Adjusted ElGamal for Files

Cryptographic Implementation Analysis & Experimental Framework

This repository contains an enhanced implementation of the ElGamal public-key cryptosystem, specifically redesigned to support efficient file encryption, decryption, and large prime generation.
It provides a modular framework for evaluating different primality-testing algorithms, optimized key generation, file chunking, and multithreading techniques.

---

🚀 Introduction

Classical ElGamal suffers from limitations such as:

• Slow key generation
• Inefficient ciphertext expansion
• Inability to encrypt large files directly
• Slow encryption/decryption for large data

This project implements Adjusted ElGamal, improving both security and performance through:

• Fast large-prime generation (Miller–Rabin, AKS, Solovay–Strassen)
• Extended Euclidean Algorithm for faster key generation
• File block processing (chunk-based encryption)
• Multithreading to accelerate computation
• Preprocessing optimizations using ByteArrayIO and StringIO

This framework allows researchers to compare performance across multiple configurations.

---

📦 Installation Requirements

Before running the program, ensure your Python environment includes the following packages (all are standard libraries except DebugTimer):

math
random
time
logging
sys
pickle
hashlib
functools
queue
io
threading
DebugTimer # custom / external helper (included in project if provided)

No additional third-party dependencies are required.

---

▶️ How to Run

Run the full experimental pipeline:

python3 testAll.py

This will:

• Generate large primes using three different algorithms

• Run Adjusted ElGamal encryption & decryption

• Compare single-thread vs multithread performance

• Show the processed-data optimization results

📁 Project Structure & File Explanation

---

🔢 Part 1 — Large Prime Generation

Algorithm	File
Miller–Rabin	Large_Prime_Generation.py
AKS primality test	AKS_algorithm.py
Solovay–Strassen test	solovay_strassen.py

These modules allow consistent comparison of prime length, performance, and accuracy.

---

🔐 Part 2 — Adjusted ElGamal Encryption & Decryption

Component	Description	File
Original ElGamal + Large Prime	Baseline implementation	OurMethod.py
File Encryption (single thread)	Supports .txt and images	File_encryption.py
File Encryption (multithreading)	Enhanced speed, .txt only	improved_plus.py

Enhancements include:

• Chunk-based processing
• Fast modular inverse using Extended Euclid
• Multithreaded block-level encryption
• Data preprocessing (ByteArrayIO / StringIO) to minimize conversion cost

---

🧪 Test Suite

File	Description
testAll.py	Runs all examples, experiments, and comparisons

This is the recommended entry point.

---

📊 Experimental Summary

Based on our implementation:

---

🔢 Large Prime Generation Performance

Algorithm	Performance
Miller–Rabin	Fastest (≈309-digit prime in ~0.15s)
AKS	Deterministic but slower
Solovay–Strassen	Correct but extremely slow for large primes

---

⚡ Adjusted ElGamal Performance

Multithreading + data preprocessing significantly accelerates both encryption and decryption:

Method	Encryption Time	Decryption Time
Single-threaded	~10.5s	~8.0s
Multithreaded + Preprocessing	~5.7s	~5.8s

---

🧩 Techniques Used

• Modular arithmetic & primality testing
• Public-key cryptography
• Multithreaded task queues
• Python byte-stream optimization
• Performance benchmarking