Determination of the acoustical diffusion and impedance characteristics of surfaces by Bayesian inversion of a modified Helmholtz equation

A surface diffusion coefficient is used in architectural-acoustics to evaluate the effectiveness of diffusing surfaces. The inclusion of the diffusion characteristics is also important for the accuracy of room prediction models. Another important parameter is the absorption or impedance of a surface. In settings with significant low-frequency noise, phase effects are important; consequently impedance values of surfaces are necessary for accurate modeling. A review of existing models for specular and diffuse reflection is made. A new diffusion coefficient is defined and included in three new forward models for predicting the steady-state sound-pressure level above a finite-impedance plane in an otherwise free field. Data are collected for several typical architectural surfaces in an anechoic chamber. Inverse methods are utilized in order to estimate the diffusion coefficient for surfaces given each of the models. This is done without knowledge of the surface impedance, which is simultaneously estimated. The models are compared with each other and with independently measured values of the surface impedance and diffusion. Inversion is found to be a reasonable way of determining the diffusion properties of a surface.Applied Science, Faculty ofMechanical Engineering, Department ofGraduat