Improving Channel Utilization in VANETs Using Q-Learning-Based Data Rate Congestion Control

Vehicular Ad-Hoc Network(VANET) is an emerging wireless technology vital to the Intelligent Transportation System(ITS), which aims to mitigate traffic problems and improve road safety. Many VANET safety applications rely on the periodic broadcast of vehicle status information in the form of Basic Safety Messages (BSMs). When the vehicle density increases, the wireless channel faces congestion resulting in unreliable safety applications. Various decentralized congestion control algorithms have been proposed to effectively decrease channel congestion by controlling transmission parameters such as message rate, transmission power, and data rate. This paper proposes a data rate-based congestion control technique using the Q-Learning algorithm to maintain the channel load below the target threshold. The congestion problem is formulated as a Markov Decision Process (MDP) and solved using a Q-learning algorithm. The goal is to select the most appropriate data rate when transmitting a BSM such that the channel load remains at an acceptable level. Data obtained from a simulated dynamic traffic environment is used to train the Q-Learning algorithm. Our results indicate that the proposed algorithm is able to achieve the target channel load while reducing packet loss compared to existing data rate-based approaches.