Adaptive Selecting in Clustered LEO Systems: Direct or Cooperative Communication?

Satellite clustering has the potential to enhance inter-satellite cooperation, resist satellite malfunction, and enable more agile space-air-ground integrated applications. This paper investigates a promising model for clustered low Earth orbit (LEO) systems, in which one typical uncrewed aerial vehicle (UAV) can assist one satellite cluster to serve one random terrestrial user. Particularly, intra-cluster satellites can communicate user, while inter-cluster satellites are regarded as interference. Two types of satellites and users are randomly deployed at three visible spherical spaces by adopting three independent spherical Poisson point processes. In the modeling, an adaptive selecting mechanism is proposed to pick the strongest received signal between direct and cooperative transmissions. To facilitate a simpler analysis, we firstly transform the three spaces into the three planes through modifying their respective density. Next, assuming that the shadowed-Rician fading is employed in the satellite channel, two Gamma random variables are utilized to approximately express the aggregated power of interference and noise received by the UAV and user, respectively. Subsequently, the exact conditional user association and approximate Laplace transform of the accumulated signal power are derived to further investigate the conditional coverage probability. Finally, simulation results illustrate that: 1) Moderate satellite cluster sizes combined with a UAV altitude of about 200m are beneficial for achieving higher coverage probability; and 2) The adaptive selection mechanism generally achieves comparable or better performance than traditional transmissions by leveraging spatial diversity.