Detection of weak electromagnetic interference attacks based on fingerprint in IIoT systems

Kai Fang, Tingting Wang, Xiaochen Yuan, Chunyu Miao, Yuanyuan Pan, Jianqing Li

研究成果: Article同行評審

31 引文 斯高帕斯(Scopus)

摘要

In Industrial Internet of Things (IIoT) systems, the intelligent devices are vulnerable to be attacked by weak Electromagnetic Interference (EMI), thereby threatening the security of the systems. Therefore, it is of great significance to investigate the weak EMI attack of IIoT systems. The different manufacturing processes and deployment environments make the intelligent devices carry different noises, called fingerprints, which are unchanged unless these intelligent devices are attacked. Hence, we can detect weak EMI attacks by judging whether the fingerprint of intelligent device has been changed, which is different from using professional detection equipment as in other methods. Based on the fingerprint of intelligent device, this paper proposes a highly efficient weak EMI attack detection method which is divided into three steps. First, the fingerprint is extracted by Linear Time-Invariant (LTI) model and Kalman algorithm. Second, according to the extracted fingerprint, a fusion model is designed to determine whether the device is attacked by weak EMI. In the fusion model, Feature Extraction Unit (FEU) combines with Long Short-Term Memory (LSTM) to improve the detection accuracy. Finally, an edge computing framework is proposed to enhance the efficiency of the method. The experimental results show that the detection accuracy and the efficiency of the proposed method are 5.2% and 42.2% higher than those of the state-of-the-art method, respectively.

原文English
頁(從 - 到)295-304
頁數10
期刊Future Generation Computer Systems
126
DOIs
出版狀態Published - 1月 2022
對外發佈

指紋

深入研究「Detection of weak electromagnetic interference attacks based on fingerprint in IIoT systems」主題。共同形成了獨特的指紋。

引用此