This letter deals with microwave imaging via a migration-like inversion method. The aim is the determination of a spatial sampling scheme which allows to collect as low as possible spatial measurements and to preserve the point-spread function features and hence, the achievable performance in the reconstructions. To keep math simple and to more easily illustrate the theoretical arguments, the study is developed for a 2-D monostatic scalar configuration and a homogeneous background medium. It is shown that the new sampling scheme requires much fewer samples than the standard literature results and that the samples must be nonuniformly deployed across the measurement domain. This letter deals with microwave imaging via a migration-like inversion method. The aim is the determination of a spatial sampling scheme which allows to collect as low as possible spatial measurements and to preserve the point-spread function features and hence, the achievable performance in the reconstructions. To keep math simple and to more easily illustrate the theoretical arguments, the study is developed for a 2-D monostatic scalar configuration and a homogeneous background medium. It is shown that the new sampling scheme requires much fewer samples than the standard literature results and that the samples must be nonuniformly deployed across the measurement domain.
