Machine Learning at the Mobile Edge: The Case of Dynamic Adaptive Streaming over HTTP (DASH)Show others and affiliations
2022 (English)In: IEEE Transactions on Network and Service Management, E-ISSN 1932-4537, Vol. 19, no 4, p. 4779-4793Article in journal (Refereed) Published
Abstract [en]
Dynamic Adaptive Streaming over HTTP (DASH) is a standard for delivering video in segments and adapting each segment’s bitrate (quality), to adjust to changing and limited network bandwidth. We study segment prefetching, informed by machine learning predictions of bitrates of client segment requests, implemented at the network edge. We formulate this client segment request prediction problem as a supervised learning problem of predicting the bitrate of a client’s next segment request, in order to prefetch it at the mobile edge, with the objective of jointly improving the video streaming experience for the users and network bandwidth utilization for the service provider. The results of extensive evaluations showed a segment request prediction accuracy of close to 90% and reduced video segment access delay with a cache hit ratio of 58%, and reduced transport network load by lowering the backhaul link utilization by 60.91%.
Place, publisher, year, edition, pages
Institute of Electrical and Electronics Engineers Inc. , 2022. Vol. 19, no 4, p. 4779-4793
Keywords [en]
5G, Bandwidth, Bit rate, Caching, DASH, Machine Learning, Measurement, MEC, Prediction algorithms, Prefetching, Servers, Streaming media, Video Streaming, 5G mobile communication systems, Forecasting, HTTP, Mobile edge computing, Bit rates, Dynamic Adaptive Streaming over HTTP, Machine-learning, Streaming medium, Video-streaming
National Category
Computer Engineering
Identifiers
URN: urn:nbn:se:ri:diva-60201DOI: 10.1109/TNSM.2022.3193856Scopus ID: 2-s2.0-85135765992OAI: oai:DiVA.org:ri-60201DiVA, id: diva2:1698155
Note
This work has been funded by the EU’s Horizon 2020project 5G-CARMEN (grant no. 825012), by the H2020AI@Edge project (grant. no. 101015922), and by the SwedishFoundation for Strategic Research (SSF) Time Critical Cloudsproject (grant. no. RIT15-0075).
2022-09-222022-09-222024-07-04Bibliographically approved