Network Security Inference Method Based on Dempster-Shafer Theory

Abstract

The network environment of the edge layer of the Internet of Things in power systems (IOTIPS) is complex in an open environment, and the security threats are diverse and dynamic. It is difficult for the traditional network security system based on firewall to effectively deal with the emerging security problems of the IOTIPS. This paper uses the data-driven idea to dynamically judge the network behavior of edge devices. Based on the characteristics of the IOTIPS, by using the three dimensions of network log, domain name generated by domain generation algorithm, and network traffic as criteria, a basic probability assignment method based on confusion matrix is proposed. Multi-source information fusion is performed by using the Dempster-Shafer theory, and the network security of IOTIPS is distinguished.

Keywords

• internet of things in power systems (iotips)
• dempster-shafer theory
• network security
• evidence fusion