FRACTAL PROPERTIES OF NETWORK TRAFFIC IN A CLIENT-SERVER INFORMATION SYSTEM

Abstract

This paper uses three statistical analyses methods, R/S statistic, variance-time plots and periodogram-based analysis, to estimate the Hurst parameter of the wireless network traffic. The fractal statistical characteristics of wireless network traffic are studied based on the Hurst parameter. Simulation results compare the parameter estimation values of fractal wireless network traffic by utilizing above three statistical approaches. The simulation results demonstrate that wireless network traffic exhibits fractal characteristics. These results can be very useful for the performance improvement and the system design optimization of wireless network.