Downlink NOMA for Coexistence of Aerial and Terrestrial Users: Stochastic Geometry Analysis

Connecting aerial users (AU)s to cellular networks expands the potential of many applications. Nevertheless, existing cellular networks are not designed to efficiently serve AUs and terrestrial users (TU)s. Limited spectrum and high interference make the task even more challenging. Thus, in this paper, we examine the feasibility of fixed-power non-orthogonal multiple access (NOMA) scheme in meeting the demands of AUs and TUs by leveraging stochastic geometry. To this end, we propose a tractable framework for evaluating the coverage of AU and TU in cellular networks, where BSs are distributed using Poisson Point Process (PPP). We then derive the coverage probability of AU and TU. Using these analytical expressions, we further analyze the impact of different network parameters such as AU's altitude, TU's distance, and power allocation, and obtain key insights for designing an efficient NOMA scheme. Our results show that i) increasing AU's altitude does not always degrade the signal-to-interference ratio (SIR), ii) fixed-power NOMA scheme cannot be employed solely to serve AU's control and command (C&C) link, iii) pairing a near TU with a typical AU may prevent significant performance degradation, and iv) mitigating the interference of TU could improve the performance of AU.

I. Introduction

Connecting aerial users (AU)s to cellular networks, also known as cellular-connected UAV, has gained significant interest in recent years [1]. Such connections enable AUs to operate beyond-visual line-of-sight (LOS), and therefore enhances the potential of many drone-based applications. Nevertheless, existing cellular networks are primarily designed to serve terrestrial users (TU)s only. For instance, terrestrial base stations (BS)s usually tilt their antennas downward to enhance the desired signal and reduce the interference of TUs. This causes AUs to be served by the BS's sidelobe and therefore might suffer significant performance degradation [2].