Compressive classification for Through-the-Wall Radar Imaging

Through-the-Wall Radar Imaging (TWRI) is an emerging technology, desirable for a variety of military and civilian applications. Some of these applications require the ability to make a rapid decision on the contents of a target scene, rather than reconstruct it perfectly. In this work, we address this problem by modifying the smashed filter proposed by Davenport et al., which is based on compressive sensing (CS) theory. We introduce two compressive classification methods, which allow for an efficient hardware implementation as well as deliver accurate classification results in the considered scenarios using only a fraction of the available data.