Compressive sensing for through wall radar imaging of stationary scenes using arbitrary data measurements

Abstract

In this paper, we deal with removal of wall EM reflections prior to image reconstruction using step-frequency radars. The goal is to enable behind-the-wall target detection and localization from reduced data measurements. In the underlying problem, few frequency observations are available and they differ from one antenna position to another in a SAR imaging system. Because of using a different set of frequencies for different antennas, direct applications of wall clutter mitigation methods, such as subspace and spatial filtering, prove ineffective. To provide these methods with the response measured at the same set of frequencies, a compressive sensing approach is used to reconstruct the range profiles. We use prior knowledge of the wall standoff distance to speed up the convergence of the Orthogonal Matching Pursuit for sparse data reconstruction.