Multiclass Intrusion Detection in IIoT Environments Using Attention Mechanisms and Optimized LSTM Networks

View at publisher Summarise with VIRAT
Authors: Bighnaraj Naik, Geetanjali Bhoi, Manohar Mishra, Rajat Kumar Sahu, Subhashree Mohapatra, Monalisa Biswal
Cited by: 0
References: 25

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

The rapid expansion of Industrial Internet of Things (IIoT) networks has improved automation and operational efficiency but has also increased vulnerability to sophisticated cyberattacks. Traditional intrusion detection systems often fail to detect evolving threats due to their static design and limited adaptability. This paper proposes an optimized Attention-based Long Short-Term Memory (LSTM) model for multiclass attack detection in IoT environments. The approach leverages temporal sequence learning with an attention mechanism to focus on critical features, enhancing detection accuracy across 19 distinct attack types. To improve model generalization, hyperparameters such as the number of LSTM units and dropout rate are optimized using a randomized search strategy. Experiments conducted on the X-IIoTID dataset demonstrate that the proposed method achieves superior performance compared to traditional machine learning models and conventional LSTM architectures, achieving an F1-score, precision, recall, and accuracy exceeding 0.995. These results confirm the effectiveness of integrating attention mechanisms and automated hyperparameter tuning for robust IIoT intrusion detection, making it well-suited for real-world, dynamic industrial environments.

Fields of Study & Subjects

Network Security and Intrusion Detection Internet of Things and AI Artificial Immune Systems Applications

Publication details

Journal / source
Biennial International Conference on Nascent Technologies in Engineering (ICNTE)
Pages
1–6
Publication date
16 January 2026
Language
EN