摘要
针对电力通信网络频繁遭受Sybil攻击的问题,提出了一种基于K-means边聚类的Sybil攻击团体检测算法.通过优化边聚类和边介数的计算方法,提出了传统K-means聚类算法的改进方法,计算了通信网络中的聚类系数,根据合法用户的真实数量,建立更加精确的攻击边集合与真实边集合,从而初步检测出所有可疑的攻击边,并使用标签传播算法检测Sybil攻击行为所在的恶意团体.仿真结果表明,与经典的SybilLimit算法相比,在所有的攻击路径数量下,该Sybil攻击检测算法具有更加优秀的检测性能.
In order to solve the problem that the power communication network had been frequently attacked by Sybil,a Sybil attackgroup detection algorithm based on K-means edge clustering was proposed.By optimizing the calculation methods of edge clustering and edge betweenness,an improved method in terms of traditional K-means clustering algorithm was suggested,and the clustering coefficient in the communication network was calculated.According to the real number of legitimate users,a more accurate set of attack edges and real edges was established so as to initially detect all suspicious attack edges.In addition,the label propagation algorithm was used to detect the evil group committing Sybil attack.The simulation results show that the as-proposed Sybil attack detection algorithm has better detection performance under all attack paths compared with the classic SybilLimit algorithm.
作者
党晓婧
张林
吕启深
彭浩
张柏松
DANG Xiao-jing;ZHANG Lin;Lü Qi-shen;PENG Hao;ZHANG Bai-song(Shenzhen Electric Power Science Research Institute,China Southern Power Grid,Shenzhen 518000,China;Business Research Center,Shenzhen Comtop Information Technology Co.Ltd.,Shenzhen 518034,China)
出处
《沈阳工业大学学报》
CAS
北大核心
2022年第5期502-506,共5页
Journal of Shenyang University of Technology
基金
国家自然科学基金项目(61033004)
中国南方电网有限公司深圳供电局有限公司科技项目(090000GS62161590).