一种基于ITSEH的音频水印算法的研究与仿真

Research and Simulation on Audio Watermarking Using Improved Time Spread Echo Hiding

针对版权保护问题,提出了一种基于改进型时域扩展回声隐藏(Improved Time Spread Echo Hiding,ITSEH)的盲音频水印算法。运用混沌映射对水印序列加密生成水印信号,使用改进的PN(Pseudo Noise)序列构造幅频响应较小的回声核,基于心理声学模型实现水印嵌入强度的自适应,利用复倒谱分析中相关函数的峰值特性改进水印信号的提取方法,并通过混沌映射解密为表征版权的水印序列。理论分析和仿真表明改进算法提高了水印的不可感知性和鲁棒性。

This paper presented a blind audio watermarking algorithm based on Improved Time Spread Echo Hiding （ITSEH）. The application of chaotic mapping results in watermark. The improved Pseudo Noise （PN） sequence yields an echo kernel whose magnitude response has smaller magnitudes in the perceptually significant region. The psychoacoustics model ensures that the embedded watermark in the original audio is self-adaptive. The correlation function has more large peaks from which a modified decoding function was derived, then watermark bits were decrypted to derive the actual watermark sequence by utilizing chaotic mapping. Theoretical analysis and simulation examples illustrate that the algorithm improves both the imperceptibility and robustness of watermarks.