An efficient channel selection based on task classification

In this Paper we are going to select the Efficient Channel based on Feature Extraction for Motor Imagery Task Classification in the context of Electroencephalography (EEG) — based brain — computer interface (BCI). This method uses LP-SVD (Linear Prediction Singular Value Decomposition) for feature extraction. The proposed approach was first benchmarked against two related state-of-art feature extraction approaches, namely, discrete cosine transform (DCT) and adaptive autoregressive (AAR) — based methods. By achieving an accuracy of 67.35%, the LP-SVD approach outperformed the other approaches by large margins(25% compared with DCT and
^Λ
% compared with AAR-based methods). To improve the discriminatory capability of the extracted features and reduce the computational complexity, we introduced two extra features, namely Q- and the Hotel ling's T2. Our method came with an average accuracy of 81.38%