Volume 13, Issue 11 · IJARST · ISSN 2319-1783 / 2320-1126

A Federated Learning Based Privacy Preserving Dynamic Access Control Framework for Distributed Systems

A Federated Learning Based Privacy Preserving Dynamic Access Control Framework for Distributed Systems is an open-access, peer-reviewed research paper by PAUL NDUDIRI OHIA, EMMANUEL WARIBO OTITI, published in Volume 13, Issue 11 of the International Journal of Advanced Research in Science and Technology (IJARST), a UGC-approved journal (Print ISSN 2319-1783, Online ISSN 2320-1126).

Author

PAUL NDUDIRI OHIA, EMMANUEL WARIBO OTITI

Abstract

Traditional access control systems usually depend on centralized authentication, authorization policies, and user activity monitoring. Although this approach is useful in small environments, it becomes risky in distributed systems where institutions, branches, or edge nodes generate sensitive behaviour data. Sending all access logs to a central server may expose private user activities, device information, login patterns, protected-resource usage, and institutional security operations. This study developed and evaluated a Federated Learning-Based Privacy-Preserving Dynamic Access Control Framework for Distributed Systems. The completed web application simulated three institutions: University A, University B, and University C. Each institution maintained its own local access logs, trained a local Logistic Regression model, and submitted model updates rather than raw logs to a central federated aggregator. The aggregator applied Federated Averaging to generate a global model, which was redistributed to the participating institutions for dynamic access-risk prediction. The system used behavioural features such as login time, device fingerprint, IP/location risk, requested resource, action type, failed attempts, request rate, session age, user role, and institution code. The implemented system generated 2,404 access-control records, completed six FedAvg rounds, recorded 18 local model-update submissions, and produced graduated access decisions including Allow Access, Allow with Monitoring, Step-up Verification, Restricted Access, and Deny and Alert Admin. The results showed that the framework supported collaborative access-control intelligence while keeping raw institutional access logs local.

Keywords: Federated Learning, Dynamic Access Control, Privacy Preservation, FedAvg, FedProx, Distributed Systems, Behaviour Analytics, Edge Security

DOI: https://doi.org/10.62226/ijarst20262844

 

REFERENCES:

[1] S. Agrawal, S. Sarkar, O. Aouedi, G. Yenduri, K. Piamrat, M. Alazab, S. Bhattacharya, P. K. R. Maddikunta, and T. R. Gadekallu, “Federated Learning for Intrusion Detection System: Concepts, Challenges and Future Directions,” Computer Communications, vol. 195, pp. 346–361, 2022. https://doi.org/10.1016/j.comcom.2022.09.012

[2] E. M. Campos, P. F. Saura, A. González-Vidal, J. L. Hernández-Ramos, J. Bernal Bernabé, G. Baldini, and A. Skarmeta, “Evaluating Federated Learning for Intrusion Detection in Internet of Things: Review and Challenges,” Computer Networks, vol. 203, article 108661, 2022. https://doi.org/10.1016/j.comnet.2021.108661

[3] N. Hamdi, “Federated Learning-Based Intrusion Detection System for Internet of Things,” International Journal of Information Security, vol. 22, pp. 1937–1948, 2023. https://doi.org/10.1007/s10207-023-00727-6

[4] M. Sarhan, S. Layeghy, N. Moustafa, and M. Portmann, “Cyber Threat Intelligence Sharing Scheme Based on Federated Learning for Network Intrusion Detection,” Journal of Network and Systems Management, vol. 31, article 3, 2023. https://doi.org/10.1007/s10922-022-09691-3

[5] D. Javeed, M. S. Saeed, M. Adil, P. Kumar, and A. Jolfaei, “A Federated Learning-Based Zero Trust Intrusion Detection System for Internet of Things,” Ad Hoc Networks, vol. 162, article 103540, 2024. https://doi.org/10.1016/j.adhoc.2024.103540

[6] M. A. T. Ayedh, A. W. A. Wahab, and M. Y. I. Idris, “Enhanced Adaptable and Distributed Access Control Decision Making Model Based on Machine Learning for Policy Conflict Resolution in BYOD Environment,” Applied Sciences, vol. 13, no. 12, 2023, p. 7102.

DOI

10.62226/ijarst20262844

PAGES : 1521-1527 | 6 VIEWS | 2 DOWNLOADS

How do you cite this paper?

PAUL NDUDIRI OHIA, EMMANUEL WARIBO OTITI — “A Federated Learning Based Privacy Preserving Dynamic Access Control Framework for Distributed Systems.” International Journal of Advanced Research in Science and Technology (IJARST), Volume 13, Issue 11. DOI: 10.62226/ijarst20262844.

Read / Download Full Article