EUVIMEDEUVIMED entdecken
Crossref

Multi-Functional RIS-Assisted Task Offloading and Model Inference in Low-Altitude Networks With Hybrid RSMA-NOMA

Jie Zhang, Junjie Li, Liang Yang, Lexi Xu, Kai Sun, Qingqing Wu

IEEE Transactions on Communications · 2026 · 74 · 12591-12607

DOI: 10.1109/tcomm.2026.3721656 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Communications, 2026. Verfasst von Jie Zhang und 5 weiteren Autor:innen. Die Arbeit nennt 35 Literaturangaben. Gefördert von: National Natural Science Foundation of China, science and technology innovation Program of Hunan Province, Hainan Provincial Natural Science Foundation of China.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

As low-altitude economic applications continue to expand, traditional network paradigms are being reshaped, necessitating solutions for integrated communication, computation, and control among multiple nodes. This work focuses on minimizing energy consumption for partial task offloading and model inference assisted by the UAV in low-altitude networks (LANs), introducing multi-functional RIS (MF-RIS) to address terrestrial non-line-of-sight (NLoS) link issues. We decompose the non-convex optimization problem into multiple subproblems and propose a two-layer optimization framework for solution. In the upper layer, the block coordinate descent (BCD) is used to iteratively optimize the offloading ratio, computation frequency, power allocation, and RIS configuration. Meanwhile, in the lower layer, the convex decomposition (CD) based cross attention (CA) enhanced deep reinforcement learning (DRL) is applied to guide the UAV to learn the flight trajectory. Finally, the experimental results reveal the effectiveness of MF-RIS in link reconstruction when terrestrial links fail. Meanwhile, they verify that the convex decomposition based method can reduce the state space and action space for DRL, and the CA mechanism can enhance the understanding of network information by DRL algorithms.

Förderung

  • National Natural Science Foundation of China
  • science and technology innovation Program of Hunan Province
  • Hainan Provincial Natural Science Foundation of China
  • Natural Science Foundation of Inner Mongolia Autonomous Region
  • Program for Innovative Research Team in Universities of Inner Mongolia Autonomous Region

Zitieren

APA

Zhang, J., Li, J., Yang, L., Xu, L., Sun, K., & Wu, Q. (2026). Multi-Functional RIS-Assisted Task Offloading and Model Inference in Low-Altitude Networks With Hybrid RSMA-NOMA. IEEE Transactions on Communications, 74, 12591-12607. https://doi.org/10.1109/tcomm.2026.3721656

Vancouver

Zhang J, Li J, Yang L, Xu L, Sun K, Wu Q. Multi-Functional RIS-Assisted Task Offloading and Model Inference in Low-Altitude Networks With Hybrid RSMA-NOMA. IEEE Transactions on Communications. 2026;74:12591-12607. doi:10.1109/tcomm.2026.3721656

BibTeX
@article{Zhang2026,
  author = {Zhang, Jie and Li, Junjie and Yang, Liang and Xu, Lexi and Sun, Kai and Wu, Qingqing},
  title = {Multi-Functional RIS-Assisted Task Offloading and Model Inference in Low-Altitude Networks With Hybrid RSMA-NOMA},
  journal = {IEEE Transactions on Communications},
  year = {2026},
  volume = {74},
  pages = {12591--12607},
  doi = {10.1109/tcomm.2026.3721656}
}
Kostenlos starten →