Effect of the window length on the EMG spectral estimation through the Blackman-Tukey method

This paper presents the application of spectral analysis methods that are based on the FFT algorithm to the study of the surface EMG power spectrum compression. Some preferred characteristic frequencies used to monitor the spectral compression are the mean and median frequencies. In this work, we show that the spectral density estimation by the Blackman-Tukey method associated with the weighting windows permits a better track of the compression with regard to the periodogram. The effect of the type and the size of windows is examined in order to determine the appropriate window that well describes the spectral compression; and thus the best process descriptor of the muscular fatigue.