Parametric modeling of EMG signals

The authors deal with the application to EMG signals of parametric modeling and performance evaluation. Two main topics are discussed: the parametric spectral analysis of EMG signals as compared to FFT-based spectral estimation, and the deconvolution of EMG, i.e. retrieval of the MUAPs using the multipulse excitation modeling. It is shown that parametric models give more accurate spectrum estimates than the FFT. What is more important, they give parameters which are relevant. Multipulse modeling seems to be of great interest in the retrieval of the MUAPs.<>