A Malicious Node Identification Method Based on Edge Computing

Abstract

Edge computing can meet the needs of many industries in real-time business control, security and privacy protection. In the process of massive heterogeneous terminals accessing the network through edge devices, it is very challenging to achieve fast and reliable authentication. The physical layer authentication technology authenticates through the channel characteristics, which has the characteristics of lightweight, and can well adapt to the authentication scenario of massive heterogeneous terminal access. In order to identify malicious nodes, this paper proposes an attack identification scheme of malicious nodes under edge computing. The new channel response information vector is constructed by using the correlation between the channel information of consecutive frames. Two or more time slot channel frequency response vectors are averaged to obtain a new channel response vector. It has the advantages of low computational complexity and high recognition accuracy. And combined with the channel frequency response based on the deep neural network to identify malicious nodes, the data set in the factory environment is simulated.