Resource Allocation in Software Defined Fog Vehicular Networks

Abstract

Vehicular network is an important application scenario of the fifth generation (5G) mobile communications. Due to the increasing number of vehicles and the users' various requirements, resource allocation problem in vehicular networks becomes more serious and has attracted researchers' attention. In this paper, we investigate the resource allocation in software defined fog vehicular networks where we formulate the problem as a mean-field game (MFG). We present a state space function by considering both interference factors and energy availability. Then the problem can be modeled as a cost minimization with the constraint of state space. Through the mean-field approximation method, we derive the corresponding Hamilton-Jacobi-Bellman (HJB) and Fokker-Planck-Kolmogorov (FPK) equations. A joint finite difference algorithm is proposed to solve the coupled HJB and FPK equations. The numerical results are presented to show the effectiveness of the proposed algorithm.