摘要
针对光纤光栅传感网络结构复杂,入侵行为检测难度较高的问题,研究基于集成学习的光纤光栅传感网络入侵行为检测方法。选取支持向量机作为集成学习算法的基分类器,计算各基分类器分类光纤光栅传感网络入侵行为样本的误差率,依据基分类器的误差率确定基分类器的重要程度。利用AdaBoost集成学习算法,依据各基分类器的重要程度集成各基分类器,构建最终的集成分类器,利用所构建集成分类器,输出光纤光栅传感网络入侵行为检测结果。实验结果表明,该方法可以精准检测光纤光栅传感网络的远程入侵、拒绝服务入侵等入侵行为,数据丢弃量较低,提升了光纤光栅传感网络的通信性能。
In view of the complex structure of fiber Bragg grating sensor network and the difficulty of intrusion detection,the intrusion detection method of fiber Bragg grating sensor network based on ensemble learning is studied.The support vector machine(SVM)was selected as the base classifier of the integrated learning algorithm,and the error rate of each base classifier in the classification of FBG sensor network intrusion behavior samples was calculated,and the importance of the base classifier was determined according to the error rate of the base classifier.The AdaBoost ensemble learning algorithm is used to integrate each base classifier according to the importance of each base classifier,and the final ensemble classifier is constructed.The constructed ensemble classifier is used to output the detection results of fiber Bragg grating sensor network intrusion behavior.The experimental results show that the proposed method can accurately detect the intrusion behaviors such as remote intrusion and denial of service intrusion in the fiber Bragg grating sensor network,and the amount of data discarded is low,which improves the communication performance of FBG sensor network.
作者
要丽娟
郭银芳
YAO Lijuan;GUO Yinfang(Taiyuan University,Taiyuan 030032,China)
出处
《激光杂志》
CAS
北大核心
2023年第11期147-151,共5页
Laser Journal
基金
山西省教育厅创新项目(No.J20221192)。
