基于蛋白质相互作用网络的DDoS攻击检测  

DDoS Detection Which Bases on Protein Interaction Networks

作　　者：康松林[1] 詹煜[1] 樊晓平[1] 施荣华[1] 

机构地区：[1]中南大学信息科学与工程学院,长沙410083

出　　处：《小型微型计算机系统》2015年第6期1283-1290,共8页Journal of Chinese Computer Systems

基　　金：国家自然科学基金项目(60773013)资助

摘　　要：DDo S攻击一直严重威胁着网络的安全,针对DDo S攻击的检测与防御是网络安全研究中的一个重要课题.利用蛋白质相互作用网络中的功能预测特点,设计了一个DDo S攻击检测及追踪方案.该方案在分时统计的基础上创建RTCT值,并在服务器端根据不同的RTCT值来生成不同的个体,利用特征熵值构建个体蛋白质相互作用网络,并通过与目标蛋白质相互作用网络对比来检测是否有攻击发生;当有攻击发生的时候,分解RTCT值并锁定攻击源.该方案对DDo S攻击的检测较为敏感,在复杂的网络拓扑结构中能够准确的检测出DDo S攻击并锁定攻击源,模拟实验研究结果表明了该机制是高效的.DDoS attacks have been a critical threat to the Internettand the detection and defense of DDoS attacks is an important sub- ject in network security research. This paper designs a detection and tracking method which bases on time-sharing statistics by using the forecast function of protein interaction networks. Firstly, RTCT values are created in the routers; Secondly, the server uses RTCT values to generate different individuals and build protein interaction networks for each individual depending on features entropy value; Thirdly, the server compares the individual protein interaction networks with the target protein interaction networks to determine wheth- er the attacks occurred. If the attacks exist,the server decomposes the RTCT value and locates the attacks source. This method is more sensitive to detect DDoS attacks,it can also detect DDoS attacks exactly and locate the attacks source in the complex network topolo- gy. The results of extensive experimental and simulation studies demonstrate the effectiveness and efficiency of this method.

关 键 词：DDOS攻击 蛋白质相互作用网络 RTCT值 分时统计 特征熵值 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]