Compressive tomographic radar imaging with total variation regularization

We consider the problem of compressive imaging of a three-dimensional (3D) scene using multiple observations collected from parallel baselines, formed by monostatic sensors moving in space. In particular, we present a novel iterative imaging method based on the Omega-K algorithm with edge-preserving 3D total variation (TV) regularization. The method combines joint processing of multi-baseline data with TV minimization in a computationally efficient way, thus enabling high-resolution imaging of the reflectivity map of the scene. We demonstrate the potential of our method through numerical evaluations on simulated data with noise.