EUVIMEDEUVIMED entdecken
Crossref

Energy-Efficient Link Adaptation for Underwater Acoustic Communications Based on Meta Deep Reinforcement Learning

Xiaowen Ye, Liqun Fu, Xianxin Song, Yi Wu

IEEE Internet of Things Journal · 15.11.2025 · 12 (22) · 46648-46660

DOI: 10.1109/jiot.2025.3608752 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Internet of Things Journal, erschienen am 15.11.2025. Verfasst von Xiaowen Ye und 3 weiteren Autor:innen. Die Arbeit wurde bisher 6-mal zitiert (Crossref) und nennt 43 Literaturangaben. Gefördert von: National Natural Science Foundation of China, National Social Science Fund of China, Fujian Province Special Fund Project for Promoting High-Quality Development of Marine and Fisheries Industries, China.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Due to the harsh channel conditions and operational difficulties in battery recharging, energy-efficient transmission is critical in underwater acoustic communications (UACs). This article investigates a new link adaptation technique for UACs that jointly optimizes transmission frequency, power, and rate to maximize energy efficiency. Conventional optimization-based approaches typically require real-time and perfect channel state information (CSI) and have high computational complexity, making them difficult to implement in realistic systems. To circumvent this problem, we put forth meta deep reinforcement learning transmission (MetaDT), a model-free link adaptation technique combining deep reinforcement learning (DRL) with meta-learning. To enable powerful reasoning and fast decision-making, we further propose a dueling echo state network (ESN) with separate output architecture for incorporation into MetaDT. Besides, to enable MetaDT to quickly adapt to diverse new/unseen environments, a low-complexity meta-learning is developed to find the optimal meta-parameters of the dueling ESN architecture. Numerical results show that compared with various benchmarks, MetaDT attains significant energy efficiency gains and is more robust against different transmission distances and numbers of multipaths. In comparison to conventional neural networks, dueling ESN shortens the run-time of MetaDT by more than 89.58% and is more efficient for temporal inference. In addition, we demonstrate the generalization capability of MetaDT with meta-learning to new/unseen environment configurations.

Förderung

  • National Natural Science Foundation of China
  • National Social Science Fund of China
  • Fujian Province Special Fund Project for Promoting High-Quality Development of Marine and Fisheries Industries, China
  • Fujian Provincial Direct Unit Special Project for Education and Research, Fujian Financial Referral (2023), China
  • Fujian Provincial Central Guiding Local Science and Technology Development Fund Project, China

Zitieren

APA

Ye, X., Fu, L., Song, X., & Wu, Y. (2025). Energy-Efficient Link Adaptation for Underwater Acoustic Communications Based on Meta Deep Reinforcement Learning. IEEE Internet of Things Journal, 12(22), 46648-46660. https://doi.org/10.1109/jiot.2025.3608752

Vancouver

Ye X, Fu L, Song X, Wu Y. Energy-Efficient Link Adaptation for Underwater Acoustic Communications Based on Meta Deep Reinforcement Learning. IEEE Internet of Things Journal. 2025;12(22):46648-46660. doi:10.1109/jiot.2025.3608752

BibTeX
@article{Ye2025,
  author = {Ye, Xiaowen and Fu, Liqun and Song, Xianxin and Wu, Yi},
  title = {Energy-Efficient Link Adaptation for Underwater Acoustic Communications Based on Meta Deep Reinforcement Learning},
  journal = {IEEE Internet of Things Journal},
  year = {2025},
  volume = {12},
  number = {22},
  pages = {46648--46660},
  doi = {10.1109/jiot.2025.3608752}
}
Kostenlos starten →