Network Traffic Prediction Based on Particle Swarm Optimization

Predicting the network traffic flow for large-scale network can significantly improve the quality of service security, and this problem has been attracted more and more researches. In this paper, we study on forecast network traffic by a hybrid Flexible neural tree and Particle swarm optimization model. Framework of the particle swarm optimization based network traffic forecasting is made up of three steps: 1) Obtaining network flow data, 2) Constructing the network flow and 3) Building a flexible neural tree to implement the network traffic prediction system. As flexible neural tree based network traffic prediction is greatly influence by parameters selection, We utilize particle swarm optimization to optimize parameters for the proposed algorithm. Experimental results demonstrate that the proposed algorithm can effectively forecast network traffic with lower error rate.