基于社交关系的DHT网络Sybil攻击防御  

Sybil defenses in DHT networks based on social relationships

作　　者：韩心慧[1] 肖祥全 张建宇[1] 刘丙双[1] 张缘[1] 

机构地区：[1]北京大学计算机科学与技术研究所,北京100871

出　　处：《清华大学学报（自然科学版）》2014年第1期1-7,共7页Journal of Tsinghua University(Science and Technology)

摘　　要：Sybil攻击通过恶意伪造大量虚假身份,破坏对等网络(P2P)网络中正常节点的寻路过程,是分布式Hash表网络(distributed Hash table,DHT)中主要的安全威胁。该文利用社交网络中社交关系的高可信度以及伪造难度大等特点,设计了Social-DHT方法以缓解DHT网络中Sybil攻击的影响。该方法采用基于社交关系的随机游走策略以构建相对可信的路由表,继而可以有效抵御Sybil恶意节点的影响,实现安全、高效的寻路过程。此外对该方法建立模型,对路由表的可信性和寻路阶段的成功概率进行了理论分析。仿真实验表明:在有10 000条攻击边的情况下节点路由表中Sybil节点比例不超过3%,搜索成功率则能够达到99%,并且在搜索速度和带宽开销等方面优于已有的算法。The Syhil attack, which creates a large amount of fake node identities to break the normal routing process in the peer-to-peer （P2P） networks, is the main threat faced by distributed networks. A Social-DHT protocol was developed using the properties of social relationships to mitigate Sybil attacks in distributed Hash table （DHT） networks using random walks over the social relationships. In addition, a model is given using a formalized definition to analyze the successful probability of searches. Simulations show that the Social-DHT routing table includes less than 3% of the Sybil nodes when there are 10000 attack edges and the successful search ratio reaches 99 %, which is better than existing methods.

关 键 词：对等网络(P2P) 分布式Hash表(DHT) SYBIL攻击 社交网络 

分 类 号：TP309[自动化与计算机技术—计算机系统结构]