In this paper, we study the path planning for a cellular-connected unmanned
aerial vehicle (UAV) to minimize its flying distance from given initial to
final locations, while ensuring a target link quality in terms of the
large-scale channel gain with each of its associated ground base stations
(GBSs) during the flight. To this end, we propose the use of radio map that
provides the information on the large-scale channel gains between each GBS and
uniformly sampled locations on a three-dimensional (3D) grid over the region of
interest, which are assumed to be time-invariant due to the generally static
and large-size obstacles therein (e.g., buildings). Based on the given radio
maps of the GBSs, we first obtain the optimal UAV path by solving an equivalent
shortest path problem (SPP) in graph theory. To reduce the computation
complexity of the optimal solution, we further propose a grid quantization
method whereby the grid points in each GBS's radio map are more coarsely
sampled by exploiting the spatial channel correlation over neighboring grids.
Then, we solve the approximate SPP over the reduced-size radio map (graph) more
efficiently. Numerical results show that the proposed solutions can effectively
minimize the flying distance of the UAV subject to its communication quality
constraint. Moreover, a flexible trade-off between performance and complexity
can be achieved by adjusting the quantization ratio for the radio map.

本文针对基于蜂窝网络的无人机路径规划问题，提出了使用无线电信号映射技术和图论算法进行路径规划的方法，旨在最小化从起始点到终点的飞行距离，并同时保证在飞行过程中与地面基站之间的通信质量达到预期要求。该方法在降低计算复杂度的同时，能够有效地实现路径规划任务，且可以通过调整无线电信号映射的量化比率来权衡算法的性能和复杂度。