EEMD在超宽带雷达生命信号提取中的应用

Applications of EEMD in vital signal detection for UWB radar

超宽带生命探测雷达回波信号具有非线性、非平稳特性,由于心跳信号能量较弱,且受到呼吸谐波的干扰,传统的数字滤波方法无法有效地提取心跳信号。鉴于此采用一种从时域上提取生命信号的新方法。应用聚类经验模式分解(EEMD)将生命信号分解成有限个固有模态函数(IMF),再依据模式判别准则从时域上重构呼吸和心跳信号。实验结果表明,相比于经验模式分解(EMD),EEMD能有效提取呼吸信号和心跳信号。

For the vital signal detection using the UWB through wall radar, echo signals are non linear and non stationary. It is difficult to separate the heart and respiration using the digital filtering method, because heart signal is very weak comparing with respiration and heartbeat frequency is interfered with respiration harmonic frequencies. In view of this,a new method is proposed to extract life signal in the time domain. The vital signal decomposes into a finite number of intrinsic mode function （IMF） using ensemble empirical mode decomposition （EEMD）, then we can reconstruct the respiration signal and heart signal in the time domain according to modes extract criterion. EEMD can extract effectively the respiratory signal and heartbeat signal comparing with empirical mode decomposition （EMD）, because EEMD can eliminate the mode mixing problem existed in EMD.