ECG Signal Real-Time Detection Of The Remote Monitoring Terminal

The heart disease has a high death rate and often takes by a sudden chance, butthe medical resources are scarce and the distribution is extremely uneven. In order tosatisfy people’s need for medical care, improve the life quality and reduce the medicalcosts, the preventing, analyzing and diagnosing of the heart disease take an importantsocial significance. Aming to detecte the physiological condition easily and timely, itis useful to develop a portable, practical and intelligent remote electrocardiogram（abbreviated as ECG） monitoring terminals to realize self-discovery, self-diagnosisand self-care for cardiac patients.But most of the ECG terminal have not the automatic diagnosis function and cannot meet the needs for real-time processing and automatic analysis. Especially it can’tidentify the ECG parameters which are life-threatening for people. Based on thisdemandings, the wavelet transform is taken to explore the detecting algorithms of thecharacteristic parameters with the ECG signals which were carried out in the DSPplatform.Firstly, we design an ECG signal acquisition and conditioning circuits based onphysiological ECG and its characteristics, take the TMS320VC5509DSP as the mainprocessor, set up the terminal hardware of the remote ECG monitoring system,complete the ECG signal acquisition, amplification conditioning, analysis processingand feature parameters detection.Secondly, the characteristics of the ECG signasl and noises are analysisedaccording to the nature of wavelet transform which is providing the sufficienttheoretical fact for denoising the ECG signals. Summarizing the steps of the ECGdenoising and the selection of threshold. A more reasonable for the ECG signal de-noising algorithm is promoted on the basis of quadratic B-splint wavelet. Carrying outthe Matlab simulations and proving the effectiveness of the Wavelet transform method for the ECG signal denoising.Finally, we introduce the outstanding advantages of quadratic B-splint wavelet inthe parameters identification after analysis the analysis of singularity which takingwavelet transform for ECG signal detecting. Taking the8-layer ECG decomposition,identification and locating the R wave to find the starting and ending position,determining the location of P wave and T wave. We carry out the simulation of thisproposed algorithm based on the MIT-BIH database. At the end, we verify theaccuracy and effectiveness of the algorithms which are carried on the DSP platform ofTMS320VC5509to test parts of five ECG signals. The test results of profiler andhardware system effectively show that this algorithm can satisfy the requirements ofreal-time for the remote ECG monitoring system.