Research Article
An Anomaly-based Intrusion Detection System for IoT environments using autoencoder neural networks and TinyML
|
International Journal of Computer Applications
Foundation of Computer Science (FCS), NY, USA
|
| Volume 187 - Issue 59 |
| Published: November 2025 |
| Authors: Hiba Kandil, Wiam Bouimejane, Mohammed Mouhcine, Hafssa Benaboud |
10.5120/ijca2025926000
|
Hiba Kandil, Wiam Bouimejane, Mohammed Mouhcine, Hafssa Benaboud . An Anomaly-based Intrusion Detection System for IoT environments using autoencoder neural networks and TinyML. International Journal of Computer Applications. 187, 59 (November 2025), 9-15. DOI=10.5120/ijca2025926000
@article{ 10.5120/ijca2025926000,
author = { Hiba Kandil,Wiam Bouimejane,Mohammed Mouhcine,Hafssa Benaboud },
title = { An Anomaly-based Intrusion Detection System for IoT environments using autoencoder neural networks and TinyML },
journal = { International Journal of Computer Applications },
year = { 2025 },
volume = { 187 },
number = { 59 },
pages = { 9-15 },
doi = { 10.5120/ijca2025926000 },
publisher = { Foundation of Computer Science (FCS), NY, USA }
}
%0 Journal Article
%D 2025
%A Hiba Kandil
%A Wiam Bouimejane
%A Mohammed Mouhcine
%A Hafssa Benaboud
%T An Anomaly-based Intrusion Detection System for IoT environments using autoencoder neural networks and TinyML%T
%J International Journal of Computer Applications
%V 187
%N 59
%P 9-15
%R 10.5120/ijca2025926000
%I Foundation of Computer Science (FCS), NY, USA
Abstract
The scalable nature of IoT systems leads to continually evolving security challenges, threats, and device vulnerability to cyberattacks. The traditional Intrusion Detection Systems (IDS) struggle with the resource-limited nature of IoT devices. However, Machine Learning (ML) techniques have appeared as a promising solution for IDS, offering several benefits. In this paper, we introduce an unsupervised Deep Learning model combined with TinyML principles for efficient deployment of Intrusion Detection Systems on IoT networks. The model is trained exclusively on normal network traffic and detects anomalies through reconstruction error. To enable deployment on constrained devices, the model is quantized and converted to Lite format, resulting in a lightweight version suitable for TinyML environments. Evaluation was conducted using the IoT-23 dataset and NS-3-based traffic simulation. The proposed system enables real-time, on-device threat detection while operating within the strict memory, latency, and energy constraints typical of embedded IoT environments.
References
- H. Kandil and H. Benaboud, ”Using Machine Learning to Deal with Privacy and Confidentiality in Internet of Things: An Overview,” In: Innovations in Smart Cities Applications Volume 8. SCA 2024. Lecture Notes in Networks and Systems, vol 1310. 2025. Springer, Cham. https://doi. org/10.1007/978-3-031-88653-9_73
- H. Kandil and H. Benaboud, ”About Scalability and End-to-End Security in Internet of Things: An Overview,” In Proceedings of the 7th International Conference on Networking, Intelligent Systems and Security (NISS’24). Association for Computing Machinery, New York, NY, USA, Article 6, 1–7. 2024. https://doi.org/10.1145/ 3659677.3659690
- J. Lin, L. Zhu,W.-M. Chen,W.-C.Wang, and S. Han, ”Tiny Machine Learning: Progress and Futures,” arXiv preprint arXiv:2403.19076, Mar. 2024. [Online]. Available:https: //arxiv.org/abs/2403.19076
- Stratosphere Laboratory, ”IoT-23: A labeled dataset with malicious and benign IoT network traffic,” Jan. 2020. [Online]. Available:https://www.stratosphereips. org/datasets-iot23
- R. Butt, N. Tariq, M. Humayun, and A. Ishaq, “Malware Detection in Network Traffic Data for Internet of Things,” in 2024 26th International Multi-Topic Conference (INMIC), Dec. 2024, pp. 1–6. doi:10.1109/ INMIC64792.2024.11004366.
- N. Wei et al., “An Autoencoder-Based Hybrid Detection Model for Intrusion Detection With Small-Sample Problem,” IEEE Transactions on Network and Service Management, vol. 21, pp. 2402–2412, 2024, doi:10.1109/TNSM.2023.3334028.
- A. Ahli, A. Raza, K. O. Akpinar, and M. Akpinar, “Binary and Multi-Class Classification on the IoT-23 Dataset,” in 2023 Advances in Science and Engineering Technology International Conferences (ASET), Feb. 2023, pp. 1–7. doi:10.1109/ASET56582.2023.10180848.
- N. Abdalgawad, A. Sajun, Y. Kaddoura, I. A. Zualkernan, and F. Aloul, “Generative Deep Learning to Detect Cyberattacks for the IoT-23 Dataset,” IEEE Access, vol. 10, pp. 6430–6441, 2022, doi:10.1109/ACCESS.2021. 3140015.
- C. V. Oha et al., “Analysis of IoT-23 datasets and machine learning models for malicious traffic detection,” ERA. Accessed: Jul. 16, 2025. Available:https://era.library.ualberta.ca/ items/28700e1c-3afc-4892-b12e-5c229df9e056.
- A. K. Sahu, S. Sharma, M. Tanveer, and R. Raja, “Internet of Things attack detection using hybrid Deep Learning Model,” Computer Communications, vol. 176, pp. 146–154, Aug. 2021, doi:10.1016/j.comcom. 2021.05.024.
- M. Wo´zniak, J. Siłka, M. Wieczorek, and M. Alrashoud, “Recurrent Neural Network Model for IoT and Networking Malware Threat Detection,” IEEE Transactions on Industrial Informatics, vol. 17, no. 8, pp. 5583–5594, Aug. 2021, doi:10.1109/TII.2020.3021689.
- A. Gueriani, H. Kheddar and A. C. Mazari, ”Adaptive Cyber-Attack Detection in IIoT Using Attention-Based LSTM-CNN Models,” 2024 International Conference on Telecommunications and Intelligent Systems (ICTIS), Djelfa, Algeria, 2024, pp. 1-6, doi: 10.1109/ICTIS62692.2024.10894509.
- F. Chollet, ”An Introduction to Autoencoders,” in Deep Learning with Python, Manning Publications, 2017. https://blog.keras.io/ building-autoencoders-in-keras.html
- A. Chandola, A. Banerjee, and V. Kumar, ”Anomaly detection: A survey,” ACM Computing Surveys, vol. 41, no. 3, pp. 1–58, 2009. https://doi.org/10.1145/ 1541880.1541882
- NS-3 Consortium, “ns-3 network simulator,” [Online]. Available:https://www.nsnam.org
Index Terms
Keywords