Related Experiment Video
Updated: Jun 19, 2026
08:48
Demonstration of Spin-Multiplexed and Direction-Multiplexed All-Dielectric Visible Metaholograms
Published on: September 25, 2020
Deep learning-enabled temporal sequencing of metasurface for rewritable and customizable electromagnetic illusions
Haoran Han1,2,3, Jiwei Zhao1,2,3, Huan Lu1,2,3
1State Key Laboratory of Extreme Photonics and Instrumentation, ZJU-Hangzhou Global Scientific and Technological Innovation Center, Zhejiang University, Hangzhou 310027, China.
National Science Review
|June 18, 2026
Summary
This study introduces a modular metasurface framework for programmable electromagnetic illusions. It enables complex, customizable wave control by sequencing temporal modulation units, achieving high-fidelity results in radar imaging.
Area of Science:
- Electromagnetics
- Metasurface Technology
- Wave Manipulation
Background:
- Precise spatiotemporal control of electromagnetic waves is crucial for advanced applications like electromagnetic illusions.
- Conventional metasurfaces lack the flexibility for complex, dynamic wave manipulation due to static, periodic protocols.
- Existing methods require extensive redesign for each new illusion pattern, limiting adaptability.
Purpose of the Study:
- To develop a novel framework for rewritable and customizable electromagnetic illusions.
- To enable dynamic and complex wave control beyond the limitations of conventional metasurfaces.
- To establish a practical method for translating target illusion specifications into executable metasurface control signals.
Main Methods:
- Introduction of a modular metasurface time-domain programming framework using a library of discrete temporal modulation waveforms.
- Development of a deep generative model to map target illusions directly to specific time-domain modulation sequences.
- Validation on a synthetic aperture imaging testbed, executing user-defined signals across distinct pulses.
Main Results:
- Successful synthesis of diverse electromagnetic illusions by flexibly selecting and sequencing modulation units.
- High fidelity achieved between intended objectives and measurements (structural similarity index ≥0.91) in radar imaging scenarios.
- Demonstration of rewriting periodic baselines and synthesizing representative aperiodic illusion patterns.
Conclusions:
- The developed framework provides a practical and scalable paradigm for task-driven wave manipulation.
- This approach enables precise spatiotemporal control for creating sophisticated electromagnetic illusions.
- The study establishes a direct route from target-scene specification to executable metasurface control for radar imaging.
