應用於心電訊號偵測系統之小波轉換數位訊號處理電路

Abstract

由於許多先進國家包括台灣皆進入老年化社會，老人生活的照護將是一個很大問題。隨著半導體製程的蓬勃發展，使得低成本、長時間使用、小型且可靠的無線生醫系統可以實現，藉由身上的感測裝置，人們的健康可以隨時受到照護，且行動不也再受到限制，不但擴大了慢性病及老人的生活圈，也減輕其家人照料的負擔。 我們使用台積電1P6M 0.18 um 標準CMOS製程來實現我們所設計之無線心電訊號感測系統，其內部電路包含類比前端放大器、類比數位轉換器、數位訊號處理電路及無線收發器。數位訊號處理電路由我負責，其它則由實驗室其它學長及同學們負責設計。 一般量測到的心電訊號常會受到環境中的許多干擾，像是肌電訊號、受測者移動或電極貼片形變所造成的基線漂移，以及電源線的60赫茲雜訊。為了濾除掉這些眾多的干擾，我們應用了小波轉換的方法來達成，且藉由其多重解析度分析訊號的特性，能夠更加準確地偵測R波。此設計中也採用了非同步數位信號接收及傳送模組當作它對外的界面，以方便和電腦、個人行動助理和藍芽等裝置連結。此外，我們亦完成LabVIEW程式作為和晶片溝通的人機介面，其除了可即時顯示心電波形外，還能計算出心跳速率及指示心電訊號狀態。

Since many developed countries include Taiwan become aging society, the life nursing of the old will be a big problem. Depending on the growing development of semiconductor process, it makes this possible to achieve low-cost, small and reliable wireless medical systems. By placing the sensors on human body, people’s health can be cared at any time and any where. It can not only expand the living community of the old and chronic-illness patients, but also reduce the burden of their family. We achieved the wireless ECG monitoring system on chip (SoC) by TSMC 1P6M 0.18 um standard CMOS process. The chip includes an analog front end (AFE), a 10 bit successive approximation register analog-to-digital converter (SAR ADC), a digital signal processing (DSP) circuit, an on-off keying transmitter (OOK TX), and an on-off keying receiver (OOK RX). The DSP circuit is designed by me, and the other circuits are designed by the other members in our laboratory. The ECG in body surface has some strong random and background noises, which mainly include electromyography (EMG), baseline-drift caused by motion artifact and electrode motion, and power-line interference. In order to remove these noises, we apply the discrete wavelet transform based method to denoise the ECG signals. And by its multi-resolution feature, we can also increase the accuracy of R-wave detection. The design also includes UART as the interface to communicate with other device, such as computer, PDA, Bluetooth, and so on. Besides, we accomplish a LabVIEW program to be the human-machine interface connecting with our chip. Except for the real-time waveform display, it also indicates the heart rate and status of ECG signal.