Spatiotemporal Modeling of Cellular-Connected UAV Networks with NOMA scheme

The integration of Unmanned Aerial Vehicles (UAVs), a new type of Aerial User (AU), into terrestrial cellular networks to form a cellular-connected UAV network has emerged as a significant research topic in the field of communications. However, the introduction of AUs competes for channel resources with Terrestrial Users (TUs), which can lead to a deterioration in the performance of TUs. Taking into account the channel differences between AUs and TUs, this paper utilizes a Non-Orthogonal Multiple Access (NOMA) scheme to enhance the spectral efficiency of the network. Utilizing tools from stochastic geometry and queueing theory, we propose a spatiotemporal analytical model to analyze the network performance from both spatial and temporal perspectives. Based on this framework, we first calculate the average queue length and average waiting time to quantify the system's temporal performance. Following this, we derive the successful transmission probability for both AUs and TUs, and subsequently obtain the average ergodic rate for the entire network. This comprehensive approach enables a thorough understanding of the network's performance across both spatial and temporal domains.