🧬 Genetic Cryptanalysis Project

📋 Overview

This project uses a Genetic Algorithm to break Vigenere cipher encryption. It evolves populations of candidate keys to find those that produce decrypted text closest to English language patterns, measured by letter frequency analysis.

🔍 Key Features:

• Parallel GA execution with realtime visualization
• Multiple crossover operators (Uniform & PMX)
• Live plotting with progress tables
• Automated result saving and analysis

🚀 Getting Started

📦 Requirements

• 🐍 Python 3.7+
• 📊 matplotlib
• 🔢 numpy

Install dependencies:

pip install matplotlib numpy

🎯 Runnable Files

All commands run from the project root directory.

src/main.py

Command: python src/main.py
Description: Runs parallel GA experiments with realtime plotting. Tests 5 parameter sets across uniform/PMX crossover and both datasets. Opens matplotlib window with live fitness curves and progress tables. Saves results to timestamped folder in results/.
Runtime: ⏱️ ~10-30 minutes depending on hardware.

src/test.py

Command: python src/test.py
Description: Runs a single GA experiment with easily customizable parameters at the top of the file. Shows realtime plotting of fitness evolution (best and average curves) plus a live table of current generation, best, and average fitness. Saves results and plot to results/Tests/single_[timestamp]/.
Runtime: ⏱️ ~1-5 minutes depending on parameters.

📁 Project Structure

genetic-cryptanalysis/
├── 📂 data/                    # Encrypted datasets
│   ├── Data1.txt              # 26-char key ciphertexts
│   └── Data2.txt              # 40-char key ciphertexts
├── 📂 src/                     # Source code
│   ├── main.py                # 🎯 Main experiments
│   └── util/                  # Utility modules
│       ├── evaluation.py      # 🔍 Fitness functions
│       ├── ga.py              # 🧬 GA implementation
│       └── rt_plot.py         # 📊 Realtime plotting
├── 📂 results/                 # Output files
│   └── experiment_*/          # Timestamped results
├── 📂 references/              # Assignment materials
└── .gitignore

📖 Usage

🧪 Running Experiments

Execute the main experiment suite:

python src/main.py

What happens:

• 🔄 Loads both datasets
• 🧬 Runs GA with 5 parameter sets (a-e)
• 🔀 Tests uniform and PMX crossover
• 📊 Displays realtime plots with fitness evolution
• 💾 Saves results to results/experiment_[timestamp]/

📊 Results Interpretation

🏆 Fitness Score

Lower values = better keys (closer to English frequencies)

📋 Parameter Sets

Set	Crossover Rate	Mutation Rate	Notes
a	1.0	0.0	High crossover, no mutation
b	1.0	0.1	High crossover, moderate mutation
c	0.9	0.0	Moderate crossover, no mutation
d	0.9	0.1	Moderate crossover, moderate mutation
e	0.6	0.1	⭐ Optimal (from search)

🔄 Crossover Types

• Uniform: Random gene swapping
• PMX: Preserves relative ordering

📄 Datasets

• Data1: Shorter 26-character keys
• Data2: Longer 40-character keys

📈 Sample Output

Experiment results include:

• 🏅 Best chromosomes (decrypted keys)
• 📈 Fitness evolution over generations
• 📊 Comparative performance tables
• 🖼️ Final plot PNGs

---

Built by Alaqmar Gandhi for Brock University (COSC 3P71)