Salvato in:
Dettagli Bibliografici
Autori principali: Du, Yutong, Liu, Zicheng, Huang, Yi, Matkerim, Bazargul, Qi, Bo, Zong, Yali, Han, Peixian
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:https://arxiv.org/abs/2602.13805
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866917274773880832
author Du, Yutong
Liu, Zicheng
Huang, Yi
Matkerim, Bazargul
Qi, Bo
Zong, Yali
Han, Peixian
author_facet Du, Yutong
Liu, Zicheng
Huang, Yi
Matkerim, Bazargul
Qi, Bo
Zong, Yali
Han, Peixian
contents Untrained neural networks (UNNs) offer high-fidelity electromagnetic inverse scattering reconstruction but are computationally limited by high-dimensional spatial-domain optimization. We propose a Real-Time Physics-Driven Fourier-Spectral (PDF) solver that achieves sub-second reconstruction through spectral-domain dimensionality reduction. By expanding induced currents using a truncated Fourier basis, the optimization is confined to a compact low-frequency parameter space supported by scattering measurements. The solver integrates a contraction integral equation (CIE) to mitigate high-contrast nonlinearity and a contrast-compensated operator (CCO) to correct spectral-induced attenuation. Furthermore, a bridge-suppressing loss is formulated to enhance boundary sharpness between adjacent scatterers. Numerical and experimental results demonstrate a 100-fold speedup over state-of-the-art UNNs with robust performance under noise and antenna uncertainties, enabling real-time microwave imaging applications.
format Preprint
id arxiv_https___arxiv_org_abs_2602_13805
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Fast Physics-Driven Untrained Network for Highly Nonlinear Inverse Scattering Problems
Du, Yutong
Liu, Zicheng
Huang, Yi
Matkerim, Bazargul
Qi, Bo
Zong, Yali
Han, Peixian
Machine Learning
Computational Physics
Untrained neural networks (UNNs) offer high-fidelity electromagnetic inverse scattering reconstruction but are computationally limited by high-dimensional spatial-domain optimization. We propose a Real-Time Physics-Driven Fourier-Spectral (PDF) solver that achieves sub-second reconstruction through spectral-domain dimensionality reduction. By expanding induced currents using a truncated Fourier basis, the optimization is confined to a compact low-frequency parameter space supported by scattering measurements. The solver integrates a contraction integral equation (CIE) to mitigate high-contrast nonlinearity and a contrast-compensated operator (CCO) to correct spectral-induced attenuation. Furthermore, a bridge-suppressing loss is formulated to enhance boundary sharpness between adjacent scatterers. Numerical and experimental results demonstrate a 100-fold speedup over state-of-the-art UNNs with robust performance under noise and antenna uncertainties, enabling real-time microwave imaging applications.
title Fast Physics-Driven Untrained Network for Highly Nonlinear Inverse Scattering Problems
topic Machine Learning
Computational Physics
url https://arxiv.org/abs/2602.13805