Reinforcement Learning Based Channel Selection for Design of Routing Protocol in Cognitive Radio Network

Cognitive Radio Network (CRN) is a next generation of wireless communication technology for efficient spectrum utilization. A cognitive Radio (CR) is able to recognize the idle spectrum. It solves the problem of spectrum scarcity in CRN. Due to intermittent channel usage by Primary Users (PUs), it is difficult to perform routing task in CRN. We are proposing a solution for optimal channel selection and routing in Cognitive Radio Network. Due to uncertainty in the number of active users, Monte Carlo method is performed for probabilistic outcome. During spectrum access process, Reinforcement Learning (RL) is applied to select best frequency band for routing. From the Simulation results, it is observed that the proposed routing protocol outperform in terms of throughput, packet delivery ratio, delay, dropping ratio & jitter compared to the routing protocol without machine learning assisted routing decision.