Theoretic and experimental comparison of root-mean-square and mean-absolute-value electromyogram amplitude detectors

It is typically assumed that the probability density of the surface electromyogram (EMG) is Gaussian. This assumption leads to root-mean-square (RMS) processing as the maximum likelihood estimator of the EMG amplitude. Contrary to this theoretical formulation, recent experimental work has found mean-absolute-value (MAV) processing to be superior to RMS. This paper reviews RMS processing with the Gaussian model and then derives the expected (inferior) performance of MAV processing with the Gaussian model. Next, a new model for the surface EMG waveform, using a Laplacian density, is presented. It is shown that the MAV processor is the maximum likelihood estimator of the EMG amplitude for the Laplacian model. Lastly, experimental data from isometric, constant-force, non-fatiguing contractions were examined. On average, the Laplacian density best fit the experimental data (although results varied with subject). For amplitude estimation, MAV processing was clearly superior to RMS processing.