Artificial intelligence is facilitating human life in many aspects. Previous artificial intelligence has been mainly focused on computer algorithms (e.g. deep-learning and extremelearning) and integrated circuits. Recently, all-optical diffractive deep neural networks (D 2 NN) were realized by using passive structures, which can perform complicated functions designed by computer-based neural networks at the light speed. However, once a passive D 2 NN architecture is fabricated, its function will be fixed. Here, we propose a programmable artificial intelligence machine (PAIM) that can execute various intellectual tasks by realizing hierarchical connections of brain neurons via a multi-layer digital-coding metasurface array. Integrated with two amplifier chips in each meta-atom, its transmission coefficient covers a dynamic range of 35 dB (from -40 dB to -5 dB), which is the basis to construct the reprogrammable physical layers of D 2 NN, in which the digital meta-atoms make the artificial neurons alive. We experimentally show that PAIM can handle various deep-learning tasks for wave sensing, including image classifications, mobile communication coder-decoder, and real-time multi-beam focusing. In particular, we propose a reinforcement learning algorithm for on-site learning and discrete optimization algorithm for digital coding, making PAIM have autonomous intelligence ability and perform self-learning tasks without the support of extra computer.
Intelligent coding metasurface is a kind of information-carrying metasurface that can manipulate electromagnetic waves and associate digital information simultaneously in a smart way. One of its widely explored applications is to develop advanced schemes of dynamic holographic imaging. By now, the controlling coding sequences of the metasurface are usually designed by performing iterative approaches, including the Gerchberg–Saxton (GS) algorithm and stochastic optimization algorithm, which set a large barrier on the deployment of the intelligent coding metasurface in many practical scenarios with strong demands on high efficiency and capability. Here, we propose an efficient non-iterative algorithm for designing intelligent coding metasurface holograms in the context of unsupervised conditional generative adversarial networks (cGANs), which is referred to as physics-driven variational auto-encoder (VAE) cGAN (VAE-cGAN). Sharply different from the conventional cGAN with a harsh requirement on a large amount of manual-marked training data, the proposed VAE-cGAN behaves in a physics-driving way and thus can fundamentally remove the difficulties in the conventional cGAN. Specifically, the physical operation mechanism between the electric-field distribution and metasurface is introduced to model the VAE decoding module of the developed VAE-cGAN. Selected simulation and experimental results have been provided to demonstrate the state-of-the-art reliability and high efficiency of our VAE-cGAN. It could be faithfully expected that smart holograms could be developed by deploying our VAE-cGAN on neural network chips, finding more valuable applications in communication, microscopy, and so on.
The computational time to evaluate the physical optics (PO) expression by numerical integration increases rapidly with the increase of electrical size of scattering surfaces. However, the computational time of PO integrals for electrically large object can be greatly reduced by using the stationary phase method, which is independent of the wavenumber. For this method, the theory and numerical implementations for isolated critical points have been well developed. However, for cases of nearby critical points, there are still a few issues to be considered, especially, in numerical implementations. In this paper, we mainly study the numerical implementations for several most common cases of nearby critical points. In particular, the cases of two nearby inner stationary phase points and complex inner stationary phase points are discussed in more details. Such cases occur frequently when the scattering surface includes convex-concave parts, but the numerical implementations to such cases have not been reported to our knowledge. The difficulty lies in how to identify whether two inner stationary points and complex inner stationary points on the surfaces with arbitrary shapes are close to each other or not. A strategy is designed to solve this difficulty. By validation in some typical examples, we find that the stationary phase method is robust enough to evaluate the PO integrals accurately. Finally, some interesting phenomena observed in numerical validations are interpreted.Index Terms-Boundary stationary point, complex boundary stationary point, complex inner stationary point, corner point, critical point, inner stationary point, uniform stationary phase method.
Recent years, reconfigurable metasurfaces become research hotspots since they can manipulate electromagnetic (EM) waves dynamically and flexibly. Achieving cost-effective and energy-efficient wireless communications via reconfigurable intelligent surfaces (RISs) are attractive in various application sceneries. In this work, a +45°-polarized dual-band reconfigurable metasurface with beam steering ability is proposed to combat with signal attenuation for wireless communications. Through varying the bias voltages of the integrated varactors in the specially designed meta-atom, 0 and π phase responses have been excited around both 2.4 GHz and 5.8 GHz. Taking advantage of this unique feature, we proposed a 1-bit coding RIS device which provides dual-band application. To validate the proposed concepts, a reconfigurable metasurface composed of 16×16 elements was fabricated and measured. The measurement results agree well with simulation, which promise the proposed prototype a good candidate for future intelligent communications.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.