A Wavelet-Based Steganographic Method for Text Hiding in an Audio Signal

Area of Science:

• Digital Signal Processing
• Information Security
• Steganography

Background:

• Steganography faces challenges from signal manipulation.
• Existing methods embedding data in high-frequency components are vulnerable to compression.
• Robustness against audio signal manipulation is crucial for secure data hiding.

Purpose of the Study:

• To develop a more robust steganographic method for audio signals.
• To increase the resilience of stego-systems against signal compression and distortion.
• To preserve the integrity of hidden text information despite unauthorized manipulations.

Main Methods:

• Utilizing a multilevel discrete wavelet transform.
• Employing adaptive block normalization for text information.
• Implementing recursive embedding in the low-frequency component of audio signals.
• Applying scalar product with Daubechies wavelet filters.

Main Results:

• The developed method recursively embeds data in the low-frequency component.
• Experimental results show a significant increase in the average power of hidden data.
• Achieved a compression ratio (CR) of 20, compared to 6 for existing high-frequency methods.
• Demonstrated a 3.3 times increase in stego-system resistance to audio signal compression.

Conclusions:

• Recursive embedding in the low-frequency component enhances data hiding robustness.
• The proposed wavelet-based steganography method offers superior resistance to signal compression.
• This technique effectively preserves text information integrity in manipulated audio signals.