Geostationary Earth Orbit (GEO) satellites, as a crucial infrastructure for
global communication and emergency support systems, are highly valued due to
their wide coverage, long communication distances, and large transmission
capacity. However, the scarcity of orbital slots and the difficulty in quickly
identifying available satellites in actual user scenarios have become the main
bottlenecks restricting the communication efficiency of multi-satellite relay
systems in geostationary orbit. To address these issues, this paper proposes an
accurate and efficient method for selecting the set of available satellites for
multi-satellite relay applications. We construct a spatial coverage model based
on the H3 hierarchical discrete global grid system and a geometric model that
characterizes the visible orbital arc of the user station. By comprehensively
considering the geographic locations of gateway stations and user stations,
satellite beam coverage boundaries, and the visibility constraints imposed by
user terminal antenna pointing, the proposed method can accurately screen out
qualified satellite sets that satisfy the end-to-end link condition of
“gateway–satellite–user station”. Simulation experiments are conducted in the
Ku-band to validate the effectiveness of the proposed scheme, with performance
metrics including coverage probability, accessible satellite quantity
distribution, and selection efficiency evaluated. The results show that the
H3-based modeling approach effectively reduces computational complexity while
ensuring high accuracy in visibility judgment. The proposed method significantly
improves the accuracy and efficiency of satellite selection, and supports
dynamic updates in mobile user scenarios. This work presents a reliable
satellite selection algorithm, which can be used to support efficient resource
scheduling and provide a strong guarantee for GEO multi-satellite relay
communication.