Title: Reinforcement learning framework based on hybrid honey badger-cat swarm optimisation for media access control protocol in WSN

Authors: B. Ramesh; A. Rajani

Addresses: Department of ECE, CVR College of Engineering, Hyderabad, Telangana 501510, India; Department of ECE, JNTU Hyderabad, Kukatpally, Hyderabad, Telangana 500085, India ' Department of ECE, Jawaharlal Nehru Technological University, Hyderabad, Kukatpally, Hyderabad, Telangana 500085, India

Abstract: Adaptive models that modify a network's response over time are required for wireless networks. This paper establishes Parameter Optimised-Based Reinforcement Learning Media Access Control (PORL-MAC), a new MAC protocol for Wireless Sensor Networks (WSN) using a hybrid optimisation strategy. The latest protocols utilise adaptive duty cycles for later optimisation of energy utilisation. In this research, the nodes actively infer other node states by utilising an optimised reinforcement learning-based controlling mechanism to maximise throughput for a large number of traffic scenarios. In reinforcement learning, the optimisation of parameters in RL takes place by utilising the hybrid algorithm named Hybrid Honey Badger-Cat Swarm Optimisation (HHB-CSO). The experimental result indicates reduced computational complexity for practical applications in WSN. The throughput analysis is validated thus: it shows 12.5%, 15.8%, 3.7%, 7.07% and 3.29% better performance than DHOA-PORL-MAC, HHO-PORL-MAC, CSO-PORL-MAC, HBA-PORL-MAC and DQN-RL.

Keywords: wireless sensor network; media access control protocol; hybrid honey badge-cat swarm optimisation; parameter optimised based reinforcement learning media access control.

DOI: 10.1504/IJWMC.2025.147620

International Journal of Wireless and Mobile Computing, 2025 Vol.29 No.1, pp.68 - 82

Received: 05 Jan 2024
Accepted: 20 Jun 2024
Published online: 24 Jul 2025 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article