ECE Professor Lutz Lampe, along with his MASc Student Lazar Atanackovic have been working towards securing marine infrastructure through the reliability of underwater communications via acoustic signaling. The purpose is to be able to detect, identify and localize various threats underwater. This could be for military or civilian environments, such as submerged mines, oil rigs or simply to detect scuba divers.
The project has been sponsored by the North Atlantic Treaty Organization (NATO), under the Science for Peace and Security program. The UBC researchers collaborate closely with Professor Paolo Casari at IMDEA Networks in Spain and Professor Roee Diamant from the The University of Haifa in Israel. Professor Diamant is a UBC alumni and former NSERC Vanier scholar, who completed his PhD in underwater communications using acoustics in our department. As the project is split into different aspects all pertaining to securing infrastructure in marine environments, Lampe’s team focuses on the signal processing side, combining theory, simulation and actual experimental work. This research has enjoyed kind support from Ocean Networks Canada (ONC), which provided data from its underwater listening stations.
Acoustic signaling works similar to how marine animals, like whales or dolphins communicate, where acoustic sounds are sent out and the reflections are listened to and analyzed. The same principles are applied to track divers or to scan the bottom of the ocean for stationary mines. Lampe and his team are focusing on communication reliability, specifically in harsh environments which means there are interferences present, in the form of external noise such as from ships. To remove these interference signals, the noise is first identified, then reconstructed and then it is subtracted, allowing for the intended communication signals to be heard more clearly. Applying directional hearing using arrays of hydrophones (underwater microphones) and algorithms to identify signal directions is another important part of the scope. This is alike the directional hearing you can do using your two ears, to listen such that you don’t hear the interference.
To determine if it is possible to zone out another signal and process it, in May 2019, Lazar took a 3-week trip to Israel where he prepared tests, participated in sea trials, and collected data. He notes that it is rare to get to do these kinds of experiments in this field, as their main focus is mostly models and computer simulations. During the trials, led by Professor Diamant, an autonomous underwater vehicle with attached cameras and sensors were deployed at sea, where the area was scanned to find suspected targets. Concurrently, multiple hydrophones were submerged, which facilitated the collection of acoustic data for directional hearing. During this time another boat was present, generating the external source interference. The experimental setup is shown in the video linked here.
The UBC researchers have shown that their interference-aware processing improves underwater acoustic transmission reliability. The collaborators Professors Diament and Casari have also proved that it is possible to track divers and identify underwater structures like mines. The next steps are to explore different machine learning techniques to build upon these ideas before the project comes to an end in May 2020. After that, the hope is for these techniques to be adopted for commercial projects used in military or civilian environments.
Learn more:
Threat-Defect: NATO SPS Project
Statistical Shipping Noise Characterization and Mitigation for Underwater Acoustic Communications
Low Probability of Detection for Underwater Acoustic Communication: A Review