Fei Guo
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Fei Guo has a total of 8 co-authors. Here are the top co-authors by number of shared publications:
- Bingbing Song
State Key Laboratory of Millimeter waves, School of Cyber Science and Engineering, Southeast University, Nanjing, China
3 shared publications - Tian Liu
State Key Laboratory of Millimeter Waves, School of Information Science and Engineering, Southeast University, Nanjing, China
3 shared publications - Wei-Bing Lu
State Key Laboratory of Millimeter Waves, School of Information Science and Engineering, Southeast University, Nanjing, China
3 shared publications - Wu Yang
State Key Laboratory of Millimeter Waves, School of Information Science and Engineering, Southeast University, Nanjing, China
3 shared publications - Hao Chen
State Key Laboratory of Millimeter Waves, School of Information Science and Engineering, Southeast University, Nanjing, China
2 shared publications - Jian Wei You
State Key Laboratory of Millimeter Wave Southeast University Nanjing China
2 shared publications
Latest Publications
- 1 October 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Antennas and Propagation
- Vol. 74 (10) , 9886-9899
Intelligent Subentire-Domain Basis Function Method Based on Adaptive Physics-Informed Transformer Framework
- 1 October 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Antennas and Propagation
- Vol. 74 (10) , 9886-9899
Intelligent Subentire-Domain Basis Function Method Based on Adaptive Physics-Informed Transformer Framework
- 27 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Antennas and Wireless Propagation Letters
- Vol. 25 (1) , 319-323
Efficient Analysis of Electromagnetic Scattering From Composite Finite Periodic Structures by Prior Physics-Driven Graph Neural Network
In this letter, a novel prior physics-driven graph neural network (PPDGNN) is proposed firstly to rapidly compute the far-field scattering in the composite large-scale finite periodic structures (LFPSs). Each composite cell is considered as a graph node and all nodes are prior physically connected according to the mutual coupling relationship. The current coefficients of the metal cell are taken as a priori physical information, which together with the structural parameters forms the feature matrix. The PPDGNN can effectively capture the scattering field of the composite structure driven by the prior information, instead of filling and solving impedance matrix equations. Additionally, transfer learning is applied to reduce the time cost by nearly 80%, while enabling the model to analyze structures with larger scales and complex metal cells. Numerical experiments demonstrate that the PPDGNN exhibits accuracy, generalization, and efficiency.
- 27 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Antennas and Wireless Propagation Letters
- Vol. 25 (1) , 319-323
Efficient Analysis of Electromagnetic Scattering From Composite Finite Periodic Structures by Prior Physics-Driven Graph Neural Network
In this letter, a novel prior physics-driven graph neural network (PPDGNN) is proposed firstly to rapidly compute the far-field scattering in the composite large-scale finite periodic structures (LFPSs). Each composite cell is considered as a graph node and all nodes are prior physically connected according to the mutual coupling relationship. The current coefficients of the metal cell are taken as a priori physical information, which together with the structural parameters forms the feature matrix. The PPDGNN can effectively capture the scattering field of the composite structure driven by the prior information, instead of filling and solving impedance matrix equations. Additionally, transfer learning is applied to reduce the time cost by nearly 80%, while enabling the model to analyze structures with larger scales and complex metal cells. Numerical experiments demonstrate that the PPDGNN exhibits accuracy, generalization, and efficiency.
- 6 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Antennas and Wireless Propagation Letters
- Vol. 24 (12) , 4970-4974
Efficient Analysis of Electromagnetic Scattering in Large-Scale Finite Periodic Structures Using Characteristic Modes
Periodic characteristic mode (CM) analysis (PCMA) has proven to be an efficient method for analyzing large-scale finite periodic structures (LFPSs). However, its CMs are obtained solely from the reference cell, the solution process neglects the coupling effects among cells. To address this issue, this letter proposes an accurate method for analyzing the scattering properties of LFPSs. On one hand, by introducing dummy cells, the computation of CMs incorporates the coupling effects of surrounding cells, thereby improving accuracy. On the other hand, the compressive sensing (CS) technique is utilized to partially fill the impedance matrix to rapidly construct an overdetermined equation instead of a reduced matrix equation. Numerical examples are given to verify the accuracy and efficiency of the proposed method.
- 6 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Antennas and Wireless Propagation Letters
- Vol. 24 (12) , 4970-4974
Efficient Analysis of Electromagnetic Scattering in Large-Scale Finite Periodic Structures Using Characteristic Modes
Periodic characteristic mode (CM) analysis (PCMA) has proven to be an efficient method for analyzing large-scale finite periodic structures (LFPSs). However, its CMs are obtained solely from the reference cell, the solution process neglects the coupling effects among cells. To address this issue, this letter proposes an accurate method for analyzing the scattering properties of LFPSs. On one hand, by introducing dummy cells, the computation of CMs incorporates the coupling effects of surrounding cells, thereby improving accuracy. On the other hand, the compressive sensing (CS) technique is utilized to partially fill the impedance matrix to rapidly construct an overdetermined equation instead of a reduced matrix equation. Numerical examples are given to verify the accuracy and efficiency of the proposed method.