A Novel Long-Time Coherent Integration Algorithm for Doppler-Ambiguous Radar Maneuvering Target Detection

A long-time coherent integration could effectively improve the detection ability of radar for maneuvering targets. Nevertheless, the Doppler ambiguity and frequency migration caused by the high speed and acceleration severely degrade the detection performance. In this regard, a novel coherent integration algorithm is proposed, particularly for maneuvering targets with Doppler ambiguity. Specifically, the acceleration is firstly estimated by scaled non-uniform fast Fourier transform (SNuFFT). Then, the scaled periodic discrete Fourier transform (SPDFT), which periodically extends the observable Doppler scope of discrete Fourier transform (DFT), is proposed to estimate the unambiguous Doppler frequency. Finally, the grating lobes are significantly suppressed via product operation, and coherent integration is achieved after phase compensation. To alleviate the computational burden and eliminate the brute force searching procedure, an efficient implementation based on chirp-z transform (CZT) is also derived. Analysis shows that the proposed algorithm achieves a good balance between computational complexity and anti-noise performance. Extensive simulations and real measured radar data are conducted to verify the proposed algorithm.