Cover Image

Rate limiting algorithm on video traffic with Pareto traffic model in broadband network / Rafidah Samsuri

Bandwidth management is one of the major issues for Quality of Service (QoS) for network traffic. Most network administrators are looking at providing best QoS and reliable traffic performances especially on video traffic. Thus, monitoring network traffic activity is a crucial task in providing bett...

Full description

Saved in:
Bibliographic Details
Main Author: Samsuri, Rafidah
Format: Thesis
Language:English
Published: 2016
Online Access:https://ir.uitm.edu.my/id/eprint/80486/1/80486.pdf
https://ir.uitm.edu.my/id/eprint/80486/
Tags: Add Tag
No Tags, Be the first to tag this record!
id my.uitm.ir.80486
record_format eprints
spelling my.uitm.ir.804862023-09-21T04:26:38Z https://ir.uitm.edu.my/id/eprint/80486/ Rate limiting algorithm on video traffic with Pareto traffic model in broadband network / Rafidah Samsuri Samsuri, Rafidah Bandwidth management is one of the major issues for Quality of Service (QoS) for network traffic. Most network administrators are looking at providing best QoS and reliable traffic performances especially on video traffic. Thus, monitoring network traffic activity is a crucial task in providing better bandwidth usage. This research presents an analysis on real time video network traffic in a Broadband Network in Malaysia. Real data is captured and collected at one of a telecommunications service company based for Business and Home Streamyx users in Johor. Statistical analysis is derived and new traffic model and characterizations are presented. Goodness of fit (GoF) and Kolmogorov Smirnov (KS) test is used to fit the real data to get a best Traffic distribution model. Results present four top video used in the network traffic which are You Tube, MPEG, TV on Streamyx and Dailymotion using identified video protocol. Fitted traffics presents Pareto model is best fitted on video traffic and Empirical Cumulative Distribution function (CDF) derived a Generalized Pareto (GP) distribution is the best fitting model for video traffic. GP characterization presents three important parameters which are shape, scale and location. By increase value of shape parameter is helps in controlling some burst traffic over identified bandwidth threshold. A new mathematical formulation is derived with present policing and shaping process by using the GP traffic model. As a result, policed and shaped make 60%, 35%, 47% and 45% usage to reach full utilization of 100Kbyte, 10Kbyte, 9Kbyte and 5Kbyte. Thus, traffic policing and shaping helps to optimize bandwidth utilization and to avoid frame loss of bandwidth. 2016 Thesis NonPeerReviewed text en https://ir.uitm.edu.my/id/eprint/80486/1/80486.pdf Rate limiting algorithm on video traffic with Pareto traffic model in broadband network / Rafidah Samsuri. (2016) Masters thesis, thesis, Universiti Teknologi MARA (UiTM).
institution Universiti Teknologi Mara
building Tun Abdul Razak Library
collection Institutional Repository
continent Asia
country Malaysia
content_provider Universiti Teknologi Mara
content_source UiTM Institutional Repository
url_provider http://ir.uitm.edu.my/
language English
description Bandwidth management is one of the major issues for Quality of Service (QoS) for network traffic. Most network administrators are looking at providing best QoS and reliable traffic performances especially on video traffic. Thus, monitoring network traffic activity is a crucial task in providing better bandwidth usage. This research presents an analysis on real time video network traffic in a Broadband Network in Malaysia. Real data is captured and collected at one of a telecommunications service company based for Business and Home Streamyx users in Johor. Statistical analysis is derived and new traffic model and characterizations are presented. Goodness of fit (GoF) and Kolmogorov Smirnov (KS) test is used to fit the real data to get a best Traffic distribution model. Results present four top video used in the network traffic which are You Tube, MPEG, TV on Streamyx and Dailymotion using identified video protocol. Fitted traffics presents Pareto model is best fitted on video traffic and Empirical Cumulative Distribution function (CDF) derived a Generalized Pareto (GP) distribution is the best fitting model for video traffic. GP characterization presents three important parameters which are shape, scale and location. By increase value of shape parameter is helps in controlling some burst traffic over identified bandwidth threshold. A new mathematical formulation is derived with present policing and shaping process by using the GP traffic model. As a result, policed and shaped make 60%, 35%, 47% and 45% usage to reach full utilization of 100Kbyte, 10Kbyte, 9Kbyte and 5Kbyte. Thus, traffic policing and shaping helps to optimize bandwidth utilization and to avoid frame loss of bandwidth.
format Thesis
author Samsuri, Rafidah
spellingShingle Samsuri, Rafidah
Rate limiting algorithm on video traffic with Pareto traffic model in broadband network / Rafidah Samsuri
author_facet Samsuri, Rafidah
author_sort Samsuri, Rafidah
title Rate limiting algorithm on video traffic with Pareto traffic model in broadband network / Rafidah Samsuri
title_short Rate limiting algorithm on video traffic with Pareto traffic model in broadband network / Rafidah Samsuri
title_full Rate limiting algorithm on video traffic with Pareto traffic model in broadband network / Rafidah Samsuri
title_fullStr Rate limiting algorithm on video traffic with Pareto traffic model in broadband network / Rafidah Samsuri
title_full_unstemmed Rate limiting algorithm on video traffic with Pareto traffic model in broadband network / Rafidah Samsuri
title_sort rate limiting algorithm on video traffic with pareto traffic model in broadband network / rafidah samsuri
publishDate 2016
url https://ir.uitm.edu.my/id/eprint/80486/1/80486.pdf
https://ir.uitm.edu.my/id/eprint/80486/
_version_ 1778165899677663232
score 13.211869

Similar Items