Energy Efficiency Optimization for UAV Distribution and Resource Allocation in NOMA and Multi-UAV Assisted Wireless Networks

When natural disasters, such as earthquake, flood and forest fire, occur, it is important to deploy emergency wireless infrastructures to recover the destroyed communication system. Considering this fact, this paper studies the non-orthogonal multiple access (NOMA) and multiple unmanned aerial vehicles (UAVs) assisted emergency communication system, in which multiple UAVs act as air relays to help the uplink information transmission from ground users (GUs) to the base station. The K-means clustering algorithm is employed to cluster GUs into U groups. At the same time, each GU group is assigned a serving UAV based on improved particle swarm optimization (PSO) algorithm. A system energy efficiency (EE) maximization problem is then formulated by jointly optimizing the UAV space locations as well as the transmit powers of GUs and UAVs. To cope with the non-convex problem, we successfully decompose the original problem into two tractable sub-problems. For the sub-problem for determining the space locations of UAVs, the improved PSO algorithm is used to obtain the three dimensional locations. The sub-problem for resource allocation is solved by the successive convex approximation and Dinkelbach methods. Simulation results show that the proposed scheme can effectively improve the system EE, as compared to three baselines. At the same time, it is proved that optimizing the transmit power of GUs and UAVs has a significant impact on improving the system EE.