True-African/Terrain-aware-campus-backhaul-planning

Research project for selecting antenna height and gain for a 5.8 GHz campus backhaul link using link-budget analysis, Radio Mobile simulations, and fade-margin validation. # Terrain-Aware Campus Backhaul Planning Dataset This project evaluates a 7.159 km point-to-point 5.8 GHz campus backhaul link between the Nyarugenge and Remera sites (in Rwanda). It uses Radio Mobile simulation outputs and a link-budget analysis script to compare antenna-height and antenna-gain configurations under fade-margin constraints. The repository is organized as a reproducibility package: raw simulation exports, processed datasets, analysis code, and generated figures are kept together so the results can be inspected or regenerated. ## Preprint Find full paper at: papers.ssrn.com ## Structure - analyze_radio_mobile_results.py: processes the Radio Mobile data, computes validation metrics, and regenerates the figures. - data/radio_mobile_expanded_results.csv: processed Radio Mobile results for all simulated height/gain scenarios. - data/radio_mobile_expanded_with_model.csv: processed results with link-budget model predictions added. - data/validation_metrics.csv: model error and feasibility-agreement metrics. - data/validation_summary.md: short text summary of the validation results. - data/raw_radio_mobile_html/: raw Radio Mobile HTML exports. - figures/: generated plots. - CODEBOOK.md: field descriptions for the main datasets. - requirements.txt: Python dependency list. ## Reproduce the Analysis Clone the repository and enter the project folder: ```bash git clone github.com cd Terrain-aware-campus-backhaul-planning ``` Create a virtual environment: ```bash python -m venv .venv ``` Activate the virtual environment. Windows PowerShell: ```powershell .\.venv\Scripts\Activate.ps1 ``` macOS/Linux: ```bash source .venv/bin/activate ``` Install the Python dependency: ```bash pip install -r requirements.txt ``` Run the analysis script: ```bash python analyze_radio_mobile_results.py ``` The script regenerates the model-augmented dataset, …