Wavelet Neural Network Based Emg Signal Classifier


Classification of EMG signals is an important area in biomedical signal processing. Several algorithms have been developed for classification of EMG signals. These techniques extract features, which are either temporal or a transformed representation of the EMG waveforms. Artificial Neural Networks (ANN) trained with BP algorithm classifies the applied input EMG to an appropriate class which either normal or abnormal muscular responses. This paper shows an approach for EMG signal processing based on ANN and transform domain (Discrete Wavelet Transform (DWT) in order to perform automatic analysis using personal computers. The Neural Networks (NN) are introduced to solve different pattern recognition problems associated with EMG analysis. A Multi-Layer Perceptron (MLP) NN is used in the present work with Back Propagation (BP) algorithm to train the proposed network.