Insect 3-D Alignment Retrieval From Multiview and Multifrequency Entomological Radars Using Symmetry Constrained Estimation

Estimating the 3-D alignment of migratory insects is essential for understanding their 3-D directional behavior. The existing approach reconstructs 3-D alignment using azimuthal angle measurements from dual-view entomological radars. However, the measurement errors often propagate into the reconstructed 3-D alignment, causing significant inaccuracies, especially under low signal-to-noise ratio (SNR) conditions. To overcome this problem, this letter proposes a new method based on a multiview, multifrequency, and full-polarization entomological radar system. The method directly estimates insect 3-D alignment by integrating polarization scattering matrices (SMs) from multiple views and frequencies, leveraging the assumption of scattering symmetry in insect bodies. For symmetric targets, the off-diagonal SM elements vanish when the polarization direction aligns with the symmetry plane. Based on this theory, an optimization problem is formulated to estimate 3-D alignment by minimizing the sum of the powers of off-diagonal elements across multiview and multifrequency SMs. Both simulations and field experiments demonstrate that the proposed method achieves substantially higher accuracy than the traditional method.