Malicious AIS Spoofing and Abnormal Stealth Deviations: A Comprehensive Statistical Framework for Maritime Anomaly Detection

The automatic identification system (AIS) is an essential and economical equipment for collision avoidance and maritime surveillance. However, AIS can be subject to intentional reporting of false information, or “spoofing”. This article assumes the vessel trajectory nominally follows a piecewise mea...

Full description

Saved in:
Bibliographic Details
Published inIEEE transactions on aerospace and electronic systems Vol. 57; no. 4; pp. 2093 - 2108
Main Authors d'Afflisio, Enrica, Braca, Paolo, Willett, Peter
Format Journal Article
LanguageEnglish
Published New York IEEE 01.08.2021
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Subjects
Anomalies
Artificial intelligence
Automatic identification system (AIS)
Collision avoidance
Data models
data spoofing
Deviation
Hypothesis testing
Likelihood ratio
maritime anomaly detection
maritime security
maritime situational awareness
Model testing
model-order selection (MOS)
multiple statistical hypothesis test
Ornstein–Uhlenbeck (OU) process
Radar
Radar tracking
Spoofing
Surveillance
target tracking
Testing
Trajectory
Online AccessGet full text

Cover

More Information
Summary:The automatic identification system (AIS) is an essential and economical equipment for collision avoidance and maritime surveillance. However, AIS can be subject to intentional reporting of false information, or “spoofing”. This article assumes the vessel trajectory nominally follows a piecewise mean-reverting process; thereby, it addresses the problem of establishing whether a vessel is reporting adulterated position information through AIS messages in order to hide its current planned route and a possible deviation from the nominal route. Multiple hypothesis testing suggests a framework to enlist reliable information from monitoring systems (coastal radars and space-born satellite sensors) in support of detection of anomalies, spoofing, and stealth deviations. The proposed solution involves the derivation of anomaly detection rules based on the generalized likelihood ratio test and the model-order selection methodologies. The effectiveness of the proposed anomaly detection strategy is tested for different case studies within an operational scenario with simulated data.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0018-9251
1557-9603
DOI:10.1109/TAES.2021.3083466