Multi-service traffic models of heterogeneous wireless communication networks

Abstract

A good traffic model can greatly improve the spectrum efficiency of wireless networks. Conventional traffic models, such as Poisson distribution, only suit for traditional traffics, such as Voice. With the emergency of various kinds of traffics, especially the packet-switched data services of 3G/B3G in Heterogeneous Wireless communication networks, new traffic models are brought to apply increasingly. Long-range dependence (LRD) or self-similarity and multi-fractality are the distinct characteristics of these new traffics, accordingly kinds of traffic models were brought out subsequently. This paper analyzes the principle and the differences of these models, and proposes an optimal traffic modeling algorithm, named Infinitely Divisible Cascades Model (IDC). Further topics on the multi-service traffic modeling for the heterogeneous wireless communication networks are also presented in this paper.