Age of Information Minimization Through Reinforcement Learning in IoT Enabled Cognitive Radio Networks

The rapid growth of Internet of Things (IoT) applications has led to an unprecedented demand for efficient wireless communication, particularly in scenarios where devices compete for limited spectrum resources. Cognitive Radio Networks (CRNs) provide a promising solution by enabling dynamic spectrum access, yet maintaining timely information updates remains a challenge. Age of Information (AoI), a metric that measures the freshness of status updates, has become a key performance indicator in such networks. This paper proposes a reinforcement learning (RL)-based scheduling framework designed to minimize AoI in IoT-enabled CRNs. The framework models the dynamic environment of spectrum availability and device activity, and uses an adaptive RL agent to optimize transmission scheduling decisions in real time. A mathematical model of the AoI minimization problem is elaborated and after that, the proposed learning algorithm is designed. Large-scale simulations are done in Python in Google Colab and the performance of these simulations is compared with four state-of-the-art scheduling strategies. Findings demonstrate that the suggested RL-based system is able to generate lower AoI with different network scales, channel characteristics, and traffic dynamics. The results indicate the promise of RL-based solutions to the creation of scalable and reactive IoT communication systems.