Laser & Optoelectronics Progress, Volume. 61, Issue 4, 0424001(2024)
Optimized Design of Terahertz Coding Frequency Selection Surface Based on Deep Learning
Fig. 1. Schematic of the FSS simulation structure. (a) 3D simulation structure; (b) FSS topological code
Fig. 6. Illustration of CNN training process. (a) Loss value variation; (b) RMSE variation
Fig. 7. Comparison of prediction results and simulation results of CNN. (a)‒(d) Training set; (e)‒(h) testing set
Fig. 9. Narrowband bandpass FSS design result. (a) Optimal result; (b) topological encoding matrix
Fig. 10. Broadband bandpass FSS design result. (a) Optimal result; (b) topological encoding matrix
Fig. 11. Narrowband bandstop FSS design result. (a) Optimal result; (b) topological encoding matrix
Fig. 12. Broadband bandstop FSS design result. (a) Optimal result; (b) topological encoding matrix
Fig. 13. Dual-band narrowband bandpass FSS design result. (a) Optimal result; (b) topological encoding matrix
Get Citation
Copy Citation Text
Pan Zhou, Lei Gong, Zhiqiang Yang, Liguo Wang, Lihong Yang, Yao Li, Haibin Wang, Jie Yu. Optimized Design of Terahertz Coding Frequency Selection Surface Based on Deep Learning[J]. Laser & Optoelectronics Progress, 2024, 61(4): 0424001
Category: Optics at Surfaces
Received: May. 6, 2023
Accepted: Jun. 20, 2023
Published Online: Feb. 27, 2024
The Author Email: Lei Gong (gonglei@xatu.edu.cn)
CSTR:32186.14.LOP231229