Stackelberg Game-Based Optimization of Resource Allocation for IoV Edge-Cloud Cooperation

Resource allocation is the foundation of Internet of Vehicles in the edge cloud system. Based on that, a joint optimization problem is considered in a multi-layer vehicular architecture. To model the interaction between the vehicles and the edge servers, a framework of the Stackelberg game is introduced. The vehicles, which are viewed as distributed intelligent bodies, act as the followers and the edge server acts as the leader. However, due to the unlimited authorities, followers ignore the existence of the system, leading the excessive game. Therefore, to strengthen the authorities of the leader, a new offloading strategy matrix is proposed to replace the original offloading vector. Then, with the cooperation of the Chaos game, an improved Stackelberg game-based distributed optimization algorithm (ISGDOA) is proposed. Simulation verifies the efficient performances of ISGDOA and shows that besides the prevention of the excessive game, it still achieves an effective solution regardless of the performances of other sub-problems.