Revolutionizing Vanets with Graph Neural Networks using Dynamic Traffic Management

The increasing movement from rural areas to urban areas, along with the widening gap in population, has resulted in metropolitan areas becoming extremely overpopulated. As a result of the high volume of traffic that occurs in these areas, traffic monitoring is an extremely important activity. According to the findings of this study, an improved authentication and communication protocol that is based on clusters could be implemented for Intelligent Transportation Systems in Vehicular Ad Hoc Networks (VANETs). Our number one objective is to enhance the sharing of resources amongst vehicles through improved communication. Cluster-based routing protocols allowed us to increase the scalability, stability, and dependability of fast-moving VANETs. This was accomplished in the context of vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) communications. To easing concerns regarding privacy and safety, we arranged for the vehicles to be certified by an independent contractor. Through the utilization of Graph Neural Networks (GNNs), we can reduce the number of instances in which links fail, as well as minimize end-to-end (E2E) delays and route requests. Our approach has resulted in several important benefits, including enhancements to throughput, reductions in the amount of time required for TCP socket initialization, acceleration of TCP handshake response, and DNS lookup. Shortrange peer-to-peer wireless communication is the focus of the protocols that are used within a cluster that is 400 meters in radius. Utilizing new peer-to-peer wireless communications over VANET is what is meant by the term resource-conserving in this context. Within the framework of the suggested protocol secure authentication method, a certifying authority is responsible for the generation of a secure authentication key for the vehicle, which is subsequently provided to the vehicle.