Chongfan Technology
News
12
2026
-
06
Inverse Design of Phase-Change Reconfigurable Terahertz Metasurfaces Based on Deep Learning
Author:
The teams led by Zhen Tian at Tianjin University, Ranjan Singh at the University of Notre Dame, and Tun Cao at Dalian University of Technology have proposed a deep-learning–based inverse-design framework capable of fabricating dynamically reconfigurable terahertz metamaterial devices and enabling adaptive multidimensional encryption. By leveraging a residual neural network trained to directly map a target electromagnetic response to the device’s geometry—thus spanning continuous material phase transitions—the authors’ approach overcomes the bottlenecks of conventional iterative design and achieves rapid, high‑precision generation of multifunctional metamorphic architectures. The resulting devices enable multiplexed control over polarization, depth, and phase transition in Ge₂Sb₂Te₅ (GST), realizing an eight‑channel encrypted hologram with minimal crosstalk and fidelity approaching the diffraction limit. Furthermore, the authors have demonstrated a reconfigurable diffractive terahertz neuromorphic metamaterial that performs universal logic operations under a dual‑key security protocol, requiring both physical hardware and a digital key sequence for precise decryption. The integration of intelligent design automation with physical‑layer encryption establishes a new paradigm for secure, high‑capacity, and adaptive terahertz communications, paving the way for dynamic, reconfigurable wireless architectures in 6G and beyond.
The research findings were published on June 5, 2026, in Advanced Materials, under the title “Deep Learning Inverse Design of Phase-Change Reconfigurable Terahertz Metadevices for Multidimensional Secure Communication.”
Figure 1: Inverse Design of a Dynamic Encryption Terahertz Reconfigurable Metasurface Based on Deep Learning
Figure 2: Architecture and results of the forward prediction network, along with the inverse design framework for multidimensional multiplexing terahertz metadevices.
Figure 3: Simulation and experimental characterization of a terahertz dynamic reconfigurable metamaterial device.
Figure 4: Terahertz Reconfigurable Logic Computing Platform
Source: Optics World
LATEST NEWS
Thank you for visiting the official website of Chongfan Technology. If you have cooperation intentions or suggestions, please contact us through the following methods, and we will reply as soon as possible, thank you!
Address: Room 403, Building 6, Phase III of R&D, No. 36 Xiyong Avenue, High tech Zone, Chongqing, China.
Telephone: +86-13658337211
E-mail: Sales@cfkeji.net
Website: www.cfkeji.net
Mobile Version
