• Proceedings of the Korean Society for Emotion and Sensibility Conference
• /
• 2003.05a
• /
• pp.128-133
• /
• 2003

본 연구에서는 심전도 신호의 전처리 과정에서 원신호의 왜곡을 최소화하여 기저선을 제거 할 수 있는 웨이브렛 모함수를 결정하기 위하여, European S-T T database의 심전도 신호에 다양한 웨이브렛 모함수를 적용하여 기저선을 제거하였으며 제거효율을 평가하기 위하여 SNR과 RSE를 계산하였다. 실험결과 가장 우수했던 웨이브렛 모함수는 db8(diff: 27.12), coif5(diff: 25.32), sym7(diff: 25.13)이었으며, diff(meanSNR-meanRSE)의 값이 23미만으로는 심전도의 진단 파라미터까지 왜곡시키므로 사용할 수 없다는 것을 알 수 있었다.

• Progress in Superconductivity
• /
• v.5 no.1
• /
• pp.26-30
• /
• 2003

Magneto-cardiogram(MCG) signals may be highly distorted by the environmental noise, such as power-line interference, broadband white noise, surrounding magnetic noise, and baseline wondering. Several kinds of digital filters and noise cancellation methods have been designed and realized by many researchers, but these methods gave some problems that the original signal may be distorted by digital filter due to the wideband characteristics of background noise. To eliminate noise effectively without distortion of MCG signals, we performed multi-level frequency decomposition using wavelet packets and local adaptive noise cancellation in each local frequency range. In addition to the proposed wavelet filter to eliminate these various non-stationary noise elements, the local adaptive filter using the least mean square(LMS) algorithm and the soft threshold do-noising method are introduced in this paper. The signal to noise ratio(SNR) and the reconstruction square error(RSE) are calculated to evaluate the performance of the proposed method and compared with the results of the conventional wavelet filter and adaptive filter. The experimental results show that the proposed local adaptive filtering method is better than the conventional methods.