基于攻击行为主动跟踪算法的气象观测信息共享网络入侵检测OA
Intrusion Detection of Meteorological Observation Information Sharing Network Based on Active Tracking Algorithm of Attack Behavior
多源气象观测信息线性的关联会增加数据量对检测结果的影响,导致检测结果准确度较低.因此,提出基于攻击行为主动跟踪算法的气象观测信息共享网络入侵检测方法.引入多层感知模型实时感知气象观测信息共享网络访问数据.通过隐藏层神经元识别异常访问数据,计算感知数据的异常值并设定异常范围以筛选相应的数据,降低数据量影响.分析特征关联相关性,基于攻击行为主动跟踪算法对识别的异常数据进行网络行为追踪,实现入侵检测过程.实验结果表明,所提出的方法应用后得到的AUC值和检测准确度较高,满足了气象观测信息共享网络安全保障工作的现实需求.
The linear correlation of multi-source meteorological observation information increases the impact of data volume on detection results,resulting in lower accuracy of detection results.Therefore,a meteorological observation information sharing network intrusion detection method based on active tracking algorithm of attack behavior is proposed.The article introduces a multi-layer perception model to perceive meteorological observation information sharing network access data in real time.Ab-normal access data is identified through the hidden layer neurons,the outliers of the perceived data are calculated,and the ab-normal range is set to screen out the corresponding data,reducing the impact of data volume.Analyze the correlation of feature associations,and based on the attack behavior active tracking algorithm,track the network behaviors of the identified abnormal data to achieve the intrusion detection process.The experimental results show that the AUC value and detection accuracy ob-tained after the application of the proposed method are relatively high,meeting the practical needs of the security guarantee work for meteorological observation information sharing network.
刘浩;王东岳
黑龙江省气象数据中心,黑龙江,哈尔滨 150030黑龙江省气象数据中心,黑龙江,哈尔滨 150030
信息技术与安全科学
气象观测信息信息共享网络网络入侵攻击行为主动跟踪算法入侵检测
meteorological observation informationinformation sharing networknetwork intrusionactive tracking algorithm of attack behaviorintrusion detection
《微型电脑应用》 2026 (7)
46-50,5
黑龙江省气象局2021竞争性科技攻关项目(HQGG202108)
评论