Instantaneous Velocity Estimation Based on Envelope Cross Correlation and Joint Frequency-Phase Estimation for Wideband LFM Radar

To increase the probability of detection, long-term integration (LTI) is often used in the signal processing of the wideband linear-frequency modulation (LFM) radar. LTI produces the Doppler-frequency migration (DFM) and range migration (RM), thereby distorting parameter estimation of maneuvering targets. Several methods are proposed to tackle the distortion of velocity estimation. However, the performance of these methods is negatively affected by signal-to-noise ratio (SNR) degradation. To tackle the problem, an envelope cross correlation (ECC)-based method is proposed, where the parameter search is evicted. First, envelopes of adjacent stretched echoes are extracted for phase removal. Then the conjugate multiplication is performed on the obtained envelopes to decrease the bandwidth. With the frequency peak by the ECC as the input, a joint frequency-phase estimation is developed for more accurate velocity estimation. The simulation shows that the SNR threshold and root-mean-square error (RMSE) of the velocity estimation are improved extensively. The accuracy of estimated velocity on real radar data implies that the proposed ECC is superior to the comparison methods.