Concerning the problem of low accuracy of human activity recognition, a recognition method combining Support Vector Machine (SVM) with context-awareness (actual logic or statistical model of human motion state transition) was proposed to identify six types of human activities (walking, going upstairs, going downstairs, sitting, standing, lying). Logical relationships existing between human activity samples were used by the method. Firstly, the SVM model was optimized by using the Improved Particle Swarm Optimization (IPSO) algorithm. Then, the optimized SVM was used to classify the human activities. Finally, the context-awareness was used to correct the error recognition results. Experimental results show that the classification accuracy of the proposed method reaches 94.2% on the Human Activity Recognition Using Smartphones (HARUS) dataset of University of California， Irvine (UCI), which is higher than that of traditional classification method based on pattern recognition.

The existing Convolutional Neural Network (CNN) models have a large number of redundant parameters. In order to address this problem, two lightweight network models named Fnet1 and Fnet2, based on the SqueezeNet core structure Fire module, were proposed. Then, in the view of the characteristics of distributed data collection and processing of mobile terminals, based on Fnet2, a new network model named FnetDNN, with Fnet2 integrated with Deep Neural Network (DNN), was proposed according to Dempster-Shafer (D-S) evidence theory. Firstly, a neural network named Cent with four convolutional layers was used as the benchmark, and Mel Frequency Cepstral Coefficient (MFCC) as the input feature. From aspects of the network structure characteristics, calculation cost, number of convolution kernel parameters and recognition accuracy, Fnet1, Fnet2 and Cent were analyzed. Results showed that Fnet1 only used 10.3% parameters of that of Cnet, and had the recognition accuracy of 86.7%. Secondly, MFCC and the global feature vector were input into the FnetDNN model, which improved the recognition accuracy of the model to 94.4%. Experimental results indicate that the proposed Fnet network model can compress redundant parameters as well as integrate with other networks, which has the ability to expand the model.