Local dependencies and Poissonification: a case study

Research output: Contribution to journalArticle

Abstract

Traffic characterization is of great importance in the analysis and dimensioning of modem telecommunication networks. In this paper, the microdynamics of some sets of traffic are investigated. The approach we use to study the microdynamics is Poissonification, which is away of transforming a point process into a Poisson process locally. We want to gain basic knowledge of how the traffic behaves on smaller time scales, but we also want to see if there are any effects when the traffic is fed into a queue. As an application, Poissonification can be used when modeling is done with a doubly stochastic Poisson process to find the time scale where we no longer have the doubly stochastic behavior. Three different traffic traces were analyzed in this paper. The results show that provided the Poissonification is carried out at a sufficiently small time scale, the performance characteristics are not decisively changed and this gives a guide to choosing the time scale at which to start traffic modeling.

Details

Authors
  • Sofia Andersson
  • Tobias Rydén
Organisations
Research areas and keywords

Subject classification (UKÄ) – MANDATORY

  • Probability Theory and Statistics

Keywords

  • analysis, performance, traffic analysis, microdynamics, Poissonification, point process spectrum
Original languageEnglish
Pages (from-to)41-58
JournalPerformance Evaluation
Volume52
Issue number1
Publication statusPublished - 2003
Publication categoryResearch
Peer-reviewedYes