Solving Electromagnetic Scattering from 3-D Scatterers with Incomplete Profiles by Deep Learning

Ji Yuan Wang; Zhi Yuan Chen; Liang Zhang; Xin Yue Lou; Yun Chuan Wang; Chen Tao Zhao; Xiao Min Pan

doi:10.1109/ICCEM68495.2026.11628900

Solving Electromagnetic Scattering from 3-D Scatterers with Incomplete Profiles by Deep Learning

Ji Yuan Wang
, Zhi Yuan Chen
, Liang Zhang
, Xin Yue Lou
, Yun Chuan Wang
, Chen Tao Zhao
, Xiao Min Pan^*

^*Corresponding author for this work

Beijing Institute of Technology

Research output: Contribution to journal › Conference article › peer-review

Abstract

A deep learning (DL)-based scheme is proposed to solve electromagnetic (EM) scattering from a three-dimensional (3-D) dielectric scatterer whose profile is incomplete. As compensation, a limited amount of scattering data can be acquired. The measurement data from the Fresnel database are employed to validate the proposed scheme.

Original language	English
Journal	IEEE International Conference on Computational Electromagnetics, ICCEM
Issue number	2026
DOIs	https://doi.org/10.1109/ICCEM68495.2026.11628900
Publication status	Published - 2026
Externally published	Yes
Event	2026 IEEE International Conference on Computational Electromagnetics, ICCEM 2026 - Shanghai, China Duration: 10 Apr 2026 → 12 Apr 2026

Keywords

deep learning (DL)
Electromagnetic (EM) scattering
EM inverse scattering

Access to Document

10.1109/ICCEM68495.2026.11628900

Cite this

@article{351d9eaf0fc549f3b3425b3185fd53bd,

title = "Solving Electromagnetic Scattering from 3-D Scatterers with Incomplete Profiles by Deep Learning",

abstract = "A deep learning (DL)-based scheme is proposed to solve electromagnetic (EM) scattering from a three-dimensional (3-D) dielectric scatterer whose profile is incomplete. As compensation, a limited amount of scattering data can be acquired. The measurement data from the Fresnel database are employed to validate the proposed scheme.",

keywords = "deep learning (DL), Electromagnetic (EM) scattering, EM inverse scattering",

author = "Wang, \{Ji Yuan\} and Chen, \{Zhi Yuan\} and Liang Zhang and Lou, \{Xin Yue\} and Wang, \{Yun Chuan\} and Zhao, \{Chen Tao\} and Pan, \{Xiao Min\}",

note = "Publisher Copyright: {\textcopyright} 2026 IEEE.; 2026 IEEE International Conference on Computational Electromagnetics, ICCEM 2026 ; Conference date: 10-04-2026 Through 12-04-2026",

year = "2026",

doi = "10.1109/ICCEM68495.2026.11628900",

language = "English",

journal = "IEEE International Conference on Computational Electromagnetics, ICCEM",

issn = "2768-2552",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

number = "2026",

}

TY - JOUR

T1 - Solving Electromagnetic Scattering from 3-D Scatterers with Incomplete Profiles by Deep Learning

AU - Wang, Ji Yuan

AU - Chen, Zhi Yuan

AU - Zhang, Liang

AU - Lou, Xin Yue

AU - Wang, Yun Chuan

AU - Zhao, Chen Tao

AU - Pan, Xiao Min

PY - 2026

Y1 - 2026

N2 - A deep learning (DL)-based scheme is proposed to solve electromagnetic (EM) scattering from a three-dimensional (3-D) dielectric scatterer whose profile is incomplete. As compensation, a limited amount of scattering data can be acquired. The measurement data from the Fresnel database are employed to validate the proposed scheme.

AB - A deep learning (DL)-based scheme is proposed to solve electromagnetic (EM) scattering from a three-dimensional (3-D) dielectric scatterer whose profile is incomplete. As compensation, a limited amount of scattering data can be acquired. The measurement data from the Fresnel database are employed to validate the proposed scheme.

KW - deep learning (DL)

KW - Electromagnetic (EM) scattering

KW - EM inverse scattering

UR - https://www.scopus.com/pages/publications/105050190631

U2 - 10.1109/ICCEM68495.2026.11628900

DO - 10.1109/ICCEM68495.2026.11628900

M3 - Conference article

AN - SCOPUS:105050190631

SN - 2768-2552

JO - IEEE International Conference on Computational Electromagnetics, ICCEM

JF - IEEE International Conference on Computational Electromagnetics, ICCEM

IS - 2026

T2 - 2026 IEEE International Conference on Computational Electromagnetics, ICCEM 2026

Y2 - 10 April 2026 through 12 April 2026

ER -

Solving Electromagnetic Scattering from 3-D Scatterers with Incomplete Profiles by Deep Learning

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this