Network Traffic Prediction in Network Security Based on EMD and LSTM

Abstract

With the rapid development of the Internet, the scale of the network continues to expand, and the situation of network security is getting more and more severe. Network security requires more reliable information to support, and the prediction of network traffic is an important part of network security. Network traffic prediction data can provide important data reference for network security, especially for reliable data transmission and network monitoring. In fact, network traffic data is affected by a variety of complex and random factors, so network traffic data is a nonlinear data sequence. This paper analyzes the characteristics of network traffic data and proposes an EMD-LSTM model for network traffic data prediction. Firstly, the complex and variable network data traffic is decomposed into several smooth data sequences, and then, the LSTM neural network model, which is suitable for data sequence prediction, is used to predict. The results of the comparison experiments show that the proposed network traffic prediction method reduces the prediction root mean square error in network traffic prediction.