Application of Support Vector Machines in upper limb motion classification using myoelectric signals

This paper presents a novel support vector machine (SVM) approach to upper limb motion classification using myoelectric signals. The main purpose of this paper is to compare SVM-based classifiers with LDA and MLP. SVM demonstrates exceptional classification accuracy and results in a robust way of limb motion classification with low computational cost. The validity of entropy, as an index to measure correctness of classification, is also examined. Experimental results show that entropy is a reliable measure for online training in myoelectric control systems.