A robust audio steganography algorithm based on differential evolution

Abstract

Audio steganography is to hide secret information into the audio carrier and has become a research hotspot in the field of information hiding.Most of the existing studies focus on minimizing distortion at the expense of steganography capacity, and it is often difficult for them to extract secret information correctly after some common signal processing attacks.Therefore, based on the spread spectrum technology, firstly, the relationship between steganography parameters (i.e., segmented scaling parameters and steganography capacity) and imperceptibility as well as robustness was analyzed.Next, a multi-objective optimization model of audio steganography was presented, in which segmented scaling parameters and steganography capacity were decision variables, imperceptibility and steganography capacity were optimization objectives, and the signal-to-noise ratio was a constraint.Then, a robust audio steganography algorithm based on differential evolution was proposed, including the corresponding encoding, fitness function, crossover and mutation operators.Finally, comparative experimental results show that the proposed steganography algorithm can achieve better robustness against common signal processing attacks on the premise of ensuring imperceptibility and anti-detection.

Keywords

• audio steganography
• steganography parameters
• differential evolution
• robustness