Volume 44, Issue 4, February 2025, Pages 1022–1029
Kamagaté Beman Hamidja1, Kanga Koffi2, Coulibaly Kpinnan Tiekoura3, and Konaté Adama4
1 Ecole Supérieure Africaine des Technologies de l’Information et de la Communication (ESATIC), Abidjan, Côte d’Ivoire
2 Ecole Supérieure Africaine des Technologies de l’Information et de la Communication (ESATIC), Abidjan, Côte d’Ivoire
3 Ecole Supérieure Africaine des Technologies de l’Information et de la Communication (ESATIC), Abidjan, Côte d’Ivoire
4 Ecole Supérieure Africaine des Technologies de l’Information et de la Communication (ESATIC), Abidjan, Côte d’Ivoire
Original language: English
Copyright © 2025 ISSR Journals. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
The rapid growth of Internet of Things (IoT) networks, fuelled by advancements in Low-Power Wide-Area Networks (LPWANs) and 5G technologies, has transformed industries such as healthcare, smart cities, and manufacturing. However, this expansion has also exposed IoT systems to cybersecurity vulnerabilities, making them prime targets for network intrusions and cyberattacks. Addressing these threats requires effective Intrusion Detection Systems (IDS) capable of identifying and classifying malicious traffic patterns. This paper proposes a hybrid IDS framework that integrates a 1D Convolutional Neural Network (1D CNN) and a Bidirectional Long Short-Term Memory (BiLSTM) model. The 1D CNN serves as a feature extractor, capturing spatial patterns in network traffic, while the BiLSTM leverages temporal dependencies in both forward and backward directions to enhance classification accuracy. Experiments assess the model’s performance in both binary and multi-class classification tasks. The results demonstrate that the 1D CNN+BiLSTM outperforms traditional methods, including SVM, XGBoost, and CNN+LSTM, achieving the highest accuracy (95.03%), recall (94.80%), and F1-score (94.90%). These findings highlight the model’s ability to minimize false positives and false negatives, making it highly suitable for real-time intrusion detection in IoT environments.
Author Keywords: Network security, Deep learning, anomaly detection, multi-class classification, IoT.
Kamagaté Beman Hamidja1, Kanga Koffi2, Coulibaly Kpinnan Tiekoura3, and Konaté Adama4
1 Ecole Supérieure Africaine des Technologies de l’Information et de la Communication (ESATIC), Abidjan, Côte d’Ivoire
2 Ecole Supérieure Africaine des Technologies de l’Information et de la Communication (ESATIC), Abidjan, Côte d’Ivoire
3 Ecole Supérieure Africaine des Technologies de l’Information et de la Communication (ESATIC), Abidjan, Côte d’Ivoire
4 Ecole Supérieure Africaine des Technologies de l’Information et de la Communication (ESATIC), Abidjan, Côte d’Ivoire
Original language: English
Copyright © 2025 ISSR Journals. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Abstract
The rapid growth of Internet of Things (IoT) networks, fuelled by advancements in Low-Power Wide-Area Networks (LPWANs) and 5G technologies, has transformed industries such as healthcare, smart cities, and manufacturing. However, this expansion has also exposed IoT systems to cybersecurity vulnerabilities, making them prime targets for network intrusions and cyberattacks. Addressing these threats requires effective Intrusion Detection Systems (IDS) capable of identifying and classifying malicious traffic patterns. This paper proposes a hybrid IDS framework that integrates a 1D Convolutional Neural Network (1D CNN) and a Bidirectional Long Short-Term Memory (BiLSTM) model. The 1D CNN serves as a feature extractor, capturing spatial patterns in network traffic, while the BiLSTM leverages temporal dependencies in both forward and backward directions to enhance classification accuracy. Experiments assess the model’s performance in both binary and multi-class classification tasks. The results demonstrate that the 1D CNN+BiLSTM outperforms traditional methods, including SVM, XGBoost, and CNN+LSTM, achieving the highest accuracy (95.03%), recall (94.80%), and F1-score (94.90%). These findings highlight the model’s ability to minimize false positives and false negatives, making it highly suitable for real-time intrusion detection in IoT environments.
Author Keywords: Network security, Deep learning, anomaly detection, multi-class classification, IoT.
How to Cite this Article
Kamagaté Beman Hamidja, Kanga Koffi, Coulibaly Kpinnan Tiekoura, and Konaté Adama, “Hybrid 1DCNN+BiLSTM Architecture for Network Intrusion Detection Systems,” International Journal of Innovation and Applied Studies, vol. 44, no. 4, pp. 1022–1029, February 2025.
