A novel SAR sparse imaging algorithm based on low-rank matrix completion

Due to the complex electromagnetic environment, the airborne Synthetic Aperture Radar (SAR) system is faced with the challenge of missing part of the echo data, leading to the degradation on SAR imaging performance. To solve this problems, a novel SAR sparse imaging algorithm based on low rank matrix completion is proposed in this paper. Based on the mechanism of joint aperture and bandwidth synthesis, a two-dimensional BP-based SAR sparse imaging model is constructed at first. And then, on the basis of the above model, the concept of low rank matrix completion is introduced in this paper, to reconstruct the missing elements of the SAR signals in the frequency domain, due to the sparse sampling. Specifically, the sparse signal reconstruction problem is transformed into a regularized least squares optimization problem. Which can be solved iteratively by using the imprecise augmented Lagrange multiplier (IALM) method alternately. Therefore, the issue of failure for compressed sensing based sparse imaging under low sampling rate can be resolved effectively. Numerical results demonstrate the effectiveness of the proposed method, and it also shows that the Integrated Side Lobe Ratio (ISLR) of the proposed algorithm is lower than that of the existed methods.