Stackelberg Game-Based Deployment Design and Radio Resource Allocation in Coordinated UAVs-Assisted Vehicular Communication Networks

Abstract

This paper studies optimum deployment design and radio resource allocation in a coordinated unmanned aerial vehicle (UAVs)-assisted vehicular communication network. The scarcity of radio resources and the lack of appropriate communication paths have been challenging in the existing vehicular networks. Although employing UAVs with well-designed deployments can support vehicular communications by providing line-of-sight (LoS) links, however, limited energy and spectrum scarcity issues motivate us to apply non-orthogonal multiple access (NOMA) with successive-interference-cancellation (SIC) as a solution to enhance radio resource efficiency for downlink infrastructure-to-vehicle (I2V) links. Meanwhile, due to the high complexity of the overall optimization problem that jointly handles deployment design and resource allocation for the studied UAVs-assisted vehicular network, we propose a low-complexity hierarchical suboptimal solution method. First, to efficiently find the optimum deployment of UAVs, we propose Stackelberg game with competitive nature as a game theory-based method, where UAVs are divided into leaders and followers groups that compete with each other to obtain the optimal position. Subsequently, we perform a low-complexity dynamic method in which radio resources are allocated in a sub-channel assignment algorithm and in a power allocation problem, respectively. Finally, the closed-form expressions of the optimal power allocation are derived using the KKT optimality conditions. Simulation results demonstrate improved performance for deployment design using the Stackelberg game. It is shown that NOMA improves the achievable sum rate compared to conventional orthogonal multiple access (OMA). Also, fast convergence of the proposed resource allocation and deployment design is obtained.