EEG SIGNALS CLASSIFICATION METHOD DURING THE MOTOR ACTIVITY IMAGINATION IN THE UNTRAINED OPERATOR

摘要

FIELD: computer equipment.;SUBSTANCE: invention relates to the field of data digital processing and analysis, and is intended for the multi-channel electroencephalograms processing for the associated with the motor activity imagination in untrained operators the brain electrical activity characteristic patterns selection in real time. EEG signals classification method with the untrained operator motor activity imagination is, that with the help of sensors, recording the EEG signals are from the occipital, central and frontal regions, for which in the time-frequency analysis unit calculating the continuous wavelet transform value with the base Morlaille wavelet, calculating the wavelet spectrum energy average value in the alpha 8–12 Hz range for the frontal, central and occipital regions and the wavelet spectrum energy average value in the delta 1–5 Hz range for the frontal region, next, in the adaptive filtering unit performing obtained by the empirical modes the averaged values decomposition and extracting these dependencies low-frequency component, highlighting the fourth-order empirical modes, then in the classification unit performing the obtained empirical modes behavior over time analysis, at that, the time moments, for which the empirical mode amplitude, calculated on the EEG signals alpha rhythm basis for the frontal, central and occipital regions, increases, and the empirical mode amplitude, calculated on the frontal EEG delta rhythm basis, decreases, classified as the physical activity imagination episodes.;EFFECT: invention enables associated with the motor activity imagination brain electrical activity patterns reliable detection, in untrained operators in real time mode.;1 cl, 2 dwg