An SDN-aided low-latency live video streaming over HTTP


Ozcelik İ. M., Ersoy C.

Multimedia Tools and Applications, cilt.81, sa.16, ss.23145-23162, 2022 (SCI-Expanded) identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 81 Sayı: 16
  • Basım Tarihi: 2022
  • Doi Numarası: 10.1007/s11042-022-12389-y
  • Dergi Adı: Multimedia Tools and Applications
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, FRANCIS, ABI/INFORM, Applied Science & Technology Source, Compendex, Computer & Applied Sciences, INSPEC, zbMATH
  • Sayfa Sayıları: ss.23145-23162
  • Anahtar Kelimeler: Adaptive video bitrate control, CMAF, DASH, Live streaming, Low-latency, QoE
  • İstanbul Gelişim Üniversitesi Adresli: Hayır

Özet

Dynamic adaptive streaming over HTTP (DASH) is the crucial factor in the rapid penetration of over-the-top (OTT) service providers for on-demand video streaming. It can also be used for live video streaming by the OTT providers. The recent advancements of the HTTP chunked transfer, and the Common Media Application Format (CMAF) echo this tendency, which introduces the possibility to deliver a video segment by small chunks before the full segment is generated. It can deliver live latency of three seconds or less on a conventional DASH player with a small buffer capacity less than the target live latency. However, legacy bitrate adaptation mechanisms inaccurately measure the available bandwidth due to idle times between the chunks at the encoder side. To resolve this problem, we utilize the Software-Defined Networking (SDN) paradigm that directly provides the network statistics with the available bandwidth. We, then, propose an SDN-assisted bitrate adaptation mechanism for live streaming with HTTP 1.1 Chunked Transfer of CMAF packages while keeping the coexistence with the legacy DASH clients. Our SDN-based central framework asynchronously sends the video bitrate levels by continuously monitoring the background traffic flows and the available capacity for DASH clients on the same shared bottleneck link. Results show that our proposed mechanism achieves a lower video freeze rate and provides a better quality-of-experience while reducing the live latency down to about three seconds in the existence of varying background traffic.