An Approach for Text Steganography Based on Markov Chains

TL;DR

This paper presents a Markov chain-based text steganography method that generates natural-looking texts for hiding information, improving upon simpler models that produce less realistic outputs.

Contribution

It introduces a novel Markov chain approach for text steganography that maintains language naturalness and reduces detectability compared to previous methods.

Findings

Produces more natural and less detectable steganographic texts

Maintains high fidelity to the original language model

Provides a reference implementation for practical use

Abstract

A text steganography method based on Markov chains is introduced, together with a reference implementation. This method allows for information hiding in texts that are automatically generated following a given Markov model. Other Markov - based systems of this kind rely on big simplifications of the language model to work, which produces less natural looking and more easily detectable texts. The method described here is designed to generate texts within a good approximation of the original language model provided.