LAUSR

Recently, we have proposed the coprime tomographic synthetic aperture radar (TomoSAR) technique, whose baseline configuration conforms to the coprime array geometry. This technique is devoted to reducing the required number of acquisitions in the practical application where the number of flight passes is usually restricted due to cost consideration and temporal decoherence. This letter extends the tomographic reconstruction of the coprime TomoSAR to the atomic norm minimization (ANM) framework to pursue super-resolution. A compact ANM approach is proposed in this letter for the tomographic reconstruction of coprime TomoSAR in the presence of multiple looks data. Compared with the conventional ANM approach, the proposed approach compresses the dimension of the ANM model to a smaller size by two operations. One operation is that the equivalent covariance matrix is constructed to be conformed with the covariance matrix of the real acquisition data. The other operation is adopting the singular value decomposition (SVD) technique to reduce the look dimension of acquisition data. As a result, the compact approach reduces the computation complexity without performance loss. It is confirmed by simulation experiments.