An Integrated Architecture Combining Blockchain, Artificial Intelligence, and Machine Learning for Decentralised Identity Management in the Internet of Things
Journal of Emerging Technologies for Society · 2026-05 · 27
DOI: 10.57017/jets.v1.iss1.02 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Emerging Technologies for Society, 2026-05. Verfasst von Marius Iulian MIHĂILESCU und 2 weiteren Autor:innen.
Abstract (Originalsprache)
The identity-management challenge addressed in this study emerges after device enrolment, when an Internet of Things (IoT) device may continue to present valid credentials even though its software state, behavioural patterns, or operating conditions have changed. To address this limitation, the study proposes an integrated architecture that treats cryptographic identity and observed behaviour as complementary but distinct sources of trust evidence. A permissioned ledger anchors decentralised identifiers, credential states, policy versions, model attestations, and decision evidence, while behavioural analysis is performed close to the device and federated learning enables collaborative model development without centralising raw telemetry. At the authorisation stage, credential validity, behavioural risk, and request context are combined into a dynamic trust state, enabling graduated responses as evidence evolves. Low-level observation and temporary restrictions may be automated, whereas suspension is subject to review and permanent credential revocation requires human authorisation. The study adopts a Design Science Research approach, treating the proposed architecture as an artefact evaluated through requirements-based assessment, threat-model analysis, and feasibility evidence derived from the underlying technological mechanisms. The evaluation identifies both design-level capabilities and unresolved dependencies related to secure key custody on legacy devices, federated coordination at scale, poisoning by credentialled participants, and regulatory conformity. The proposed architecture therefore extends decentralised identity management beyond static credential verification by integrating continuous behavioural assessment, auditable decision-making, and human oversight into a unified IoT trust-management framework.Copyright: ©The Author(s) 2026. This article is distributed under the terms and conditions of the license Creative Commons Attribution (CC BY) license. Article info: Received: 9 March, 2026 Revised: 30 March, 2026 Accepted: 19 May, 2026 Published: 22 May, 2026
Zitieren
APA
MIHĂILESCU, M. I., MĂRĂSCU, V., & NIȚĂ, Ș. L. (2026). An Integrated Architecture Combining Blockchain, Artificial Intelligence, and Machine Learning for Decentralised Identity Management in the Internet of Things. Journal of Emerging Technologies for Society, 27. https://doi.org/10.57017/jets.v1.iss1.02
Vancouver
MIHĂILESCU MI, MĂRĂSCU V, NIȚĂ ȘL. An Integrated Architecture Combining Blockchain, Artificial Intelligence, and Machine Learning for Decentralised Identity Management in the Internet of Things. Journal of Emerging Technologies for Society. 2026:27. doi:10.57017/jets.v1.iss1.02
BibTeX
@article{MIHILESCU2026,
author = {MIHĂILESCU, Marius Iulian and MĂRĂSCU, Valentina and NIȚĂ, Ștefania Loredana},
title = {An Integrated Architecture Combining Blockchain, Artificial Intelligence, and Machine Learning for Decentralised Identity Management in the Internet of Things},
journal = {Journal of Emerging Technologies for Society},
year = {2026},
pages = {27},
doi = {10.57017/jets.v1.iss1.02}
}
EUVIMED