Title: Study on abnormal data acquisition method of industrial internet of things communication based on node clustering

Authors: Yan Cui

Addresses: School of Information Engineering, Jiaozuo University, Jiaozuo, Henan, China

Abstract: In order to overcome the problems of low efficiency and poor accuracy of traditional abnormal data capture, this paper proposes an industrial IoT communication abnormal data capture method based on node clustering. First, analyse the structure of Industrial Internet of Things (IIoT), collect communication data and extract the characteristics of industrial communication data; then, the node subordination function is constructed, and the abnormal data types are divided by common neighbour node analysis. Finally, the matching degree between features is extracted through node clustering calculation, and the matching degree is input into the classifier function and the communication exception data capture result is output. The results show that the malicious node capture rate of this method is always higher than 92%, the capture time is always less than 3.6 seconds and the effect of abnormal data capture of IoT communication is improved.

Keywords: node clustering; subordinate degree function; co-adjacent nodes; abnormal data capture; unit overlap.

DOI: 10.1504/IJDMB.2022.130344

International Journal of Data Mining and Bioinformatics, 2022 Vol.27 No.1/2/3, pp.107 - 117

Received: 12 Aug 2022
Accepted: 15 Dec 2022
Published online: 17 Apr 2023 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article