Intelligent Traffic Management: SARSA Learning Approach for Congestion Control in VANETs

Abstract

VANETs (Vehicular Ad-hoc Networks) are an intelligent type of vehicle network that is utilized for successful vehicular communication. The primary goal of VANETs is to send basic safety messages (BSMs) to On-Board Units (OBUs) inside vehicles, allowing for improved network traffic monitoring. Vehicles can interact in a complicated, dynamically changing environment to ensure consistent delivery. The limited channel capacity and dynamic nature of the network might cause channel congestion in VANETs, posing the most significant issue in the research. We presented a SARSA-based framework for intelligently determining optimum transmission settings while adhering to recent channel circumstances. The necessary implementation results have been demonstrated, demonstrating that RL approaches provide an efficient solution for adaptive congestion control.