A Weighted L1-Norm Method for the Fast Real-Aperture Imaging of Spatial-Expanded Targets

The sparse reconstruction-based real-aperture imaging has significant application value in the field of radar forward-looking imaging. It is independent of target motion and capable of acquiring target image with few snapshots. However, the high computational complexity of super-resolution algorithms and the low geometric contour precision for spatially expanded targets limit the practical application of real-aperture imaging. This paper proposes a fast real-aperture imaging method of spatial-expanded targets based on weighted L1-norm, aiming to improve both imaging accuracy and efficiency. For the problem of blurred target contour, rough imaging results are used to assist fine angle measurement based on weighted Ll-norm, which could enhance the angle estimation precision. In order to tackle the challenge of high computational complexity, the array manifold dictionary is decomposed into a two-dimensional Kronecker product form. With the combination of the Woodbury identity and LADMM algorithm, the algorithm reduces matrix operation dimension, leading to the decreasing of the iteration numbers and the improvement of computational efficiency. Comparative experiments demonstrate that the proposed algorithm enhances both computational efficiency and spatial target imaging accuracy for real-aperture imaging.