Abstract
The evolution of Vehicular Ad Hoc Networks (VANETs) into data-centric ecosystems has generated unprecedented data volumes from sensors, vehicles, and roadside units. Managing, transmitting, and exploiting this vehicular big data efficiently are essential for realizing the vision of smart and connected mobility. This paper presents a comprehensive study of Big-Data-Driven Vehicular Networks (BDVNs)—an emerging paradigm that merges cloud computing, edge intelligence, and AI-based analytics to optimize communication, storage, and decision-making. We examine data-collection models, cloud/edge integration frameworks, and distributed learning approaches, emphasizing scalability, latency, and privacy. Comparative evaluation shows that hybrid cloud-edge architectures reduce end-to-end delay by ≈ 35 % and increase data-processing efficiency by ≈ 40 % relative to cloud-only baselines. Finally, we identify open research challenges in standardization, interoperability, and trust management, paving the way for autonomous big-data-enabled vehicular networks in the 6G era.