Spatio-Temporal Service Analysis in Multi-Layer Non-Terrestrial Networks

In the advent of the 6G era, non-terrestrial networks (NTN) with expansive coverage are being increasingly recognized as a vital supplement to cellular networks for facilitating seamless communication. The intricate interplay between network performance and service quality necessitates a thorough investigation into the modeling and analysis of services for efficient construction of NTN. Previous studies on service analysis, predominantly focused on terrestrial networks, fall short in addressing the unique challenges posed by NTN, particularly those related to platform distribution and antenna gain modeling. This deficiency in research, coupled with the varying preferences of users for different network types, forms the basis of this study. This paper explores the spatio-temporal characteristics of services within a multi-layered NTN framework. In this context, the spatial distribution of the platforms is modeled using a binomial point process, and the antennas are characterized by a sectorized beam pattern. We derive the closed-form expressions for the association probability, the number of accessed users, and the arrival rate of services with certain delay requirements towards different types of NTN. Simulation results are provided to evaluate the influence of various parameters on the association probability, the number of accessed users, and the total arrival rate of services. The number of satellites can be determined to achieve the optimal system utility, balancing the accessed services, offloading effects, and launching costs. This initial investigation lays the groundwork for further theoretical progress in the service analysis and platform deployment of NTN.