3D Urban UAV Relay Placement: Linear Complexity Algorithm and Analysis
3D Urban UAV Relay Placement: Linear Complexity Algorithm and Analysis
复制标题
DOI：
10.1109/twc.2021.3066429
复制
发表时间：
2021-08
影响因子：
10.4
通讯作者：
Junting Chen;U. Mitra;D. Gesbert
中科院分区：
文献类型：
--
作者：
Junting Chen;U. Mitra;D. Gesbert
本文研究了在 3 维 (3D) 空间中的最佳无人机 (UAV) 放置，以在基站 (BS) 和地面用户之间建立连接。一个关键的挑战是避免由于障碍物而导致信号传播阻塞。先前的工作大多使用概率地形模型，其模型参数是从大范围的统计中学习到的，因此，针对小局部区域中的特定用户的优化效果较差。相比之下，本文在实际和细粒度地形上寻求最佳无人机位置，并开发适应飞行中测量的位置相关视距（LOS）条件程度的高效无人机定位策略。事实证明，可以根据所提出的搜索轨迹确定 3D 中的全局最佳无人机位置，该搜索轨迹仅具有目标区域直径的线性长度。因此，所提出的策略是可以实际实施的。数值实验是在现实世界的城市拓扑上进行的，并证明了与基于概率模型的现有策略相比具有卓越的性能增益。
Optimal unmanned aerial vehicle (UAV) placement in a 3-dimensional (3D) space to build a connection between a base station (BS) and a ground user is studied herein. A key challenge is to avoid signal propagation blockage due to obstacles. Much prior work uses probabilistic terrain models with model parameters learned from the statistics over a large area, and therefore, the optimization for a specific user in a small local area is poor. In contrast, this paper seeks the optimal UAV position over actual and fine-grained terrain, and develops efficient UAV positioning strategy adaptive to the degree of location-dependent line-of-sight (LOS) condition measured on the fly. It is proven that the globally optimal UAV position in 3D can be determined from the proposed search trajectory which has merely linear length in the diameter of the target area. Therefore, the proposed strategy can be practically implemented. Numerical experiments are performed over a real-world urban topology and demonstrate superior performance gain over existing strategies based on probabilistic models.