码分复用数字音频水印嵌入方案  

Digital Audio Watermarking Based on Code Division Multiplexing

作　　者：贾骏[1] 王朔中[1] 张新鹏[1] 

机构地区：[1]上海大学通信与信息工程学院,上海200072

出　　处：《上海交通大学学报》2004年第12期2073-2077,2081,共6页Journal of Shanghai Jiaotong University

基　　金：国家自然科学基金资助项目(60372090)

摘　　要：结合人耳听觉特性,用扩频方法在音频信号中的听觉不敏感区域嵌入多路数字水印,从而实现码分复用.在对嵌入信息的检测中,除了少量系统参数外,数字水印的提取不需要原始信号,并且毋需精确同步.实验表明,该水印具有很好的隐蔽性;尤其是对常见攻击,MP3编码具有良好的稳健性.This paper introduced a new method of audio watermarking based on the property of human auditory system (HAS) and spread spectrum technique. Some imperceptive spectral components are replaced with watermarks using code division multiplexing. Neither the availability of host signal nor precise synchronization is needed on the extraction side. The experimental results show that the proposed scheme can achieve high imperceptibility and satisfactory robustness against attacks, MP3 coding in particular.

关 键 词：码分复用 音频水印 扩频 心理听觉模型 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]