摘要
针对基于数据包间隔时间流水印技术中存在的同步机制弱、且缺乏可靠自纠错方法的问题,提出基于包间隔时间交叉分组的流水印技术.本方法通过将传输时间影响分散化,抵抗数据包在传输中遭遇的抖动干扰问题;通过约束数据包的传输时间满足网络中常规数据包的传输时间分布,提高水印的抗检测性;通过利用卷积码将水印序列进行扩展后再传输,实现流水印的有限自纠错;通过基于滑窗的算法动态判定数据包分组边界,实现流水印的盲追踪、并可抵御数据包在传输过程中可能出现的丢失或合并影响.理论分析与实验数据均表明:本方法可在较高阈值范围情况下准确检测出水印的存在性,抵御一般性的第三方试探性检测,具有较强的鲁棒性与隐蔽性.
To mitigate the weakness of synchronization and error-correction in packets flow watermarks based on inter-packets delays grouping,a new scheme that uses the centroid of cross inter-packet delays grouping was proposed.The new method scattered the influence of a single packet′s transmission time to dwarf the consequence of packets transmission jitters,restricted the overall packets transmission time to fulfill the normal packets transmission time to resist watermarks identification,applied the convolutional code to expand the data before transmission to implement self-correction at the point of watermark′s detection,and used packets sliding counting window to identify the borders of grouped packets to achieve the watermark blind detection and also overcome the negative consequences of packets loss and aggregation.Compared with the known methods,both theoretical analysis and experimental results show that our approach can defeat the try detection from a third party and also has better accuracy with the characteristics of robustness and hiddenness.
出处
《华中科技大学学报(自然科学版)》
EI
CAS
CSCD
北大核心
2015年第5期84-88,共5页
Journal of Huazhong University of Science and Technology(Natural Science Edition)
基金
国家自然科学基金资助项目(61300228)
江苏省科技支撑计划资助项目(BE2013103)
江苏省六大人才高峰资助项目(1631170006)
江苏大学拔尖人才计划资助项目(1213000013)
关键词
数据包流水印
交叉分组
包间隔
盲检测
卷积码
packets flow watermarking
cross-packet grouping
inter-packets delay
blind identification
convolutional code