural Networks for Signal Processing - Proceedings of the IEEE Workshop. IEEE, Piscataway, NJ, United States, pp. 326-335, Proceedings of the 1997 7th IEEE Workshop on Neural Networks for Signal Processing, NNSP'97, Amelia Island, FL, USA, 97/9/24.
Kohlmorgen J, Muller K, Pawelzik K. Segmentation and identification of drifting dynamical systems. In Neural Networks for Signal Processing - Proceedings of the IEEE Workshop. Piscataway, NJ, United States: IEEE. 1997. p. 326-335
Kohlmorgen, J. ; Muller, Klaus ; Pawelzik, K. / Segmentation and identification of drifting dynamical systems. Neural Networks for Signal Processing - Proceedings of the IEEE Workshop. Piscataway, NJ, United States : IEEE, 1997. pp. 326-335
@inproceedings{fdabb248777d455baecb8bf20fcd53ef,
title = "Segmentation and identification of drifting dynamical systems",
abstract = "A method for the analysis of nonstationary time series with multiple operating modes is presented. In particular, it is possible to detect and to model a switching of the dynamics and also a less abrupt, time consuming drift from one mode to another. This is achieved by an unsupervised algorithm that segments the data according to inherent modes, and a subsequent search through the space of possible drifts. An application to physiological wake/sleep data demonstrates that analysis and modeling of real-world time series can be improved when the drift paradigm is taken into account. In the case of wake/sleep data, we hope to gain more insight into the physiological processes that are involved in the transition from wake to sleep.",
author = "J. Kohlmorgen and Klaus Muller and K. Pawelzik",
year = "1997",
month = "12",
day = "1",
language = "English",
pages = "326--335",
booktitle = "Neural Networks for Signal Processing - Proceedings of the IEEE Workshop",
publisher = "IEEE",

}

TY - GEN

T1 - Segmentation and identification of drifting dynamical systems

AU - Kohlmorgen, J.

AU - Muller, Klaus

AU - Pawelzik, K.

PY - 1997/12/1

Y1 - 1997/12/1

N2 - A method for the analysis of nonstationary time series with multiple operating modes is presented. In particular, it is possible to detect and to model a switching of the dynamics and also a less abrupt, time consuming drift from one mode to another. This is achieved by an unsupervised algorithm that segments the data according to inherent modes, and a subsequent search through the space of possible drifts. An application to physiological wake/sleep data demonstrates that analysis and modeling of real-world time series can be improved when the drift paradigm is taken into account. In the case of wake/sleep data, we hope to gain more insight into the physiological processes that are involved in the transition from wake to sleep.

AB - A method for the analysis of nonstationary time series with multiple operating modes is presented. In particular, it is possible to detect and to model a switching of the dynamics and also a less abrupt, time consuming drift from one mode to another. This is achieved by an unsupervised algorithm that segments the data according to inherent modes, and a subsequent search through the space of possible drifts. An application to physiological wake/sleep data demonstrates that analysis and modeling of real-world time series can be improved when the drift paradigm is taken into account. In the case of wake/sleep data, we hope to gain more insight into the physiological processes that are involved in the transition from wake to sleep.

UR - http://www.scopus.com/inward/record.url?scp=0031333231&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0031333231&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:0031333231

SP - 326

EP - 335

BT - Neural Networks for Signal Processing - Proceedings of the IEEE Workshop

PB - IEEE

CY - Piscataway, NJ, United States

ER -