A Wearable ECG Monitoring Device

—A wearable ECG monitoring device with a customized SoC is proposed. The ECG signal sensed with passive electrodes is amplified, digitized and transformed into wavelet coefficients by this dedicated SoC. Following with it, a low power microcontroller (MCU) and a radio frequency module (RF) in the device resolve and transfer the coefficients to mobile platform, which can use machine learning algorithm to improve the performance of signal denoising and data compression by exploiting sensed data, resulting in reduced power consume d by the wearable device. In practical measurements, the device has resolved ECG data from MIT-BIH arrhythmia database and the tester’s real ECG signal. After the procession of the ECG data with different noise models, device can realize high improvements of signal noise ratio (SNR) and mean square error (MSE) are 23.8 dB and 88.9%, while the representative compression ratio (CR) is 4.8:1 with 1.56% percentage root mean square difference (PRD) in the condition of resolving real tester’s ECG signal. The customized SoC is fabricated in TSMC 0.18μm technology, and consumes 45μw for different applications.