Validation of EEG Personal Authentication with Multi-channels and Multi-tasks

We investigate a feature extraction method that is effective for biometric identification using brain waves in this paper. We extract power spectrums of theta waves, alpha waves, beta waves and gamma waves for the quantity of personal characteristic. We measure brain waves with five tasks and multi-channel electroencephalograph to analyze each error rate. As a result, the error rate of the alpha wave is the lowest; the authentication rate by a single channel / a single task, sixteen channels / a single task, and sixteen channels and five tasks are 90%, 92%, and 97%, respectively.