Abstract
Vehicular ad-hoc Networks (VANETs) are an integral part of intelligent transportation systems (ITS) that facilitate communications between vehicles and the internet. More recently, VANET communications research has strayed from the antiquated DSRC standard and favored more modern cellular technologies, such as fifth generation (5G). The ability of cellular networks to serve highly mobile devices combined with the drastically increased capacity of 5G, would enable VANETs to accommodate large numbers of vehicles and support range of applications. The addition of thousands of new connected devices not only stresses the cellular networks, but also the computational and storage requirements supporting the applications and software of these devices. Autonomous vehicles, with numerous on-board sensors, are expected to generate large amounts of data that must be transmitted and processed. Realistically, on-board computing and storage resources of the vehicle cannot be expected to handle all data that will be generated over the vehicles lifetime. Cloud computing will be an essential technology in VANETs and will support the majority of computation and long-term data storage. However, the networking overhead and latency associated with remote cloud resources could prove detrimental to overall network performance. Edge computing seeks to reduce the overhead by placing computational resources nearer to the end users of the network. The geographical diversity and varied hardware configurations of resource in a edge-enabled network would require careful management to ensure efficient resource utilization. In this paper, we introduce an architecture which evaluates available resources in real-time and makes allocations to the most logical and feasible resource. We evaluate our approach mathematically with the use of a multi-criteria decision analysis algorithm and validate our results with experiments using a test-bed of cloud resources. Results demonstrate that an algorithmic ranking of physical resources matches very closely with experimental results and provides a means of delegating tasks to the best available resource.
Highlights
Advancement in autonomous vehicle technologies has created a demand for higher-bandwidth, improved availability, and ubiquitous networking technologies
We propose computationally-enabled RSUs (CERSUs) that act as the base station for V2I communication, but can provide limited computational resources
We introduced the CAMEVAN architecture for vehicular networks that supports edge-computing and employs a multiple-criteria decision analysis method to choose an optimal resource for delegation of computational tasks
Summary
Advancement in autonomous vehicle technologies has created a demand for higher-bandwidth, improved availability, and ubiquitous networking technologies. Given sufficient storage space at edge nodes, the network has the potential to cache the contents and ensure a timely delivery to end users This has added value for any location-based service that provides data to vehicles. Vehicular networks with appropriate infrastructure can utilize edge computing by off-loading some tasks to the RSUs, base stations, cell towers, or other infrastructure that provides connection to the network backbone. Vehicular networks can further leverage FC by employing powerful on-board computational hardware that is expected to reside on future vehicles The inclusion of such hardware in the network creates a new architecture that could contextually execute, store, or transmit any data in the most logical and efficient fashion. Creating a distinction between task types ensures availability of sufficient resources for time-critical jobs as they arise and allows delay-tolerant jobs to be completed as resources become available
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
