A new methodology is proposed to deceive an anomalous trajectory detector by designing ship paths that deviate from the nominal traffic routes in an optimized way. The route planning is formalized as a min-max problem (with respect to surveillance system acquisition instants) focusing on the Kullback-Leibler (KL) divergence between the statistical hypotheses of the nominal and the anomalous trajectories as key performance measure. Modeling the vessel's dynamic according to the Ornstein-Uhlenbeck (OU) mean-reverting stochastic process, physical, practical, and kinematic requirements are also accounted for forcing several constraints at the design stage. A computationally efficient technique is proposed to handle the resulting non-convex optimization problem, and some case studies are reported to assess its effectiveness.
Optimal Stealth Trajectory Design to Deceive Anomaly Detection Process / Enrica d'Afflisio. - ELETTRONICO. - (2019), pp. 0-0. (Intervento presentato al convegno OCEANS 2019 - Marseille tenutosi a Marseille, France nel 17-20 June 2019) [10.1109/OCEANSE.2019.8867147].
Optimal Stealth Trajectory Design to Deceive Anomaly Detection Process
Enrica d'Afflisio
2019
Abstract
A new methodology is proposed to deceive an anomalous trajectory detector by designing ship paths that deviate from the nominal traffic routes in an optimized way. The route planning is formalized as a min-max problem (with respect to surveillance system acquisition instants) focusing on the Kullback-Leibler (KL) divergence between the statistical hypotheses of the nominal and the anomalous trajectories as key performance measure. Modeling the vessel's dynamic according to the Ornstein-Uhlenbeck (OU) mean-reverting stochastic process, physical, practical, and kinematic requirements are also accounted for forcing several constraints at the design stage. A computationally efficient technique is proposed to handle the resulting non-convex optimization problem, and some case studies are reported to assess its effectiveness.File | Dimensione | Formato | |
---|---|---|---|
08867147.pdf
accesso aperto
Tipologia:
Pdf editoriale (Version of record)
Licenza:
Open Access
Dimensione
856.42 kB
Formato
Adobe PDF
|
856.42 kB | Adobe PDF |
I documenti in FLORE sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.