The Research On Technology Of Processing Ecg Signal Based On Wavelet Transform

The heart attack is a common disease that endangers people's health. It is very necessary to do research on the technology of curing and diagnosing the heart disease. The Electrocardiogram Signal, as an important basis of diagnosing heart disease, is always a research emphasis in the field of heart disease diagnosis. After analyzing methods of biomedical signals processing at home and abroad, we find that the existing methods which process the ECG Signals have some drawbacks. This paper chooses the subject—the research on technology of processing ECG signals based on wavelet transform.Wavelet transform has well time-frequency localization characteristic which make its theory and method can be widely used in some spheres of signal and image processing. Int his paper, We study the electrocardiogram signal using the wavelet transform theory.Firstly, it introduces the Wavelet transformation elementary theory and denoising principle based on the traditional hard tand soft threshold value.In order to overcome the Pseudo-Gibbs phenomenon which is produced in the process of using the threshold value to deal with ECG Noise.This article combines method of the wavelet transformation translation invariance with threshold value denoising algorithm to solve the question. On this basis, we improve the ECG signal de-noising method. First , an adaptive notch filter is used to filter out frequency interference. Then, the threshold value method and the decomposition and reconstruction method combined with shift-invariant method are used to filter out interference and baseline EMG drift taking MIT / BIH ECG arrhythmia database as its simulation and verification basis.Secondly, we study how to detect the QRS wave and give the theory related to the discontinuity point of signal. Then the relationship between signal singularity and signal zero-crossing point of the wavelet transformed modulus maximum pairs is discussed. Similarly we can get the experimental results from the MIT/BIH arrhythmia database. This method can be used to separate the QRS wave from noise and extract the R wave crest points and the QRS wave initial points.Finally, we discuss methods to compress the electrocardiogram signal. The auto-adapted wavelet transformation code compression method is analyzed to handle the electrocardiogram signal. The result indicates that using the algorithm can acquire the good compression ratio and small degree of distortion of the restructured signal.