Abstract: An audio watermarking algorithm based on mean quantization and qualitative mapping wavelet is presented.And the Harr wavelet is attributed to the difference of two qualitative mapping.The main idea of the method is that a binary image is embeded into the original audio signals.In the preprocessing step,an image processing algorithm is used based on polarization identity toimprove the security of the method.According to the characteristics of human auditory system and using the wavelet induced by qualitative mapping,the watermarksignals are embedded into the treated lower coefficient.Simulation results showthat the algorithm is of good imperceptibility and robustness,and can resist various signal processing manipulations.

Key words: digital audio watermarking, qualitative mapping, discrete wavelet transform (DWT), mean quantization