Calibration-Aware 3-D Polarimetric Coupling Suppression for Cylindrical Near-Field Imaging and RCS Recovery

Cylindrical near-field radar cross-section (RCS) measurement can be biased by floor/support-induced target-background coupling when this coupling survives ordinary background subtraction, thereby degrading both focused 3-D images and co-polarized RCS recovery. This paper presents a calibrated recovery workflow for sequential full-polarimetric cylindrical sensing. Channel-consistent amplitude-phase equalization, drift compensation, and polarization-axis correc-tion first place the HH, HV, VH, and VV channels on a common complex reference. The calibrated residual data are then re-constructed by a copolar-fused coarse-to-fine 3-D back-projection scheme. When target scattering becomes complex and prior geometric cues are insufficient, focused im-ages alone do not reliably identify coupling sources. Voxel-wise coherency analysis and four-component decomposi-tion are therefore used only to construct a mechanism-sensitive coupling mask, after which component-weighted at-tenuation reduces the coupled power while preserving the principal target response. Metallic spheres and a metallic cylinder provide measured validation cases in both the image domain and the RCS domain, and a matched-configuration scaled-aircraft simulation is used to examine the same workflow on a more complex geometry. Under the studied chamber conditions, the results show cleaner focused images and reduced deviations between the recov-ered co-polarized RCS curves and their corresponding references.