HHT在生物雷达回波信号噪声抑制中的应用  

Application of Hilbert-Huang Transformation for Noise Suppression in Ultra Wideband Bio-radar Echo Signal

作　　者：张洋[1] 王健琪[1] 荆西京[1] 吕昊[1] 

机构地区：[1]第四军医大学生物医学工程学院,西安710032

出　　处：《医疗卫生装备》2012年第5期14-17,共4页Chinese Medical Equipment Journal

基　　金：国家自然科学专项基金项目(60927003)

摘　　要：目的:利用Hilbert-Huang Transform(HHT)的频率区分特性,对生物雷达回波信号进行处理,去除高频噪声成分,提高生物雷达对生命信息的识别能力。方法:首先对经过预处理的雷达回波信号进行经验模态分解获取其固有模态函数,并根据能量分析方法将固有模态函数区分为有用信号占主导模态与噪声占主导模态2个部分,然后用小波软阈值的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,实现时目标的识别。结果:HHT有效地去除了生物雷达回波信号中的高频噪声成分,提取出了人体呼吸信号。结论:与传统傅里叶去噪方法相比,HHT分辨率更高,能保留信号中的细节特征,为生命体目标的识别提供了更多的信息。Objective To diminish the high frequency noise components in bio -radar echo signal using the frequency differentiating property of Hilbert-Huang transform（HHT） so as to improve the ability of ultra wideband（UWB） bio-radar＇s identification for life information.Methods The intrinsic mode function（IMF） of the pre-treated bio-radar echo signal was gotten by the way of empirical mode decomposition（EMD）.the IMF was divided into two parts,of which the dominant component was noise and useful signal separately,and then the useful information of those IMFs in which noise took the dominant position was extracted,which was combined with the IMFs in which useful signal was dominant to acquire the life information in UWB bio-radar echo signal in order to achieve the target recognition.Results HHT could diminish the noise components of high frequency in UWB bio-radar echo signal and draw the respiratory signal of human subjects effectively. Conclusion HHT has high resolution in comparison with conventional way of noise suppression by Fourier transform,and it can reserve the details in signal and provide more information in identification of human subjects.

关 键 词：生物雷达 希尔伯特黄变换 经验模态分解 固有模态函数 

分 类 号：TN957.51[电子电信—信号与信息处理]