Back to the future: Throughput prediction for cellular networks using radio KPIs

Simple item page
dc.contributor.authorRaca, Darijo
dc.contributor.authorZahran, Ahmed H.
dc.contributor.authorSreenan, Cormac J.
dc.contributor.authorSinha, Rakesh K.
dc.contributor.authorHalepovic, Emir
dc.contributor.authorJana, Rittwik
dc.contributor.authorGopalakrishnan, Vijay
dc.contributor.funderScience Foundation Irelanden
dc.date.accessioned2017-10-27T08:54:35Z
dc.date.available2017-10-27T08:54:35Z
dc.date.issued2017-10-16
dc.description.abstractThe availability of reliable predictions for cellular throughput would offer a fundamental change in the way applications are designed and operated. Numerous cellular applications, including video streaming and VoIP, embed logic that attempts to estimate achievable throughput and adapt their behaviour accordingly. We believe that providing applications with reliable predictions several seconds into the future would enable profoundly better adaptation decisions and dramatically benefit demanding applications like mobile virtual and augmented reality. The question we pose and seek to address is whether such reliable predictions are possible. We conduct a preliminary study of throughput prediction in a cellular environment using statistical machine learning techniques. An accurate prediction can be very challenging in large scale cellular environments because they are characterized by highly fluctuating channel conditions. Using simulations and real-world experiments, we study how prediction error varies as a function of prediction horizon, and granularity of available data. In particular, our simulation experiments show that the prediction error for mobile devices can be reduced significantly by combining measurements from the network with measurements from the end device. Our results indicate that it is possible to accurately predict achievable throughput up to 8 sec in the future where 50th percentile of all errors are less than 15% for mobile and 2% for static devices.en
dc.description.statusPeer revieweden
dc.description.urihttps://dl.acm.org/citation.cfm?id=3127882&picked=prox&CFID=999211294&CFTOKEN=83163054en
dc.description.versionAccepted Versionen
dc.format.mimetypeapplication/pdfen
dc.identifier.citationRaca, D., Zahran, A. H., Sreenan, C. J., Sinha, R. K., Halepovic, E., Jana, R. and Gopalakrishnan, V. (2017) 'Back to the Future: Throughput Prediction For Cellular Networks using Radio KPIs', HotWireless '17 Proceedings of the 4th ACM Workshop on Hot Topics in Wireless, Snowbird, Utah, USA, 16 October, New York: ACM, pp. 37-41. doi:10.1145/3127882.3127892en
dc.identifier.doi10.1145/3127882.3127892
dc.identifier.endpage41en
dc.identifier.isbn978-1-4503-5140-9
dc.identifier.journaltitleHotWireless '17 Proceedings of the 4th ACM Workshop on Hot Topics in Wirelessen
dc.identifier.startpage37en
dc.identifier.urihttps://hdl.handle.net/10468/4934
dc.language.isoenen
dc.publisherAssociation for Computing Machinery (ACM)en
dc.relation.projectinfo:eu-repo/grantAgreement/SFI/SFI Investigator Programme/13/IA/1892/IE/An Internet Infrastructure for Video Streaming Optimisation (iVID)/en
dc.rights© 2017 Association for Computing Machinery. This is the author's version of the work. It is posted here for your personal use. Not for redistribution. The definitive Version of Record was published in HotWireless '17 Proceedings of the 4th ACM Workshop on Hot Topics in Wireless, http://dx.doi.org/10.1145/3127882.3127892en
dc.subjectCelluar networken
dc.subjectThroughput guidanceen
dc.subjectMachine learningen
dc.subjectNetworksen
dc.subjectNetwork performance evaluationen
dc.subjectNetwork typesen
dc.subjectMobile networksen
dc.subjectComputing methodologiesen
dc.titleBack to the future: Throughput prediction for cellular networks using radio KPIsen
dc.typeArticle (peer-reviewed)en
dc.typeConference itemen
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
4050.pdf
Size:
693.19 KB
Format:
Adobe Portable Document Format
Description:
Accepted version
Download
License bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
license.txt
Size:
2.71 KB
Format:
Item-specific license agreed upon to submission
Description:
Download
Collections
Computer Science - Conference Items