Prothesis movements pattern recognition of surface myoelectric signals

In this paper, an Auto-regressive (AR) Model was presented to analysis of surface myoelectric (MES) signals which recorded on the ulnar flexor carpi and extensor carpi region of the right hand in resting position. Then BP neural network classifier, RBF neural network classifier and Wavelet Neural Network (WNN) classifier were used to study the correlation between SMES and wristwork. Four types of prosthesis movements are recognized by extracting four-order AR coefficient and construct them as eigenvector into different Neural Network. The experimental results show that the proposed WNN method achieved a mean classification accuracy of 93.75% of four movements (hand open (HO), hand close (HC), forearm intorsion (FI) and forearm extorsion (FE)), which outperformed the method using BP Neural Network (with mean classification accuracy 86.25%) and the RBF Neural Network method (87.5%). This paper also finds the WNN have many advantages such as self-learning, self-adaptive, robust, fault-tolerance and generalization ability. It's better than BP neural network and RBF neural network.