Network Traffic Prediction Model Based on Auto-regressive Moving Average

With the development of Internet and computer science, computer network is changing people’s lives. Meanwhile, Network traffic prediction model itself becomes more and more complex. It is an important research direction to quickly and accurately detect the intrusions or attacks. The performance efficiency of a network intrusion detection system is dominated by pattern matching algorithm. However, In view of the complex non-linear and chaotic network traffic, and combined with its time-series properties, this paper propose a network traffic prediction model based on the auto-regressive moving average (ARMA). This model adopts the third party inspection systems which aim to save the network resources and predict the network traffic in high efficiency. By taking the data flow measurements in 16 channel analyzer to initialize the model, the simulation results show that the proposed model can effectively detect network intrusions and attacks, which improves the performance of the entire network and prolongs the network lifetime.