Font Size: a A A

The Research On Optimal De-noising Of ECG Signals Based On Wavelet Transform

Posted on:2015-07-05Degree:MasterType:Thesis
Country:ChinaCandidate:H S HeFull Text:PDF
GTID:2298330431456056Subject:Control Engineering
Abstract/Summary:PDF Full Text Request
The electrocardiogram (ECG) signal was a kind of biomedical signal which had been applied in evaluating and detecting the state of heart function earlier. Based on the Noninvasiveness and simplicity of diagnosis technology, ECG signal was used in clinical diagnosis widely, and it could provide cardiac electrophysiology information in the diagnosis of cardiovascular diseases. Because of ECG signal’s faintness and randomness, a little noise will cause the distortion of ECG signal, which leaded to the difficulty of getting effective signal, so the research on de-noising of ECG signal had been a widespread-attention subject.In order to eliminate the interference in ECG signal, the optimal De-noising method of ECG signals based on wavelet transform has been proposed in the paper. What has been done in this paper is shown as follows:1. This paper explained the work principle of ECG signal and compared varieties of de-noising methods. Based on human body physiology characteristic, it analyzed the waveform characteristic and noise source of ECG signal, and then compared commonly-used noise reduction methods, finally put forward a new de-noising method. Compared with common noise reduction method, the new one would play well in noise reduction while making a good retention of ECG signal waveform characteristics.2. Based on the theory of wavelet transform, this paper made research on the wavelet de-noising method. It established a basic mathematical model for noise signal, and compared different wavelet de-noising methods by analyzing different characteristics of signal and noise, then concluded that:the traditional wavelet de-noising method of ECG signal was to decompose the noise signal into any details, then set the threshold to distinguish signal and wavelet coefficients of noise based on the difference between signal and noise which appeared after wavelet transform, finally made the de-noising come true. It made simulation by using the traditional wavelet de-noising method of ECG signal, and expounded the deficiency.3. This paper proposed an optimal de-noising method aiming at the shortcomings of the traditional ones. It analyzed the mathematical characteristics of commonly-used wavelet and selection rules of wavelet basis function, expounded the selection and the applying way of threshold, decomposition layers, the de-noising evaluation, then designed the selection method of wavelet basis function, optimal threshold and heavy tone value.4. According to the evaluation of de-noising, this paper chose the optimal parameters to eliminate noise, and then judge the de-noising effect of ECG signal. Based on Electrical signals from the MIT-BIH international standard database, ones collected from human body and ones simulated by computer program, it used the optimal de-noising method to simulate and applied in metrological verification. Results showed that:for ECG signals collected from the same environment, the optimal de-noising method was superior to traditional one, also showed the unique advantages in the removal of a variety of mixed noise.
Keywords/Search Tags:Electrocardiogram Signal, Wavelet Transform, Optimal De-noising, Signal-to-noise Ratio, Minimum Mean Square Error
PDF Full Text Request
Related items