Microwave Imaging of Small Scatterers: Linking and Comparing the Beam-Forming and the Orthogonality Sampling Method

Advancements in sensor array technology and radar imaging applications continue to trigger the research toward the development of new imaging algorithms. The orthogonality sampling method (OSM) is one of the most recently proposed imaging methods in the literature. In this article, we are concerned with studying how the OSM is linked to and compared with classical beam-forming (BF) methods. In particular, herein, we consider two versions of the BF algorithm: the windowed and the nonwindowed schemes. This goal is pursued in the framework of small scatterer imaging, by considering a multiview/multistatic/multifrequency configuration. We succeed in analytically estimating the point-spread functions (PSFs) of the methods under comparison and this allows us to establish a clear connection between the different methods in terms of the achievable resolution and, at the same time, to highlight the link between the configuration parameters and the performance. The role of noise, the cross-terms, the multiple scattering, and the uncertain antenna frequency response are also addressed by a combination of analytical and numerical arguments. It is shown that the nonwindowed BF scheme exhibits the best tradeoff between achievable performance and robustness against noise and antenna uncertainties.