多维动态网络端口侧信道攻击快速定位仿真

Fast localization simulation of port side channel attacks in multi-dimensional dynamic networks

多维动态网络中的端口侧信道攻击具有隐蔽性和随机性,攻击源多且攻击路径模糊,使得快速准确定位攻击源较为困难。为此,提出多维动态网络端口侧信道攻击快速定位方法。利用端口侧信道发射信号的频偏获取其分布特征,采用密度聚类方法(DBSCAN)展开聚类分析,检测端口侧信道攻击。将攻击所在区域网格化,结合迭代软阈值算法与奇异值分解(SVD)算法获取各个节点的分解格式,构建信道攻击快速定位模型,采用群稀疏整体最小二乘算法对其求解,快速定位信道攻击。仿真实验结果表明,所提方法可以获取高精度的信道攻击快速定位结果,虚警概率仅为0.02%,检测耗时仅为1.56 ms,CPU利用率处于10%以下,可确保多维动态网络的稳定运行。

Port side channel attacks in multi-dimensional dynamic networks are concealed and random,and there are many attack sources and the attack paths are vague,which make it difficult to locate the attack sources quickly and accurately.Therefore,a fast location method for port side channel attacks in multi-dimensional dynamic networks is proposed.The frequency offset of the signals transmitted on the port side channel is used to obtain the distribution characteristics of the channel attacks,and the density-based spatial clustering of applications with noise(DBSCAN)is used to perform cluster analysis to detect the port side channel attacks.The region where the attack occurs is gridded.The iterative soft threshold algorithm and singular value decomposition(SVD)algorithm are combined to obtain the decomposition format of each node,and a fast channel attack location model is constructed,which is solved by a group sparse global least squares algorithm to quickly locate the channel attack.Simulation results show that the proposed method can obtain high-precision fast channel attack location results.The false alarm probability of the proposed method is only 0.02%,its detection time is only 1.56 ms,and its CPU utilization rate is no more than 10%,so the method can ensure stable operation of multi-dimensional dynamic networks.