Time frequency analysis of the electromyogram during fatigue

Spectral parameters obtained from the surface electromyogram (EMG) signal have been used as indicators of fatigue during a sustained contraction. A new technique, time frequency analysis, was applied to the EMG signal. This technique generates a continuous representation of the changing spectrum of the signal through time. Three types of time frequency distributions were applied to the EMG signal. As predicted, differences existed between the distributions. The amplitude differential from the first time slice of the distribution to the last was the smallest for the short time Fourier transform. The Wigner-Ville distribution was spread out across the most frequencies. Walls appeared in the Choi-Williams distribution, but otherwise it was the most compressed. All the distributions displayed the expected spectral compression; however, more work is necessary to clarify the results.<
>