3D Deployment of UAVs Based on An Improved Genetic Algorithm

Unmanned Aerial Vehicles (UAVs) are useful tools for providing communication services nowadays, while the strategy of deploying UAVs varies depending on environments. In this paper, communication services assisted by UAVs in a dynamic urban environment is considered, and the problem is focused on how to deploy UAVs in a way that minimizes energy consumption taking into account path loss, signal-to-noise ratio, and users’ movement. In particular, a segmentation strategy for user movements is proposed. Based on this strategy, the user’s motion trajectory can be segmented, and the energy consumption can be optimized in real time under the premise of satisfying the average channel capacity. A genetic algorithm improved in the initialization and mutation crossover stages is proposed, where the reverse numerical method and adaptive probability factor to the dynamic UAV deployment problem are combined and adapted in an innovative way to solve the problem. Simulation results demonstrate the effectiveness of the improved genetic algorithm for achieving a lower energy cost and reasonable average channel capacity.