Detecting chaotic structures in noisy pulse trains based on interspike interval reconstruction

  • Authors:

  • Takashi Kanamaru;Masatoshi Sekine

  • Affiliations:

  • Department of Basic Engineering in Global Environment, Faculty of Engineering, Kogakuin University, 2665-1 Nakano, 192-0015, Hachioji, Tokyo, Japan;Department of Electrical and Electronic Engineering, Faculty of Technology, Tokyo University of Agriculture and Technology, 2-24-16 Naka-cho, 184-8588, Koganei-city, Tokyo, Japan

  • Venue:
  • Biological Cybernetics
  • Year:
  • 2005

Quantified Score

Hi-index 0.00

Visualization

Abstract

The nonlinear prediction method based on the interspike interval (ISI) reconstruction is applied to the ISI sequence of noisy pulse trains and the detection of the deterministic structure is performed. It is found that this method cannot discriminate between the noisy periodic pulse train and the noisy chaotic one when noise-induced pulses exist. When the noise-induced pulses are eliminated by the grouping of ISI sequence with the genetic algorithm, the chaotic structure of the chaotic firings becomes clear, and the noisy chaotic pulse train could be discriminated from the periodic one.