作者: Shang-Ming Zhou , F. Sepulveda , J. Q. Gan
DOI:
关键词:
摘要: In order to characterise the non-Gaussian information contained within EEG signals, a new feature extraction method based on bispectrum is proposed and applied classification of right left motor imagery for developing EEG-based brain-computer interface (BCI) systems. The experimental results Graz BCI data set have shown that features LDA classifier outperforms winner 2003 competition same in terms criterion.