Recognizing arm motions by fluctuation analysis of EEG signals

We consider the task of recognition of fragments of multichannel electroencephalogram (EEG) records corresponding to motions of the human arm and to mental representation of these motions. It is shown that the problem of recognition can be solved by processing short EEG segments by the method of fluctuation analysis. The obtained results suggest that fluctuation analysis can be used as an algorithm of the digital signal processing in development of the neurointerface software.