Abstract
The current transportation architectures are heavily populated with smart devices and entities due to unfolded technological evolutions in Intelligent Transportation Systems (ITS). The ITS ecosystem, when introduced to Internet of Things (IoT) makes every object active and brings them online. Such devices generate data deluge that demand scalable storage and computational resources. Though centralized cloud-based solutions significantly circumvent those demands, but the current deployments still have silos and cease to meet the analytics and computational exigencies for such dynamic ITS subsystems. In this work, we investigate the current state of cloud-based solutions for fulfilling the mission-critical store and compute requirements of IoT-aided ITS architectures and revisit the motivations for adopting edge-centered fog computing paradigms. We also proposed a fog computing topology customized to ITS architectures. Further, the viability of proposed fog framework is demonstrated through a smart parking case study. Results show a significant improvement performance in terms of probabilistic QoS guarantees for private parking land owners, at the expense of a relatively small number of reserve premium spaces.
