Online sequential pattern analysis for anomaly detection in resource constrained edge devices
Global Media and Social Sciences Research Journal · 10.05.2026 · 7 (1) · 166-172
DOI: 10.71465/gmssrj197 ↗Kurz gefasst
Fachartikel in der Zeitschrift Global Media and Social Sciences Research Journal, online erschienen am 10.05.2026. Verfasst von Wei Ming Tan und 2 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0).
Abstract (Originalsprache)
Edge devices deployed in wearable computing and extended reality systems generate continuous streams of system logs and operational signals. Detecting anomalies in such environments requires lightweight algorithms capable of processing sequential data under strict computational and memory constraints. This study proposes an online sequential pattern analysis framework for anomaly detection in resource-constrained edge devices. The approach combines incremental log template extraction, probabilistic sequence modeling, and adaptive threshold estimation to identify abnormal runtime behaviors in real time. A lightweight gated recurrent architecture was designed to capture temporal dependencies in system events while maintaining minimal computational overhead. The proposed framework was evaluated on 89.6 million log records collected from 312 edge devices, including wearable sensors, XR controllers, and embedded processing modules. Experimental results show that the proposed model achieves an F1-score of 0.887, outperforming rule-based monitoring and statistical baselines by 15.4% and 12.1%, respectively. In addition, the method reduces memory consumption by 48% compared with conventional deep learning approaches and maintains detection latency below 2 seconds for over 92% of anomalous events. These results demonstrate that sequential pattern analysis provides an effective and efficient solution for anomaly detection in edge computing environments.
Zitieren
APA
Tan, W. M., Chen, L. N., & Lim, J. J. (2026). Online sequential pattern analysis for anomaly detection in resource constrained edge devices. Global Media and Social Sciences Research Journal, 7(1), 166-172. https://doi.org/10.71465/gmssrj197
Vancouver
Tan WM, Chen LN, Lim JJ. Online sequential pattern analysis for anomaly detection in resource constrained edge devices. Global Media and Social Sciences Research Journal. 2026;7(1):166-172. doi:10.71465/gmssrj197
BibTeX
@article{Tan2026,
author = {Tan, Wei Ming and Chen, Li Na and Lim, Jun Jie},
title = {Online sequential pattern analysis for anomaly detection in resource constrained edge devices},
journal = {Global Media and Social Sciences Research Journal},
year = {2026},
volume = {7},
number = {1},
pages = {166--172},
doi = {10.71465/gmssrj197}
}
EUVIMED