基于EEMD和HOC的超宽带雷达生命探测算法研究  

A Study on UWB Vital Signal Detection Method Based on EEMD and HOC

作　　者：蒋留兵[1,2] 韦洪浪 管四海 车俐[1,2] 

机构地区：[1]桂林电子科技大学信息与通信学院,广西桂林541004 [2]桂林电子科技大学广西无线宽带通信与信号处理重点实验室,广西桂林541004

出　　处：《现代雷达》2015年第5期25-30,共6页Modern Radar

基　　金：国家自然科学基金资助项目(61162007);广西自然科学基金资助项目20134GXNSFAA019323;广西科学研究与技术开发计划项目(桂科攻14122006-6)

摘　　要：针对超宽带生命探测雷达回波信号属于非线性、非平稳等特点,提出了一种基于EEMD和HOC的超宽带雷达生命探测算法。通过对雷达回波信号进行EEMD分解,将信号自适应分解为若干个本征模态函数(IMF),然后计算各个IMF分量在呼吸和心跳频带内的能量百分比重构呼吸和心跳信号,最后对重构的呼吸和心跳信号的四阶累积量进行FFT变换,获得呼吸和心跳的频率。实验结果表明,文中提出的算法比EEMD重构后直接进行FFT变换具有更高的信噪比和频率估计精度,可有效应用于生命探测雷达人体信号检测中,具有广阔的研究价值和应用前景。For the characteristics of the echo signal's nonlinearity and nonstationarity in UWB life-detection radar,the life-detection algorithm of UWB radar based on EEMD and HOC is proposed. By decomposing radar echo signal with EEMD and decomposing ra- dar echo signal into several intrinsic mode functions,then calculating energy percentage of every IMF component within the respira- tory and heartbeat band to reconstruct respiratory and heartbeat signal, finally, the respiratory and heartbeat rate can be obtained through FFT on the fourth order cumulant of reconstructed respiratory and heartbeat signal. The experimental results show that the algorithm proposed in the paper has higher SNR and higher frequency estimation precision than direct FFT after EEMD reconstruc- tion. The algorithm can be availably applied in human signals detecting of life-detection radar and it has great research value and application prospect.

关 键 词：生命探测雷达 整体平均经验模态分解 快速傅里叶变换 高阶累积量 

分 类 号：TN957.51[电子电信—信号与信息处理]