Behavioral Anomaly Detection for IoT Devices Using Lightweight Models
Research Digest on Engineering Management and Social Innovations · 13.04.2026 · 2 (4)
DOI: 10.46647/rdems0204008 ↗Kurz gefasst
Fachartikel in der Zeitschrift Research Digest on Engineering Management and Social Innovations, online erschienen am 13.04.2026. Verfasst von Dr. Suneetha Mulumudi und 4 weiteren Autor:innen.
Abstract (Originalsprache)
Behavioral anomaly detection has become an essential approach to securing Internet of Things (IoT) environments, particularly due to the limited computational capabilities and lack of built-in security mechanisms in IoT devices. Traditional intrusion detection systems rely on signature-based techniques, which are ineffective against unknown or evolving threats. This paper proposes a lightweight anomaly detection framework designed for edge deployment using a Raspberry Pi. The system passively collects network traffic and extracts statistical features to model normal device behavior. An unsupervised machine learning approach is employed to detect deviations using percentile-based thresholding. The proposed framework ensures privacy by avoiding payload inspection and minimizes computational overhead. Experimental evaluation demonstrates that the system effectively detects anomalous behavior while maintaining low resource consumption. The results indicate that the proposed solution is suitable for real-time deployment in smart environments and resource-constrained IoT networks.
Zitieren
APA
Mulumudi, D. S., Juluru, V., Thorne, M., Tejavath, S., & Macharla, S. (2026). Behavioral Anomaly Detection for IoT Devices Using Lightweight Models. Research Digest on Engineering Management and Social Innovations, 2(4). https://doi.org/10.46647/rdems0204008
Vancouver
Mulumudi DS, Juluru V, Thorne M, Tejavath S, Macharla S. Behavioral Anomaly Detection for IoT Devices Using Lightweight Models. Research Digest on Engineering Management and Social Innovations. 2026;2(4). doi:10.46647/rdems0204008
BibTeX
@article{Mulumudi2026,
author = {Mulumudi, Dr. Suneetha and Juluru, Varun and Thorne, Morarji and Tejavath, Subhadra and Macharla, Sowmya},
title = {Behavioral Anomaly Detection for IoT Devices Using Lightweight Models},
journal = {Research Digest on Engineering Management and Social Innovations},
year = {2026},
volume = {2},
number = {4},
doi = {10.46647/rdems0204008}
}
EUVIMED