Sparse autofocus via Bayesian learning iterative maximum and applied for LASAR 3-D imaging

Abstract

Linear array SAR (LASAR) is a promising 3-D radar imaging technology. As 3-D radar images usually exhibit strong sparsity, compressed sensing sparse recovery algorithms can be used for LASAR imaging even if the echoes are under-sampled. However, most of the existing sparse recovery algorithms assume exact knowledge of the signal acquisition model, which is impractical for LASAR due to the phase errors are inevitable caused by uncertainties. In this paper, a novel sparse autofocus algorithm is proposed for LASAR imaging via Bayesian learning iterative maximum. In the scheme, the sparse scatterering coefficients are treated as exponential distribution and the phase errors are assumed as uniform distribution. Exploiting the Bayesian learning and maximum likelihood estimation, the approach solves a joint optimization problem to achieve phase errors estimation and image formation simultaneously. Simulation and experimental results are presented to confirm the effectiveness of the algorithm.