Audio Watermarking for Covert Communication through Telephone System

In this paper, a novel sewice that uses audio watermarking for covert communication through telephone system is presented. This system is suitable for ISDN and PSTN networks and it can be modified for using in mobile systems. We use a Direct sequence spread spectrum algorithm with perceptual modeling of HAS for embedding watermark into the audio or speech signals. The watermark is robust against telephone channel attacks such as AWGN, Low pass filtering, DIA-A/D conversion, Downsampling down to 64 kbps and A-Law or pLaw conversion. This system is also robust against audio format conversion such as wave (windows PCM) format to mp3 (down to 64 kbps) compression.