Vehicular Ad Hoc Networks (VANETs) play a crucial role in enabling intelligent transportation systems by facilitating real-time communication between vehicles and infrastructure; however, traditional VANET models face challenges such as high mobility, dynamic topology changes, network congestion, and increased latency, which reduce communication efficiency and reliability. To address these limitations, this paper proposes BEACON-V, a Behavior-Enhanced Artificial Intelligence-based communication framework designed to optimize vehicular network performance. The proposed system integrates machine learning techniques with behavioral analysis to improve communication decision-making in dynamic environments by continuously collecting real-time data such as vehicle speed, traffic density, and movement patterns. This data is processed using AI models to predict network conditions and dynamically optimize routing strategies, ensuring efficient data transmission and reduced packet loss. By incorporating behavior-aware communication, the system adapts to changing traffic scenarios and enhances overall network performance. Additionally, BEACON-V supports Vehicle-to-Vehicle (V2V) and Vehicle-to-Infrastructure (V2I) communication, enabling seamless coordination across the transportation ecosystem. Experimental evaluation shows that the proposed framework significantly improves key performance metrics such as latency, throughput, and packet delivery ratio when compared to conventional VANET approaches. Overall, BEACON-V provides a scalable, adaptive, and intelligent solution for next-generation vehicular communication systems, with future enhancements focusing on integration with edge computing, 5G networks, and autonomous vehicle technologies to further improve efficiency and real-world applicability.

Keywords- VANET, Artificial Intelligence, Vehicular Networks, Machine Learning, Intelligent Transportation Systems, Adaptive Communication