APPLICATION OF THE BAYESIAN INFERENCE FOR ESTIMATION OF THE LONG-TERM NOISE INDICATORS AND THEIR UNCERTAINTY

摘要

The problem of estimation of the environmental noise hazard indices and their uncertainty is presented in the hereby paper. The main attention is focused on the estimation process of the long-term noise indicators and their type A standard uncertainty defined by the standard deviation of the mean of the measurement results. The rules given in the ISO/IEC Guide 98 are used in the calculations. It is usually determined by means of the classic variance estimators, at the assumption of the normality of measurements results. However, such assumption in relation to the acoustic measurements is rather questionable. This is the reason, that the authors indicated the necessity of implementation of non-classic statistic solutions. There was formulated the estimation idea of seeking density function of long-term noise indicators distribution by Bayesian inference, which doesn't generate limitations for form and properties of analyzed statistics. There was presented theoretical basis of the proposed method, and the example of calculation process which make possible determining searching estimators of expected value and variance of long-term noise indicators L_(DEN) and L_N. The illustration for indicated solutions and usefulness analysis was constant monitoring results of a traffic noise recorded on one of the main arteries of Krakow, Poland.