An Efficient ISAR Autofocus Imaging Algorithm Utilizing the Block Sparse Structure

In this paper, an efficient phase error correction algorithm for inverse synthetic aperture radar (ISAR) imaging is proposed based on the structured sparse Bayesian learning. Considering the block-sparse structure of the targets, a modified pattern-coupled prior model is developed to encourage the continuity of the nonzero scatterers in the autofocus imaging problem. Meanwhile, a new phase error estimation method utilizing the weighted least-squares (WLS) criteria is presented in our algorithm. As a result, the estimation accuracy of the scattering coefficients and the phase errors has been improved significantly. Additionally, the generalized approximate message passing (GAMP) technique is used to reduce the computational complexity of the Bayesian inference. Experimental results, based on both the simulated and the measured data, demonstrate the effectiveness of the proposed algorithm.