Proactive Threat Hunting and Anomaly Detection for Networked Consumer Maritime Electronics Using a Variational Bayesian Adaptive IMM
IEEE Transactions on Consumer Electronics · 2026-05 · 72 (2) · 4693-4704
DOI: 10.1109/tce.2026.3682771 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Consumer Electronics, 2026-05. Verfasst von Qiuyang Dai und 2 weiteren Autor:innen. Die Arbeit nennt 34 Literaturangaben. Gefördert von: University Independent Research Project.
Abstract (Originalsprache)
Next-generation intelligent consumer electronics, including autonomous recreational vessels and autonomous surface vehicles, require reliable situational awareness during abrupt maneuvers, sensor faults, and disturbance-like interference. However, traditional centralized tracking pipelines suffer from communication bottlenecks and have limited capabilities for proactive anomaly awareness. This study proposed a threat-hunting-oriented tracking framework for consumer maritime electronics based on a collaborative edge inference paradigm and a Variational Bayesian Adaptive Interacting Multiple Model (VBTA-IMM) engine for local inference. Proactive anomaly indication was achieved through a variational Bayesian update of the transition probability matrix. Mahalanobis-distance-based prediction errors rapidly down-weighted mismatched motion models and accelerated mode switching. In addition, a quasi-uniform B-spline fitting strategy was introduced to adaptively identify the turn rate and refine the constant-turn model set for unknown maneuvers. A square-root cubature information filter was further integrated into the IMM framework to enhance numerical stability and multi-sensor fusion on resource-constrained consumer devices. Simulation and field tests showed that the proposed method improved position and velocity accuracy by 28.64% and 44.28%, respectively, compared with conventional IMM variants. These results demonstrate its effectiveness for proactive tracking and anomaly detection in networked consumer systems. Nevertheless, broader real-world embedded deployments still require further validation under stricter onboard computing and communication constraints and more heterogeneous sensing conditions.
Förderung
- University Independent Research Project
Zitieren
APA
Dai, Q., Lu, F., & Xu, J. (2026). Proactive Threat Hunting and Anomaly Detection for Networked Consumer Maritime Electronics Using a Variational Bayesian Adaptive IMM. IEEE Transactions on Consumer Electronics, 72(2), 4693-4704. https://doi.org/10.1109/tce.2026.3682771
Vancouver
Dai Q, Lu F, Xu J. Proactive Threat Hunting and Anomaly Detection for Networked Consumer Maritime Electronics Using a Variational Bayesian Adaptive IMM. IEEE Transactions on Consumer Electronics. 2026;72(2):4693-4704. doi:10.1109/tce.2026.3682771
BibTeX
@article{Dai2026,
author = {Dai, Qiuyang and Lu, Faxing and Xu, Junfei},
title = {Proactive Threat Hunting and Anomaly Detection for Networked Consumer Maritime Electronics Using a Variational Bayesian Adaptive IMM},
journal = {IEEE Transactions on Consumer Electronics},
year = {2026},
volume = {72},
number = {2},
pages = {4693--4704},
doi = {10.1109/tce.2026.3682771}
}
EUVIMED