Vehicle-Assisted Data Sensing in Vehicle Edge Metaverse: A Game Theory Approach

Quality of metaverse service depends on massive sensing data from real world. In this paper, we propose a vehicle-assisted data sensing framework in Vehicle Edge Metaverse (VEM) from an economic perspective, which incentivizes vehicles to contribute sensing data to metaverse service and thus improves the driving experience for vehicle users. We divide the data sensing problem among multiple operators and multiple vehicles into two subproblems which are the many-to-one matching subproblem and the resource pricing and power selection subproblem. Then, the two subproblems are transformed into game problems which can be modeled as Coalition Formation game (CFG) and Stackelberg game (SG), respectively. Moreover, we prove that the CFG game is Nash-stable and converges, and we analyze the SG game based on backward induction and prove that the game has a unique Nash equilibrium. Based on theoretical analysis, we propose an algorithm to solve these games, which aims to achieve an outcome that satisfies all participants through their strategic interactions. Numerical experiments show that the proposed algorithm exhibits good convergence and is able to improve the utilities of both vehicles and operators.