A Campus-Scale 5G NR Radio Propagation Dataset
Description
This dataset focuses on 5G NR radio propagation in the mountainous campus environment of the Chenggong Campus of Yunnan University. It primarily comprises drive-test measurements collected along campus roads and ray-tracing radio maps generated from a 3D environmental model. In addition to these two primary data categories, the dataset is supplemented by a DEM, a 3D building model, physical base-station locations, PCI-to-base-station associations, and matched measurement–simulation data. All data components share a common spatial reference and can be linked across files by spatial location, PCI, or physical base-station identifier.
Files
Steps to reproduce
The dataset was generated through a unified workflow including measurement preprocessing, spatial alignment, ray-tracing simulation, calibration, map generation, and quantitative evaluation. First, the 12 Cellular-Pro drive-test CSV files were processed to extract 5G NR measurements and were spatially aligned by converting WGS84 coordinates to EPSG:3857 and then to the local Blender coordinate system. Receiver height was set to DEM + 1.5 m. Next, the measurements were expanded into PCI–RSRP observations and split into nine calibration trajectories and three held-out evaluation trajectories. The calibration trajectories were used to estimate station parameters and generate per-station Sionna RT radio maps, a raw joint best-server map, and a calibration-adjusted joint map. Finally, matched measurement–simulation tables, provenance masks, and held-out evaluation summaries were exported using the accompanying code package.
Institutions
- Yunnan UniversityYunnan, Kunming