The application of artificial neural networks to the characterization of interior noise booming in passenger cars

Abstract

The booming index has been developed recently to evaluate the sound characteristics of passenger cars. Previous work maintained that booming sound quality is related to loudness and sharpness -the sound metrics used in psychoacoustics-and that the booming index is developed by using the loudness and sharpness for a signal within whole frequencies between 20 Hz and 20 kHz. In the present paper, the booming sound quality was found to be effectively related to the loudness at frequencies below 200 Hz; thus the booming index is updated by using the loudness of the signal filtered by the low pass filter at frequencies under 200 Hz. The relationship between the booming index and the sound metric is identified by an artificial neural network (ANN). Interior sounds of 10 passenger cars were measured, and 21 persons subjectively evaluated the booming sound qualities for these interior sounds. Throughout this research, it was found that there is a high correlation between the results of these evaluations and the output of a neural network. The updated booming index has been successfully applied to the objective evaluation of the booming sound quality of massproduced passenger cars.