适用于声学传输的鲁棒交叉扩频水印算法

Robust cross spread spectrum audio watermarking for acoustic transmission

为使音频水印技术更好地应用于移动互联网络中以实现广播电视节目与观众的实时互动,提出一种基于扩频水印技术(SS)的交叉扩频水印算法(CSS),利用音频信号交叉帧之间的高度相关性减小水印检测统计量的方差,降低检测错误率,进一步提出交叉扩频水印算法的改进算法(ICSS)以提高水印鲁棒性。理论分析与仿真结果表明,ICSS可明显降低原始音频信号对水印的干扰,对MP3压缩以及由Stirmark Benchmark引起的常见失真有较强的鲁棒性,能抵抗室内混响和环境噪声,适用于移动互联网络中的声学传输。

To improve the application of audio watermarking technology in mobile Internet and realize the real-time interaction between the radio or television programs and the audience,a cross spread audio watermarking algorithm(CSS)based on spread spectrum audio watermarking(SS)was proposed.The high correlation between the cross frames of the audio signal was used to decrease the variance of the decision statistic and lead to a small error probability.An improved cross spread audio watermarking algorithm(ICSS)was proposed to further improve the robustness of watermarking.Theoretical analysis and simulation results show that ICSS can significantly reduce the host interference.It is more robust not only to MP3 compression and a variety of ot-her distortions in Stirmark Benchmark for audio,but also to room reverberation and environmental noises.It can be successfully applied to the acoustic transmission in mobile Internet.