Analysis of the compression of the spectral density of a surface EMG signal by methods based on the FFT

This article presents the application of methods of spectral analysis based on the FFT algorithm to the compression of the spectral density of the power of a surface EMG signal. This change is frequently generalised by the average and median frequencies. The modal frequency is another characteristic able to be calculated, but, it is also demonstrated that it does not permit credible evaluations of the spectral density. In this work we show that the estimation of the spectral density by the correlogram method associated with the windows of balance permits control of the compression by the modal frequency. What is more, we propose to reduce the fluctuations of the periodogram by a smoothing method based on the regression polynomial of order 6. This suggestion permits pursuit of the spectral compression by the three characteristic frequencies.