Identifying interactions in mixed and noisy complex systems

Guido Nolte, Frank C. Meinecke, Andreas Ziehe, Klaus Muller

Research output: Contribution to journalArticle

35 Citations (Scopus)

Abstract

We present a technique that identifies truly interacting subsystems of a complex system from multichannel data if the recordings are an unknown linear and instantaneous mixture of the true sources. The method is valid for arbitrary noise structure. For this, a blind source separation technique is proposed that diagonalizes antisymmetrized cross-correlation or cross-spectral matrices. The resulting decomposition finds truly interacting subsystems blindly and suppresses any spurious interaction stemming from the mixture. The usefulness of this interacting source analysis is demonstrated in simulations and for real electroencephalography data.

Original languageEnglish
Article number051913
JournalPhysical Review E - Statistical, Nonlinear, and Soft Matter Physics
Volume73
Issue number5
DOIs
Publication statusPublished - 2006 Jun 1
Externally publishedYes

Fingerprint

complex systems
Complex Systems
Subsystem
Electroencephalography
Blind Source Separation
Cross-correlation
Interaction
Instantaneous
electroencephalography
interactions
Valid
Decompose
Unknown
cross correlation
Arbitrary
recording
decomposition
Simulation
matrices
simulation

ASJC Scopus subject areas

  • Physics and Astronomy(all)
  • Condensed Matter Physics
  • Statistical and Nonlinear Physics
  • Mathematical Physics

Cite this

Identifying interactions in mixed and noisy complex systems. / Nolte, Guido; Meinecke, Frank C.; Ziehe, Andreas; Muller, Klaus.

In: Physical Review E - Statistical, Nonlinear, and Soft Matter Physics, Vol. 73, No. 5, 051913, 01.06.2006.

Research output: Contribution to journalArticle

Nolte, Guido ; Meinecke, Frank C. ; Ziehe, Andreas ; Muller, Klaus. / Identifying interactions in mixed and noisy complex systems. In: Physical Review E - Statistical, Nonlinear, and Soft Matter Physics. 2006 ; Vol. 73, No. 5.
@article{e71b03c903cb422889890cd7ef45f4bd,
title = "Identifying interactions in mixed and noisy complex systems",
abstract = "We present a technique that identifies truly interacting subsystems of a complex system from multichannel data if the recordings are an unknown linear and instantaneous mixture of the true sources. The method is valid for arbitrary noise structure. For this, a blind source separation technique is proposed that diagonalizes antisymmetrized cross-correlation or cross-spectral matrices. The resulting decomposition finds truly interacting subsystems blindly and suppresses any spurious interaction stemming from the mixture. The usefulness of this interacting source analysis is demonstrated in simulations and for real electroencephalography data.",
author = "Guido Nolte and Meinecke, {Frank C.} and Andreas Ziehe and Klaus Muller",
year = "2006",
month = "6",
day = "1",
doi = "10.1103/PhysRevE.73.051913",
language = "English",
volume = "73",
journal = "Physical Review E",
issn = "2470-0045",
publisher = "American Physical Society",
number = "5",

}

TY - JOUR

T1 - Identifying interactions in mixed and noisy complex systems

AU - Nolte, Guido

AU - Meinecke, Frank C.

AU - Ziehe, Andreas

AU - Muller, Klaus

PY - 2006/6/1

Y1 - 2006/6/1

N2 - We present a technique that identifies truly interacting subsystems of a complex system from multichannel data if the recordings are an unknown linear and instantaneous mixture of the true sources. The method is valid for arbitrary noise structure. For this, a blind source separation technique is proposed that diagonalizes antisymmetrized cross-correlation or cross-spectral matrices. The resulting decomposition finds truly interacting subsystems blindly and suppresses any spurious interaction stemming from the mixture. The usefulness of this interacting source analysis is demonstrated in simulations and for real electroencephalography data.

AB - We present a technique that identifies truly interacting subsystems of a complex system from multichannel data if the recordings are an unknown linear and instantaneous mixture of the true sources. The method is valid for arbitrary noise structure. For this, a blind source separation technique is proposed that diagonalizes antisymmetrized cross-correlation or cross-spectral matrices. The resulting decomposition finds truly interacting subsystems blindly and suppresses any spurious interaction stemming from the mixture. The usefulness of this interacting source analysis is demonstrated in simulations and for real electroencephalography data.

UR - http://www.scopus.com/inward/record.url?scp=33646880239&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33646880239&partnerID=8YFLogxK

U2 - 10.1103/PhysRevE.73.051913

DO - 10.1103/PhysRevE.73.051913

M3 - Article

VL - 73

JO - Physical Review E

JF - Physical Review E

SN - 2470-0045

IS - 5

M1 - 051913

ER -