Journal
Scientific and technical journal of information technologies, mechanics and optics
UDK
Issue:1 (83)
Download PDF0 Kbyte
A method of electromyogram signals processing and identification for implementation in rehabilitation devices control is given. The method is based on the high-frequency components filtration which improves the signal/noise ratio; also it is based on the wavelet analysis for signal preprocessing and motion type classification by taught artificial neural network. Obtained accuracy of motion type classification is 94%.