The LAMA system: A “smart” magnetometer network for harbour protection

This work describes the development of an underwater anti-intrusion system based on a magnetometer self-informed network, whose purpose is to detect the presence of threats in the proximity of critical infrastructures (e.g, terrorist divers in harbours). In this context, the magnetic network fills the gaps of sonar systems at the critical boundaries of the water volume to be controlled (sea bed, docks, …), where acoustic performances deteriorate due to reflections and attenuations. The system operates in a port-protection scenario, characterized by a medium-high environmental magnetic noise that can hide the diver signal (a diver is a weak, quasi-point-like, moving source). The magnetometer network processes two inputs: the environmental magnetic noise and a signal including the target magnetic signal superimposed to the same noise; the frequencies of a diver signal lie within the noise band, hence frequency filtering proves inadequate for noise removal. The basic idea underlying the system is to measure and use the noise itself to filter the overall signal; measuring noise supports a background-subtraction process that allows to extract the target signal and therefore detect the threat presence. The effectiveness of the procedure depends on the positions of magnetometers: sensors must be close enough to one another to measure the common background noise, and, at the same time, should be distant enough from one another so that just one sensor can measure the target signal. To generate alarms when a threat is detected, a real-time software application processes data and activates a visual and acoustic alarm upon identification of a magnetic anomaly. Sea trials carried out in port areas provided extremely satisfactory results in the detection of intruders. The paper presents experimental results obtained during the method validation tests, when intruders were moving in the surrounding undersea environment.