基于高效胶囊网络和LSTM的流量分类研究

With the rapid development of communication technology and the Internet, new appearing network applications are emerging, and the demand for traffic is increasingly strong. In such an environment, a reasonable classification of network traffic is especially important to ensure that network managers can effectively monitor and allocate network resources.In this paper, we propose a traffic classification model based on efficient capsule neural network and long short-term memory. Based on the efficient capsule neural network capsule, the traffic flow spatial characteristics are extracted, which improves the overall efficiency of the model, and the performance of the classification model is further improved by using the time series characteristics extracted by the long-term short-term memory modules. Finally, compared with other deep learning methods, the experimental results show that the proposed model has better network traffic classification effect.