Multipath Exploitation in Highly Reflective Environments for Enhanced Microwave Imaging via Inverse Source Reconstruction

A team led by Quanfeng Wang (with Mei Song Tong and Thomas F. Eibert) has published a new method on arXiv that turns a common imaging nuisance—multipath reflections—into a resolution-enhancing asset. In highly reflective environments like metal-walled rooms or urban canyons, traditional microwave imaging suffers from degraded quality due to signals bouncing off surfaces. The researchers show that by modeling these reflections as virtual image sources (using image theory), they can effectively double the measurement aperture without physically enlarging the sensor array. A single-frequency inverse source solver reconstructs both the original and virtual source distributions, separates them, and then applies coherent phase-corrected imaging to both sets. This combination acts as an expanded synthetic aperture, delivering sharper resolution while suppressing interference artifacts that typically plague such environments.

Simulation results confirm the approach outperforms a ray-tracing enhanced backprojection algorithm, with theoretical backing showing resolution improvement scales with the number of exploitable reflections. The work bridges inverse scattering and source reconstruction, offering a practical path to better through-wall radar, non-destructive testing, and medical microwave imaging. Since the method works at a single frequency and doesn't require complex hardware changes, it could be deployed on existing systems with only software updates. The paper (arXiv:2605.03888) is currently under submission and available open access.