Research on the detection and de-noising algorithm of wearable ECG signal based on capacitive coupling electrode
In most traditional electrocardiogram (ECG) detection procedures, wet electrodes may cause problems of inconvenience and glue dehydrates over time. The paper designs a kind of wearable capacitive coupling electrode based on the principle of coupling capacity. Due to this kind of wearable capacitive coupling electrode, an improved wavelet threshold de-noising algorithm is proposed. The algorithm uses the improved threshold function to deal with wavelet coefficients after decomposition and reconstruct ECG signal combining the characteristics of wavelet coefficients of ECG signal and noise. The MIT-BIH database was used to validate the algorithm and it indicates that the algorithm can effectively eliminate the noise. The SNR increased by 10.72% and the RMSE reduced by 27.29% compared to the other methods, such as, smoothing filtering, morphological filtering and empirical mode decomposition. The results of the experiment show that the system can accurately detect the main characteristics of the ECG signal.