Context-Aware Mobile Edge Computing in Vehicular Ad-Hoc Networks

Vehicular Ad-Hoc Networks (VANETs) are an integral part of Intelligent Transportation Systems (ITS) and provide a means for vehicles to communicate with one another as well as to the Cloud and Internet. The recent push toward 5G networking has laid the foundation for a VANET architecture that can accommodate the high bandwidth, high speed, and high availability expected by consumers. Next-Generation technologies will support future autonomous vehicle applications and define the infrastructure necessary to facilitate ubiquitous communications. With future networking technologies set to support much higher bandwidth and high numbers of users, our focus shifts to management of the compute resources that will support the back-end computations required for autonomous vehicle applications. The numerous on-board sensors expected to be available on autonomous vehicles, makes each vehicle an invaluable source of data. Gathering large amounts of data in real-time will incur a significant cost on networking infrastructure as well as the compute resources which process the data. In this paper, we introduce an architecture which seeks to execute computational tasks on the most logical and feasible resource. In our proposed technique, resource allocation takes into account the context of the job and the real-time availability of resources.