Vessel Trajectory Route Spoofed Points Detection Using AIS Data: A Bi-LSTM Approach

The Automatic Identification System (AIS), which provides real-time vessel information for collision avoidance and marine domain awareness, is vital to maritime navigation and safety. However, AIS is vulnerable to GPS spoofing attacks, where malicious actors transmit false GPS signals to mislead vessels about their location. These attacks pose significant risks to maritime safety and security. In this paper, a novel approach to detect spoofed points within vessel trajectory routes using AIS data is proposed. The methodology leverages the power of Bidirectional Long Short-Term Memory (Bi-LSTM) networks, a deep learning architecture adept at capturing temporal dependencies in sequential data. By analysing AIS data streams, the proposed model identifies anomalies and deviations from expected patterns, effectively pinpointing instances of spoofing. Numerous tests using real-world AIS datasets were carried out, which showed that the suggested Bi-LSTM model outperformed other spoofing detection techniques. The work advances the realm of marine cybersecurity by offering a more reliable and accurate method of AIS spoofing attack detection.