EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Blockchain and federated Q-learning-based secure, fault tolerant, and energy efficient framework for ad hoc networks

Righa Tandon iD, P K Gupta, Xiaochun Cheng

PLOS One · 18.03.2026 · 21 (3) · e0342008

DOI: 10.1371/journal.pone.0342008 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PLOS One, online erschienen am 18.03.2026. Verfasst von Righa Tandon und 2 weiteren Autor:innen. Beteiligte Einrichtung: Department of Computer Science and Engineering, Chitkara University Institute of Engineering and Technology, Chitkara University, Rajpura, Punjab, In…. Hauptthemen laut MEDLINE: Blockchain, Computer Communication Networks, Computer Security. Die Arbeit nennt 27 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Doctoral Training Centre at Swansea University, Swansea University.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

An ad hoc network plays a critical role in enabling communication in environments where deploying fixed infrastructure is impractical or infeasible. However, their dynamic topology and decentralized nature make them highly susceptible to failures and security threats. This paper proposes a robust and intelligent framework that addresses these challenges by integrating secure communication, fault tolerance, and energy efficiency. The proposed model makes use of blockchain technology to encourage trust between nodes, works with several nodes in the network to spot their unusual behavior, and uses federated Q-learning for adaptive threat response. The key components of the proposed framework, i.e., identity validation, trust scoring, distributed anomaly detection, and autonomous role management, make the system stable, robust, and energy-efficient. Simulation results of 500 nodes in a dynamic network show that the proposed model provides better performance in packet delivery, fewer false detections, and shorter recovery time in comparison to other systems. Furthermore, the proposed system holds significant promise for critical applications such as battlefield communication, disaster recovery, and remote monitoring, where reliable and secure networking is essential. The novelty of the work is the combination of a lightweight blockchain (MicroChain), Adaptive Cryptographic Engine (ACE), and federated Q-learning into one framework of ad hoc networks. The proposed framework provides high-security, effective resource usage, and responsiveness to the current network environment unlike earlier solutions, which focus on security, power usage, or fault tolerance separately.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

BlockchainComputer Communication NetworksComputer SecurityFederated LearningAlgorithmsComputer Simulation

Autor:innen und Institutionen

  • Department of Computer Science and Engineering, Chitkara University Institute of Engineering and Technology, Chitkara University, Rajpura, Punjab, India.
  • Department of Computer Science and Engineering, Jaypee University of Information Technology, Solan, Himachal Pradesh, India.
  • Department of Computer Science, Swansea University, Bay Campus, Fabian Way, Swansea, SA1 8EN, Wales, UK.

ORCID-Profile: Righa Tandon

Förderung

  • Doctoral Training Centre at Swansea University
  • Swansea University

Zitieren

APA

Tandon, R., Gupta, P. K., & Cheng, X. (2026). Blockchain and federated Q-learning-based secure, fault tolerant, and energy efficient framework for ad hoc networks. PLOS One, 21(3), e0342008. https://doi.org/10.1371/journal.pone.0342008

Vancouver

Tandon R, Gupta PK, Cheng X. Blockchain and federated Q-learning-based secure, fault tolerant, and energy efficient framework for ad hoc networks. PLOS One. 2026;21(3):e0342008. doi:10.1371/journal.pone.0342008

BibTeX
@article{Tandon2026,
  author = {Tandon, Righa and Gupta, P K and Cheng, Xiaochun},
  title = {Blockchain and federated Q-learning-based secure, fault tolerant, and energy efficient framework for ad hoc networks},
  journal = {PLOS One},
  year = {2026},
  volume = {21},
  number = {3},
  pages = {e0342008},
  doi = {10.1371/journal.pone.0342008}
}
Kostenlos starten →