Multiscale connection-level analysis of network traffic

Shriram Sarvotham; Rudolf Riedi; Richard Baraniuk

Multiscale connection-level analysis of network traffic

Shriram Sarvotham, Rudolf Riedi, Richard Baraniuk

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

Network traffic exhibits drastically different statistics, ranging from nearly Gaussian marginals and long range dependence at very large time scales to highly non-Gaussian marginals and multifractal scaling on small scales. This behavior can be explained by decomposing traffic into two components according to the connection bandwidth: the small bandwidth component absorbs most traffic and is Gaussian, while large bandwidth component constitutes virtually all of the small scale bursts. Based on this understanding, we propose a novel traffic model that parsimoniously accounts for user behavior, network topology, and the heterogeneous distribution of network bandwidths.

Original language	English (US)
Title of host publication	Conference Record of the Asilomar Conference on Signals, Systems and Computers
Editors	M.B. Matthews
Pages	29-33
Number of pages	5
Volume	1
State	Published - 2002
Event	The Thirty-Sixth Asilomar Conference on Signals Systems and Computers - Pacific Groove, CA, United States Duration: Nov 3 2002 → Nov 6 2002

Other

Other	The Thirty-Sixth Asilomar Conference on Signals Systems and Computers
Country	United States
City	Pacific Groove, CA
Period	11/3/02 → 11/6/02

ASJC Scopus subject areas

Electrical and Electronic Engineering
Signal Processing
Hardware and Architecture

Access to Document

Fingerprint Dive into the research topics of 'Multiscale connection-level analysis of network traffic'. Together they form a unique fingerprint.

Cite this

@inproceedings{f4eb12636aa842a5962a7704d94b80d8,

title = "Multiscale connection-level analysis of network traffic",

abstract = "Network traffic exhibits drastically different statistics, ranging from nearly Gaussian marginals and long range dependence at very large time scales to highly non-Gaussian marginals and multifractal scaling on small scales. This behavior can be explained by decomposing traffic into two components according to the connection bandwidth: the small bandwidth component absorbs most traffic and is Gaussian, while large bandwidth component constitutes virtually all of the small scale bursts. Based on this understanding, we propose a novel traffic model that parsimoniously accounts for user behavior, network topology, and the heterogeneous distribution of network bandwidths.",

author = "Shriram Sarvotham and Rudolf Riedi and Richard Baraniuk",

year = "2002",

language = "English (US)",

volume = "1",

pages = "29--33",

editor = "M.B. Matthews",

booktitle = "Conference Record of the Asilomar Conference on Signals, Systems and Computers",

note = "The Thirty-Sixth Asilomar Conference on Signals Systems and Computers ; Conference date: 03-11-2002 Through 06-11-2002",

}

TY - GEN

T1 - Multiscale connection-level analysis of network traffic

AU - Sarvotham, Shriram

AU - Riedi, Rudolf

AU - Baraniuk, Richard

PY - 2002

Y1 - 2002

N2 - Network traffic exhibits drastically different statistics, ranging from nearly Gaussian marginals and long range dependence at very large time scales to highly non-Gaussian marginals and multifractal scaling on small scales. This behavior can be explained by decomposing traffic into two components according to the connection bandwidth: the small bandwidth component absorbs most traffic and is Gaussian, while large bandwidth component constitutes virtually all of the small scale bursts. Based on this understanding, we propose a novel traffic model that parsimoniously accounts for user behavior, network topology, and the heterogeneous distribution of network bandwidths.

AB - Network traffic exhibits drastically different statistics, ranging from nearly Gaussian marginals and long range dependence at very large time scales to highly non-Gaussian marginals and multifractal scaling on small scales. This behavior can be explained by decomposing traffic into two components according to the connection bandwidth: the small bandwidth component absorbs most traffic and is Gaussian, while large bandwidth component constitutes virtually all of the small scale bursts. Based on this understanding, we propose a novel traffic model that parsimoniously accounts for user behavior, network topology, and the heterogeneous distribution of network bandwidths.

UR - http://www.scopus.com/inward/record.url?scp=0038643829&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0038643829&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:0038643829

VL - 1

SP - 29

EP - 33

BT - Conference Record of the Asilomar Conference on Signals, Systems and Computers

A2 - Matthews, M.B.

T2 - The Thirty-Sixth Asilomar Conference on Signals Systems and Computers

Y2 - 3 November 2002 through 6 November 2002

ER -