Comparative Study of Different EMG Signal decomposition Techniques

EMG signals are electromyogram signals generated by firing of MUs (motor units) in muscle fibers. The decomposition of EMG signal of a muscle provides useful information for the diagnosis of neuro-muscular diseases by physician and neurologist. In decomposition of EMG signal different MUAPs (Motor Unit Action Potentials) are classified into different categories. This paper gives a review of different techniques used for decomposition of EMG signal. The techniques discussed are the decomposition of surface EMG signal based on blind source separation of convolved mixtures[11] , An artificial neural network (ANN) technique based on unsupervised learning, using the self-organizing feature maps (SOFM) algorithm and learning vector quantization (LVQ)[1] and high precision EMG signal decomposition using communication techniques[5].