A multipath suppression technique for through-the-wall radar

Through-the-wall (TTW) radar images would be deteriorated by multipath artefacts resulting from the reflection of enclosed walls. This paper presents an innovative multipath ghost suppression technique based on the compressive sensing theory. The principle of the proposed technique is that: radar echoes are the projection of target space with the operator that defines the transfer function of environment and the image of the target space can be reconstructed by a minimum-norm inversion once the echoes and the operator are known. Conventional imaging technique fails to deal with complicated multipath scenario without prior information. The proposed technique, however, makes use of prior information of the location of walls and establishes a reflection multipath model to depict the scattering response of walls. Based on the model, an overcomplete dictionary that account for interaction of the radar, targets and environment can be constructed. The image reconstruction is implemented with the CS technique by solving a convex optimization problem. The proposed technique can achieve excellent multipath interference suppression effect and can significantly reduce data recording time of TTW radar. The new technique is validated by numeric simulation.