Analysis of surface EMG signal based on empirical mode decomposition

Abstract

In this paper, we propose a combination method based on the empirical mode decomposition and largest Lyapunov exponent technique for the feature extraction of surface EMG signals. Subsequently, the BP neural network is used as a classifier to identify the pattern category of upper limb motions. By the recognition analysis of the surface EMG signals, the data of the single channel contain some useful information of multi-category motions, such as the channel corresponding to the extensor digitorum muscle. And for all four channels, the better classification rates verify the usefulness of the presented method for six motions of hand and wrist.