Privacy Preserving Federated Intrusion Detection in Distributed Networks
IJAICET - International Journal of Artificial Intelligence, Cybersecurity and Emerging Technologies · 2026
DOI: 10.68307/5ercgy27 ↗Kurz gefasst
Fachartikel in der Zeitschrift IJAICET - International Journal of Artificial Intelligence, Cybersecurity and Emerging Technologies, 2026. Verfasst von Ram Chouhan und 2 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Collaborative network intrusion detection requires diverse training data while preserving organisational control over traffic records, therefore this study examines a federated detector that combines a convolutional recurrent classifier with update clipping and Gaussian perturbation alongside secure aggregation, using the original study tables to compare centralised training and ordinary federated averaging with the protected configuration on NSL-KDD and CICIDS2017, where reported protected model accuracies are 97.3% and 97.6% respectively compared with centralised accuracies of 98.6% and 99.1% and ordinary federated accuracies of 97.9% and 98.2%, while the reported communication totals increase from 142 MB to 148 MB corresponding to a calculated increase of approximately 4.23%, with the reported membership inference indicator decreasing from 34.1% to 9.7% although its attack definition and metric are unspecified, consequently these measurements provide descriptive evidence of a modest reported detection gap but do not establish statistical significance or a verified privacy guarantee, because run logs and the noise multiplier are unavailable and the corrected weighted aggregation formulation has not been rerun, while the accompanying analysis identifies the sensitivity and accounting conditions required to validate the target privacy budget before deployment claims can be justified.
Zitieren
APA
Chouhan, R., Jain, D. K., & Bagwani, M. K. (2026). Privacy Preserving Federated Intrusion Detection in Distributed Networks. IJAICET - International Journal of Artificial Intelligence, Cybersecurity and Emerging Technologies. https://doi.org/10.68307/5ercgy27
Vancouver
Chouhan R, Jain DK, Bagwani MK. Privacy Preserving Federated Intrusion Detection in Distributed Networks. IJAICET - International Journal of Artificial Intelligence, Cybersecurity and Emerging Technologies. 2026. doi:10.68307/5ercgy27
BibTeX
@article{Chouhan2026,
author = {Chouhan, Ram and Jain, Dr. Kirti and Bagwani, Mahesh Kumar},
title = {Privacy Preserving Federated Intrusion Detection in Distributed Networks},
journal = {IJAICET - International Journal of Artificial Intelligence, Cybersecurity and Emerging Technologies},
year = {2026},
doi = {10.68307/5ercgy27}
}
EUVIMED