3D Deployment of Multi-UAV for Energy-Saving: A Game-Based Learning Approach

In this paper, we investigate the problem of three-dimensional (3D) coverage deployment of unmanned aerial vehicles (UAVs). The downlink scenario of air-to-ground communication networks is considered, where UAVs are supposed to maximize the coverage utility of mission area with minimal transmission power. We decompose the problem into two sub-problems, which are UAV deployment and power allocation, respectively. In the first sub-problem, we adjust the 3D locations of UAVs to maximize the coverage utility. In the second sub-problem, we adjust the transmission power to maximize the energy efficiency. By constructing marginal utility function, we prove that these two problems are exact potential games (EPG) and exist Nash equilibrium (NE) points. After that, an algorithm based on binary log-linear learning is proposed to achieve the NE point. The simulation results verify the efficiency of the algorithm and the stability of the model.