Paradox: Recursive Visual Entropy Key Derivation Engine (RVE-KDE)

1. Project Overview & Motivation

Paradox is an experimental cryptographic key derivation engine that maps high-dimensional visual media (images) to symmetric keys deterministically.

Why Visual Entropy?

Traditional key derivation functions (like PBKDF2 or bcrypt) process low-entropy, linear inputs (passwords). In contrast, images represent high-dimensional physical entropy source matrices. Paradox leverages this visual media matrix to establish a deterministic "visual factor" for key agreement, file archiving, and cover-medium key setup.

2. Core Architecture

The pipeline processes input visual data and nonces through seven discrete layers:

1. Initial Seed Generator: Formulates Seed0 = SHA3-512(ImageHash + Nonce + Timestamp + Version).
2. Recursive Walker: Slices coordinates modulo dimensions and traverses local adjacent neighbors.
3. Luminance & Contrast Extractor: Extracts color states, perceived brightness, and local grid standard deviations.
4. Hash Chain Evolved State: Formulates sequential hash linkages of coordinates, pixel bytes, and neighbor values.
5. Multi-Layer Recursion Manager: Runs dependent walks in sequence (Layer $n$ seed depends on Layer $n-1$'s final state) to diffuse spatial dependencies.
6. Entropy Pool Collector: Aggregates SHA3-256 slices from each step into a master entropy pool.
7. KDF Compressor: Squeezes the pool into 128/256/512-bit keys using HKDF-SHA256 or BLAKE3-KDF.

3. Installation & Developer Setup

Clone the repository, install Maturin/build dependencies, and build the Rust native extension in release mode:

python3 -m venv .venv
source .venv/bin/activate
pip install maturin patchelf
export PYO3_USE_ABI3_FORWARD_COMPATIBILITY=1
maturin develop --release

4. Quick Start & Examples

Text Encryption

import paradox

img = paradox.useImage("sample.png")

# Encrypts message and returns metadata envelope
encrypted = paradox.encrypt_text("Confidential message", image=img, security_level="low")

# Decrypts payload deterministically
decrypted = paradox.decrypt_text(encrypted, image=img)
print(decrypted) # "Confidential message"

File Encryption

import paradox

img = paradox.useImage("sample.png")
paradox.encrypt_file("doc.pdf", "doc.pdf.enc", image=img, security_level="low")
paradox.decrypt_file("doc.pdf.enc", "doc_dec.pdf", image=img)

5. API Reference

6. Experimental Validation Results

• Entropy Density: Derived keys exhibit a Shannon entropy of 7.99946 bits/byte (squeezed via HKDF-SHA256).
• Chi-Square Goodness-of-Fit: Passed ($p = 0.75 > 0.01$), confirming uniform byte distributions.
• Avalanche Effect: 50.09% mean bit difference following a 1-bit input offset.
• Collision Rate: 0.00% collision count over 3,000 keys.

For detailed plots and analysis, see newresults.md and changes.md.

7. Comparative Benchmark Metrics

Symmetric key (256-bit) latency and memory footprints compared to traditional standards:

8. Citation & Acknowledgements

If you use Paradox in your cryptographic research or academic publications, please cite it using:

@software{paradox_kdf2026,
  author = {Chirag Ferwani},
  title = {Paradox: Recursive Visual Entropy Key Derivation Engine},
  url = {https://github.com/chiragferwani/paradox},
  version = {1.0.2},
  year = {2026},
  doi = {10.5281/zenodo.20811708}
}

Acknowledgements

• Matplotlib and NumPy contributors for diagnostic utilities.
• The Google DeepMind pair-programming agent workspace.
• Established standard KDF authors (Argon2, PBKDF2, BLAKE3).

9. License

Licensed under the MIT License - see LICENSE for details.