Maritime Traffic Networks: From Historical Positioning Data to Unsupervised Maritime Traffic Monitoring

The large maritime traffic volume and its implications in economy, environment, safety, and security require an unsupervised system to monitor maritime traffic. In this paper, a method is proposed to automatically produce synthetic maritime traffic representations from historical self-reporting positioning data, more specifically from automatic identification system data. The method builds a two-layer network that represents the maritime traffic in the monitored area, where the external layer presents the network's basic structure and the inner layer provides precision and granularity to the representation. The method is tested in a specific scenario with high traffic density, the Baltic Sea. Experimental results reveal a decrease of over 99% storage data with a negligible precision drop. Finally, the novel method presents a light and structured representation of the maritime traffic, which sets the foundations to real-time automatic maritime traffic monitoring, anomaly detection, and situation prediction.