论文部分内容阅读
本文首先描述了一套在人的封闭耳道内记录耳声发射的装置的设计和制作。其中包括声学系统和电学系统以及存一台Z80单板微机上进行数字信号处理的计算机程序。声探头是真实记录的关键部件。诱发性耳声发射的记录主要依靠时域信号平均技术。在声信号测试过程中,存在着两类噪声。分析指出,在程序中采用限幅筛选的办法来去除含第二类噪声(非记录系统本底噪声)的样本,减小了随机过程的非平稳用素,可以有效提高信号处理质量,频域处理以FFT为基础。介绍了提高FFT效率的程序设计要点。用周期图迭加法可以获得自发性耳声发射的频谱估计。对于系统的定标,也作了说明。
This article first describes the design and fabrication of a set of devices for recording otoacoustic emissions in a closed human ear canal. These include acoustical and electrical systems as well as computer programs for storing digital signals on a single Z80 single-board microcomputer. Acoustic probe is a real record of the key components. Evoked otoacoustic emissions mainly rely on time-domain signal averaging techniques. There are two types of noise during acoustic signal testing. Analysis pointed out that the use of clipping in the program to remove the second type of noise (non-recording system noise floor) samples, reducing the non-stationary random factors, can effectively improve the quality of signal processing, frequency domain Processing is based on FFT. Introduced the key points to improve the efficiency of FFT programming. Spectral estimation of spontaneous otoacoustic emissions can be obtained using the periodic map superposition method. For the system calibration, also made a description.