In order to eliminate the noise in Surface Electromyography(sEMG) signals for live working, a sEMG signal denoising method based on MEEMD and DFA was proposed. The sEMG signals of right arm biceps after wearing insulating gloves were collected under typical working conditions, and the Detrend Fluctuation Analysis (DFA) was used as a filtering index to enhance the recognition ability of Modified Ensemble Empirical Mode Decomposition (MEEMD) on the effective information in sEMG signals, so as to improve the effect of secondary noise reduction. The results show that the MAE, MSE and SNR of DFA-MEEMD secondary denoising method are 4.40×10^-3, 5.80×10^-5 and 25.4dB respectively for typical sEMG signals with strong noise, which can provide a method for e1xtracting useful information from sEMG signals.

Live working,insulating gloves,Surface Electromyography (sEMG),Modified Ensemble Empirical Mode Decomposition (MEEMD),Detrended Fluctuation Analysis (DFA)