Synthetic aperture inversion with sparsity constraints

In this paper, we introduced a novel approach to SAR image formation with sparsity constraints. We formulate the sparse SAR image reconstruction problem as an L^p-norm constrained quadratic inversion problem and approximate it with a sequence of L²-norm constrained problems. We addressed each of L²-norm constrained problem analytically using microlocal techniques. Our inversion method can be implemented efficiently at every iteration using the fast computation of Fourier Integral Operators.