Reinforcement Learning for Optimized Resource Allocation in 5G URLLC

Abstract

With the rapid expansion of applications requiring ultra-reliable low-latency communication (URLLC) in 5G networks, optimizing resource allocation has become more critical than ever. In this paper, we propose an innovative approach that combines dual connectivity and reinforcement learning to enhance resource optimization in 5G URLLC scenarios. Leveraging the Open RAN (O-RAN) architecture, we integrate machine learning (ML) and artificial intelligence (AI) to intelligently manage radio access network (RAN) resources. Our dual connectivity setup involves a master gNB and a secondary gNB working collaboratively to boost performance and reliability by enabling simultaneous connections that reduce latency and enhance reliability. By employing Q-learning, a model-free reinforcement learning algorithm, we dynamically adapt resource allocation based on real-time network conditions, ensuring optimal throughput and minimal congestion. Simulation results reveal significant improvements in key performance metrics, demonstrating the effectiveness of our approach in meeting the stringent demands of 5 G URLLC applications.