基于数据分析的网络安全威胁检测分析

Network Security Threat Detection and Analysis Based on Data Analysis

物联网与边缘计算的结合,是智能家居中网络连接的重要方式。为了保证系统的安全性,需要采用一系列的网络安全分析方法,将数据划分为不同的类别,用于解决网络攻击的识别问题。文中基于阿里云搭建了虚拟的智能家居系统,并基于ECC,RSA和MAC的数据传输模拟出数据集,并对支持向量机进行了训练和测试。结果表明,基于RBF分类器的支持向量机在攻击检测任务中表现最好,在ECC,RSA和MAC等3种数据传输接方式中,对网络威胁的识别准确率分别为91.64%,98.14%和98.267%。

The combination of Internet of Things and edge computing is an important way of network connection in smart homes.In order to ensure the security of the system,a series of cyber security analysis methods need to be used to divide the data into different categories to solve the problem of network attack identification.In this paper,a virtual smart home system is built based on Alibaba Cloud,and a dataset is simulated based on ECC,RSA and MAC data transmission,and support vector machines are trained and tested.The results show that the support vector machine based on the RBF classifier performs best in the attack detection task.Among the three data transmission modes including ECC,RSA and MAC,the recognition accuracy of network threats is 91.64%,98.14%and 98.267%,respectively.