Wideband super-resolution imaging in Radio Interferometry via low rankness and joint average sparsity models (HyperSARA)

We propose a new approach within the versatile framework of convex optimization to solve the radio-interferometric wideband imaging problem. Our approach, dubbed HyperSARA, leverages low rankness, and joint average sparsity priors to enable formation of high-resolution and high-dynamic range image cubes from visibility data. The resulting minimization problem is solved using a primal-dual algorithm. The algorithmic structure is shipped with highly interesting functionalities such as preconditioning for accelerated convergence, and parallelization enabling to spread the computational cost and memory requirements across a multitude of processing nodes with limited resources. In this work, we provide a proof of concept for wideband image reconstruction of megabyte-size images. The better performance of HyperSARA, in terms of resolution and dynamic range of the formed images, compared to single channel imaging and the clean-based wideband imaging algorithm in the wsclean software, is showcased on simulations and Very Large Array observations. Our matlab code is available online on github.