SVD-RPCA-Based Clutter Suppression Method Used for Through-Wall Radar Signal

When using through-wall radar to locate a target behind the wall, to suppress the noise associated with the environment such as the wall is necessary which helps to improve the accuracy of target location. In this paper, a clutter suppression algorithm based on the singular value decomposition (SVD) and the robust principal component analysis (RPCA) is proposed to process the step-frequency through-wall radar data. According to the feature of SVD filtering which does not need to pre-save background information and has low computational complexity, the SVD algorithm is first used to filter the most of the wall clutter signals from the echoes. Then, the RPCA algorithm is also introduced to further suppress the residual wall-noise and other clutter signals so as to improve the target localization accuracy. Finally, a simple step-frequency through-wall radar device based on a vector network analyzer is built, and the acquired data is processed utilizing the proposed method. The results prove this method can improve the quality of the range profile well.