Clutter Suppression Based on ICUR-RPCA for Random Missing Data

The performance of clutter suppression in synthetic aperture radar (SAR) systems is significantly compromised by random missing data. Typically, traditional methods frequently encounter a balance between accuracy and computational complexity. To address the problem, a clutter suppression method based on inverse CUR-robust principal component analysis (ICUR-RPCA) is proposed for the random missing data. Taking advantage of the low-rank property, key nodes entries are completed with reorganizing the whole echo data, first. Subsequently, RPCA is combined with the recovered key nodes and ICUR decomposition, so that the clutter suppression performance with the completed whole data can be achieved by only completing a portion of the data. Finally, the superiority of our ICUR-RPCA method is shown in numerical experiments.