摘要
针对大规模网络流量数据特征提取性能差、无界DoS入侵检测覆盖率低等问题,提出基于重引力搜索的大规模网络DoS入侵检测方法。提取大规模网络流量数据特征——流量统计特征、流量速率特征、流量分布特征与时间序列特征,以深度学习模型为工具,构建无界DoS入侵检测模型,通过重引力搜索算法对典型特征自相关参数进行寻优,进而提升无界DoS入侵检测的精准度。实验结果表明,应用该方法后,流量速率曲线与实际流量速率曲线几乎一致,无界DoS入侵检测覆盖率最大值达到95%。大规模网络无界DoS入侵检测效果优异,性能良好。
Aiming at the problems of poor feature extraction performance and low coverage of unbounded DoS intrusion detection in large-scale network traffic data,a large-scale network DoS intrusion detection method based on reweighted gravity search is proposed.Extract large-scale network traffic data features-traffic statistical features,traffic rate features,traffic distribution features,and time series features.Using deep learning models as tools,construct an unbounded DoS intrusion detection model.Through reweighted gravity search algorithms,optimize typical feature autocorrelation parameters to achieve accurate detection of unbounded DoS intrusions.The experimental results show that after the application of this method,the traffic rate curve is almost consistent with the actual traffic rate curve,and the maximum coverage of unbounded DoS intrusion detection reaches 95%.The detection effect of large-scale network boundaryless DoS intrusion is perfect,and the application performance is relatively good.
作者
杨云帆
黄浩
古振威
马腾腾
王依云
YANG Yunfan;HUANG Hao;GU Zhenwei;MA Tengteng;WANG Yiyun(Power Dispatching and Control Center,Guangdong Power Grid Co.,Ltd.,Guangzhou 510000,China;China Southern Power Grid Digital Power Grid Research Institute Co.,Ltd.,Guangzhou 510000,China)
出处
《电子设计工程》
2026年第4期151-154,共4页
Electronic Design Engineering
基金
南方电网基于国产密码的统一服务平台及数据安全示范应用项目(GDKJXM20198529)。
关键词
大规模网络
入侵检测
无界DoS攻击
重引力搜索算法
large-scale network
intrusion detection
unbounded DoS attack
reweighted gravity search algorithm
