Database Reference
In-Depth Information
vessels. Devices from these two classes broadcast information at different time
intervals (Table
11.1
) and at different ranges (typically 20-40 miles for class A
and generally 5-10 miles for class B).
Enhanced Worldwide Positioning Systems
are emerging especially to address drawbacks of both systems, which are com-
plementary but imperfect. On one hand, ARPA is useful to detect and track ves-
sels that might not have AIS devices onboard. On the other hand, it brings limited
information and cannot identify a mobile object, and its coverage includes blind
areas. The automatic identification system is useful to obtain more complete
information, but devices are not available on all ships and data can be falsified.
The most important issue that guides evolutions concerns the limited tracking
range of both systems, which is insufficient to follow ships engaged on interna-
tional journeys. Satellite communications systems are going to be more inten-
sively employed, in particular to enhance or replace the AIS. For instance, Long-
Range Identification and Tracking (LRIT) reports vessels' positions to their flag
administrations at least four times a day. Satellite-based AIS-monitoring service
(S-AIS) uses satellite communications to broadcast AIS information. Nowadays,
position reports for European coasts reach almost 1.5 million positions per day
(about 72,000 ships). The ever-increasing data flows provided by this evolution
are going to emphasize issues on maritime data integration, fusion, filtering,
processing, and analysis.
Location-Based Data
While radar data are limited to a tuple composed of an identifier, a position,
and a related time, the automatic identification system broadcasts a wide range
of richer information. Information systems onboard or in vessel traffic services
generally merge AIS and radar positions into a single accurate one. When a
ship is not fitted with an AIS (typically small boats), the reported informa-
tion for data analysis is only limited to the aforementioned tuple. From our
perspective, this does not impact the data mining process and therefore moti-
vate an analysis focusing on the AIS data more easily accessible. Transmitted
AIS data come from twenty-seven different messages, each providing specific
information either related to the behavior of the AIS system or to a ship's
locations and characteristics. Positioning data defines point-based trajectories
describing 2D routes on the sea surface. That is, an ordered series of locations
expressed in WGS84 format (latitude
λ
, longitude
ϕ
, time
t
) of a given mobile
object with
t
indicating the timestamp of the location (
λ, ϕ
). Among all the
received data, meaningful information that can be considered in a purpose of
movement discovery and understanding can be classified in the three following
categories:
