Parameter Estimation of Accelerating Moving Targets at Low SNR for MEO SAR

This paper proposes a new method for parameter estimation in the azimuth frequency domain, which can accommodate lower signal-to-noise ratios (SNR). By compensating for azimuth compression, range walk, and range curvature simultaneously, the SNR of signal is improved, so as to complete the range cell migration correction. Previously, we first compensated for the ambiguity velocity in the azimuth time domain to avoid aliasing and ambiguity in the frequency domain. Finally, we use cubic phase function and Radon transform to estimate the polynomial phase signal. Numerical results validate the proposed method.