The sign tale of hurst parameter: a revisit

Network traffic measurement studies have shown the presence of self similar behavior in both local area and wide area traffic traces. In this study we investigated the magnitude of the self similarity behavior of network traffic and its Hurst measurement, H. Our study has revealed several characteri...

Full description

Saved in:
Bibliographic Details
Main Authors: Salleh, Mazleena, Bakar, Ahmad Zaki Abu
Format: Book Section
Language:en
Published: Institute of Electrical and Electronics Engineering (IEEE) 2007
Subjects:
Online Access:http://eprints.utm.my/9641/1/MazleenaSalleh2007_TheSignTaleOfHurst.pdf
http://eprints.utm.my/9641/
http://dx.doi.org/10.1109/ICCIT.2007.207
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1845472202562994176
author Salleh, Mazleena
Bakar, Ahmad Zaki Abu
author_facet Salleh, Mazleena
Bakar, Ahmad Zaki Abu
author_sort Salleh, Mazleena
building UTM Library
collection Institutional Repository
content_provider Universiti Teknologi Malaysia
content_source UTM Institutional Repository
continent Asia
country Malaysia
description Network traffic measurement studies have shown the presence of self similar behavior in both local area and wide area traffic traces. In this study we investigated the magnitude of the self similarity behavior of network traffic and its Hurst measurement, H. Our study has revealed several characteristics of H based on the network traffic changes. From the finding, H can be used as traffic load indicated only if the sampling size is large. On the other hand for a smaller sampling size, H can be used to indicate the change in traffic pattern. However network traffic does not always exhibit self similarity property. Activities such as peer-to-peer can steer the traffic to lose this property.
format Book Section
id my.utm.eprints-9641
institution Universiti Teknologi Malaysia
language en
publishDate 2007
publisher Institute of Electrical and Electronics Engineering (IEEE)
record_format eprints
spelling my.utm.eprints-96412017-08-15T01:35:57Z http://eprints.utm.my/9641/ The sign tale of hurst parameter: a revisit Salleh, Mazleena Bakar, Ahmad Zaki Abu QA75 Electronic computers. Computer science Network traffic measurement studies have shown the presence of self similar behavior in both local area and wide area traffic traces. In this study we investigated the magnitude of the self similarity behavior of network traffic and its Hurst measurement, H. Our study has revealed several characteristics of H based on the network traffic changes. From the finding, H can be used as traffic load indicated only if the sampling size is large. On the other hand for a smaller sampling size, H can be used to indicate the change in traffic pattern. However network traffic does not always exhibit self similarity property. Activities such as peer-to-peer can steer the traffic to lose this property. Institute of Electrical and Electronics Engineering (IEEE) 2007-11 Book Section PeerReviewed application/pdf en http://eprints.utm.my/9641/1/MazleenaSalleh2007_TheSignTaleOfHurst.pdf Salleh, Mazleena and Bakar, Ahmad Zaki Abu (2007) The sign tale of hurst parameter: a revisit. In: International Conference on Convergence Information Technology 2007. Institute of Electrical and Electronics Engineering (IEEE), pp. 2169-2174. ISBN 0-7695-3038-9 http://dx.doi.org/10.1109/ICCIT.2007.207 doi : 10.1109/ICCIT.2007.207
spellingShingle QA75 Electronic computers. Computer science
Salleh, Mazleena
Bakar, Ahmad Zaki Abu
The sign tale of hurst parameter: a revisit
title The sign tale of hurst parameter: a revisit
title_full The sign tale of hurst parameter: a revisit
title_fullStr The sign tale of hurst parameter: a revisit
title_full_unstemmed The sign tale of hurst parameter: a revisit
title_short The sign tale of hurst parameter: a revisit
title_sort sign tale of hurst parameter: a revisit
topic QA75 Electronic computers. Computer science
url http://eprints.utm.my/9641/1/MazleenaSalleh2007_TheSignTaleOfHurst.pdf
http://eprints.utm.my/9641/
http://dx.doi.org/10.1109/ICCIT.2007.207
url_provider http://eprints.utm.my/