Compressed Sensing experiments in a noise SAR

The paper presents a practical application of Compressed Sensing paradigm to a noise synthetic aperture radar. Nonuniform sampling in spatial dimension followed by Compressed Sensing reconstruction of SAR image allows to lower the number of spatial (along-track) domain samples taken in a SAR radar. When the spatial sampling frequency is limited by external constraints, the image taken with classical approach would suffer from spatial aliasing. For sparse targets, Compressed Sensing approach leads to correct image reconstruction. The described technique is applied to the data from an experimental ground-based noise SAR built in Warsaw University of Technology.