EUVIMEDEUVIMED entdecken
StudienprotokollOpen Access · CC BY 4.0

A Resilient Routing Protocol to Reduce Update Cost by Unsupervised Learning and Deep Reinforcement Learning in Mobile Ad Hoc Networks

BeomKyu Suh, Ismatov Akobir, Jian Kim, YongBeom Park, Ki-Il Kim

Electronics · 03.01.2025 · 14 (1) · 166

DOI: 10.3390/electronics14010166 ↗

Kurz gefasst

Studienprotokoll in der Zeitschrift Electronics, online erschienen am 03.01.2025. Verfasst von BeomKyu Suh und 4 weiteren Autor:innen. Die Arbeit wurde bisher 13-mal zitiert (Crossref) und nennt 20 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Ministry of Agriculture, Food and Rural Affairs (MAFRA), Korean government MSIT.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Reinforcement Learning (RL)-based routing protocol has been proposed to establish paths in mobile ad hoc networks. However, due to the overhead of updating reward values according to frequent topology changes, existing protocols based on RL suffer from scalability problems with a large number of state and action spaces. To defeat this problem, in this paper, we propose a new resilient routing protocol by applying Unsupervised Learning (UL) prior to Deep Reinforcement Learning (DRL). In the former scheme, each node is clustered by mobility-resilient parameters. A reliable path that consists of only robust nodes in UL is decided by DRL with reasonable weight value through Multi-Objective Decision Making (MCDM). This approach leads to a reduction in update cost for reward value by excluding nodes that are considered severely affected by mobility. The comparative simulation results demonstrated that the proposed scheme outperformed the existing scheme in the aspects of Packet Delivery Ratio (PDR) and energy consumption. Our protocol demonstrates up to 35% higher PDR and reduces energy usage by approximately 20% under high-mobility conditions compared to Q-Learning-based protocols.

Förderung

  • Ministry of Agriculture, Food and Rural Affairs (MAFRA)
  • Korean government MSIT

Zitieren

APA

Suh, B., Akobir, I., Kim, J., Park, Y., & Kim, K. (2025). A Resilient Routing Protocol to Reduce Update Cost by Unsupervised Learning and Deep Reinforcement Learning in Mobile Ad Hoc Networks. Electronics, 14(1), 166. https://doi.org/10.3390/electronics14010166

Vancouver

Suh B, Akobir I, Kim J, Park Y, Kim K. A Resilient Routing Protocol to Reduce Update Cost by Unsupervised Learning and Deep Reinforcement Learning in Mobile Ad Hoc Networks. Electronics. 2025;14(1):166. doi:10.3390/electronics14010166

BibTeX
@article{Suh2025,
  author = {Suh, BeomKyu and Akobir, Ismatov and Kim, Jian and Park, YongBeom and Kim, Ki-Il},
  title = {A Resilient Routing Protocol to Reduce Update Cost by Unsupervised Learning and Deep Reinforcement Learning in Mobile Ad Hoc Networks},
  journal = {Electronics},
  year = {2025},
  volume = {14},
  number = {1},
  pages = {166},
  doi = {10.3390/electronics14010166}
}
Kostenlos starten →