FFT-based auto-correlation estimation (FACE) for extended radar pulse integration subject to large doppler change

In such applications as detection and tracking of stealthy targets, it is often necessary to integrate radar pulses over an extended period so as to accumulate the strength of target returns while averaging out noise. The autocorrelation estimate (ACE) method was shown to be less sensitive to changes in Doppler frequency than the fast Fourier transform (FFT)-based coherent integration method [1]. However, when the Doppler frequency is subject to large changes, the ACE method still suffers from performance degradation. In this paper, we set forth a new method for extended radar pulse integration that combines the FFT and ACE methods, which we call FACE. Among various existing techniques analyzed in this paper, the connections of the new method to the Wigner-Ville transform and the cyclic autocorrelation function are emphasized. For the case of a quadratic phase function or a linear frequency modulation (LFM) where the change rate in frequency is constant, simulations are presented to demonstrate the functionality of the new method for estimation of chirp signal parameters as compared to the ACE method.