Cellular Automata denoising based Crypt Analysis - CACA.
This project combines cellular automata techniques with cryptanalysis, specifically using techniques from image denoising research to analyze patterns in encrypted data.
#######
CACA (Cellular Automata-based Cryptanalysis) is a project that explores the use of Cellular Automata (CA) techniques for potential cryptanalysis applications. The project adapts methods from image denoising research to analyze patterns in encrypted data.
This project combines two main approaches:
- NIST Statistical Test Suite (NIST STS) for evaluating randomness in bit sequences
- Cellular Automata (CA) for pattern detection and noise filtering
The idea is to test whether CA-based denoising techniques that are effective for recovering signals from noisy images (particularly salt-and-pepper noise in biometric applications) can also reveal patterns or weaknesses in encrypted data.
- Implementation of the NIST STS randomness testing suite (15 statistical tests)
- Various random number generator implementations for comparison
- Cellular Automata processing with Moore neighborhood rules
- Performance-optimized C++ implementation with multi-threading support
- Frequency (Monobit) Test
- Block Frequency Test
- Runs Test
- Longest Run Test
- Binary Matrix Rank Test
- Discrete Fourier Transform (Spectral) Test
- Non-overlapping Template Matching Test
- Overlapping Template Matching Test
- Maurer's Universal Statistical Test
- Linear Complexity Test
- Serial Test
- Approximate Entropy Test
- Cumulative Sums Test
- Random Excursions Test
- Random Excursions Variant Test
- Linear Congruential Generator (LCG)
- Blum-Blum-Shub (BBS)
- SHA-1 based generator
- XOR shift register
- Quadratic Congruential (two types)
- Cubic Congruential
- Micali-Schnorr
- Modular Exponentiation
The project is inspired by research showing that mirrored CA with Moore neighborhood (8-cell) rules can effectively denoise images, particularly salt-and-pepper noise in biometric applications. These CA techniques have shown superior performance compared to traditional methods, achieving better PSNR (Peak Signal-to-Noise Ratio), MSE (Mean Squared Error), and SSIM (Structural Similarity) metrics.
The hypothesis is that if encryption has minor imperfections, CA-based filtering might help reveal underlying patterns or biases.
Multi-threading support Configurable test parameters Efficient big number arithmetic implementations
Usage: << programName << [options] std::cerr << Options: std::cerr << -f, --file Input file to analyze std::cerr << -a, --ascii Treat input as ASCII (default: binary) std::cerr << -o, --output Output file prefix std::cerr << -g, --generator Test specific random number generator std::cerr << -G, --all-generators Test all available generators std::cerr << -l, --list List available generators std::cerr << -i, --iterations Number of CA iterations (default: 5) std::cerr << -L, --length Squence length for generator tests (default: 1000000) std::cerr << -r, --ca-rules <r1,r2> Comma-separated CA rules (default: 30,82,110,150) std::cerr << -v, --verbose Verbose output std::cerr << -h, --help Show this help message
:: Run the analysis on the file encrypted_data.bin with 10 iterations of CA processing.
caca_app.exe -f encrypted_data.bin -i 4
:: Analyze an encrypted binary file with default CA rules ./caca_app -f encrypted.bin
:: Analyze an ASCII file with specific CA rules ./caca_app -f input.txt -a -r 30,110
:: Test a specific random number generator ./caca_app -g 'Linear Congruential' -L 500000
:: Test all available generators ./caca_app -G
:: List available generators ./caca_app -l
This project is based on research from several key papers:
- Suresh A, et al. "An improved cellular automata (ca) based image denoising method for biometric applications." Biomedical Research, 2018.
- Rosin PL. "Image processing using 3-state cellular automata." Computer Vision Image Understanding, 2010.
- Various papers on the NIST Statistical Test Suite (NIST STS) for randomness testing
This project is open source and available under the MIT License. Feel free to use, modify, and distribute it as per the terms of the license.
Claude 3.7 Sonnet, Claude 3.5 Sonnet, Claude 3.5 Haiku, chatGPTo1, chatGPT4.5preview.