We found a match
Your institution may have rights to this item. Sign in to continue.
- Title
Long-range dependence and performance in telecom networks.
- Authors
Conti, Pier Luigi; Lijoi, Antonio; Ruggeri, Fabrizio
- Abstract
Telecommunications systems have recently undergone significant innovations. These call for suitable statistical models that can properly describe the behaviour of the input traffic in a network. Here we use fractional Brownian motion (FBM) to model cumulative traffic network, thus taking into account the possible presence of long-range dependence in the data. A Bayesian approach is devised in such a way that we are able to: (a) estimate the Hurst parameter H of the FBM; (b) estimate the overflow probability which is a parameter measuring the quality of service of a network: (c) develop a test for comparing the null hypothesis of long-range dependence in the data versus the alternative of short-range dependence. In order to achieve these inferential results, we elaborate an MCMC sampling scheme whose output enables us to obtain an approximation of the quantities of interest. An application to three real datasets, corresponding to three different levels of traffic, is finally considered. Copyright © 2004 John Wiley & Sons, Ltd.
- Subjects
TELECOMMUNICATION systems; TELEPHONE systems; QUALITY control; BAYESIAN analysis; STATISTICS; PROBABILITY theory
- Publication
Applied Stochastic Models in Business & Industry, 2004, Vol 20, Issue 4, p305
- ISSN
1524-1904
- Publication type
Article
- DOI
10.1002/asmb.542