Sound power level estimation – choice of the prior distribution

Bayesian inference is one of the methods used to determine the sound power level of sound sources. This method requires knowledge of two probability distributions. The first is the sampling density, while the second is the prior distribution. In this study, the effect of the prior distribution on the sound power level estimation results was investigated. For this purpose, three prior distributions were used: 1) a normal distribution, 2) a distribution determined using the kernel density estimator, 3) a uniform distribution. The sound power level results determined by the engineering method were used to illustrate the proposed solutions and carry out the analysis. The results of the experiment were compared with the results of the sound power level determined using the precision method in the hemi-anechoic room according to ISO 3745:2012. The statistical inference has been carried out based on results of non-parametric statistical tests at the significance level α = 0.05.