Open Access

REVIEW

A Comprehensive Survey of Contemporary Anomaly Detection Methods for Securing Smart IoT Systems

Chaimae Hazman^1,2, Azidine Guezzaz², Said Benkirane², Mourade Azrour^3,*, Vinayakumar Ravi⁴, Abdulatif Alabdulatif ⁵

1 Research Team LAMIGEP, EMSI, Marrakech, 40000, Morocco
2 SISAR Team, LaRTID Laboratory, Higher School of Technology, Cadi Ayyad University, Marrakech, 40000, Morocco
3 IMIA Laboratory, MSIA Team, Faculty of Sciences and Techniques, Moulay Ismail University of Meknes, Errachidia, 52000, Morocco
4 Center for Artificial Intelligence, Prince Mohammad Bin Fahd University, Khobar, 31952, Saudi Arabia
5 Department of Computer Science, College of Computer, Qassim University, Buraydah, 52571, Saudi Arabia

* Corresponding Author: Mourade Azrour. Email:

Computers, Materials & Continua 2025, 85(1), 301-329. https://doi.org/10.32604/cmc.2025.064777

Received 24 February 2025; Accepted 22 July 2025; Issue published 29 August 2025

Abstract

Attacks are growing more complex and dangerous as network capabilities improve at a rapid pace. Network intrusion detection is usually regarded as an efficient means of dealing with security attacks. Many ways have been presented, utilizing various strategies and focusing on different types of visitors. Anomaly-based network intrusion monitoring is an essential area of intrusion detection investigation and development. Despite extensive research on anomaly-based network detection, there is still a lack of comprehensive literature reviews covering current methodologies and datasets. Despite the substantial research into anomaly-based network intrusion detection algorithms, there is a dearth of a research evaluation of new methodologies and datasets. We explore and evaluate 50 highest publications on anomaly-based intrusion detection using an in-depth review of related literature techniques. Our work thoroughly explores the technological environment of the subject in order to help future research in this sector. Our examination is carried out from the relevant angles: application areas, data preprocessing and threat detection approaches, assessment measures, and datasets. We select unresolved research difficulties and underexplored research areas from every viewpoint recommendation of the study. Finally, we outline five potentially increased research areas for the future.

---

Keywords

---