EUVIMEDEUVIMED entdecken
Crossref

SuperABR: Super-Resolution-Assisted QoE Optimization for Mobile Adaptive Video Streaming With Queue-Driven Deep Reinforcement Learning

Yongyi Ran, Wenshu Huang, Tianyi Chen, Shuangwu Chen, Jiangtao Luo, Jian Yang

IEEE Transactions on Cognitive Communications and Networking · 2026 · 12 · 8506-8521

DOI: 10.1109/tccn.2026.3694941 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Cognitive Communications and Networking, 2026. Verfasst von Yongyi Ran und 5 weiteren Autor:innen. Die Arbeit nennt 42 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Science and Technology Research Program of Chongqing Municipal Education Commission.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

High and ultrahigh definition videos are becoming increasingly popular due to their superior visual and immersive experience, yet it demands substantially higher network bandwidth to accommodate the increased volume of video data. The existing widely-used adaptive bitrate approaches struggle to deliver high-quality videos in scenarios with relatively low network bandwidth. In this paper, we propose an edge super-resolution assisted mobile adaptive video streaming scheme, named SuperABR, to improve viewers’ Quality of Experience (QoE) while mitigating the influence of poor network conditions. First, SuperABR introduces Video Super-Resolution (VSR) into video transmission systems, and reconstructs the low-resolution videos into high-resolution ones at the edge server. Second, to optimize the video quality, rebuffering time and video quality jitter, we formulate the optimization model by fully considering the states of all caching queues in SuperABR and imposing probability constraints on the playback queue. Third, a queue-driven deep reinforcement learning (Q-DRL) algorithm is proposed to determine source transmission resolution (STR) for the video streaming server and VSR-reconstructed resolution (VRR) for the edge server. Finally, extensive experiments driven by real-world datasets are carried out to demonstrate that our proposed algorithm can improve viewers’ QoE by 16%–160% and 14%–176% in low-bandwidth and high-bandwidth scenario, respectively.

Förderung

  • National Natural Science Foundation of China
  • Science and Technology Research Program of Chongqing Municipal Education Commission

Zitieren

APA

Ran, Y., Huang, W., Chen, T., Chen, S., Luo, J., & Yang, J. (2026). SuperABR: Super-Resolution-Assisted QoE Optimization for Mobile Adaptive Video Streaming With Queue-Driven Deep Reinforcement Learning. IEEE Transactions on Cognitive Communications and Networking, 12, 8506-8521. https://doi.org/10.1109/tccn.2026.3694941

Vancouver

Ran Y, Huang W, Chen T, Chen S, Luo J, Yang J. SuperABR: Super-Resolution-Assisted QoE Optimization for Mobile Adaptive Video Streaming With Queue-Driven Deep Reinforcement Learning. IEEE Transactions on Cognitive Communications and Networking. 2026;12:8506-8521. doi:10.1109/tccn.2026.3694941

BibTeX
@article{Ran2026,
  author = {Ran, Yongyi and Huang, Wenshu and Chen, Tianyi and Chen, Shuangwu and Luo, Jiangtao and Yang, Jian},
  title = {SuperABR: Super-Resolution-Assisted QoE Optimization for Mobile Adaptive Video Streaming With Queue-Driven Deep Reinforcement Learning},
  journal = {IEEE Transactions on Cognitive Communications and Networking},
  year = {2026},
  volume = {12},
  pages = {8506--8521},
  doi = {10.1109/tccn.2026.3694941}
}
Kostenlos starten →