Marine Soundscape Monitoring Enabled by Artificial Intelligence-Aided Integrated Distributed Sensing and Communications

Optical fibers, the backbone of global transoceanic communications infrastructure, have the potential to support distributed acoustic sensing (DAS) enhanced by pattern recognition algorithms for oceanic soundscape monitoring and animal conservation. This work enhances subsea telecommunication single...

Full description

Saved in:
Bibliographic Details
Main Authors: Juan M. Marin, Wahyu Hendra Gunawan, Alaaeddine Rjeb, Islam Ashry, Baoshi Sun, Talha Ariff, Chun Hong Kang, Tien Khee Ng, Shinkyu Park, Carlos M. Duarte, Boon S. Ooi
Format: Article
Language:English
Published: American Association for the Advancement of Science (AAAS) 2025-01-01
Series:Advanced Devices & Instrumentation
Online Access:https://spj.science.org/doi/10.34133/adi.0089
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Optical fibers, the backbone of global transoceanic communications infrastructure, have the potential to support distributed acoustic sensing (DAS) enhanced by pattern recognition algorithms for oceanic soundscape monitoring and animal conservation. This work enhances subsea telecommunication single-mode fibers (SMFs) by using wavelength division multiplexing (WDM) to concurrently transmit communication and DAS signals. We identified orthogonal frequency division multiplexing (OFDM) as the optimal modulation scheme, minimizing interference between DAS and communications. OFDM achieved a signal-to-noise ratio (SNR) of 10.71 dB for DAS, compared to 8.76 dB for on-off keying (OOK), and enabled a data rate of 7.59 Gbps with a bit error rate (BER) of 2.93 × 10−3 over 1 km of SMF. The system, tested in a water tank with marine animal sounds, achieved over 90% classification accuracy when integrated with a convolutional neural network (CNN), and maintained 85% accuracy in real-world conditions in the Red Sea. These results demonstrate the potential for enhanced oceanic monitoring without the need for installing new subsea monitoring tools.
ISSN:2767-9713