mdh.sePublications
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Modeling, Implementation and Evaluation of IP Network Bandwidth Measurement Methods
Mälardalen University, Department of Computer Science and Electronics.
2007 (English)Doctoral thesis, comprehensive summary (Other scientific)
Abstract [en]

Internet has gained much popularity among the public since the mid 1990's and is now an integrated part of our society. A large range of high-speedbroadband providers and the development of new and more efficient Internet applications increase the possibilities to watch movies and live TV, use IP-telephony and share files over the Internet. Such applications demand high data transmission rates, which in turn consume network bandwidth. Since several users must share the common bandwidth capacity on the Internet, there will be locations in the network where the demand is higher than the capacity. This causes network congestion, which has negative impact on both the data transmission rate and transmission quality.

This thesis is about methods for measuring the available bandwidth of a network path between two computers. The available bandwidth can be interpreted as the maximum transfer rate possible without causing congestion. By deploying the methods studied in this thesis the available bandwidth can be measured without previous knowledge of the network topology. When an estimate of the available bandwidth is obtained, the transfer rate when sending messages between computers can be set to the measured value in order to avoid congestion.

In the thesis an active end-to-end available bandwidth measurement method called "Bandwidth Available in Real Time" (BART for short) is evaluated. BART measures the available bandwidth by injecting probe packets into the network at a given rate and then analysing how this rate has changed on the receiving side. A Kalman filter is used to update the current estimate of the available bandwidth using the new measurement sample.

The focus of the thesis is on how methods, such as BART, function in wireless 802.11 networks, which are very popular in work as well as in home environments. Wireless networks have a different construction compared to many other types of networks and this can affect the accuracy of the measurement methods discussed in this thesis. The effects must be analyzed and understood in order to obtain accurate available bandwidth estimates. Since wireless links are often parts of the network path between a sender and a receiver on the Internet, it is important to study how these links affect the estimates of the available bandwidth.

Place, publisher, year, edition, pages
Institutionen för datavetenskap och elektronik , 2007. , p. 170
Series
Mälardalen University Press Dissertations, ISSN 1651-4238 ; 42
Keywords [en]
Computer networks, Bandwidth measurement, Wireless networks
National Category
Computer Sciences
Research subject
Datavetenskap
Identifiers
URN: urn:nbn:se:mdh:diva-195ISBN: 978-91-85485-41-3 (print)OAI: oai:DiVA.org:mdh-195DiVA, id: diva2:120641
Public defence
2007-04-27, Gamma, U, Rosenhill, Västerås, 13:15
Opponent
Supervisors
Available from: 2007-03-16 Created: 2007-03-16 Last updated: 2018-01-13
List of papers
1. On the Analysis of Packet-Train Probing Schemes
Open this publication in new window or tab >>On the Analysis of Packet-Train Probing Schemes
2004 (English)Conference paper, Published paper (Other academic)
Abstract [en]

With a better understanding of how probe packets and cross-traffic packets interact with each other, more accurate measurement methods based on active probing can be developed. Several existing measurement methods rely on packet-train probing schemes. In this article, we study and describe the interactions between probe packets and cross-traffic packets. When one packet within a packet train is delayed, the dispersion (i.e. packet separation) of at least two (and possibly more) probe packets will change. Furthermore, the dispersions are not independent, which may bias calculations based on statistical operations. Many methods use dispersion averages, such as the mean, in the calculation of bandwidth estimates and predictions. We describe cross traffic effects on packet trains. The interaction results in mirror, chain and quantifi- cation patterns. Experiments have been performed in a testbed to explore these patterns. In histograms of delay variations for adjacent probe packets, these patterns are manifested as different identifiable signatures. Finally, we also discuss the effect of these patterns on the mean and median operations.

National Category
Computer Engineering
Identifiers
urn:nbn:se:mdh:diva-4071 (URN)1-932415-36-X (ISBN)
Conference
International Conference on Communications in Computing, CIC'04, June 21-24, 2004, Las Vegas, Nevada, USA
Available from: 2007-03-16 Created: 2007-03-16 Last updated: 2018-01-13Bibliographically approved
2. An Analysis of Active End-to-end Bandwidth Measurement Methods in Wireless Networks
Open this publication in new window or tab >>An Analysis of Active End-to-end Bandwidth Measurement Methods in Wireless Networks
2006 (English)In: E2EMON 06: 4th IEEE/IFIP Workshop on End-to-End Monitoring Techniques and Services: REAL-TIME MONITORING OF INTERNET PATHS, 2006, p. 75-82Conference paper, Published paper (Other academic)
Abstract [en]

For active, probing-based bandwidth measurements performed on top of the unifying IP layer, it may seem reasonable to expect the measurement problem in wireless networks to be no different than the one in wired networks. However, in networks with 802.11 wireless bottleneck links we show that this is not the case. The results from the experiments presented in this paper show that the measured available bandwidth is dependent on the probe packet size (contrary to what is observed in wired networks). Another equally important finding is that the measured link capacity, using the well known TOPP model, is dependent on the probe packet size and on the cross-traffic intensity. The underlying reasons for the observed differences are analyzed by incorporating the characteristics of 802.11 wireless networks into the TOPP model. The extended model is applicable to other end-to-end bandwidth measurement methods as well, such as BART, Pathload and PTR.

National Category
Engineering and Technology
Identifiers
urn:nbn:se:mdh:diva-4072 (URN)10.1109/E2EMON.2006.1651282 (DOI)000238288300010 ()978-142440145-1 (ISBN)
Conference
4th IEEE/IFIP Workshop on End-to-End Monitoring Techniques and Services, E2EMON; Vancouver; 3 April 2006 through 3 April 2006;
Available from: 2007-03-16 Created: 2007-03-16 Last updated: 2013-12-03Bibliographically approved
3. Real-time Measurement of End-to-End Available Bandwidth Using Kalman Filtering
Open this publication in new window or tab >>Real-time Measurement of End-to-End Available Bandwidth Using Kalman Filtering
Show others...
(English)Manuscript (Other academic)
Identifiers
urn:nbn:se:mdh:diva-4073 (URN)
Available from: 2007-03-16 Created: 2007-03-16 Last updated: 2013-12-03Bibliographically approved
4. Measuring the Impact of Active Probing on TCP
Open this publication in new window or tab >>Measuring the Impact of Active Probing on TCP
2006 (English)Conference paper, Published paper (Other academic)
Abstract [en]

Available bandwidth measurement methods have be- ome more and more a epted to be used when seeking the status of a network path. To measure the end-toend available bandwidth without a ess to the path routers, these methods inje t UDP based probe pa kets into the network path. The probe-pa ket load an transiently be high and thus it is important to study the impa t on the existing network ows. In this paper, we show and dis uss our simulation results on how the TCP ows are ae ted when inje ting probe pa kets with dierent ight patterns into the network path. We investigate the relation between the amount of inje ted probe pa kets and the redu tion in TCP performan e. Further, we suggest a quantitative denition of the term network friendly probing.

National Category
Computer Engineering
Identifiers
urn:nbn:se:mdh:diva-4074 (URN)
Conference
2006 International Symposium on Performance Evaluation of Computer and Telecommunication Systems (SPECTS'06), July 31-August 2, 2006 Calgary, Canada
Available from: 2007-03-16 Created: 2007-03-16 Last updated: 2018-01-13Bibliographically approved
5. On measuring the available bandwidth in wireless 802.11b networks
Open this publication in new window or tab >>On measuring the available bandwidth in wireless 802.11b networks
(English)Manuscript (Other academic)
National Category
Computer Engineering
Identifiers
urn:nbn:se:mdh:diva-4075 (URN)
Available from: 2007-03-16 Created: 2007-03-16 Last updated: 2018-01-13Bibliographically approved

Open Access in DiVA

fulltext(384 kB)1706 downloads
File information
File name FULLTEXT01.pdfFile size 384 kBChecksum SHA-1
2a263dce19e910d8f7f5fa19d6182767f4f3a5cf00f7fe005d481b43be2a0a652a96e04e
Type fulltextMimetype application/pdf

By organisation
Department of Computer Science and Electronics
Computer Sciences

Search outside of DiVA

GoogleGoogle Scholar
Total: 1706 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

isbn
urn-nbn

Altmetric score

isbn
urn-nbn
Total: 609 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf