基于级联滤波的物联网入侵行为检测方法  

A Method for Internet of Things Intrusion Behavior Detection Based on Cascaded Filtering

作　　者：王燕红[1] 吴昌钱 WANG Yanhong;WU Changqian(College of Information and Electronic Engineering,Liming Vocational University,Quanzhou Fujian 362007,China;Department of Computer Science and Technology,Minnan Science and Technology College,Quanzhou Fujian 366200,China)

机构地区：[1]黎明职业大学信息与电子工程学院,福建泉州362007 [2]闽南科技学院计算机信息学院,福建泉州366200

出　　处：《淮阴师范学院学报（自然科学版）》2024年第4期314-318,328,共6页Journal of Huaiyin Teachers College(Natural Science Edition)

基　　金：福建省中青年教师教育科研项目(JAT171085)。

摘　　要：提出基于级联滤波的物联网入侵行为检测方法.建立拒绝服务攻击的多尺度的样本集,利用朴素贝叶斯网络进行拒绝服务攻击特征挖掘,得到物联网入侵行为取证的模糊统计特征量;利用级联滤波分析方法逐步筛选出与入侵行为最相关的特征,增强相关入侵信息;结合物联网入侵行为的信息浓度、行为分布和包络幅值,实现物联网入侵行为检测.实验结果表明,所提方法可以有效提取出拒绝服务攻击的特征,并较为准确的检测出物联网遭受拒绝服务攻击的次数,响应时间在1.5s以内.说明该方法具有较强的自适应性,可以有效提高物联网入侵检测能力.A method for detecting intrusion behavior in the Internet of Things based on cascaded filtering has been proposed.A multi-scale sample set for denial of service attacks was established,and Naive Bayes network was used to mine the characteristics of denial of service attacks,obtaining fuzzy statistical feature quantities for Internet of Things intrusion behavior forensics.The cascaded filtering analysis method was used to gradually screen out the most relevant features to intrusion behavior,enhancing the relevant intrusion information.By combining the information concentration,behavior distribution,and envelope amplitude of Internet of Things intrusion behavior,Internet of Things intrusion behavior detection has been achieved.The experimental results show that the proposed method can effectively extract the characteristics of denial of service attacks and accurately detect the number of times the Internet of Things has been subjected to denial of service attacks.The response time is within 1.5 s.This method has strong adaptability and can effectively improve the intrusion detection capability of the Internet of Things.

关 键 词：朴素贝叶斯 物联网 入侵行为 拒绝服务攻击 级联滤波 特征提取 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]