Cognitive Radio-Enabled Internet of Things Communications: A Reinforcement Learning-Based Routing Method

Cognitive radio (CR) features allow for dynamic spectrum allocation, and IoT devices are increasingly finding usage in a variety of smart applications. Achieving faster data rates and minimizing end-to-end routing latencies in CR-enabled Internet of Things (IoT) communication are our aims in this effort to optimize throughput. In an Internet of Things (IoT) setting based on cognitive radio networks (CRNs), we suggest a routing method based on reinforcement learning (RL). With the goal of reducing EED and packet collisions, the channel choice decision capacity is proposed to be included into the network layer. By modeling the cognitive radio cognitive network (CRCN) simulators as a cognitive radio-enabled Internet of Things (CR-IoT) interactions environment, we compare the network performance of our proposed RL-IoT routing mechanism to that of the recently developed AODV-IoT and ML-based routing methods. Based on our testing, the RL-IoT model outperforms the competition in terms of throughput and packet delivery ratio.