Fast parameter estimation method for maneuvering target by using non-uniformly resampling reducing order technique

Abstract

Abstract In radar systems, prolonging the integration time can improve the performance of motion parameter estimation for weak targets. However, when the target is maneuvering, the long integration time may lead to rang migration (RM) and Doppler frequency migration (DFM) problems. In traditional methods, multi-dimensional search of motion parameters is often used, but the computational cost is high. In this paper, a fast long-time coherent integration method for maneuvering target is proposed. By using a newly defined cross correlation function based on a non-uniform resampling technique, we eliminate the high-order RM and DFM simultaneously in the range frequency-slow time domain. Then, the target energy is integrated efficiently by performing inverse fast Fourier transform (IFFT) and fast Fourier transform (FFT) along the range and azimuth directions. The proposed method has lower computational complexity than traditional methods, because it uses FFT and IFFT instead of direct multi-dimensional search. In addition, the proposed method is applicable for multi-targets and fast moving target with ambiguities. The simulations and theoretical analysis are provided to demonstrate the effectiveness of the proposed method.