网络多属性安全态势感知方法仿真研究  

Simulation Research on Network Multi-Attribute Security Situation Awareness Method

作　　者：何璘琳 何月顺[1] 罗亚宾 刘志锋[1] HE Lin-lin;HE Yue-shun;LUO Ya-bin;LIU Zhi-feng(School of Information Engineering,East China University of Technology of China,Nanchang Jiangxi 330013,China)

机构地区：[1]东华理工大学信息工程学院,江西南昌330013

出　　处：《计算机仿真》2024年第10期333-336,366,共5页Computer Simulation

基　　金：江西省重点研究计划项目(20224BBC41001)。

摘　　要：由于网络安全要素属性类型较多,在开展网络安全态势感知时,若提取的网络安全要素精度较差会直接影响后续的网络实际安全态势感知效果,为此提出网络多属性安全态势感知方法。对多源数据实施数据清洗处理,清除数据集合中的冗余数据,利用深度学习方法对网络中多源数据实施特征提取处理,通过特征分类结果获取网络中属性多样的安全态势要素集合;针对网络多属性进一步提取网络态势要素,建立轨迹态势矩阵,完成态势感知。实验结果表明,上述方法的感知效果较好,感知误差与感知时延较低。At present,there are many types of network security attributes.When conducting network security situational awareness,if the accuracy of the extracted security elements is not ideal,the subsequent network security situation awareness effect will be directly affected.Therefore,a multi-attribute security situation awareness method was proposed.Firstly,the multi-source data was cleaned,and then the redundant data was cleared from the data set.Secondly,the multi-source data in the network was extracted by the deep learning method.Based on the feature classification results,the set of security situation elements with various attributes was obtained from the network.Furthermore,network situation elements were extracted.Finally,a situational matrix of trajectory was constructed.Thus,the situational awareness was completed.Experimental results show that this method has a good perception effect and low perception error and delay.

关 键 词：多源数据 数据分类 数据清洗 网络安全 态势感知 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]