Microsecond Federated SVD on Grassmann Manifold for Real-time IoT Intrusion Detection

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Nguyen, Tung-Anh, Bui, Van-Phuc, Pandey, Shashi Raj, Ta, Kim Hue, Tran, Nguyen H., Popovski, Petar
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866915566578565120
author Nguyen, Tung-Anh
Bui, Van-Phuc
Pandey, Shashi Raj
Ta, Kim Hue
Tran, Nguyen H.
Popovski, Petar
author_facet Nguyen, Tung-Anh
Bui, Van-Phuc
Pandey, Shashi Raj
Ta, Kim Hue
Tran, Nguyen H.
Popovski, Petar
contents This paper introduces FedSVD, a novel unsupervised federated learning framework for real-time anomaly detection in IoT networks. By leveraging Singular Value Decomposition (SVD) and optimization on the Grassmann manifolds, FedSVD enables accurate detection of both known and unknown intrusions without relying on labeled data or centralized data sharing. Tailored for deployment on low-power devices like the NVIDIA Jetson AGX Orin, the proposed method significantly reduces communication overhead and computational cost. Experimental results show that FedSVD achieves performance comparable to deep learning baselines while reducing inference latency by over 10x, making it suitable for latency-sensitive IoT applications.
format Preprint
id arxiv_https___arxiv_org_abs_2510_18501
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Microsecond Federated SVD on Grassmann Manifold for Real-time IoT Intrusion Detection
Nguyen, Tung-Anh
Bui, Van-Phuc
Pandey, Shashi Raj
Ta, Kim Hue
Tran, Nguyen H.
Popovski, Petar
Signal Processing
This paper introduces FedSVD, a novel unsupervised federated learning framework for real-time anomaly detection in IoT networks. By leveraging Singular Value Decomposition (SVD) and optimization on the Grassmann manifolds, FedSVD enables accurate detection of both known and unknown intrusions without relying on labeled data or centralized data sharing. Tailored for deployment on low-power devices like the NVIDIA Jetson AGX Orin, the proposed method significantly reduces communication overhead and computational cost. Experimental results show that FedSVD achieves performance comparable to deep learning baselines while reducing inference latency by over 10x, making it suitable for latency-sensitive IoT applications.
title Microsecond Federated SVD on Grassmann Manifold for Real-time IoT Intrusion Detection
topic Signal Processing
url https://arxiv.org/abs/2510.18501