EEG-P300 extraction using neural network based adaptive recursive filter and adaptive autoregressive models

Abstract

In this paper, an adaptive feature extraction for EEG-based P300 signals is presented by combining the adaptive recursive (AR) filter and adaptive autoregressive (AAR) model. The extracted signals are then classified using multilayer neural networks (MNNs). It was found that the application of the proposed methods are improving and strengthening the EEG signal according to the small-amplitude of the P300 component in the EEG signals. The experimental results on the EEG raw data show that the implementation of the proposed method achieves a statistically significant improvement.