MVDR Beamforming for Through-the-Wall Radar Imaging

We present high-definition imaging for targets behind walls and enclosed structures based on constrained minimization RF multisensor processing. Minimum variance distortionless response (MVDR) beamforming is used on both sensor-frequency raw data radar returns and spatial spectrum data, which is obtained by the Fourier transform of the delay and sum beamformer image. We compare both methods for near-field and far-field scenes. The paper considers both cases of known and unknown wall parameters and uses manifold constraints to allow target localization in high-definition imaging in the presence of wall errors. Also, through analyses and simulations, we show how to effectively use the spatial spectrum to improve covariance matrix estimation and subsequently enhance image quality in the sense of lower sidelobes.