Performance improvement in multi-ship imaging for ScanSAR based on sparse representation

Abstract

There is always a compromise between unambiguous wide-swath imaging and high cross-range resolution owing to the constraint of minimum antenna area for conventional single-channel spaceborne synthetic aperture radar (SAR) imaging. To overcome the inherent systemic limitation, multi-channel SAR imaging has been developed. Nevertheless, this still suffers from various problems such as high system complexity. To simplify the system structure, a novel algorithm for high resolution multi-ship ScanSAR imaging based on sparse representation is proposed in this paper, where the SAR imaging model is established via maximum a posterior estimation by utilizing the sparsity prior of multi-ship targets. In the scheme, a wide swath is generated in the ScanSAR mode by continuously switching the radar footprint between subswaths. Meanwhile, high cross-range resolution is realized from sparse subapertures by exploiting the sparsity feature of multi-ship imaging. In particular, the SAR observation operator is constructed approximately as the inverse of conventional SAR imaging and then high resolution SAR imaging including range cell migration compensation is achieved by solving the optimization. Compared with multi-channel SAR imaging, the system complexity is effectively reduced in the ScanSAR mode. In addition, enhancement of the cross-range resolution is realized by incorporating the sparsity prior with sparse subapertures. As a result, the amount of data is effectively reduced. Experiments based on measured data have been carried out to confirm the effectiveness and validity of the proposed algorithm.