Fuzzy logic, edge enabled underwater video surveillance through partially wireless optical communication.

Stewart, Craig; Fough, Nazila; Prabhu, Radhakrishna

Abstract

Underwater surveillance is inherently tricky to achieve. Even in the clearest waters, the visibility tends to be in the range of tens of meters. Normally, tethered Remotely Operated Vehicles (ROVs) with underwater cameras are used for underwater imaging at closer ranges. Currently, detailed visible light imaging can be achieved utilising green laser technology, and this is limited to close ranges due to the inherent properties of light attenuation in water. The alternative is to utilise sonar based imaging which is capable of visualising distances, however, this technique is vulnerable to noise that interferes with the operating frequency, rendering the applications somewhat limited. The emergence of high data-rate, wireless, optical communication could allow for dense placement of short-range imaging equipment to monitor areas of strategic interest to extend the range, however, there needs to be a reliable method of wirelessly communicating this data to the sea surface regardless of the localised environmental conditions that may interfere with a visible light transmission. This paper proposes a fuzzy logic, edge computing enabled routing algorithm for optical networks that utilises a wired connection among source nodes to "pass" video data around among themselves to decide which seafloor node is best placed to transmit the data according to relative local turbidity, light intensity and sea-life activity, the main factors that hamper a well-considered wireless optical network. From there, a selected node can theoretically transmit the data from the source to the sea-surface through the wireless optical relay network implemented above. This mechanism shows promise in improving link reliability and throughput compared to alternative systems.

Citation

STEWART, C., FOUGH, N. and PRABHU, R. 2023. Fuzzy logic, edge enabled underwater video surveillance through partially wireless optical communication. In Bouma, H., Dijk, J., Prabhu, R., Stokes, R.J. and Yitzhaky, Y. (eds.) Artificial intelligence for security and defence applications: proceedings of the 2023 SPIE joint Remote sensing and Security + defence conference (SPIE Sensor + Imaging 2023), 3-7 September 2023, Amsterdam, Netherlands. Proceedings of SPIE, 12742. Bellingham, WA: SPIE [online], paper 127420A. Available from: https://doi.org/10.1117/12.2685278

• Presentation Conference Type: Conference Paper (published)
• Conference Name: 2023 SPIE (International Society for Optics and Photonics) joint Remote sensing and Security + defence conference (SPIE Sensor + Imaging 2023)
• Start Date: Sep 3, 2023
• End Date: Sep 6, 2023
• Acceptance Date: Jun 15, 2023
• Online Publication Date: Oct 17, 2023
• Publication Date: Dec 31, 2023
• Deposit Date: Nov 14, 2023
• Publicly Available Date: Nov 14, 2023
• Publisher: Society of Photo-optical Instrumentation Engineers
• Peer Reviewed: Peer Reviewed
• Series Title: Proceedings of SPIE
• Series Number: 12742
• Series ISSN: 0277-786X; 1996-756X
• Book Title: Artificial intelligence for security and defence applications: proceedings of the 2023 SPIE (International Society for Optics and Photonics) joint Remote sensing and Security + defence conference (SPIE Sensor + Imaging 2023), 3-7 September 2023, Amsterda
• ISBN: 9781510667136
• DOI: https://doi.org/10.1117/12.2685278
• Keywords: Optical networks; Remote sensing; Underwater imaging; Remotely operated vehicles (ROVs)
• Public URL: https://rgu-repository.worktribe.com/output/2072360