Novel HHT-Based Features for Biometric Identification Using EEG Signals

In this paper we present a novel approach for biometric identification using electroencephalogram (EEG) signals based on features extracted with the Hilbert-Huang Transform (HHT). The instantaneous amplitude and the instantaneous frequency were computed after the HHT, and these were then used to generate the features for classification. The proposed system was evaluated using two publicly available databases in scenarios where only a single electrode is used to provide biometric information. One database (with 122 subjects) has the users viewing a series of pictures while the other one (with 109 subjects) has the users performing motor/imagery tasks. Average identification accuracies of 96% and 99% were reached for these two databases respectively using only a single electrode. These compare favourably with previously published results employing a variety of other features and classification approaches.