A multi-frame hybrid integration method combined with differential evolution for maneuvering target detection with GNSS-based passive radar

Current target detection methods tailored for global navigation satellite system (GNSS)-based passive radar are primarily put forward for non-maneuvering targets and utilize a second-order polynomial model to correct motion migrations of the target return during the long integration time. However, the detection performances of these current approaches diminish when applied to maneuvering targets, as they fail to address the high-order motion migrations. This study concentrates on maneuvering target detection issue. First, we evaluate the applicability of a third-order polynomial model (TPM) for approximating the bistatic range history of the maneuvering target. Then, we analyze which high-order motion migration corrections are necessary in terms of the TPM and the range and Doppler resolutions of GNSS-based passive radar. Based on this analysis, we propose a multi-frame hybrid integration method to detect maneuvering target. The realization of the proposed method is formulated as a constrained optimization problem in the bistatic parameter searching space, for which differential evolution is employed to improve processing efficiency. Both simulated and real experimental results confirm the effectiveness of the proposed method. Monte Carlo trials demonstrate that the proposed method needs a signal-to-noise ratio threshold at least 2 dB lower than the existing methods to achieve the same detection probability of 0.9.