High Precision Doppler Estimation Method for LFM Signals Based on Chirplet Transform

The traditional Doppler estimation methods of LFM signals can not give consideration to both estimation accuracy and calculation speed. In this paper, aiming at the problem of high-precision Doppler estimation of LFM signals in underwater acoustic communication, a high-precision Doppler estimation method using LFM signals as Chirplet transform kernel function is proposed based on the characteristics of frequency shear distribution of LFM signals. Methods First, Chirplet transform the LFM signals before and after Doppler, and convert the time domain signals to the frequency shear domain. On this basis, linearly fit the coordinates of the energy peak area of the frequency shear distribution of the LFM signals before and after Doppler, respectively, to obtain the peak ridge straight line and parabola. The intersection of the two lines is the frequency and shear value of the LFM signals after Doppler frequency shift. The simulation results show that the algorithm can realize fast frequency estimation of single frequency and FM signals, and the estimation accuracy can reach or even exceed the Cramer's lower bound of unbiased estimation at the SNR of about 10dB.