Project Overview & Problem Statement

Official Research Documentation

This project implements an innovative Hybrid Text Steganography System Using Bit Mapping & Invisible Characters, designed by Cyber Security researchers at the University of Science & Technology.

1. Background & Real-World Context

In contemporary communication networks, surveillance and eavesdropping are no longer theoretical concerns. High-profile incidents have exposed deep vulnerabilities in conventional messaging security:

  • September 2018 (Facebook Token Breach): Over 50 million access tokens were leaked, exposing private user correspondence directly to attackers.
  • CVE-2019-3568 (WhatsApp Remote Code Execution): Exploited by NSO Group to inject the infamous 'Pegasus' spyware into devices.
  • Internal Employee Abuse ('SnapLion'): Internal Snapchat tools demonstrated that privileged administrators can bypass encryption boundaries to access private media.

While standard encryption (like end-to-end messaging) shields content from plain text interception, ciphertext strings inherently arouse immediate suspicion under active monitoring. Text steganography offers a stealth alternative by concealing secrets inside natural daily text exchanges without alerting observers.

2. The Core Problem Statement

Text is mathematically the most challenging medium for steganography. Unlike images, audio, or video, text has no redundant bits. The human eye detects even minute alterations to grammar, spacing, or character appearance immediately.

Prior text steganography approaches have suffered from severe trade-off bottlenecks:

  • Low Capacity: Hiding 1 byte required 4 to 8 characters of cover text (capacity ratios below 15%).
  • Format Fragility: Reliance on font attributes, line-shifts, or spaces resulted in complete data loss when copied and pasted into modern chat inputs.
  • Language Incompatibility: Zero-width characters often invert right-to-left (RTL) Arabic scripts or corrupt emoji renderers.
  • Weak or Missing Encryption: Obsolete algorithms (XOR constants, RC4, ICE) failed to withstand cryptographic cryptanalysis.

3. Project Objectives

STEGNOLINES resolves these historical limitations through five unified technical pillars:

  1. Invisible Embedding Engine: Mapping encrypted bitstreams to Unicode Variation Selectors (U+FE00–U+FE0F & U+E0100–U+E01EF), achieving 100% human imperceptibility.
  2. WASM Brotli Stream Compression: Halving secret payload volume to boost effective hiding capacity up to 100% of cover text size.
  3. Military-Grade Cryptography: AES-256-CTR zero-overhead stream cipher driven by PBKDF2-HMAC-SHA256 (100,000 iterations) with dynamic cover-bound salts.
  4. Cross-Platform Robustness: Full survival across major social and messaging networks (WhatsApp, Telegram, X, Facebook, Instagram, TikTok, YouTube, LinkedIn, Snapchat).
  5. Zero-Backend Decentralization: 100% client-side execution in the browser sandbox, eliminating server breach risks and Man-in-the-Middle (MITM) attacks.

4. System Scope & Boundaries

The system is strictly focused on Structural Text-Steganography. All operations occur in volatile browser memory without API dependencies. While Unicode Variation Selectors expand UTF-8 byte size by 2–4 bytes per selector, social media platforms enforce limits by glyph character count, enabling full seamless transmission.