Efficient parameter estimation method for maneuvering targets in discrete randomly-modulated radar

This paper proposes an efficient parameter estimation method for maneuvering targets based on the discrete randomly-modulated (DRM) radar. To correct the range cell migration, the range frequency spectrum of the received echo is evenly divided into two parts, and a synthetic aligned signal is constructed by multiplying one with the conjugate of the other. Then, the aligned azimuth signal is modeled as a nonuniform linear frequency modulated (Nu-LFM) signal and a two-dimensional nonuniform symmetric correlation distribution (2D-NUSCD) is applied to uniquely present the Nu-LFM signal in the 2D frequency domain. The 2D-NUSCD is simple and only requires complex multiplication and nonuniform fast Fourier transform (NUFFT). Owing to the application of NUFFT, the searching processing is unnecessary for the nonuniformly spaced signal and the computational cost is reduced greatly. Furthermore, theoretical analysis illustrates that the proposed method is appropriate for multitargets and is an asymptotically efficient approach with good anti-noise performance. Compared to other representative methods, the proposed method significantly improves the computational efficiency on the premise of retaining the estimation performance. It is suitable for both DRM and determinate radar systems, providing broader application scope. Simulation results verify the effectiveness of the proposed method.