A Fast Coherent Integration Algorithm for Maneuvering Target Detection

This paper investigates the coherent integration problem in the maneuvering target detection. We propose a new coherent integration algorithm based on the short-time Fourier transform (STFT) and the cubic phase function (CPF), i.e., STFT-CPF. First, the STFT is employed to construct a special compensation function for the linear range migration. Then, the CPF is applied to estimate the target's acceleration and jerk, thereby compensating the linear Doppler frequency migration and the quadratic Doppler frequency migration. After that, the coherent integration can be achieved by the FT. The proposed algorithm involves only 1-D search. It reaches a good balance between the computational cost and detection performance in a certain range of acceleration and jerk. Simulations and real data demonstrate the effectiveness of the proposed algorithm.