An Analysis of Data Traffic Considering the Delay and Cell Loss Probability

There are many problems that must solve to construct next generation high-speed communication network. Among these, item that must consider basically is characteristics analysis of traffic that nows to network Traffic characteristics of many Internet services that is offered present have shown that network traffic exhibits at a wide range of scals-self-similarity. Self-similarity is expressed by long term dependency, this is contradictory concept with Poisson model that have relativity short term dependency. Therefore, first of all, for design and dimensioning of next generation communication network, traffic model that are reflected burstiness and self-similarity is required. Here self-similarity can be characterized by Hurst parameter. In this paper, the calculation equation is derived considering queueing delay and self-similarity of data traffic art compared with simulation results.