On the Chaotic Dynamics Analysis of Internet Traffic

In this paper, the real network traffic is analyzed about its chaotic dynamic properties and the traffic signals are reconstructed using FBM based fractal interpolation algorithm. The self-similarity of network traffic is analyzed by estimating the value of Hurst exponent and the traffic time series is reconstructed as a phase trajectory by properly choosing some parameters. Through the reconstructed phase space and corresponding data analysis, it can be found that the network traffic reveals the chaotic property and compared with classic stochastic model, the chaos-based model is found to be more suitable and precise for forecasting of network traffic.