Discriminating Mental Tasks Using EEG Represented by AR Models

|EEG signals are modeled using single-channel and multi-channel autoregressive (AR) techniques. The co-eecients of these models are used to classify EEG data into one of two classes corresponding to the mental task the subjects are performing. A neural network is trained to perform the classiication. When applying a trained network to test data, we nd that the multivariate AR representation performs slightly better, resulting in an average classiica-tion accuracy of about 91%.