Estimation of target's acceleration based on empirical wavelet transform

Abstract

Target's accelerations lead to spectrum shift and broadening of target's echo signal,resulting in the inaccuracy estimation of target's Doppler frequency with traditional pulse radar velocity measurement method. To overcome the effect of acceleration on pulse radar velocity measurement,an empirical wavelet transform( EWT) based radial acceleration estimation method was proposed. The instantaneous frequency of the echo signal can be extracted through EWT and energy-oriented principal frequency components extraction method. The high order coefficients of the phase were obtained through robust least square fitting on the instantaneous frequency,which correspond to the radial velocity and radial acceleration respectively. After compensating the echo signal with estimated accelerations,the Doppler frequency of echo signal can be accurately estimated. Simulations show that the EWT method is a fast algorithm with high estimation accuracy,and the estimation error is close to Cramer-Rao lower bound. Applying EWT method on measured pulse radar data of high speed vehicle,the estimated acceleration error is smaller than 0. 4 m / s2. EWT method is applicable in real time pulse radar acceleration estimation.