Skip to content
Conference Open access

A Hybrid Deep Learning Framework for Intelligent Cyber Threat Detection in IoT Networks

Sep 2026 · 2026 International Conference on Secure IoT and Cybersecurity (ICOSICS) · pp. 602-609 · 0 citations · 17 references

Abstract

Internet of Things (IoT) is a expertise that allows to interconnect devices to utilized in smart healthcare, intelligent transportation, industrial automation and smart cities. But, due to the heterogeneous and resource constrained nature of IoT devices, they are prone to any kind of cyberattack, such as Distributed Denial-of-Service (DDoS), Denial-of-Service (DoS), spoofing, reconnaissance, brute-force, botnet, etc., and data manipulation attacks. While traditional intrusion detection systems (IDSs) and conventional machine learning techniques fall short in identifying complex and evolving attack patterns. Traditional intrusion detection systems (IDSs) and conventional machine learning techniques have their limitations in detecting complex and evolving attack patterns. In this paper, an Autoencoder (AE), one-dimensional CNN (Convolutional Neural Network) and Bidirectional Gated Recurrent Unit (Bi-GRU) based adaptive hybrid framework using deep learning for intelligent multi-class cyber threat detection. The framework preprocesses the data, normalizes the features, and uses an Autoencoder to produce compact representations to enable efficient learning. CNN extracts spatial features and Bi-GRU extracts bidirectional temporal dependencies of network traffics. The last threat classification is done by a SoftMax classifier. The proposed model is tested on the CICIoT2023 benchmark dataset using accuracy, precision, recall, f1 score, ROC-AUC and confusion matrix. The experimental results show its effectiveness in detection accuracy, scalability and computational efficiency, serving as an effective solution for securing the next-generation IoT networks.

Read PDF

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.