Applications of Compressed Sensing for SAR Moving-Target Velocity Estimation and Image Compression

Compressed sensing (CS) is a recently introduced concept that enables the recovery of signals sampled below the Nyquist rate. A prerequisite for its application is the sparsity of the concerned signals in a certain basis. The concept has been applied to medical imaging, seismic imaging, optical remote sensing, and, recently, radar. This paper describes two possible applications of CS for synthetic aperture radar (SAR): 1) estimation of moving-target velocities in the case of high and low signal-to-clutter ratios, as well as more than one scatterer in a single pixel making use of clutter cancellation, and 2) online compression of processed SAR images with the help of curvelet sparsity-promoting offline decoding. CS for SAR offers the possibility of using lesser amount of data, as well as a convenient way of knowing motion and position parameters of multiple moving targets. The CS technique is useful for simplification of hardware or reduction of data that are useful in the case of limited onboard data storage capacity. The results are demonstrated by means of numerical experiments using simulated data for moving-target velocity estimation and MiniSAR data for SAR image compression.