音频信号的数字水印技术

电子通信论文编号:TX087       论文字数:12763,页数:32

目  录
1绪论   1
1.1 数字水印   1
1.1.1 数字水印的概念   1
1.1.2 数字水印的特性   1
1.2 音频数字水印的概念和模型   1
1.3 音频水印的基本知识和相关参数  2
1.4 音频水印算法的要求  3
1.5 音频数字水印的典型应用    4
1.6 论文主要内容  4
2数字音频水印技术  5
2.1 数字音频的基本知识   5
2.1.1 音频信号的数字化   5
2.1.2 WAVE文件数据格式   5
2.1.3 音频信号的几个参数  6
2.1.4 声压与声压级  6
2.1.5 混响和回声  6
2.1.6 人类听觉系统模型   8
2.2 常见的数字音频水印技术  9
2.2.1 变换域算法  9
2.2.2 相位编码法   10
2.2.3 最低有效位(LSB)方法   11
2.2.4 回声隐藏算法    11
2.2.5 时间标度修改  12
2.3 小结数字音频水印的发展  13
3基于回声隐藏的音频数字水印技术   14
3.1 回声隐藏的基本原理   14
3.2 回声隐藏的具体算法和实现   14
3.3 音频中水印信息的提取  15
3.4 系统仿真及分析   15
3.5改进的算法   19
3.6 攻击测试与分析  21
3.7 回声隐藏算法小结   22
3.8 仿真实验中用到的部分Matlab函数  23
4论文总结和展望  25
参考文献  26
致  谢  28

摘  要
多媒体技术和网络技术的广泛应用，使得对图像、音频、视频等多媒体内容的保护成为迫切需要解决的问题。数字水印技术作为版权保护的重要手段，正得到深入的研究。它包含了版权信息和著作权信息，保护数字产品合法的复制和传播，已成为学术界的研究热点。由于人类听觉系统比人类视觉系统更为敏感，数字音频水印比起数字图像和视频水印更面临挑战。
论文首先介绍了数字水印系统的概念，然后讨论了数字音频水印系统的特点、典型应用和要求，比较了经典的数字音频水印技术，如变换域算法、相位编码算法和最低有效位算法。接着论文介绍了音频信号的基本特征，详细论述了基于回声隐藏的数字音频水印技术的原理和实现方案。首先将二值的水印的图像转换成一维序列，接着将音频信号分帧，帧数是水印图像的长度。然后将音频信号分别与回声核卷积，完成水印嵌入过程。算法利用同态处理的复倒谱分析实现水印的提取。仿真实验证明该算法具有一定的鲁棒性。

关键词：版权保护，数字音频水印，人类听觉系统，回声隐藏，复倒谱，鲁棒性

Abstract
With the development of computer network and information, it is urgent to protect the digital products copyright. Digital watermarking as a copyright protection technology is developed in this condition. Digital watermarking contains the copyright information and authentication information, and protects legal copy and distribution of digital products, which has been the hot spot of international academia. Digital audio watermarking is especially challenging compared with digital video and image watermarking because the human auditory system (HAS) is far more sensitive than the human visual system (HVS).
The fundamental principles of digital watermarking technique are introduced in this paper. Then the demands and properties of digital audio watermarking system and classical watermarking technologies are discussed and compared，such as transformation domain, phase coding and the least significant bit algorithm. The basic acknowledges of the audio signals are also mentioned. Next, the digital audio watermark based on echo hiding technology are discussed in details. Firstly, the binary watermarking image is converted into a one-dimension sequence. Then the audio signal is enframed into several segments. The embedding process is correspondingly the convolution of the signals and the echo kernels. The complex cepstrum is used to extract the watermarking from the audio signals. Simulation experiences show that the algorithm is robust.

Keywords: copyright protection, digital audio watermarking, HVS, echo hiding, complex cepstrum, robustness