Industrial Internet Intrusion Detection Method Based on ResCNN-Attention-BISRU.

Saved in:

Bibliographic Details
Title:	Industrial Internet Intrusion Detection Method Based on ResCNN-Attention-BISRU.
Authors:	ZENG-YU CAI¹, PENG-RONG LI¹, JIAN-WEI ZHANG^2,3 mailzjw@163.com, YUAN FENG⁴, LIANG ZHU¹
Source:	Journal of Information Science & Engineering. May2025, Vol. 41 Issue 3, p675-697. 23p.
Subjects:	Computer network traffic, Internet traffic, Generative adversarial networks, Industrialism, Feature extraction, Intrusion detection systems (Computer security)
Abstract:	An industrial Internet intrusion detection method based on ResCNN-Attention BISRU is proposed in this paper. The purpose is to enhance the model's ability to identify network abnormal behaviors in industrial Internet data, accelerate the intrusion response speed, and then effectively improve the accuracy and robustness of intrusion detection to ensure the security of industrial Internet systems. This method adopts the feature extraction method of ID CNN combined with residual connection, captures the long-term dependencies in network traffic through the multi-head attention layer, and uses BISRU to obtain information from time series, which enhances the model's ability to process complex data. Experimental results on the Gas Pipeline dataset show that the model achieves an impressive 99.05% accuracy and 96.27% precision. These results represent a significant improvement of the method proposed in this paper, with an accuracy rate 4.16% higher than the traditional network model and 0.09% higher than the fusion algorithm. Due to the unique nature of industrial Internet traffic data, the limited availability of certain attack type samples affects the model's detection accuracy for such attacks. Future research will explore incorporating sample generation methods such as genetic algorithms, generative adversarial networks, or VAE to enhance the detection system's ability to recognize small-sample attacks and unknown attacks. [ABSTRACT FROM AUTHOR]
	Copyright of Journal of Information Science & Engineering is the property of Institute of Information Science, Academia Sinica and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database:	Engineering Source

Description
Abstract:	An industrial Internet intrusion detection method based on ResCNN-Attention BISRU is proposed in this paper. The purpose is to enhance the model's ability to identify network abnormal behaviors in industrial Internet data, accelerate the intrusion response speed, and then effectively improve the accuracy and robustness of intrusion detection to ensure the security of industrial Internet systems. This method adopts the feature extraction method of ID CNN combined with residual connection, captures the long-term dependencies in network traffic through the multi-head attention layer, and uses BISRU to obtain information from time series, which enhances the model's ability to process complex data. Experimental results on the Gas Pipeline dataset show that the model achieves an impressive 99.05% accuracy and 96.27% precision. These results represent a significant improvement of the method proposed in this paper, with an accuracy rate 4.16% higher than the traditional network model and 0.09% higher than the fusion algorithm. Due to the unique nature of industrial Internet traffic data, the limited availability of certain attack type samples affects the model's detection accuracy for such attacks. Future research will explore incorporating sample generation methods such as genetic algorithms, generative adversarial networks, or VAE to enhance the detection system's ability to recognize small-sample attacks and unknown attacks. [ABSTRACT FROM AUTHOR]
ISSN:	10162364
DOI:	10.6688/JISE.202505_410).0010