摘要
针对数据篡改攻击的增多以及瞬态故障问题,提出用于系统安全监控的面向服务的架构(SOA),并利用该架构构建监控缓存(M-Cache),以检查供应链风险预警与决策系统的数据完整性。实验结果表明,与系统动态篡改数据相比,对系统代码攻击和对内存数据的攻击更容易被检测,内存数据的攻击检测率最大为99.6%,最小为98.6%。且监控缓存方法的检测准确率达到99.07%,优于随机森林模型与深度学习模型,进一步降低了漏检率。当攻击功率达到6 w时,基于SOA的监控缓存方法具有出色的检测效果,检测准确率为98.03%。随着攻击功率增大,检测准确率仍呈上升趋势。
In view of the increasing number of data tampering attacks and transient failures,a service-oriented architecture(SOA)for system security monitoring was proposed,and the M-Cache was constructed by using this architecture to check the data integrity of the supply chain risk early warning and decision-making system.The experimental results showed that the attacks on system code and memory data were easier to detect than those on dynamic tampering with data in the system,and the detection rate of the attack on memory data was 99.6%and 98.6%at the maximum.And the detection accuracy of the monitoring cache method reached 99.07%,which was better than the random forest model and deep learning model,further reducing the missed detection rate.When the attack power reached 6 watts,the monitoring cache method based on SOA had excellent detection performance,with a detection accuracy of 98.03%.As the attack power increased,the detection accuracy still showed an upward trend.
作者
丁晓
周红勇
陈继宁
DING Xiao;ZHOU Hongyong;CHEN Jining(State Grid Jiangsu Electric Power Co.,Ltd.,Nanjing 210024,China)
出处
《粘接》
CAS
2024年第12期193-196,共4页
Adhesion
关键词
面向服务的架构
供应链系统
监控缓存
检测准确率
service-oriented architecture
supply chain system
monitoring cache
detection accuracy
