A Novel Method for Reconstructing 3D Scattering Centers Based on Multiple HRR Profiles and Its Performance Bounds

Abstract

Reconstructing three-dimensional scattering centers of radar targets without knowledge of viewing angles or relative radar-target motion is investigated. The performance bounds are deduced in which the parameter constraints and rotatory ambiguities are specially treated. We propose a novel method based on the signal subspace invariance in multiple one-dimensional projection sequences, which greatly outperforms that in [5] in both computing complexity and reconstructing accuracy. Simulations show that our method approaches the Cramer-Rao Lower Bounds at ordinary SNRs.