Classification of Motor Imagery Using Trial Extension in Spatial Domain with Rhythmic Components of EEG

Electrical activities of the human brain can be recorded with electroencephalography (EEG). To characterize motor imagery (MI) tasks for brain–computer interface (BCI) implementation is an easy and cost-effective tool. The MI task represented by a short-time trial multichannel EEG. In this paper, signal each channel raw EEG decomposed into finite set narrowband signals using Fourier-transformation-based bandpass filter. Rhythmic components are that related to tasks. subband arranged extend dimension in spatial domain. features extracted from extended trials common pattern (CSP). An optimum number employed classify artificial neural network. integrated approach full-band implemented derive discriminative classification. addition, subject-dependent parameter optimization scheme enhances performance proposed method. evaluation method obtained two publicly available benchmark datasets (Dataset I Dataset II). experimental results terms classification accuracy (93.88% 91.55% II) show it performs better than recently developed algorithms. enhanced very much applicable BCI implementation.