Adaptive approach for change detection in EMG recordings
Résumé
In this paper we present a new algorithm to detect abrupt changes in a signal when there is no a priori knowledge of the hypotheses on the process to be detected. This algorithm is based on the CUSUM algorithm. It can be applied in case of frequency and energy changes. This algorithm works when the samples are dependent and autoregressive modeling is needed. It is used to distinguish EMG segments from noise segments.