A new HOS-based blind source extraction method to extract µ rhythms from EEG signals

  • Authors:

  • Kun Cai;Shengli Xie

  • Affiliations:

  • ,College of Engineering, South China Agriculture University, Guangzhou, Guangdong, China;School of Electronic and Information Engineering, South China University of Technology, Guangzhou, Guangdong, China

  • Venue:
  • ICSI'10 Proceedings of the First international conference on Advances in Swarm Intelligence - Volume Part II
  • Year:
  • 2010

Quantified Score

Hi-index 0.00

Visualization

Abstract

The μ rhythm is a type of EEG rhythms, which usually occurs over the motor-sensory cortex of the brain. It is believed to reflect the limb movement and imaginary limb movement controlled by the brain, thus it is one of the important sources of BCI systems. In this paper, a new fixed-point BSE algorithm based on skewness is proposed to extract μ rhythms by the feature of asymmetric distribution. The local stability of the algorithm is also proved in this article. The results from simulations indicate that, for the μ rhythm extraction, the proposed skewness-based algorithm performs better than the negentropy-based FastICA.