Research Of A Method Of Non-contact ELectrocardiograph Measurement

Continuous electrocardiograph(ECG) monitoring has an important meaning to heartdiseases prevention, non-contact measurement is an effective method to monitor humanbody ECG signal which does not affect people normal activities,coupling capacitance isone method for designing non-contact ECG measurement systerm. Due to the measuredelectrode and the human body not fixed together, a small movement will make thedistance between the two plates of the coupling capacitance or the measured position ofhuman body skin change which leads to the ECG signal collected changes and ECGsignal distorting and weakened after signal processing and filtering. Therefore, a filteringmethod is needed to match the non-contact measurements of ECG signal.This paper is based on the theory of principal component analysis (PCA), and fromthe angle of energy, an algorithm based on the fast adaptive PCA is proposed to removethe noise in ECG signal, which provides a new filtering method for the long-termmonitoring of human ECG signal. The main research contents and methods are asfollows:A non-contact measurement system is designed using the principle of couplingcapacitance, and the ECG signal collected is first filtered (mainly 50 Hz frequencyinterference) and amplified in the hardware circuit, then converted from the analog signalto a digital signal by A/D converter.After Summarizing and analyzing the basic theory of PCA method, we get thedenoising algorithm with PCA: Firstly, an embedded matrix that contains all informationof collection signal is obtained using dynamic embedded (Dynamical embedding, DE)technology. Then principal components of signals are extracted and the required numberof principal components is determined based on the Bayesian information criterion.Finally, the selected principal components are combined linearly. Different signals takenas noise, the PCA denoising algorithm is demonstrated with simulation experiment, andthe amplitude characteristic is made a further analysis, then the accuracy of the filteringmethod is demonstrated. The human body ECG signal is collected using the non-contact measurement, andprocessed with the PCA method. The experiment proves the filtering method usingnon-contact measurement to acquire and process human body ECG signal can purifyjamming signal disposable on the premise of keeping the main characters of ECG signal.The effect stability of filtering also can be guaranteed even if the input signals changes,and it can combine with the way of non-contact measurement well.