Radio Map-Based 3D Path Planning for Cellular-Connected UAV

In this paper, we study the three-dimensional (3D) path planning for a cellular-connected unmanned aerial vehicle (UAV) to minimize its flying distance from given initial final locations, while ensuring target link quality in terms of expected signal-to-interference-plus-noise ratio (SINR) at UAV receiver with each associated ground base stations (GBSs) during flight. To exploit location-dependent and spatially varying channel as well interference over 3D space, propose new radio map based framework UAV. Specifically, consider gain GBS that provides large-scale gains uniformly sampled locations on grid, which are due static large-size obstacles (e.g., buildings) thus assumed be time-invariant. Based maps GBSs their loading factors, then construct an SINR depicts levels locations. By leveraging obtained map, proceed derive optimal by solving equivalent shortest problem (SPP) graph theory. We further grid quantization approach where points more coarsely exploiting spatial channel/interference correlation neighboring grids. Then, solve approximate SPP reduced-size (graph) reduced complexity. Numerical results show proposed solution can effectively distance/time subject communication constraint, flexible trade-off between performance complexity achieved adjusting map. Moreover, significantly outperforms various benchmark schemes without fully distribution network.