A Device Based On Adaptive Noise Cancellation For ECG Characteristics Detection And Early Warning

Recently,coronary heart disease becomes one of the major components of cardiovascular disease with the fast life rhythm.And the threat to human health is increasing gradually.The condition of abnormal heart can be found by timely detection of Electrocardiogram(ECG),which has certain value for the diagnosis of coronary heart disease and early warning of chronic diseases.At present,wearable detecting device has become the developing trend of ECG measurement.However,the development of wearable equipment is also faced with the problems of easy drop of lead electrode connection,motion interference of acquisition signal and low accuracy of feature detection.This paper addresses the existing problems from the following aspects:(1)To improve the stability of monitoring ECG signals and reducing the power consumption of the system,a wearable wireless device for detecting ECG is designed by using the tight vest.The lead electrode module embedded in the tight vest is connected with the device stably by USB interface,and the switch between charging and power supply is realized by electronic switch module.ECG signals and chest impedance of human body are detected by bioelectric capturing chip.Acceleration sensor is utilized to collect three-axis acceleration,and beidou positioning module is used to output satellite positioning information.In the software design of the device,the driving mode of state machine is utilized to control the module to run orderly,so that the device can collect signs stably.(2)Aiming at the problems of the baseline drift and motion artifact interference existing during detecting ECG signal in human motion,the adaptive filter and FIR low-pass filter are designed for ECG signal preprocessing.Chest impedance and triaxle acceleration are outputted from sensors as reference signals.Firstly,the adaptive filtering algorithms with different convergence steps are simulated.By comparing the Signal to Noise Ratio(SNR)of the filters,it is shown that the SNR of the normalized least mean square algorithm is higher,and the SNR increases with the decrease of the step size.Then an adaptive filter with a step size of 0.05 is designed to eliminate the motion interference of ECG signals.In addition,the differential square threshold method is utilized to locate R wave,and the slope mutation search algorithm is used to extract QRS wave group.The heart rate can be measured by the device in real time.A prototype test for ECG detection device is designed on the experimental platform.It is shown that the prototype size of ECG detection device is small and the power consumption is low.Its duration is more than 10 hours.The frame error rate of wireless ECG data transmission in 10 meters range of motion is less than 0.1%.By the actual acquisition of ECG signals from several subjects,the results indicate that the baseline of ECG signals which is outputed from the ECG detection device is stable,the detection accuracy of QRS wave group is more than 99%,and the relative error of measuring ambulatory heart rate in is less than 4%.The ECG detection device has the advantages of easy to wear,high accuracy and low cost.The measured parameters meet the requirements of monitoring ECG signals.