On the implications of the off periods distribution in two-state traffic models

Author

Garroppo, Rosario G. ; Giordano, Stefano ; Isopi, Marco ; Pagano, Michele

Author_Institution

Pisa Univ., Italy

Volume

3

Issue

7

fYear

1999

fDate

7/1/1999 12:00:00 AM

Firstpage

220

Lastpage

222

Abstract

The paper highlights the impact of the inactive period distribution of an on-off source on the autocorrelation structure of the model. In particular, two discrete-time on-off models, both with geometric distribution of the active period, are considered: the GeoGeo model (GGM) and the ParetoGeo model (PGM), characterized by a geometric and Pareto-like distribution of the off periods length, respectively. The two models are compared in terms of their main statistical features as well as the queueing impact. The asymptotic behavior of the autocorrelation function for the proposed PGM has been analytically evaluated. It is shown it has a hyperbolic decay related to the tail behavior of the off periods distribution.

Keywords

Markov processes; Pareto distribution; binary sequences; correlation methods; queueing theory; telecommunication traffic; GeoGeo model; Pareto-like distribution; ParetoGeo model; asymptotic behavior; autocorrelation function; autocorrelation structure; binary sequence; discrete-time on-off models; geometric distribution; hyperbolic decay; inactive period distribution; off periods distribution; off periods length; on-off source; queueing impact; statistical features; tail behavior; two-state discrete-time Markov chain; two-state traffic models; Autocorrelation; Binary sequences; Ethernet networks; LAN interconnection; Length measurement; Local area networks; Probability distribution; Solid modeling; Traffic control;

fLanguage

English

Journal_Title

Communications Letters, IEEE

Publisher

ieee

ISSN

1089-7798

Type

jour

DOI

10.1109/4234.775260

Filename

775260