A simultaneous filtering and feature extraction strategy for direct brain interfacing

Abstract

Parametric modeling strategies are explored in conjunction with Linear Discriminant Analysis (LDA) to facilitate an Electroencephalogram (EEG) based direct-brain interface. A left/right self-paced typing exercise is analysed by employing an AutoRegressive (AR) model for feature extraction and an AutoRegressive with Exogenous input (ARX) model for combined filtering and feature extraction. Modeling both the signal and noise is found to be more effective than modeling the noise alone with the former yielding a classification accuracy of 81.0% and the latter an accuracy of 57.4%.