A method for detecting intrusion behavior in the Internet of Things based on cascaded fil-tering 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,en-hancing 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
Key words
Naive Bayes/Internet of Things/invasion behavior/denial of service attacks/cascade filtering/feature extraction
维普
万方数据