Acousto-optic capture of the sound field in a room based on sparse measurement data

Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review

110 Downloads (Pure)


Capturing the spatio-temporal properties of the sound field in a room is valuable for its characterization, as it enables to analyze the specific acoustic space. Current sensing methods typically aim at inferring directional properties of the sound field (i.e., directional impulse responses, power flows, etc.) at a specific location, or a combination of independent locations in the room. This work examines the use of novel sensing principles and reconstruction methods for capturing an acoustic field over a large region of space. This is a challenging problem, as measurements of the sound field over large volumes typically require an unfeasible experimental effort, particularly in large rooms. Specifically, we present an acousto-optic tomography method aimed at reconstructing the sound field from a sparse selection of seemingly incomplete data. A set of wave basis functions is used to interpolate the field, making it possible to predict the sound pressure field over a large spatial aperture. The study presents the measurement principle and experimental results of the volumetric reconstruction of a sound field in a lightly damped room.
Original languageEnglish
Title of host publicationProceedings of the International Symposium on Room Acoustics
PublisherNederlands Akoestisch Genootschap
Publication date2019
Publication statusPublished - 2019
EventInternational Symposium on Room Acoustics 2019 - Pakhuis de Zwijger, Amsterdam, Netherlands
Duration: 15 Sep 201917 Sep 2019


ConferenceInternational Symposium on Room Acoustics 2019
LocationPakhuis de Zwijger
Internet address


  • Room acoustic measurements
  • Acousto-optic effect
  • Sensing
  • Acoustic holography
  • Tomography
  • Compressive sensing
  • Inverse problems


Dive into the research topics of 'Acousto-optic capture of the sound field in a room based on sparse measurement data'. Together they form a unique fingerprint.

Cite this