Analysis of the effect of baseline redundancy on synthetic aperture interferometric radiometers performance

In the recent years, microwave synthetic aperture interferometric radiometers have been proposed for many space-based L-band passive imaging applications. As an example, the European Space Agency SMOS mission currently embarks the MIRAS instrument (Microwave Imaging Radiometer with Aperture Synthesis), which is the first interferometric radiometer with 2-D imaging capability used for earth observations from space [1]. MIRAS is based on the so called “Y shaped” antenna array configuration and has the objective of providing global maps of soil moisture over land and surface salinity over oceans. MIRAS allows generating records of brightness temperature with accuracy of 2–5K with 30–50Km of spatial resolution and a swath of approximately 900Km depending on incidence angles. Nowadays, an important aspect which is under investigation for next generation radiometers is the improvement of spatial resolution down to 20Km in order to allow regional scales retrievals. However, an increase of spatial resolution yields to a consequent reduction of radiometric resolution if other instrument performance parameters are maintained the same (FOV, instrument complexity,…). In this respect, increasing the level of redundancy of the visibility samples as much as possible uniformly over the (u,v) plane could be a means to improve the sensitivity with no expense for spatial resolution [2]. This paper preliminary analyzes the performance of synthetic aperture interferometric radiometers for three different geometrical array configurations defined to exploit uniform baseline redundancy distribution over the (u,v) plane.