The invention discloses a myoelectricity classification method based on an improved small-world echo state network. Firstly, a small-world network is used for improving a reserve pool structure of theESN, then an edge adding probability is used for improving the small-world network, and the network is called as an improved small-world echo state network, so that the adaptability of a reserve poolis improved, and the generalization ability and stability of the ESN are improved. Then, the output weight of the network can be obtained by training the network, and the output weight is used as a corresponding feature. Electromyographic signals of six actions of falling down, walking, sitting, squatting, going upstairs and going downstairs are collected, corresponding features are extracted through ISWLESN, and then feature dimensions are reduced through PCV. And finally, the performance of the network features is represented by using the scatter diagram, the class separability index and the DBI. Results show that ISWLESN has good clustering performance, and has high precision when used for support vector machine classification.