中国物理B ›› 2005, Vol. 14 ›› Issue (11): 2176-2180.doi: 10.1088/1009-1963/14/11/006

• GENERAL • 上一篇    下一篇

Blind source separation of multichannel electroencephalogram based on wavelet transform and ICA

游荣义1, 陈忠2   

  1. (1)Department of Physics, Jimei University, Xiamen 361021, China; (2)Department of Physics, Xiamen University, Xiamen 361005, China
  • 收稿日期:2004-08-05 修回日期:2005-05-09 出版日期:2005-11-20 发布日期:2005-11-20
  • 基金资助:
    Project supported by the National Natural Science Foundation of China (Grant No 10234070) and by the Science Foundation of Educational Commission of Fujian Province of China (Grant No JAO04238).

Blind source separation of multichannel electroencephalogram based on wavelet transform and ICA

You Rong-Yi (游荣义)a, Chen Zhong (陈忠)b   

  1. a Department of Physics, Jimei University, Xiamen 361021, China; b Department of Physics, Xiamen University, Xiamen 361005, China
  • Received:2004-08-05 Revised:2005-05-09 Online:2005-11-20 Published:2005-11-20
  • Supported by:
    Project supported by the National Natural Science Foundation of China (Grant No 10234070) and by the Science Foundation of Educational Commission of Fujian Province of China (Grant No JAO04238).

PDF (PC)

883

赞

摘要： Combination of the wavelet transform and independent component analysis (ICA) was employed for blind source separation (BSS) of multichannel electroencephalogram (EEG). After denoising the original signals by discrete wavelet transform, high frequency components of some noises and artifacts were removed from the original signals. The denoised signals were reconstructed again for the purpose of ICA, such that the drawback that ICA cannot distinguish noises from source signals can be overcome effectively. The practical processing results showed that this method is an effective way to BSS of multichannel EEG. The method is actually a combination of wavelet transform with adaptive neural network, so it is also useful for BBS of other complex signals.

Abstract: Combination of the wavelet transform and independent component analysis (ICA) was employed for blind source separation (BSS) of multichannel electroencephalogram (EEG). After denoising the original signals by discrete wavelet transform, high frequency components of some noises and artifacts were removed from the original signals. The denoised signals were reconstructed again for the purpose of ICA, such that the drawback that ICA cannot distinguish noises from source signals can be overcome effectively. The practical processing results showed that this method is an effective way to BSS of multichannel EEG. The method is actually a combination of wavelet transform with adaptive neural network, so it is also useful for BBS of other complex signals.

Key words: blind source separation, electroencephalogram, wavelet transform, independent component analysis

中图分类号:  (Biological signal processing)

  • 87.85.Ng
87.19.R- (Mechanical and electrical properties of tissues and organs) 07.05.Mh (Neural networks, fuzzy logic, artificial intelligence) 02.30.Uu (Integral transforms) 02.50.Fz (Stochastic analysis)