ARCHIVES
VOL. 12, ISSUE 3 (2026)
An edge-based CNN-LSTM hybrid model for real-time intrusion detection in industrial internet of things (IIoT) environments
Authors
Yasameen A Muhsin
Abstract
Industrial Internet of Things (IIoT)
environments have witnessed increasing expansion due to the growing reliance on
connected devices, sensors, and intelligent systems in industrial monitoring,
control, and automation. Despite the advantages these environments offer, the
proliferation of devices and protocols, along with the expansion of
connectivity, leads to a wider attack surface and escalating security
challenges. This underscores the importance of developing intrusion detection
systems capable of effectively analyzing network traffic and detecting
malicious activity. This study aims to design and evaluate a hybrid model based
on Convolutional Neural Networks (CNNs) and Long Short-Term Memory Networks
(LSTMs) for intrusion detection in IIoT environments. The model leverages CNNs'
ability to extract patterns and LSTMs' ability to process temporal
relationships in sequential data. The study utilized the DNN-EdgeIIoT-dataset
associated with the Edge-IIoTset, a dataset specifically designed to evaluate
cybersecurity technologies in IoT and IIoT applications. Data processing
involved verifying timestamps and retaining valid records dating back to 2021. 43
network features were then selected for model building. After addressing
missing values and standardizing features using StandardScaler, the records
were sorted chronologically and divided into 70% for training, 15% for validation,
and 15% for testing. The results indicate that the model has a high capacity
for reducing false positives and maintaining high precision; however, low
recall is the main limitation of its current performance. Therefore, improving
the model's attack detection capabilities is a priority for future work. The
inference speed also suggests the potential for studying the model in the
context of edge applications, although future testing on actual edge devices is
necessary to confirm this.
Download
Pages:28-33
How to cite this article:
Yasameen A Muhsin "An edge-based CNN-LSTM hybrid model for real-time intrusion detection in industrial internet of things (IIoT) environments". International Journal of Research in Advanced Engineering and Technology, Vol 12, Issue 3, 2026, Pages 28-33
Download Author Certificate
Please enter the email address corresponding to this article submission to download your certificate.