A fractal modeling for network traffic

Abstract

This paper describes fractal modeling with application to broadcast traffic of a switch-based LAN. The traffic is represented by two variables, packet inter-arrival time and packet size. The series of these two variables is normalized separately and considered as a number of random points in 2-dimensional space. In the study, fractal behavior of the traffic is considered to be caused by high variability of the packet inter-arrival time. Hence measuring variability of inter-arrival time is a necessity for fractal analysis. This analysis is done using long-range dependence (LRD) of variance measurement. The work compares LRD properties between real broadcast traffic and the synthetic one. The real traffic is captured from switch-based LAN, while the synthetic one is generated using iterated function system (IFS) method.