Cellular-Connected UAV: Performance Analysis with 3D Antenna Modelling

The ever-increasing applications enabled by unmanned aerial vehicles (UAVs) calls for high-rate and ultra-reliable two-way wireless communication between UAVs and their associated ground entities. Integrating UAVs into cellular networks is regarded as a promising approach to meet such goals. However, given the unique UAV-ground channel characteristics and the fact that the existing cellular network has been mainly designed to serve ground users, it is not immediately clear whether the new aerial users can be supported in a similar manner as the ground users. In this paper, we provide the performance analysis for cellular-connected UAVs based on stochastic geometry, by studying the association probability and signal-to-interference-and-noise ratio (SINR) with two different association policies, namely nearest-distance based and maximum-power based associations. One striking difference of our work as compared to most existing analytical works is the use of the practical three dimensional (3D) antenna model at the base stations specified by 3GPP, which is important for more accurate performance analysis with UAV users of much higher altitude. Some useful observations are obtained based on our theoretical analysis and numerical results.