A shifted gamma distribution model for long-range dependent Internet traffic

Author

Kim, Sunggon ; Lee, Ju Yong ; Sung, Dan Keun

Author_Institution

Dept. of Electr. Eng. & Comput. Sci., KAIST, Daejon, South Korea

Volume

7

Issue

3

fYear

2003

fDate

3/1/2003 12:00:00 AM

Firstpage

124

Lastpage

126

Abstract

It is important to characterize the distributional property and the correlation structure of traffic arrival processes in modeling internet traffic. The conventional fractional Gaussian noise (fGn) model fails in characterizing the distributional property when the distribution of the input traffic rates is nongaussian. We propose a shifted gamma distribution model which can solve this problem. A linear-time generation algorithm is also given.

Keywords

Gaussian noise; Internet; correlation methods; gamma distribution; telecommunication traffic; transport protocols; IP communication; Internet traffic modeling; correlation structure; distributional property; fractional Gaussian noise model; input traffic rates; linear-time generation algorithm; long-range dependent Internet traffic; nonGaussian distribution; shifted gamma distribution model; traffic arrival processes; 1f noise; Autocorrelation; Communication networks; Equations; Fluctuations; Gaussian distribution; Gaussian noise; Internet; Telecommunication traffic; Traffic control;

fLanguage

English

Journal_Title

Communications Letters, IEEE

Publisher

ieee

ISSN

1089-7798

Type

jour

DOI

10.1109/LCOMM.2002.808400

Filename

1187381