A Secure and Scalable Framework for Blockchain Based Edge Computation Offloading in Social Internet of Vehicles
Uzair Javaid, Biplab Sikdar
Abstract
The integration of Internet of Vehicles (IoV) with social networks has introduced Social IoV (SIoV) that will offer new applications in vehicular networks, e.g., personalized recommendations and route planning. This will be facilitated by heterogeneous access technologies and edge computing to offload tasks from vehicles via secure resource assignment. Thus, each vehicle in SIoV acts as a social subject that manages its own network. SIoV will lead to an explosive growth in network size, and induce issues like scalability and resource discovery. Blockchain is a potential candidate to address these, however, it is not suitable for SIoV with traditional proof-of-work (PoW) consensus. In this paper, we propose a framework that uses a dynamic PoW (dPoW) consensus with a checkpoint mechanism and a resource assignment policy. The dPoW consensus has different mining difficulty levels that change according to the communication traffic, whereas the checkpoint defines an alternative mechanism to generate the next block hash. The assignment policy manages an access control list to mandate the edge modules to securely distribute resources among vehicles. To study the feasibility of our framework, we present a formal security analysis using the Access Control Logic model. For the performance analysis, we use three metrics, i.e., scalability, latency, and security. With these analyses, we demonstrate that our framework offers enhanced security and can scale with a minimal increase in computation overhead. A case study with a comparative analysis is also discussed that evaluates the network dynamics and attests the superior performance of the framework under a real-life vehicular network scenario.