Audio Digital Watermarking for Copyright Protection

When dealing with copyright protection for audio, watermarking can be considered to label the digital medium, by hiding information into the data. To make the watermark imperceptible and robust against some signal processing, we propose in this paper a new method based on frequential substitution, for which the bit error rate is lower than those obtained with existing methods. For such a purpose, frequential masking threshold is first estimated. Then a BPSK modulation is carried out with an adaptive carrier frequency that depends on the original signal features. To retrieve hidden information, a secret key is generated during the watermark process. Some of the watermarking models are also presented for allowing us to clearly establish our method in watermarking context.