Research Of Distortion Product Otoacoustic Emission Modeling And Hearing Detector

Otoacoustic emission produces in the cochlea, and it is an audio energywhich is released into the external auditory canal from ossicular chain and theperiosteum. For the discovery of otoacoustic emission, that not only can detectthe cochlea and distinguish the voice of the outside world, but also active voice.The clinical significance is that it provides a method which detects cochlearamplification function and outer hair cell function integrity. Now, there weremany studies of otoacoustic emission detector in abroad. The technology isrelatively mature and the price is relatively higher, but the hearing instrument ofhuman ears modeling are not reported. So we design a new hearing detectorwhich has Volterra kernel modeling function and reasonable price.Human ears are nonline system. In order to distortion product otoacousticemission modeling of the human ears, we further study the principle ofOtoacoustic emission mechanism, nonlinear system Volterra kernel description,Vandermode Volterra kernel and multi tone signal method. Combining with thecharacteristics of distortion product otoacoustic emission, we complete humanears modeling of Volterra kernel. We launch the same input signal to the humanear and modeling, comparing the output signal acquisition, and had proved it bysimulation, that the output signal is consistent, and the model is proved to becorrect.The Volterra kernel ear modeling is used in hearing instrument, as a newfunctional hearing detection. Each time a user detection of human ears hearingdetector, compared with the normal human ear modeling, judging whether thenormal human ear.This paper uses ATmega128microcontroller as the core controller, designsthe hardware and software of the hearing detector. In terms of hardware circuit,induced stimulation module consists of a probe, a digital to analog converter and an audio power amplifier, signal processing is composed of filter,program-controlled amplifier. In the software aspect, we use coherent averageand bandpass filtering to handle noise and use time domain windowing methodand the nonlinear average method to deal with the stimulus artifact, in order toobtain accurate distortion product otoacoustic emission data.This paper completes the design hardware modules and software program ofhearing detection.Through a lot of experiments, we prove that the detectionfunction of hearing detector Volterra kernel modeling and distortion productotoacoustic emission is effective.