Committee machines and quadratic B-spline wavelet for the P300 speller paradigm

In this study we propose a new approach to analyze data from the P300 speller paradigm using the quadratic B-spline wavelet coefficients and committee machines technique. Data set II from the BCI competition 2005 were used. The data were decomposed into five-octave frequency bands. We used coefficients between 4-8 Hz (theta) and 0.5-4 Hz (delta) frequency ranges as features. Extracted features used as inputs into committee machines (CM) based on LDA, MLP and SVM. This algorithm achieved an accuracy of 94% in P300 detection. We used only 7 electrodes of 64 recorded electrodes, Therefore it can noticeably reduce the time and cost of EEG measurement.