Investigation on error model of in-vehicle acoustic transfer function based on the human body scattering effects

Abstract

Sound field control technology enhances the passenger experience by processing signals. The effectiveness of sound field control depends on the accurate measurement of the acoustic transfer function (ATF) and the ability to resist interference from acoustic environment. Due to the difficulty in obtaining the distribution of ATF errors caused by human body scattering and the constantly changing sitting situation inside the car, a method to establish an ATF error model for the vehicle interior is proposed in this paper. Firstly, to obtain errors for the probability distribution of ATF errors inside the vehicle, the actual ATF were obtained through experimental testing. Then, combined with psychoacoustics and kernel density estimation methods, the probability distribution of the ATF errors in each Bark frequency band was calculated. Finally, the relevant parameters of the error model were determined using probability density function to establish the ATF error model, and the model was applied and verified through multizone sound field control of the in-vehicle environment. The results show that the use of the ATF error model proposed in this paper effectively improves the control effect of the in-vehicle sound field, providing an effective approach for establishing the ATF error model for the in-vehicle sound field.