EEG signal classification using nonlinear independent component analysis

One of the preprocessors can be used to improve the performance of brain-computer interface (BCI) systems is independent component analysis (ICA). ICA is a signal processing technique in which observed random data are transformed into components that are statistically independent from each other. This suggests the possibility of using ICA to separate different independent brain activities during motor imagery into separate components. However, there is no guarantee for linear combination of brain sources in EEG signals. Thus the identification of nonlinear dynamic of EEG signals should be taken into consideration. In this paper, a new method is proposed for EEG signal classification in BCI systems by using nonlinear ICA algorithm. The effectiveness of the proposed method is evaluated by using the classification of EEG signals. The tasks to be discriminated are the imaginative hand movement and the resting state. The results demonstrate that the proposed method performed well in several experiments on different subjects and can improve the classification accuracy in the BCI systems.