Cryptographic-Based Machine Learning Models for Data Security in Internet of Things - A Systematic Literature Review
International Journal of Computer Networks and Applications · 30.06.2026 · 13 (3) · 583-612
DOI: 10.22247/ijcna/2026/31 ↗Kurz gefasst
Systematische Übersicht in der Zeitschrift International Journal of Computer Networks and Applications, online erschienen am 30.06.2026. Verfasst von Mathias Agbeko und 1 weiteren Autor:innen.
Abstract (Originalsprache)
– As Internet of Things (IoT) deployments surge from 16.6 billion devices in 2023 to 21.1 billion now, legacy security measures cannot keep up with escalating data breaches, unauthorized access, and cyberattacks in resource-constrained, heterogeneous networks. This systematic literature review synthesizes 14 peer-reviewed studies (2019 – 2025) from IEEE Xplore, Scopus, Web of Science, and ScienceDirect to map how cryptographic-based machine learning (CB-ML) models can fill this gap. We analyze integrations of symmetric/asymmetric encryption, hash functions, and digital signatures with supervised learning, anomaly detection, and reinforcement learning, with emphasis on 2023 – 2025. Five CB-ML paradigms emerge: (1) reinforcement learning-driven cryptographic optimization, (2) lightweight hybrid cryptography with ML support, (3) hardware-assisted authentication via Physically Unclonable Functions, (4) biometric cryptographic authentication, and (5) anomaly-aware secure frameworks. Random Forest and k-Nearest Neighbors achieved 99% intrusion detection accuracy, yet substantial trade-offs persist among security strength, energy efficiency, and computational complexity in constrained IoT environments. We conclude that while CB-ML models significantly bolster IoT security, realizing deployable systems demands simultaneous advances in lightweight cryptography, privacy-preserving learning, post-quantum integration, and large-scale
Zitieren
APA
Agbeko, M., & Handa, D. (2026). Cryptographic-Based Machine Learning Models for Data Security in Internet of Things - A Systematic Literature Review. International Journal of Computer Networks and Applications, 13(3), 583-612. https://doi.org/10.22247/ijcna/2026/31
Vancouver
Agbeko M, Handa D. Cryptographic-Based Machine Learning Models for Data Security in Internet of Things - A Systematic Literature Review. International Journal of Computer Networks and Applications. 2026;13(3):583-612. doi:10.22247/ijcna/2026/31
BibTeX
@article{Agbeko2026,
author = {Agbeko, Mathias and Handa, Disha},
title = {Cryptographic-Based Machine Learning Models for Data Security in Internet of Things - A Systematic Literature Review},
journal = {International Journal of Computer Networks and Applications},
year = {2026},
volume = {13},
number = {3},
pages = {583--612},
doi = {10.22247/ijcna/2026/31}
}
EUVIMED