Connected and Autonomous Vehicle Networks: A Comparative Analysis of Vehicular Cloudification and VANET Modes for Freeway Mobility Management

Connected and Autonomous Vehicles (CAVs) have emerged as a significant technological advancement in recent decades, introducing various automated features that contribute to enhanced safety and improved mobility within the transportation sector. However, managing the increasingly complex network of these vehicles requires innovative solutions that consider mobility and communications perspectives. One such solution is Vehicle Cloudification, which establishes virtual clouds with vehicle clusters leveraging infrastructure-aided communication to harness the computational capabilities of intelligent vehicles on roads. This paper offers a comparative analysis between two communication models for CAVs: vehicular ad-hoc network (VANET) and vehicular cloud. Two distinct cloud models were developed: stationary and dynamic platoon-based cloudification systems. To evaluate the efficacy of the developed vehicular cloud-based network management system, simulations are conducted using the AIMSUN microscopic traffic simulator, incorporating four communication modes: DSRC, C-V2X, 4G LTE, and 5G. The results reveal significant network operational improvement with vehicular cloudification, demonstrating reduced latency and packet loss ratios by 5.1% and 8.3% respectively, compared to VANET-based systems, while maintaining freeway mobility and safety levels. Furthermore, an incident management strategy is deployed, showcasing an 8% reduction in travel time delay (TTD) compared VANET. This study underscores the potential of connected vehicular cloudification in optimizing transportation networks, by mitigating risks associated with message interference and system overload, providing valuable evidence for policymakers to consider in decisions associated with policies and infrastructure investments.