A. N. Doletskiy, D. S. Guzenko
The paper presents the general structure of the brain-computer interfaces, based on the recognition of motion types, imagined by subjects. A review of the main methods used at each stage of signal processing of EEG has been made. The availability of a wide range of approaches to quantitative EEG analysis in BCI applications has been shown. We have established that the current approaches do not provide the required classification accuracy required for the crea- tion of systems to be conveniently used in everyday life. Prospects for improving the effectiveness of the methods have been considered.
brain-computer interface, BCI, EEG