提出了一种基于数字全息技术的盲音频水印算法．水印图像经随机相位调制形成物光波，该物光波的傅里叶变换与参考光波发生干涉，形成傅里叶变换全息图．再利用离散余弦变换的能量压缩能力，通过量化算法把降维的水印，即傅里叶变换全息图嵌入音频信号中，在水印提取过程中不需要原始音频信号的参与．仿真实验表明，该算法对低通滤波、噪音干扰特别是剪裁具有一定的稳健性．

A blind audio watermark embedding algorithm by a holographic technique was proposed. The digital-watermark image to be hidden was phase modulated in a random wise in order to form the object wave. The Fourier transform of the object wave was interfered with the reference wave to form the Fourier-transformed digital hologram. It made full use of the energy compression of DCT (Discrete Cosine Transform) to embed the watermark,in the form of Fourier-transformed digital hologram,into the audio signal by quantification process. The watermark can be extracted without the original audio signal. The process of constructing and reconstructing a digital hologram were described on the basis of the theory of Fourier optics and simulated by the computer. Experimental results show that the algorithm was robust against common signal manipulations and attacks,such as Gaussian noise,low pass filter and cropping especially.