19 July 2024 in Science Advances
Diffractive neural network in electromagnetic wave–driven system has attracted great attention due to its ultrahigh parallel computing capability and energy efficiency. However, recent neural networks based on the diffractive framework still face the bottlenecks of misalignment and relatively large size limiting their further applications. Here, we propose a planar diffractive neural network (pla-NN) with a highly integrated and conformal architecture to achieve direct signal processing in the microwave frequency. On the basis of printed circuit fabrication process, the misalignment could be effectively circumvented while enabling flexible extension for multiple conformal and stacking designs. We first conduct validation on the fashion-MNIST dataset and experimentally build up a system using the proposed network architecture for direct recognition of different geometry structures in the electromagnetic space. We envision that the presented architecture, once combined with the advanced dynamic maneuvering techniques and flexible topology, would exhibit unlimited potentials in the areas of high-performance computing, wireless sensing, and flexible wearable electronics.
25 August 2025 in Nature Communications
Brain–computer interface (BCI) provides an interconnected pathway between the human brain and external devices and paves a potential route for mind manipulations. However, most existing BCI technologies are based on simple signal transmission and are independent of other interface devices, with limited consideration for the reliability and security of the human brain’s information interaction in complicated wireless environments. Here, we propose a deep fusion coding scheme that combines the BCI visual stimulation coding with metasurface space-time coding at the physical layer, enabling reliable and secure information transfers between the human brain and external devices. A brain space-time-coding metasurface platform is designed to implement a secure wireless communication system by using harmonic-encrypted beams. We design and fabricate a proof-of-principle prototype and experimentally show that the proposed wireless BCI scheme can establish a remote but safeguarded paradigm for human–machine interactions and intelligent metasurfaces, providing a potential direction in future secure wireless communications.
3 April 2023 in IEEE Transactions on Antennas and Propagation
Traditional phase shifters are mainly focused on phase performance and rarely consider the phase shift between different electromagnetic (EM) modes, limiting their widespread applications. Spoof surface plasmon polaritons (SSPPs) support EM waves with strong field confinements, giving rise to the time delay of signals. Due to the flexible dispersion behaviors, SSPPs can be engineered at will by designing the plasmonic structure parameters. Benefitting from these characteristics, here we propose a multi-mode and reconfigurable SSPP phase shifter loaded with varactor diodes. By controlling the dispersion behaviors, phase differences between different SSPP modes can be realized, and the phase shift of each mode can be reconfigured. Experimental results show that the proposed SSPP phase shifter can achieve about 240°, 241°, and 340° phase shifts among the fundamental (even) modes, the high-order (odd) modes, and the fundamental and high-order modes, respectively, by switching the bias voltages applied to the varactors. Additionally, the proposed SSPP phase shifter has the capability to continuously tune the phase difference in each mode. Simulations and measurements have a good consistency. This work lays the groundwork for SSPP applications in large-scale integrated circuits and wireless communication systems and has a promising application in Terahertz circuits.
24 November 2025 in Nature Communications
Optical diffractive neural networks are emerging for improving speed and energy efficiency in machine learning. However, the challenges of nonlinear activation functions (e.g., latency issues, high power consumption, and cascading complexity) impede their performance and practical deployment. Here, we propose a programmable multilayer full-space nonlinear neural network operating in the microwave frequency band. Its nonlinear layers are constructed using programmable metasurfaces integrated with RF components, implementing a ReLU-like activation function. The nonlinear architecture achieves a nanosecond-scale delay (17.7 ns), representing orders of magnitude improvement in speed over photoelectric conversion-based nonlinearities. Moreover, the nonlinearity is characterized by exceedingly low thresholds and reconfigurable nonlinear activation functions. The system demonstrates remarkable classification capability in image classification and real-time human posture recognition tasks. Characterized by low latency, high speed, low power consumption, and flexible nonlinear activation, this architecture holds great promise for applications in security screening, medical rehabilitation, human-computer interaction, and numerous other fields.
29 October 2025 in Nature Electronics
29 April 2026 in Nature Electronics