Sound Field Reconstruction using Bayesian Inference (vor Ort)
* Presenting author
Abstract:
Capturing the spatial properties of a sound field in a room is an important task for various applications, e.g. in sound field analysis and sound field control. Usually a large number of measurements is required to determine the sound field within an extended region of the room. In order to reduce the experimental cost, it is the goal to reconstruct the sound field within a room based on only a limited number of measurement points. In this contribution, a Bayesian approach for sound field reconstruction in the modal frequency range is presented, since Bayesian probability theory is well suited to handle this lack of experimental observations. The reconstruction method is applied to a measured sound field in a horizontal evaluation area of a lightly damped room. The Bayesian approach shows an accurate reconstruction of the sound field and in addition provides a probability distribution of the sound pressure at every reconstruction point. This additional information allows to quantify the uncertainty of the reconstruction and provides insights into its robustness.
