Performance Analysis of Massive LEO Communication Networks Using Real Geographic Data

Abstract

This thesis investigates how massive low Earth orbit (LEO) satellite constellations can provide broadband connectivity along high latitude railway corridors, motivated by the limited coverage of terrestrial networks on remote routes and the need for reliable links for future mission-critical railway communication systems. A MATLABbased simulation framework was developed to integrate satellite orbital models with real geographic data, allowing for the analysis of arbitrary transport routes. The same framework can be extended in future work to evaluate LEO network performance across different traffc corridors, such as highways, shipping lanes, or aviation routes. In this work, the framework is applied to the Helsinki–Kolari railway route as a representative high latitude case study and used to evaluate key metrics such as coverage probability, elevation angle, signal-to-noise ratio (SNR), and handover dynamics. Three constellation scenarios are analyzed: a Starlink-like Walker-Delta shell, a OneWeb-like Walker-Star shell, and a custom design. Results from snapshot and time sweep simulations show that near polar constellations (e.g., OneWeb) achieve continuous coverage and stable link quality, while low inclination constellations (e.g., Starlink) suffer coverage gaps at northern latitudes despite strong SNR where visible. Parameter sweeps reveal that inclination strongly influences coverage continuity, altitude affects SNR and footprint size; and increasing satellite density beyond moderate levels yields diminishing returns while increasing handover complexity. Overall, the thesis highlights trade-offs between coverage, link margin, and mobility management, providing information for designing LEO constellations to support railway connectivity in high latitude regions.