Adaptive Dynamic Regularization Super-Resolution Imaging Method of Forward-looking Scanning Radar Based on Data-Depended

Non-convex
$L_{p}$
regularization method has been applied in airborne forward-looking scanning radar imaging to enhance the azimuth resolution because of its superior performance in sparse target reconstruction. However, the algorithm performance is sensitive to the selection of the parameter
$p$
, the results are troublesome because it should be manually chosen. In this paper, to select an appropriate parameter
$p$
based on the data property, an adaptive dynamic regularization super-resolution method based on data-depended is proposed for airborne forward-looking scanning radar imaging. First, the iteratively reweighted algorithm is employed to solve the
$L_{p}$
norm minimization. Second, to adaptively select the norm
$p$
depending on the echo data, a data-depended adaptive dynamic regularization approach is proposed by updating the parameter
$p$
in each iteration. Simulation results demonstrate the adaptive performance of the proposed method, which can determine the regularization parameter depending on the echo data and improve the adaptivity of the traditional
$L_{p}$
regularization method.