F3SNet: A Four-Step Strategy for QIM Steganalysis of Compressed Speech Based on Hierarchical Attention Network

TL;DR

F3SNet introduces a four-step hierarchical attention-based approach utilizing Bayesian networks to improve steganalysis of compressed speech, especially for small samples and low embedding rates.

Contribution

The paper presents F3SNet, a novel four-step strategy combining embedding, encoding, attention, and classification for enhanced QIM steganalysis of compressed speech.

Findings

F3SNet outperforms existing methods in detecting low embedding rate steganography.

The hierarchical attention mechanism effectively identifies impactful codewords.

Bayesian networks help model correlations between codewords for better analysis.

Abstract

Traditional machine learning-based steganalysis methods on compressed speech have achieved great success in the field of communication security. However, previous studies lacked mathematical description and modeling of the correlation between codewords, and there is still room for improvement in steganalysis for small-sized and low embedding rates sample. To deal with the challenge, We use Bayesian networks to measure different types of correlations between codewords in linear prediction code and present F3SNet -- a four-step strategy: Embedding, Encoding, Attention and Classification for quantizaition index modulation steganalysis of compressed speech based on Hierarchical Attention Network. Among them, Embedding converts codewords into high-density numerical vectors, Encoding uses the memory characteristics of LSTM to retain more information by distributing it among all its vectors and Attention further determines which vectors have a greater impact on the final classification result. To evaluate the performance of F3SNet, we make comprehensive comparison of F3SNet with existing steganography methods. Experimental results show that F3SNet surpasses the state-of-the-art methods, particularly for small-sized and low embedding rate samples.