5G Signal-Based Localization – Path Loss & Ray Tracing

This repository contains two separate but related simulations used in a 5G signal-based localization project:

• path_loss_model.py: A path-loss optimization script using measurement data and building/vegetation geometries.
• ray_tracing_simulation.py: A 3D ray tracing setup using Sionna RT to simulate signal propagation paths.

IMPORTANT: Raw measurement data (.xlsx) and 3D object models (.obj) are not included due to privacy and competition restrictions.

---

Path_loss_model.py

An adaptive path-loss model that:

• Uses both geometric distance and Timing Advance (TA) data
• Automatically selects single or dual-slope models based on LOS/NLOS balance
• Performs grid search with Huber regression and multiple weighting strategies
• Outputs prediction results and residuals for evaluation

Requirements:

pip install pandas numpy matplotlib scipy tqdm trimesh pyproj

## ray_tracing_simulation.py
A minimal Sionna-RT setup that:

Loads ground, vegetation, building and base station meshes

Adds transmitters and receivers from Excel

Simulates ray paths and computes path loss

Visualizes path loss values for each receiver

Data Policy
All measurement files, geospatial Excel sheets, and .obj geometries used in this project are confidential and cannot be shared due to data restrictions.