CoRL: Collaborative Reinforcement Learning-Based MAC Protocol for IoT Networks

• DOI: 10.3390/electronics9010143
• Publication Date: 2020-01-13T09:05:51Z
• Authors: Taegyeom Lee; Ohyun Jo; Kyungseop Shin
• Keywords: MAC protocol; reinforcement learning; IoT networks; Q-learning; 0202 electrical engineering, electronic engineering, information engineering; 02 engineering and technology

Devices used in Internet of Things (IoT) networks continue to perform sensing, gathering, modifying, and forwarding data. Since IoT have a lot participants, mitigating reducing collisions among the participants becomes an essential requirement for Medium Access Control (MAC) protocols increase system performance. A collision occurs wireless channel when two or more nodes try access at same time. In this paper, reinforcement learning-based MAC protocol was proposed provide high throughput alleviate problem. collaboratively predicted Q-value update their value functions by using communications trial information other nodes. Our confirmed intensive level simulations that it can reduce convergence time 34.1% compared conventional Q-learning-based protocol.