Towards underwater ship and unmanned vehicle detection using distributed acoustic sensing

Kurzfassung

Within the last decade, technological advancement and research development have made it possible to exploit existing underwater fiber optic cables for monitoring their environment as well as their intrinsic properties. Besides their scientific value, those cables are critical parts of the maritime infrastructure and, thus, became targets for sabotage through vessels. Besides the pressure waves that vessels produce, the mechanical parts of vessels such as propellers and engines create acoustic waves with distinct signatures at specific frequencies. Both waves alter the scattering locations of back-scattering light within the fibers. By using the principles of distributed acoustic sensing, information about the change in scattering locations enables us to detect vessels and unmanned vehicles around the cables. This study aims to test a set of signal-based features including frequency peaks and entropy for their performance in detecting vessels and unmanned vehicles underwater. We compared signatures from different ships traveling close to existing underwater cables as well as from unmanned vehicles operating close to standard fiber optic cables deployed in a lake in Northern Germany. Besides different environmental conditions and measurement parameters we present different indicators suitable for ship and vehicle detection. This provides a complement to object detection and protection of the cable infrastructures.