Clutters caused by multipath have been widely researched in through-the-wall radar imaging (TWRI). The existing research work of multipath only consider reflections from the wall, while in the condition of a small scene, with the increasing number of targets, multipath from targets to targets, named interaction multipath, usually generates ghosts, which degrades the performance of TWRI. In order to mitigate the effect of interaction multipath, considering fast data acquisition and measurement reduction, we made use of the propagation characteristic of interaction multipath to build the sparse model of the target scene and developed a compressive sensing (CS)-based method, which is referred to as 'interaction CS'. For the number of point targets increasing from 5-8, intensive evaluation and direct comparison of the imaging results with existing methods are conducted to show that the proposed interaction CS performs better at ghost suppression in the same condition of the signal-to-noise ratio (SNR).
Keywords: compressive sensing; ghost suppression; interaction multipath; through-the-wall radar imaging.