CAJ | 학술논문

目的：基于经验模态分解（EMD）的多尺度时域滤波特性，抑制生物雷达回波信号噪声。方法将经过预处理的雷达回波信号进行EMD分解，获取固有模态函数（IMF）分量；根据噪声与信号自相关函数的特性差异将IMF区分为有用信号占主导模态与噪声占主导模态两部分；然后用基于小波软阈值的去噪方法提取噪声占主导模态部分中的有用成分，与有用信号占主导模态部分的固有模态函数叠加，获取雷达回波信号中的生命信息。结果基于EMD的噪声抑制方法有效地去除了雷达回波信号中的高频噪声成分，提取出了有用信息。结论基于EMD的噪声抑制方法能为生命体目标的识别提供更多信息。
목적：기우경험모태분해（EMD）적다척도시역려파특성，억제생물뢰체회파신호조성。방법장경과예처리적뢰체회파신호진행EMD분해，획취고유모태함수（IMF）분량；근거조성여신호자상관함수적특성차이장IMF구분위유용신호점주도모태여조성점주도모태량부분；연후용기우소파연역치적거조방법제취조성점주도모태부분중적유용성분，여유용신호점주도모태부분적고유모태함수첩가，획취뢰체회파신호중적생명신식。결과기우EMD적조성억제방법유효지거제료뢰체회파신호중적고빈조성성분，제취출료유용신식。결론기우EMD적조성억제방법능위생명체목표적식별제공경다신식。
Objective To suppress the noise of bioradar echo signal based on multiscale time-domain filtering features of empirical mode decomposition (EMD).Methods EMD decomposition of preprocessed bioradar echo signal was conducted to obtain intrinsic mode functions (IMFs), which were divided into one mode with noise as its dominant components and the other mode with useful signal as its dominant components according to the differences of the features of noise and signal autocorrelation functions. Then the useful information of MIFs in the mode with noise as its dominant components extracted through the denoising method based on wavelet soft threshold was superimposed on MIFs in the mode with useful signal as its dominant components to obtain the life information of bioradar echo signal.Results The high-frequency noise of bioradar echo signal has been ifltered and the useful information of bioradar echo signal has been extracted with the noise suppression method based on EMD. Conclusion The noise suppression method based on EMD can provide more useful information for the identify of organism objectives.