Micro-Doppler character analysis of moving objects using through-wall radar based on improved EEMD

The micro-Doppler signals of human's heartbeat breathe and arm-moving using through-wall radar is nonlinear and non-stationary, which can be analyzed by Empirical Mode Decomposition (EMD). Due to the mode mixing problem in EMD, an improved Ensemble Empirical Mode Decomposition (EEMD) is proposed in this paper, and is applied to the human micro-Doppler character analysis of the through-wall radar. The time-frequency-energy spectrum is obtained by using Hilbert-Huang Transform (HHT) to every Intrinsic Mode Functions (IMF). The analysis on simulation data and experimental results show that the improved EEMD can effectively eliminate the mode mixing problem in EMD, which means different frequency scales in human's micro-Doppler signals are decomposed in different IMF. Furthermore, this method can restrain the noise in the original signal and more detail information can be seen clearly in the time-frequency spectrum.