A Systematic Literature Review on AI-Based Intrusion Detection Systems for IoT and Sensor Networks
International Journal of Advanced Smart Sensor Network Systems · 18.06.2026 · 16 (1/2) · 01-23
DOI: 10.5121/ijassn.2026.16201 ↗Kurz gefasst
Systematische Übersicht in der Zeitschrift International Journal of Advanced Smart Sensor Network Systems, online erschienen am 18.06.2026. Verfasst von Ouechtati Hamdi.
Abstract (Originalsprache)
The Internet of Things (IoT) and sensor networks enable large-scale automation, monitoring, and control, but their heterogeneity, limited resources, and pervasive connectivity substantially enlarge the cyberattack surface. Artificial Intelligence (AI)-based Intrusion Detection Systems (IDS) have therefore become a major research direction because they can learn attack patterns, detect abnormal behavior, and adapt to evolving threats. This Systematic Literature Review (SLR), conducted according to PRISMA 2020 principles, synthesizes peer-reviewed research on AI-based IDS for IoT and sensor networks published from 2015 to early 2025. From an initial pool of 823 records, 75 studies were selected after duplicate removal, screening, eligibility assessment, and quality appraisal. The review compares the methods used in the literature, including classical machine learning, deep learning, hybrid and ensemble learning, federated learning, and edge-oriented lightweight models. It also adds a quality-assessment scheme, a method-by-method synthesis, a quantitative performance summary, and a benchmarking protocol. The findings show that deep learning and hybrid methods often achieve the highest detection performance on benchmark datasets, whereas federated and edge-based IDS provide stronger privacy, latency, and deployment advantages. Persistent gaps remain in dataset realism, cross-dataset validation, adversarial robustness, explainability, energy-aware deployment, and reproducible evaluation. The review concludes by proposing a layered conceptual framework and future directions for trustworthy, lightweight, explainable, and privacy-preserving IDS in next-generation IoT environments.
Zitieren
APA
Hamdi, O. (2026). A Systematic Literature Review on AI-Based Intrusion Detection Systems for IoT and Sensor Networks. International Journal of Advanced Smart Sensor Network Systems, 16(1/2), 01-23. https://doi.org/10.5121/ijassn.2026.16201
Vancouver
Hamdi O. A Systematic Literature Review on AI-Based Intrusion Detection Systems for IoT and Sensor Networks. International Journal of Advanced Smart Sensor Network Systems. 2026;16(1/2):01-23. doi:10.5121/ijassn.2026.16201
BibTeX
@article{Hamdi2026,
author = {Hamdi, Ouechtati},
title = {A Systematic Literature Review on AI-Based Intrusion Detection Systems for IoT and Sensor Networks},
journal = {International Journal of Advanced Smart Sensor Network Systems},
year = {2026},
volume = {16},
number = {1/2},
pages = {01--23},
doi = {10.5121/ijassn.2026.16201}
}
EUVIMED