Deep-learning Based Wall Mitigation Method for Through-the-wall Radar Imaging

Through-the-wall radar imaging is a noteworthy method for imaging targets hidden behind walls. The presence of a wall, however, serves as a spatial filter, impacting the scattering quality of the target’s field, which in turn influences the clarity of the resultant image. To address this, a variety of filtering strategies have been developed to separate the target’s signal from the wall’s response. These techniques, while effective in distinguishing the target and differentiating between multiple targets, also affect the imaging capability and the discrimination process. Here, we propose a deep-learning approach to avoid this.