文献类型：期刊文献

中文题名：移动物联网下异质信息入侵感知预测方法仿真

英文题名：Simulation of Awareness Prediction Method for Heterogeneous Information in Moving Objects

作者：李恩林[1];刘洋[2]

第一作者：李恩林

机构：[1]河南财经政法大学计算机与信息工程学院;[2]河南财经政法大学云计算与大数据研究所

第一机构：河南财经政法大学计算机与信息工程学院

年份：2018

卷号：35

期号：2

起止页码：321-324

中文期刊名：计算机仿真

外文期刊名：Computer Simulation

收录：CSTPCD;;北大核心:【北大核心2017】；

语种：中文

中文关键词：移动物联网;异质信息;入侵感知预测

外文关键词：Mobile Internet of things ; Heterogeneous information; Intrusion perception prediction

摘要：移动物联网入侵攻击环境较为复杂,具有不确定性。当前采用的异质信息入侵感知预测方法收敛性较低、难以识别出入侵形式复杂的复合攻击,导致入侵感知预测结果不可靠,精度低。提出一种移动物联网下异质信息入侵感知预测方法。利用扩展的有向图描述移动物联网入侵攻击类别以及其逻辑关系,依据向后匹配和缺项匹配的形式对入侵报警信息进行关联,按照已关联入侵攻击链的累计权值和攻击图中不同分支的权值计算其可能性。将隐马尔科夫模型应用于移动物联网下异质信息入侵感知预测中,要求移动物联网依据获得的报警信息对隐藏的入侵攻击行为进行挖掘。依据入侵攻击行为概率分布,获得初始状态矩阵、状态转移矩阵以及观测矩阵,通过引入粒子群算法对参数进行全局优化,预测下一步入侵攻击。实验结果表明,上述方法入侵感知预测结果精度较高、泛化能力强。
This paper proposes a method for predicting the intrusion perception of heterogeneous information in mobile network. Firstly, the extended directed graph is used to describe the intrusion attack category in mobile Inter- net of things and its logical relation. According to the form of backward matching and absent matching, the intrusion alarm information is correlated. According to the cumulative weight of associated intrusion attack chain and the weight of different branches in attack graph, its possibility is calculated. The hidden Markov model is applied to the predic- tion of heterogeneous information intrusion perception in mobile lnternet of things. It is required that the mobile Inter- net of things excavated the hidden intrusion attack on the basis of obtained alarm information. According to the proba- bility distribution of intrusion attack behavior, the matrix of initial state, the state transition matrix and the observa- tion matrix are obtained. The parameters are optimized through the introduction of particle swarm optimization algo- rithm, so as to predict the next intrusion attack, Simulation results show that the proposed method has high accuracy of predicted results of intrusion perception and strong generalization ability.