Energy-Efficient Super-Resolution-Assisted Adaptive Video Streaming System Over Time-Varying Networks
IEEE Access · 2026 · 14 · 67009-67022
DOI: 10.1109/access.2026.3689637 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Access, 2026. Verfasst von Hyunmin Noh. Die Arbeit nennt 35 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Institute of Information & communications Technology Planning & Evaluation, Institute of Information & Communications Technology Planning & Evaluation(IITP)-Innovative Human Resource Development for Local Intellectualization program.
Abstract (Originalsprache)
This paper proposes an energy-efficient super-resolution (SR)-assisted adaptive video streaming system to provide high-quality streaming service over unstable time-varying networks. Unlike existing dynamic adaptive streaming (DASH) systems, the proposed system employs a dual-buffer architecture that separates the playback and upscale-wait buffer and selectively applies multiple SR models via a two-step scheduling process to overcome the limitations of conventional DASH and SR. First, the proposed system determines the requested segment bitrate, whether to apply SR, and which SR model to use, based on current network conditions, available playback time, and energy efficiency. Then, the downloaded segment is moved to either the playback or upscale-wait buffer, depending on whether SR is to be applied. Furthermore, to effectively respond to unexpected network fluctuations that occur during segment download, the system performs a dual-buffer rescheduling process. This process dynamically determines whether to apply SR and which SR model to use for already received segments within the two buffers. The proposed system is validated in simulated network environments using datasets collected from fast-moving vehicles. The experimental results demonstrate that the proposed system can provide energy-efficient video streaming while maintaining higher visual quality than baseline DASH systems.
Förderung
- Institute of Information & communications Technology Planning & Evaluation
- Institute of Information & Communications Technology Planning & Evaluation(IITP)-Innovative Human Resource Development for Local Intellectualization program
Zitieren
APA
Noh, H. (2026). Energy-Efficient Super-Resolution-Assisted Adaptive Video Streaming System Over Time-Varying Networks. IEEE Access, 14, 67009-67022. https://doi.org/10.1109/access.2026.3689637
Vancouver
Noh H. Energy-Efficient Super-Resolution-Assisted Adaptive Video Streaming System Over Time-Varying Networks. IEEE Access. 2026;14:67009-67022. doi:10.1109/access.2026.3689637
BibTeX
@article{Noh2026,
author = {Noh, Hyunmin},
title = {Energy-Efficient Super-Resolution-Assisted Adaptive Video Streaming System Over Time-Varying Networks},
journal = {IEEE Access},
year = {2026},
volume = {14},
pages = {67009--67022},
doi = {10.1109/access.2026.3689637}
}
EUVIMED