Application of Second-Order Blind Identification in Motor Imagery EEG Classification

Human thinking tasks evoke Electroencephalogram (EEG) signal changes, so EEG analysis can help to design communication systems of Brain computer interface (BCI). Second-order blind identification (SOBI), a blind source separation (BSS) algorithm was applied to preprocess EEG data. Subsequently, Fisher distance was used to select the features. Finally, classification of Motor Imagery EEG evoked by a sequence of randomly mixed left and right image stimulations was performed by back-propagation neural networks, using as inputs the two SOBI-recovered SI components or the two EEG channels C3/C4. The average classification accuracy is 72.57% without SOBI preprocessing when it is 90.9% with SOBI preprocessing. The results show that classification accuracy of BCI is significantly improved by SOBI preprocessing.