Multichannel audio steganography based on MPEG surround using direct sequence spread spectrum

Abstract

Audio steganography is a technique for embedding hidden message on the audio signal. Several techniques are currently available, proposed as methods to hide secret messages on audio signals and integrated into the audio encoding system. In this paper, a data hiding technique is proposed working based on MPEG Surround (MPS), a multichannel audio encoding standard that is very popular for spatial or three-dimensional (3D) audio coding. Direct sequence spread spectrum (DSSS) is integrated with MPS to embed a secret message into the downmix signal of multichannel audio that generated from MPS encoder. The result of experiments shows that the audio that produced by the proposed system is still in acceptable quality and signal synchronization can run smoothly. The mean value of the signal to noise ratio (SNR) of the audio signal is 16.67 dB. The secret message can be successfully extracted with a mean value of bit error rate (BER) of 3.38%, and normalized correlation (NC) is 95%.