Effect of Superposition and Window Length in Fatigue Estimation By Weighted-Cumulated Methodology

The present work investigates the temporal analysis window length and the amount of overlap between successive windows in muscular fatigue estimation using the Weighted-Cumulated Methodology. The theoretical basis of the methodology is resumed, followed by a description of the weighted cumulated root mean square, weighted cumulated median frequency and weighted cumulated Fourier estimators. Analyses are made using real surface electromyography signals from isometric fatiguing contractions of the biceps bacchii of five healthy subjects. The sensibility to muscular fatigue manifestations of the three cumulated normalized parameters is evaluated using different combinations of window sizes and window overlaps. The results show a considerable influence of the window size in the values of sensibility to muscular fatigue of the cumulated normalized estimators. On the other hand, varying the amount of windows overlap is shown to have basically no effect on the sensibility descriptors.