摘要
对船舶通信网络入侵节点进行分类是保护船舶信息安全主要途径之一,但是随着入侵节点数目的增加,基于密度自适应粒子群算法的通信网络入侵节点分类方法分类误差较大,无法实现对入侵节点准确判别和分类。针对上述问题,提出基于信任值的通信网络入侵节点分类方法。首先计算节点的直接信任值,然后在此基础上计算节点的间接信任值,最后将直接信任值与间接信任值相结合,计算节点的综合信任值,同时利用异常行为检测算法对节点行为进行检测,判断节点类型,实现入侵节点的分类。结果表明:基于信任值的通信网络入侵节点分类方法与基于密度自适应粒子群算法的通信网络入侵节点分类方法相比,前者较后者分类误差降低了7%,分类准确性有了很大的提高。
The classification of intrusion nodes in ship communication network is one of the main ways to protect the information security of ships. However, with the increase of the number of intrusion nodes, the classification error of the intrusion node classification method based on the density adaptive particle swarm optimization algorithm is large, and the accurate discrimination and classification of the intrusion nodes can not be realized. In view of the above problems, a classification method based on trust value for communication network intrusion nodes is proposed. First, the direct trust value of the node is calculated, and then the indirect trust value of the node is calculated. Finally, the direct trust value and the indirect trust value are combined to calculate the comprehensive trust value of the node. At the same time, the anomaly behavior detection algorithm is used to detect the node behavior, judge the type of nodes, and achieve the classification of intrusion nodes. The results show that the classification method of the communication network intrusion nodes based on the trust value is compared with the method based on the density adaptive particle swarm optimization(PSO) based on the network intrusion node classification method, the former is 7% lower than the latter, and the classification accuracy has been greatly improved.
出处
《舰船科学技术》
北大核心
2018年第10X期163-165,共3页
Ship Science and Technology
关键词
通讯网络
节点分类
信任值
行为异常检测
communication network
node classification
trust value
behavior anomaly detection