Open-Data Cost Model

This repository contains the Open-Data Cost Model for Wireless Backhaul Networks associated with the homonym article submitted to the 18th International Conference on Network and Service Management, CNSM and "Next-Generation Wireless Backhaul Design for Rural Areas," in IEEE Transactions on Network and Service Management, vol. 22, no. 5, pp. 3921-3932, Oct. 2025.

The repository is organized as follows:

• data/ : contains the visibility graphs representing the areas of analysis together with the node attributes regarding the household and population distribution
• Sim.py : The main file for the generations of the WBN topologies
• Topology.py : the logic for the generation of the topology
• sim.yaml.example : the file containing the experimental variables
• results_110522/ : the folder containing the topologies generated for the CNSM paper
• analysis/ : the folder containing the Jupyter notebook implementing the Cost Model and the analysis of the topologies.
• processed/: the folder containing the processed results that are displayed in the paper's figures

How to replicate the results The WBN topologies generated for our research paper are available in the folder results_110522. You can either analyze those by copying them into the results folder or generate new ones by following the instruction in the next section.

To analyze the topologies you can execute the Jupyter notebook : analysis/analysis_fiber.ipynb

How to generate the WBN topologies

In order to generate the topologies, first, you need to install the python dependencies:

pip -r requirements.txt

Then, copy the sim.yaml.example file to sim.yaml, and adjust the content to your needs.

cp sim.yaml.exampe sim.yaml

Run the batch.sh file that will generate the topology for the parameters new set provided in sim.yaml:

./batch.sh

This will generate a set of topologies in the results/ folder.