Surface impedance inference via neural fields and sparse acoustic data obtained by a compact array

2026-02-11

AI Summary

This work proposes an in situ acoustic inversion method based on physics-informed neural fields to overcome the limitations of conventional sound-absorbing material characterization, which relies on idealized acoustic field assumptions and struggles in real-world environments. By leveraging sparse sound pressure measurements from a compact microphone array, the approach reconstructs near-surface broadband acoustic fields and directly infers complex surface impedance. A novel parallel multi-frequency neural field architecture enables efficient broadband impedance inversion within seconds to minutes, while low-complexity hardware integration facilitates deployment in practical scenarios. Numerical simulations and experimental validation demonstrate that the method achieves high-accuracy surface impedance reconstruction with only a few sensors and has been successfully applied in automotive cabin environments to guide optimal sensor placement.

Technology Category

• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Machine Learning: Hardware-aware ML
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Web performance, measurement, and characterization
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation

Abstract

Standardized laboratory characterizations for absorbing materials rely on idealized sound field assumptions, which deviate largely from real-life conditions. Consequently, \emph{in-situ} acoustic characterization has become essential for accurate diagnosis and virtual prototyping. We propose a physics-informed neural field that reconstructs local, near-surface broadband sound fields from sparse pressure samples to directly infer complex surface impedance. A parallel, multi-frequency architecture enables a broadband impedance retrieval within runtimes on the order of seconds to minutes. To validate the method, we developed a compact microphone array with low hardware complexity. Numerical verifications and laboratory experiments demonstrate accurate impedance retrieval with a small number of sensors under realistic conditions. We further showcase the approach in a vehicle cabin to provide practical guidance on measurement locations that avoid strong interference. Here, we show that this approach offers a robust means of characterizing \emph{in-situ} boundary conditions for architectural and automotive acoustics.

Problem

• surface impedance
• in-situ characterization
• acoustic boundary conditions
• sparse acoustic data
• absorbing materials

Innovation

• physics-informed neural field
• surface impedance inference
• sparse acoustic data
• compact microphone array
• in-situ characterization