On the performance boost for peer to peer WebRTC-based video streaming with network coding

Research output: Contribution to book/conference proceedings/anthology/reportConference contributionContributedpeer-review

Contributors

  • Patrik J. Braun - , Budapest University of Technology and Economics (Author)
  • Marton Sipos - , Budapest University of Technology and Economics (Author)
  • Peter Ekler - , Budapest University of Technology and Economics (Author)
  • Frank H.P. Fitzek - , Deutsche Telekom Chair of Communication Networks (Author)

Abstract

Video on Demand (VoD) streaming makes up an increasing portion of global IP traffic. Traditionally, data is served by content providers directly, putting a huge network load on the servers. We propose using our WebRTC-based protocols WebPeer and CodedWebPeer to create a P2P-assisted VoD streaming system with network coding. We introduce two metrics: network health to measure overall data saturation and network stability to show if the peers in the network are able to serve each other without the help of the server. We have implemented a testbed to emulate real-life network scenarios that includes a probabilistic model of peer behavior. We show through measurements that by applying network coding, network health is increased by up to 100% without changing the cache size or number of peers. Furthermore, network stability is achieved using up to half the cache compared to the uncoded approach, without increasing the servers load. Finally, to validate the scalability of our solution, we tested our protocols in a network with more than 100 tablets to deliver smooth video playback.

Details

Original languageEnglish
Title of host publication2017 IEEE International Conference on Communications, ICC 2017
EditorsMerouane Debbah, David Gesbert, Abdelhamid Mellouk
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (electronic)9781467389990
Publication statusPublished - 28 Jul 2017
Peer-reviewedYes

Publication series

SeriesIEEE International Conference on Communications
ISSN1550-3607

Conference

Title2017 IEEE International Conference on Communications, ICC 2017
Duration21 - 25 May 2017
CityParis
CountryFrance