A new modified wavelet-based ECG denoising

Figures & data

Figure 1. Schematic diagram of wavelet denoising algorithm.

Figure 2. AFR of different frequency aliasing types.

Figure 3. AFR of the ideal filter.

Table 1. Fibr wavelet coefficients.

Length	Coefficients: h(n)
8	0.001590, −0.056193, 0.056736, 0.493436, 0.493436, 0.056736, −0.056193, 0.001590

Figure 4. The result of wavelet design (fibr wavelet). From the left to right: (a) The scaling function diagram (b) The wavelet function diagram.

Table 2. Parameters of the acquisition equipment.

Parameters	Values
Defibrillation protection voltage	5000 V
DC polarization voltage	±300 mV, ±5%
Frequency response	1–75 HZ, +0.5 dB, −3.0 dB
Dynamic input range	20 mV
Calibration voltage and accuracy	1 mV, ±5％
CMRR（Common mode rejection ratio）	>60 dB
System noise	≤15 $\mu \mathrm{V}$
Nonlinear error	±10％
Baseline drift	≤1 mm (5－40 °C )
A/D resolution	13 byte, 2.44/bit
Digital sampling rate	500 Hz
Signal length/sampling point	5000

Table 3. Parameters using for generating F waves.

Parameters	Values
Frequency, $f_0$	6 Hz
Frequency variation, $\Delta f$	0.2 Hz
Frequency variation frequency, $f_f$	0.1 Hz
Harmonics, M	3
Amplitude, am [ II V1 V5 ]	[150 75 45]$\mu v$
Amplitude variation, Δam [ II V1 V5 ]	[50 25 15]$\mu v$
Amplitude variation frequency, $f_a$	0.08 Hz

Figure 5. The average denoising effects on normal ECG.

Figure 7. The denoising result under the noise with expected amplitude of 0.065 V.

Figure 8. The average denoising effects at different noise levels.

Figure 9. The average denoising effects on simulated AF signals.

Figure 10. The denoising result on the simulated AF signal.