Maneuvering target detection method based on Keystone transform and Radon local mapping sparse-modified Lv’s Distribution

This paper focuses on the range migration (RM) and Doppler frequency migration (DFM) problems, which are commonly faced in radar maneuvering target detection during the long-time integration. Considering the sparsity of target echo in the chirp rate (CR)-quadratic CR (QCR) domain, a novel coherent integration method is presented to realize the maneuvering target detection. The proposed method firstly conducts Keystone transform (KT) to remove the linear RM with respect to the target’s velocity and then performs Radon local mapping sparse-modified Lv’s distribution (RLMSMLVD) to eliminate the residual RM and DFM, thereby focusing the target in the sparse CR-QCR domain. Thanks to the adoption of two-dimensional (2-D) sparse Fourier transform (SFT), the proposed method can obtain a lower computational cost than the 2-D fast Fourier transform (FFT)-based implementation. Moreover, the local coordinates mapping is conducted in 2-D SFT to further increase the operation speed. Experimental results are given to demonstrate the high efficiency and effectiveness of the proposed method.