摘要
针对现有基于时间信道的主动网络流水印技术缺少自纠错和难以抵御熵值隐蔽性嗅探的缺陷,提出一种基于平滑分组的分组间隔时间流水印方法。利用卷积码扩展水印信息,并采用平滑分组的方法将水印信息嵌入数据分组流中,通过交替调制数据分组间隔时间,使水印数据流的分组间隔分布特征无限逼近于正常的网络流,有效降低了数据分组在传输过程中遇到的时延抖动、分组丢失、分组合并、分组分片等因素干扰。理论分析和实验结果均表明,与现有的数据分组流水印技术相比,该方法具有检测准确度高、顽健性和隐蔽性好的特点。
To improve the self-correction ability and resist the entropy-based detection, a flow watermarking approach based on the inter-packets delays with smooth crossed grouping was proposed. Such an approach extended the water-marking methods using both the convolutional code and the smooth group methods to embed the watermarks into packet flows. By adjusting the inter-packets delays of the crossed packets, the transmission time distribution of the watermarked packets can indefinitely approach to that of any normal packets transmission times. Furthermore, the approach can miti-gate the negative consequences introduced by packets transmission jitters, packets losses, packets aggregations and packets divisions for the watermarks detection. Both theoretical analysis and experimental results show that the proposed approach overweight the known watermarking methods from the aspects of identification accuracy, robustness and hiddenness.
出处
《通信学报》
EI
CSCD
北大核心
2017年第10期36-46,共11页
Journal on Communications
基金
国家自然科学基金资助项目(No.61672269)
江苏省科技成果转化基金资助项目(No.BA2015161)~~
关键词
流水印
平滑分组
间隔时间
卷积码
packets flow watermarking, smooth crossed grouping, inter-packet delay, convolutional code
