基于PCA与EMD的超宽带雷达生命信号检测算法

摘要

本文分析了脉冲超宽带(UWB)生命信号模型,提出了基于主元分析(PCA)和经验模态分解(EMD)的非接触生命信号检测方法.根据UWB信号杂波与生命目标回波特点,结合PCA去除杂波.提取适当的主元特征向量序列曲线上峰值所对应的时延,估计目标距离信息.采用EMD分解目标回波序列为有限个固有模态函数(IMF)分量,在时域上重构平滑生命特征曲线,且其在高信噪比下可实现心跳与呼吸信号的分离.实验研究表明该方法简单有效,能同时提供生命信号的频域和时域波形位置信息,且重构得到的生命信号较符合实际信号时变、非平稳特件.%A novel non-contact vital signal detection method based on Principal Component Analysis (PCA) and Empirical Mode Decomposition (EMD) is presented for the impulse ultra wideband (UWB) vital signal module. The clutter is reduced by PCA on the basis of the characteristics of the UWB clutter signal and the objective vital signal. The distance information of the vital object is estimated by the delay time, which is extracted from the appropriate principal component eigenvector peak value. The sequence of the target echo is decomposed into several IMFs by EMD, and the flatness curve of the vital sign is reconstructed in time domain. The respiration and heartbeat signals can be separated with the high signal-to-noise ratio condition. The experiment demonstrated that this signal processing algorithm is simple but feasible. The frequency spectrum characteristics,the waveform in time domain of the vital signal and the distance between the object and the antenna can be obtained simultaneously. The reconstructed vital signal is complied with the characteristics of the time variety and non-stationary for the actual signal comparatively.