New methods for estimating/forecasting link bandwidths in 802.11b WLANs

Sridharan Muthuswamy, Ivan Marsic, A. Annamalai

Research output: Contribution to journalConference article

Abstract

Accurate assessment of the available link bandwidths in wireless networks is crucial for applications to negotiate, adapt and control quality-of-service parameters. In this paper we focus on new methods for estimating/predicting saturated throughputs (link bandwidths) in 801.11b wireless local area networks (WLANs). We propose the principle component analysis (PCA) and clustering approach for the estimation problem and the pattern modeling and recognition system (PMRS) and method of moment matching (MM) for the prediction problem. A trade off between the complexity and accuracy of the estimation/prediction algorithms is discussed and compared with the Neural Networks (NN) approach.

Original languageEnglish (US)
Pages (from-to)1163-1168
Number of pages6
JournalIEEE Vehicular Technology Conference
Volume60
Issue number2
StatePublished - Dec 1 2004
Event2004 IEEE 60th Vehicular Technology Conference, VTC2004-Fall: Wireless Technologies for Global Security - Los Angeles, CA, United States
Duration: Sep 26 2004Sep 29 2004

Fingerprint

Wireless Local Area Network (WLAN)
Wireless local area networks (WLAN)
Telecommunication links
Forecasting
Bandwidth
Moment Matching
Principle Component Analysis
Prediction
Method of Moments
Method of moments
Quality of Service
Wireless Networks
Wireless networks
Quality of service
Throughput
Trade-offs
Clustering
Neural Networks
Neural networks
Modeling

All Science Journal Classification (ASJC) codes

  • Computer Science Applications
  • Electrical and Electronic Engineering
  • Applied Mathematics

Cite this

Muthuswamy, Sridharan ; Marsic, Ivan ; Annamalai, A. / New methods for estimating/forecasting link bandwidths in 802.11b WLANs. In: IEEE Vehicular Technology Conference. 2004 ; Vol. 60, No. 2. pp. 1163-1168.
@article{0f61d96ae2774c4aa4546ebcef84696e,
title = "New methods for estimating/forecasting link bandwidths in 802.11b WLANs",
abstract = "Accurate assessment of the available link bandwidths in wireless networks is crucial for applications to negotiate, adapt and control quality-of-service parameters. In this paper we focus on new methods for estimating/predicting saturated throughputs (link bandwidths) in 801.11b wireless local area networks (WLANs). We propose the principle component analysis (PCA) and clustering approach for the estimation problem and the pattern modeling and recognition system (PMRS) and method of moment matching (MM) for the prediction problem. A trade off between the complexity and accuracy of the estimation/prediction algorithms is discussed and compared with the Neural Networks (NN) approach.",
author = "Sridharan Muthuswamy and Ivan Marsic and A. Annamalai",
year = "2004",
month = "12",
day = "1",
language = "English (US)",
volume = "60",
pages = "1163--1168",
journal = "IEEE Vehicular Technology Conference",
issn = "0740-0551",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "2",

}

New methods for estimating/forecasting link bandwidths in 802.11b WLANs. / Muthuswamy, Sridharan; Marsic, Ivan; Annamalai, A.

In: IEEE Vehicular Technology Conference, Vol. 60, No. 2, 01.12.2004, p. 1163-1168.

Research output: Contribution to journalConference article

TY - JOUR

T1 - New methods for estimating/forecasting link bandwidths in 802.11b WLANs

AU - Muthuswamy, Sridharan

AU - Marsic, Ivan

AU - Annamalai, A.

PY - 2004/12/1

Y1 - 2004/12/1

N2 - Accurate assessment of the available link bandwidths in wireless networks is crucial for applications to negotiate, adapt and control quality-of-service parameters. In this paper we focus on new methods for estimating/predicting saturated throughputs (link bandwidths) in 801.11b wireless local area networks (WLANs). We propose the principle component analysis (PCA) and clustering approach for the estimation problem and the pattern modeling and recognition system (PMRS) and method of moment matching (MM) for the prediction problem. A trade off between the complexity and accuracy of the estimation/prediction algorithms is discussed and compared with the Neural Networks (NN) approach.

AB - Accurate assessment of the available link bandwidths in wireless networks is crucial for applications to negotiate, adapt and control quality-of-service parameters. In this paper we focus on new methods for estimating/predicting saturated throughputs (link bandwidths) in 801.11b wireless local area networks (WLANs). We propose the principle component analysis (PCA) and clustering approach for the estimation problem and the pattern modeling and recognition system (PMRS) and method of moment matching (MM) for the prediction problem. A trade off between the complexity and accuracy of the estimation/prediction algorithms is discussed and compared with the Neural Networks (NN) approach.

UR - http://www.scopus.com/inward/record.url?scp=17144415961&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=17144415961&partnerID=8YFLogxK

M3 - Conference article

AN - SCOPUS:17144415961

VL - 60

SP - 1163

EP - 1168

JO - IEEE Vehicular Technology Conference

JF - IEEE Vehicular Technology Conference

SN - 0740-0551

IS - 2

ER -