A Game-Theoretic Approach Based Task Offloading and Resource Pricing Method for Idle Vehicle Devices Assisted VEC
Yishan Chen,Jie Wu,Junxiao Han,Hailiang Zhao,Shuiguang Deng
DOI: https://doi.org/10.1109/jiot.2024.3381183
IF: 10.6
2024-01-01
IEEE Internet of Things Journal
Abstract:Vehicle Edge Computing (VEC), as an emerging computing paradigm, aims to achieve the high efficiencies and quality of service by distributing computation tasks to vehicles and cloud-edge servers. The resource pricing problem focuses on how to reasonably price the resources of VEC to encourage their allocation and utilization. However, VEC server overloading may lead to performance degradation, especially in urban congested areas. Meanwhile, idle resources near VEC roads, such as parked vehicles and RSUs, are underutilized and can provide additional computation and communication resources to the system. Inspired by this, this paper introduces a model to assist vehicle edge computing by attracting Idle Vehicles (IVs) to share resources. We use a two-stage Stackelberg game model to address the resource pricing and task offloading problem, analyzing the interaction between requesting vehicles and cloud-edge servers. Through a backward induction method, we transform the problem into a convex optimization problem and theoretically prove the existence of a unique Nash equilibrium. In the first stage, optimal offloading ratio strategy is solved using convex optimization theory. In the second stage, the original problem is decomposed into 2N sub-problems and solved using the Lagrangian dual method and Karush-Kuhn-Tucker (KKT) conditions for optimal resource pricing. Additionally, a price incentive mechanism and a task-vehicle stable matching game model are employed to recruit idle vehicles around the roads to spontaneously participate in the task offloading process. Finally, simulation results reveal our solution effectively reduces offloading costs, latency, energy use, and enhances task completion compared to others.