音频DFT域量化水印性能研究  

Performance analysis of quantization audio watermarking in DFT domain

作　　者：康芳[1] 周俊 杨森斌[3] 

机构地区：[1]陕西银行学校,西安710065 [2]总参陆航研究所,北京101121 [3]西安通信学院基础部电子信息研究室,西安710106

出　　处：《计算机工程与应用》2009年第30期110-112,共3页Computer Engineering and Applications

摘　　要：为研究基于量化音频DFT域的水印算法性能,分析了该算法的原理,给出了一种简便的嵌入和提取公式,重点分析了算法的透明性、鲁棒性等性能,指出了算法的缺陷。首先,利用均方误差定义得出了嵌入失真公式,认为透明性与量化步长的平方成正比,而且当载体系数较小时容易引起听觉失真;其次,依据信息提取原则推导了加性高斯白噪声和幅度规整攻击下的误比特率,认为算法抗噪性能只与量化步长有关,而算法抵抗幅度规整攻击的性能很差;最后,通过仿真实验验证了以上公式的正确性。To explore the performances of Quantization Index Modulation（QIM） audio watermarking algorithm in Discrete Fourier Transform（DFT） domain,the principle of the algorithm is analyzed firstly,and then embedding and extracting formulas of less computational complexity are given.The properties of transparency and robustness based on DFT coefficients quantization are mainly analyzed,and its disadvantages are pointed out.Firstly,the embedding distortion is educed by using the definition of mean square error.The conclusions that transparency is in proportion to the square of quantization step and little DFT coefficient leads to bad transparency are reached. Secondly,according to the extracting principle, the BER under Additive White Gaussian Noise （AWGN） and amplitude scale assaults are deduced.The robustness to AWGN is only related to the quantization step and the algorithm is sensitivity to amplitude scale. Finally,these expressions are validated by experiments.

关 键 词：音频水印 量化索引调制 离散傅里叶变换 

分 类 号：TP309[自动化与计算机技术—计算机系统结构]