Testing alpha-stable processes in modelling broadband teletraffic

The paper presents the analysis of the applicability of alpha-stable processes in traffic modelling. This study is suggested by the goodness of alpha-stable processes in capturing not only the Long Range Dependence (LRD) of actual traffic, but also the heavy tailness of its marginal distribution. The relevance of this property is proved by means of discrete event simulations carried out considering two different data traffic sets, respectively related to a LAN-to-LAN intercomnection and entertainment video service. Errors in the est.mation of Hurst parameter and in the evaluationof queueing behaviour are highlighted either analytically and empirically by simulations. In particular the queueing simulations have emphasised the improvements in the performance forecast- ing introduced by the higher flexibility of a-stable model with respect to the widely used Fractional Brownian Mo- tion (FBM).