A Novel Method for Extracting Imaginary Movement in Brain-Computer Interface

Imaginary movement extraction is one of the key and difficult tasks in brain-computer interface(BCI).In this paper,we utilize a novel classification method synthesized with continuous wavelet transform(CWT) and Bayesian neural network(BNN) to detect and recognize the characteristic EEG(electroencephalogram) information by using the event related desynchronization(ERD) caused by imaginary movement thinking.Research results show that the presented method has a higher recognition accuracy,better anti-noise ability and higher recognition speed than the generally used linear classifiers.It can meet the basic requirments of pattern recognition for realtime BCI system.