Dynamic Online Double Auction Mechanism based on Deployment Constraints in the Internet of Vehicles

Abstract

With the rapid rise of Internet of vehicles applications, a large number of time delay sensitive tasks, such as autonomous driving and virtual reality, have emerged. These tasks require the mobile terminal to have a lower transmission delay to the server. Offloading tasks to adjacent edge servers is an effective way to reduce latency, and it is also a common deployment constraint. How to optimize the allocation of edge computing resources under this constraint is a major challenge. This paper proposes a truthful dynamic online double auction mechanism, different from the traditional double auction mechanism, this paper considers multiple heterogeneous edge server nodes, each server node acts as an independent service provider, and also considers the deployment constraints of vehicles on different edge servers, that is, vehicle users only offload tasks to adjacent edge servers, and in the execution time of the task, it needs to maintain a continuous connection with the server. Then, according to the supply-demand relationship of the market, a monotonic approximate algorithm is designed to determine the winner in polynomial time. In terms of pricing, a critical-valuebased pricing strategy is proposed. Simulation results verify the effectiveness of the mechanism.