An audio digital watermarking algorithm based on QR decomposition and dither-modulation quantization is proposed. Arnold transform is utilized to scramble watermarking image in order to improve the security of the algorithm. Then chaotic sequence encrypts the watermarking which was folded before. DCT coefficients of the original audio signal are selected to embed the encrypted watermarking after QR decomposition. Dither-modulation quantization changes the largest R component values of the QR decomposition. Imperceptibility, security and robust test results show that the present method is more robust, and possesses good transparency and high security compared with the scheme based on SVD and dither-modulation quantization.
