EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Reinforcement Q-Learning-Based Adaptive Encryption Model for Cyberthreat Mitigation in Wireless Sensor Networks

Sreeja Balachandran Nair Premakumari iD, Gopikrishnan Sundaram iD, Marco Rivera iD, Patrick Wheeler iD, Ricardo E Pérez Guzmán iD

Sensors · 26.03.2025 · 25 (7) · 2056

DOI: 10.3390/s25072056 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 26.03.2025. Verfasst von Sreeja Balachandran Nair Premakumari und 4 weiteren Autor:innen. Beteiligte Einrichtung: Department of Information Technology, Karpagam College of Engineering, Myleripalayam Village, Coimbatore 641032, Tamil Nadu, India. Die Arbeit wurde bisher 26-mal zitiert (Crossref) und nennt 39 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Bill & Melinda Gates Foundation, National Research and Development Agency (ANID), National Council of Science and Technology (CONACYT).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The increasing prevalence of cyber threats in wireless sensor networks (WSNs) necessitates adaptive and efficient security mechanisms to ensure robust data transmission while addressing resource constraints. This paper proposes a reinforcement learning-based adaptive encryption framework that dynamically scales encryption levels based on real-time network conditions and threat classification. The proposed model leverages a deep learning-based anomaly detection system to classify network states into low, moderate, or high threat levels, which guides encryption policy selection. The framework integrates dynamic Q-learning for optimizing energy efficiency in low-threat conditions and double Q-learning for robust security adaptation in high-threat environments. A Hybrid Policy Derivation Algorithm is introduced to balance encryption complexity and computational overhead by dynamically switching between these learning models. The proposed system is formulated as a Markov Decision Process (MDP), where encryption level selection is driven by a reward function that optimizes the trade-off between energy efficiency and security robustness. The adaptive learning strategy employs an ϵ-greedy exploration-exploitation mechanism with an exponential decay rate to enhance convergence in dynamic WSN environments. The model also incorporates a dynamic hyperparameter tuning mechanism that optimally adjusts learning rates and exploration parameters based on real-time network feedback. Experimental evaluations conducted in a simulated WSN environment demonstrate the effectiveness of the proposed framework, achieving a 30.5% reduction in energy consumption, a 92.5% packet delivery ratio (PDR), and a 94% mitigation efficiency against multiple cyberattack scenarios, including DDoS, black-hole, and data injection attacks. Additionally, the framework reduces latency by 37% compared to conventional encryption techniques, ensuring minimal communication delays. These results highlight the scalability and adaptability of reinforcement learning-driven adaptive encryption in resource-constrained networks, paving the way for real-world deployment in next-generation IoT and WSN applications.

Schlagwörter

Stichwörter der Autor:innen

Q-learningadaptive encryptionenergy efficiencyreal-time threat detectionreinforcement learningresource-constrained networkssecurity optimizationwireless sensor networks

Autor:innen und Institutionen

  • Department of Information Technology, Karpagam College of Engineering, Myleripalayam Village, Coimbatore 641032, Tamil Nadu, India.
  • School of Computer Science and Engineering, VIT-AP University, Amaravati 522241, Andhra Pradesh, India.
  • Department of Electrical and Electronic Engineering, Faculty of Engineering, University of Nottingham, Nottingham NG7 2GT, UK.
  • Department of Computer Science, Faculty of Engineering, Universidad de Talca, Curicó 3460000, Chile.

ORCID-Profile: Sreeja Balachandran Nair Premakumari, Gopikrishnan Sundaram, Marco Rivera, Patrick Wheeler, Ricardo E Pérez Guzmán

Förderung

  • Bill & Melinda Gates Foundation
  • National Research and Development Agency (ANID)
  • National Council of Science and Technology (CONACYT)
  • University of Nottingham
  • British Council

Zitieren

APA

Premakumari, S. B. N., Sundaram, G., Rivera, M., Wheeler, P., & Guzmán, R. E. P. (2025). Reinforcement Q-Learning-Based Adaptive Encryption Model for Cyberthreat Mitigation in Wireless Sensor Networks. Sensors, 25(7), 2056. https://doi.org/10.3390/s25072056

Vancouver

Premakumari SBN, Sundaram G, Rivera M, Wheeler P, Guzmán REP. Reinforcement Q-Learning-Based Adaptive Encryption Model for Cyberthreat Mitigation in Wireless Sensor Networks. Sensors. 2025;25(7):2056. doi:10.3390/s25072056

BibTeX
@article{Premakumari2025,
  author = {Premakumari, Sreeja Balachandran Nair and Sundaram, Gopikrishnan and Rivera, Marco and Wheeler, Patrick and Guzmán, Ricardo E Pérez},
  title = {Reinforcement Q-Learning-Based Adaptive Encryption Model for Cyberthreat Mitigation in Wireless Sensor Networks},
  journal = {Sensors},
  year = {2025},
  volume = {25},
  number = {7},
  pages = {2056},
  doi = {10.3390/s25072056}
}
Kostenlos starten →