GeoRoutingAMC

A Second‑Order Absorbing Markov Chain for Geographic Routing in Aeronautical Communication Networks

---

Overview

GeoRoutingAMC implements:

• Monte Carlo simulation of geographic greedy routing using k-hop neighborhood (Greedy‑k) across varying node equipage fractions
• Second‑order absorbing Markov chain model of success ratio and hop‑stretch factor for Greedy‑k
• Entropy and conditional entropy analyses of routing uncertainty
• Distance‑vs‑hop heatmaps relating distance and hop count to ground station
• Comparison of simulation vs. Markov‑model predictions

All compute‑intensive routines support HPC/Slurm.

---

Repository Layout

• code/ — Python scripts
• results/
  • csv_files/ — CSVs
  • figures/ — PDF plots
• README.md — Usage guide

---

Installation

# 1. Clone
git clone https://github.com/ComNetsHH/geo-routing-amc.git
cd geo-routing-amc

# 2. Create & activate Conda env
conda create -n georouting python=3.10 pandas=2.3.1 scipy=1.15.3 numpy=1.26.4 matplotlib=3.10.0 networkx=3.4.2 seaborn=0.13.2 scikit-learn=1.7.1 -y
conda activate georouting

# 3. Run the full analysis pipeline
python code/py_run_all.py