PROCESS FOR POINTING OUT SELF-SIMILARITY OF INTERNET TRAFFIC

摘要

The invention relates to a process for pointing out self-similarity of traffic packages in Internet. According to the invention, the process consists of a series of steps which are taken sequentially, consisting in: making the free scale topology of the network centers, followed by making the topology of the traffic sources, afterwards there is executed a traffic simulation, the data captured after the simulation being aggregated in a time series, then there being estimated the self-similarity parameter Hurst, after which there may be determined if the traffic is of self-similar nature or not, information which is, finally, stored in a database.