Explainable Geospatial AI for Satellite Ground Station Siting Using LiDAR-Derived Terrain Intelligence

arXiv cs.AI · 6 hours ago

Researchers developed a machine learning model using LiDAR data and geospatial features to predict representative clutter height for radio propagation analysis, replacing fixed height assignments from ITU standards. The LightGBM model achieved a mean absolute error of 1.79 meters and R² of 0.765, reducing error by more than 60% compared to the ITU-R P.452-18 baseline. The improvement enables better site selection and spectrum coordination for low Earth orbit ground stations by capturing within-class terrain variation missed by current practice.