A Master-Slave Neural Network for precise recognition of the complicated hand operations based on EEG. (29th June 2009)

• Record Type: Journal Article
• Authors: Zhang, Xiao Dong; Choi, Hyouk Ryeol
• Is Part Of: International journal of advanced media and communication. Volume 3:Number 1/2(2009)
• Page Start: 55
• Page End: 79
• Publication Date: 2009-06-29
• Subjects: artificial neural networks -- ANNs -- pattern recognition -- hand operations -- EEG signals -- electroencephalography -- stability -- EEG measurements; Multimedia communications -- Periodicals; Multimedia systems -- Periodicals; 384.38
• Languages: English
• ISSNs: 1462-4613

Abstract

On the basis of excellent features of the Hopfield neural network, a new Master-Slave Neural Network (simply denoted as MSNN) model was presented in this paper. The structure of the proposed MSNN was first designed, and the corresponding training algorithm was discussed in detail, and the stability of the MSNN was analysed in detail. Finally, through a two-channel EEG measurement system set-up, and the feature of the related EEG signals extracted, some complicated hand operations were recognised by using the MSNN and BP neural network. The comparison showed that the MSNN had a better asymptotic convergence rate and a higher mapping precision, so that a higher recognition possibility was achieved than the BP network. This is an expanded version of a paper presented at the 3rd IEEE International Workshop on Medical Measurements and Applications, 9?10 May 2008, Ottawa, ON, Canada.