Assessing the stochastic error of acoustic scattering matrices using linear methods

To be able to compare the measured scattering matrices with model predictions, the quality of the measurements has to be known. Uncertainty analyses are invaluable to assess and improve the quality of measurement results in terms of accuracy and precision. Linear analyses are widespread, computationally fast and give information of the contribution of each error source to the overall measurement uncertainty; however, they cannot be applied in every situation. The purpose of this study is to determine if linear methods can be used to assess the quality of acoustic scattering matrices. The uncertainty in measured scattering matrices is assessed using a linear uncertainty analysis and the results are compared against Monte-Carlo simulations. It is shown that for plane waves, a linear uncertainty analysis, applied to the wave decomposition method, gives correct results when three conditions are satisfied. For higher order mode measurements, the number of conditions that have to be satisfied increases rapidly and the linear analysis becomes an unsuitable choice to determine the uncertainty on the scattering matrix coefficients. As the linear uncertainty analysis is most suitable for the plane wave range, an alternative linear method to assess the quality of the measurements is investigated. This method, based on matrix perturbation theory, gives qualitative information in the form of partial condition numbers and the implementation is straightforward. Using the alternative method, the measurements of higher order modes are analyzed and the observed difference in the measured reflection coefficients for different excitation conditions is explained by the disparity in modal amplitudes.