Research Article
An ECG Denoising Method Based on the Generative Adversarial Residual Network
Table 6
Model experiment results under different noise types and different noise intensities.
| SNR/noise type | MA | EM | BW | SNR | RMSE | SNR | RMSE | SNR | RMSE |
| 0 dB | 34.3956 | 0.0002 | 34.4056 | 0.0003 | 34.1810 | 0.0004 | 1 dB | 38.1268 | 0.0006 | 37.1836 | 0.0003 | 37.6323 | 0.0007 | 1.25 dB | 31.1282 | 0.0018 | 31.4057 | 0.0008 | 31.1545 | 0.0016 | 2 dB | 40.2612 | 0.0009 | 39.2438 | 0.0008 | 40.1673 | 0.0009 | 3 dB | 42.2612 | 0.0009 | 43.5693 | 0.0002 | 43.6502 | 0.0006 | 4 dB | 46.2873 | 0.0004 | 45.6912 | 0.0005 | 40.4056 | 0.0006 | 5 dB | 51.1128 | 0.0009 | 53.2246 | 0.0011 | 52.2459 | 0.0010 | Average | 40.5079 | 0.0007 | 40.6705 | 0.0005 | 41.3008 | 0.0009 |
|
|