Volume 18, Number 2

Audio Authentication using Spread Spectrum Steganography for Secure Communication and Content Ownership Protection

  Authors

Nur Izzatulain Jamain, Norhidayah Muhammad1, Nazirah Abd Hamid, Siti Dhalila Mohd Satar and Mohd Fadzil Abdul Kadir, Malaysia

  Abstract

With the current trend of protecting data and information, conventional encryption techniques aren't effective enough because information of interest can easily be detected. In this paper a robust audio steganography system based on Direct Sequence Spread Spectrum (DSSS) is proposed, which embeds DSSS-ch_used authentication identifier in the audio data. The DSSS technique spreads information over a large portion of the spectrum as opposed to traditional Least Significant Bit (LSB) directive that carries little robustness and high vulnerability to signal processing such as compression and filtering. It can "hide" as low power noise in the background, and it becomes statistically undetectable. In addition, with this research, an adaptive embedding mechanism using psychoacoustic masking technique is applied that dynamically changes the embedding strength so that the secret data is mathematically shaped under the human auditory threshold. Through experiments, a Peak Signal-to-Noise Ratio (PSNR) of 38.96 dB and a Signal-to-Noise Ratio (SNR) of 21.65 dB are obtained to acquire the high perceptual transparency. These results provide a secure ¡Èroot of trust¡É for digital forensics and voice channels including when converting between WAV and MP3 formats.

  Keywords

Audio Steganography, Spread Spectrum, DSSS, PSNR, Information Security.