Integrating Distributed Acoustic Sensing and PINN Frameworks for Enhanced Indoor Sound Source Localization

27 Sept 2024 (modified: 05 Feb 2025)Submitted to ICLR 2025EveryoneRevisionsBibTeXCC BY 4.0
Keywords: distributed acoustic sensing, physics informed neural networks, room acoustics, sound source localization
Abstract: Distributed Acoustic Sensing (DAS) is an emerging technology that transforms standard optical fibers into dense arrays of acoustic sensors, offering unprecedented opportunities for smart city applications, indoor monitoring of human activity, and surveillance without compromising privacy. In this paper, we integrate DAS with Physics-Informed Neural Networks (PINNs) for indoor sound source localization. By embedding the acoustic wave equation and impedance boundary conditions into the neural network architecture, we exploit physical laws to guide the learning process, improving accuracy and generalization. We propose two strategies for real-time sound source localization using DAS data. The first strategy involves training the PINN on all available data simultaneously, while the second strategy incrementally feeds data over time, simulating real-time data acquisition. Using real indoor DAS measurements, we demonstrate the effectiveness of our approach in deciphering complex room acoustics and accurately inferring sound source locations under both strategies. Our framework provides a novel solution for real-time indoor positioning and human activity surveillance, offering significant advantages over traditional camera-based systems by preserving individual privacy.
Primary Area: applications to physical sciences (physics, chemistry, biology, etc.)
Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.
Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2025/AuthorGuide.
Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.
No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.
Submission Number: 11052
Loading

OpenReview is a long-term project to advance science through improved peer review with legal nonprofit status. We gratefully acknowledge the support of the OpenReview Sponsors. © 2025 OpenReview