采用多流联合质心熵的网络流水印检测方法

发布时间：2018-05-19 09:42

本文选题：流量溯源 + 流水印　；参考：《小型微型计算机系统》2017年11期

【摘要】：随着跳板主机和匿名网络成为隐匿通信关系的常用手段,网络攻击流量的溯源和定位难度日益增大.网络流水印技术在网络隐私安全领域已逐渐成为了一种重要的网络流量追踪和定位手段,设计良好的网络流水印具有强大的鲁棒性和隐蔽性,使得对网络流水印的存在性检测变得异常困难,而对流水印实施有效检测是进一步实现水印移除或水印流量复制的前提.本文提出了一种基于多流联合质心熵的水印盲检测方法,其可以实现针对当前典型的时隙质心类流水印的有效检测.在实际SSH流量上的实验结果表明,所提方法在单密钥情形下可达到与当前普遍采用的多流攻击相近的检测效果,在随机多密钥情形下多流攻击方案失效而本文方案依然可以实现高效检测.
[Abstract]:As the springboard host and anonymous network become the common means to hide the communication relationship, it is increasingly difficult to trace and locate the network attack traffic. Network pipeline printing technology has gradually become an important means of network traffic tracking and location in the field of network privacy security. The well-designed network pipeline printing has strong robustness and concealment. It is very difficult to detect the existence of network stream watermark, and the effective detection of convection watermark is the premise of further realization of watermark removal or watermark traffic replication. A blind watermarking detection method based on multi-stream combined centroid entropy is proposed in this paper. It can effectively detect the current typical slot centroid stream watermarking. The experimental results on the actual SSH flow show that the proposed method can achieve a detection effect similar to that of the commonly used multi-stream attacks in the case of single secret key. In the case of random multi-key, the multi-stream attack scheme fails, but the proposed scheme can still achieve efficient detection.
【作者单位】：南京理工大学自动化学院;江苏科技大学电子信息学院;
【基金】：国家自然科学基金项目(61472188,61602247)资助江苏省自然科学基金项目(BK20150472,BK20160840)资助国家科技支撑计划项目(2014BAH41B01)资助中央高校基本科研业务费专项项目(30920140121006,30915012208)资助
【分类号】：TP393.08

【相似文献】