A New Approach for Solving Inverse Scattering Problems Based on Physics-informed Supervised Residual Learning

Tao Shan; Zhichao Lin; Xiaoqian Song; Maokun Li; Fan Yang; Shenheng Xu

A New Approach for Solving Inverse Scattering Problems Based on Physics-informed Supervised Residual Learning

Tao Shan
, Zhichao Lin
, Xiaoqian Song
, Maokun Li^*
, Fan Yang
, Shenheng Xu

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

5 Scopus citations

Abstract

In this paper, we propose a new approach for solving inverse scattering problems (ISPs) by applying the physics-informed supervised residual learning (PhiSRL) to embody the Born iterative method (BIM). Stemming from the mathematical link between the fixed-point iteration method and residual neural network (ResNet), PhiSRL fulfills the alternate iteration process of BIM by predicting the modifications of the candidate solutions regarding the calculated residuals. Thus, the proposed approach can perform the inversions of both data and models at the same time. The effectiveness of the proposed approach is further validated by synthetic data. This paper provides new insights for designing the deep learning (DL) based methods with the knowledge of traditional computational electromagnetic (EM) algorithms.

Original language	English
Title of host publication	2022 16th European Conference on Antennas and Propagation, EuCAP 2022
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9788831299046
State	Published - 2022
Externally published	Yes
Event	16th European Conference on Antennas and Propagation, EuCAP 2022 - Madrid, Spain Duration: 27 Mar 2022 → 1 Apr 2022

Publication series

Name	2022 16th European Conference on Antennas and Propagation, EuCAP 2022

Conference

Conference	16th European Conference on Antennas and Propagation, EuCAP 2022
Country/Territory	Spain
City	Madrid
Period	27/03/22 → 1/04/22

Keywords

Born iterative method
Inverse scattering problem
deep learning
fixed-point iteration method
physics-informed supervised residual learning
residual neural network

Cite this

Shan, T., Lin, Z., Song, X., Li, M., Yang, F., & Xu, S. (2022). A New Approach for Solving Inverse Scattering Problems Based on Physics-informed Supervised Residual Learning. In 2022 16th European Conference on Antennas and Propagation, EuCAP 2022 (2022 16th European Conference on Antennas and Propagation, EuCAP 2022). Institute of Electrical and Electronics Engineers Inc..

@inproceedings{b8f3dbb1330343a9bbd35135acf77415,

title = "A New Approach for Solving Inverse Scattering Problems Based on Physics-informed Supervised Residual Learning",

abstract = "In this paper, we propose a new approach for solving inverse scattering problems (ISPs) by applying the physics-informed supervised residual learning (PhiSRL) to embody the Born iterative method (BIM). Stemming from the mathematical link between the fixed-point iteration method and residual neural network (ResNet), PhiSRL fulfills the alternate iteration process of BIM by predicting the modifications of the candidate solutions regarding the calculated residuals. Thus, the proposed approach can perform the inversions of both data and models at the same time. The effectiveness of the proposed approach is further validated by synthetic data. This paper provides new insights for designing the deep learning (DL) based methods with the knowledge of traditional computational electromagnetic (EM) algorithms.",

keywords = "Born iterative method, Inverse scattering problem, deep learning, fixed-point iteration method, physics-informed supervised residual learning, residual neural network",

author = "Tao Shan and Zhichao Lin and Xiaoqian Song and Maokun Li and Fan Yang and Shenheng Xu",

note = "Publisher Copyright: {\textcopyright} 2022 European Association for Antennas and Propagation.; 16th European Conference on Antennas and Propagation, EuCAP 2022 ; Conference date: 27-03-2022 Through 01-04-2022",

year = "2022",

language = "英语",

series = "2022 16th European Conference on Antennas and Propagation, EuCAP 2022",

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

booktitle = "2022 16th European Conference on Antennas and Propagation, EuCAP 2022",

address = "美国",

}

Shan, T, Lin, Z, Song, X, Li, M, Yang, F & Xu, S 2022, A New Approach for Solving Inverse Scattering Problems Based on Physics-informed Supervised Residual Learning. in 2022 16th European Conference on Antennas and Propagation, EuCAP 2022. 2022 16th European Conference on Antennas and Propagation, EuCAP 2022, Institute of Electrical and Electronics Engineers Inc., 16th European Conference on Antennas and Propagation, EuCAP 2022, Madrid, Spain, 27/03/22.

A New Approach for Solving Inverse Scattering Problems Based on Physics-informed Supervised Residual Learning. / Shan, Tao; Lin, Zhichao; Song, Xiaoqian et al.
2022 16th European Conference on Antennas and Propagation, EuCAP 2022. Institute of Electrical and Electronics Engineers Inc., 2022. (2022 16th European Conference on Antennas and Propagation, EuCAP 2022).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - A New Approach for Solving Inverse Scattering Problems Based on Physics-informed Supervised Residual Learning

AU - Shan, Tao

AU - Lin, Zhichao

AU - Song, Xiaoqian

AU - Li, Maokun

AU - Yang, Fan

AU - Xu, Shenheng

PY - 2022

Y1 - 2022

N2 - In this paper, we propose a new approach for solving inverse scattering problems (ISPs) by applying the physics-informed supervised residual learning (PhiSRL) to embody the Born iterative method (BIM). Stemming from the mathematical link between the fixed-point iteration method and residual neural network (ResNet), PhiSRL fulfills the alternate iteration process of BIM by predicting the modifications of the candidate solutions regarding the calculated residuals. Thus, the proposed approach can perform the inversions of both data and models at the same time. The effectiveness of the proposed approach is further validated by synthetic data. This paper provides new insights for designing the deep learning (DL) based methods with the knowledge of traditional computational electromagnetic (EM) algorithms.

AB - In this paper, we propose a new approach for solving inverse scattering problems (ISPs) by applying the physics-informed supervised residual learning (PhiSRL) to embody the Born iterative method (BIM). Stemming from the mathematical link between the fixed-point iteration method and residual neural network (ResNet), PhiSRL fulfills the alternate iteration process of BIM by predicting the modifications of the candidate solutions regarding the calculated residuals. Thus, the proposed approach can perform the inversions of both data and models at the same time. The effectiveness of the proposed approach is further validated by synthetic data. This paper provides new insights for designing the deep learning (DL) based methods with the knowledge of traditional computational electromagnetic (EM) algorithms.

KW - Born iterative method

KW - Inverse scattering problem

KW - deep learning

KW - fixed-point iteration method

KW - physics-informed supervised residual learning

KW - residual neural network

UR - https://www.scopus.com/pages/publications/85130605029

M3 - 会议稿件

AN - SCOPUS:85130605029

T3 - 2022 16th European Conference on Antennas and Propagation, EuCAP 2022

BT - 2022 16th European Conference on Antennas and Propagation, EuCAP 2022

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 16th European Conference on Antennas and Propagation, EuCAP 2022

Y2 - 27 March 2022 through 1 April 2022

ER -

A New Approach for Solving Inverse Scattering Problems Based on Physics-informed Supervised Residual Learning

Abstract

Publication series

Conference

Keywords

Other files and links

Fingerprint

Cite this