Surface electromyography (sEMG) signals are commonly used in activity monitoring and rehabilitation training as they reflect effectively the motor intentions of users. This study proposed a new sEMG-based multifeature extraction and predictive model to predict knee-joint angle from multichannel sEMG. Six channels of sEMG from relevant muscles were recorded, and knee-joint angles were sampled simultaneously for six kinds of knee-joint movement models. The root–mean–square (RMS), wavelet coefficients (WC), and permutation entropy (PE) as features of sEMG were extracted. The back propagation neural network, generalized regression neural network, and least-square support vector regression machine (LS-SVR) were used as predictive models. To validate the effectiveness of the sEMG features and predictive models, twelve subjects without neural or musculoskeletal deficits participated in the experiment. Six kinds of knee-joint movement models at different speeds and different loads were respectively conducted by the subjects. Results revealed that the combination of the three features (RMS, WC, and PE) and LS-SVR performed well for the knee-joint-angle of all kinds of leg motions. The RMS error for all kinds of leg motions was <7.7°. The estimation results of joint motion state would be used to rehabilitation robot or functional electrical stimulation for active rehabilitation of spinal cord injury patients or stroke patients.

Surface electromyography (sEMG) is an electrophysiological reflection of skeletal muscle contractile activity that can directly reflect neuromuscular activity. It has been a matter of research to investigate feature extraction methods of sEMG signals. In this letter, we propose a feature extraction method of sEMG signals based on the improved small-world leaky echo state network (ISWLESN). The reservoir of leaky echo state network (LESN) is connected by a random network. First, we improved the reservoir of the echo state network (ESN) by these networks and used edge-added probability to improve these networks. That idea enhances the adaptability of the reservoir, the generalization ability, and the stability of ESN. Then we obtained the output weight of the network through training and used it as features. We recorded the sEMG signals during different activities: falling, walking, sitting, squatting, going upstairs, and going downstairs. Afterward, we extracted corresponding features by ISWLESN and used principal component analysis for dimension reduction. At the end, scatter plot, the class separability index, and the Davies-Bouldin index were used to assess the performance of features. The results showed that the ISWLESN clustering performance was better than those of LESN and ESN. By support vector machine, it was also revealed that the performance of ISWLESN for classifying the activities was better than those of ESN and LESN.