Abstract
With rapid development of holography, metasurface-based holographic communication scheme shows great potential in development of adaptive electromagnetic function. However, conventional passive metasurfaces are severely limited by poor reconfigurability, which makes it difficult to achieve wavefront manipulations in real time. Here, we propose a holographic communication strategy that on-demand target information is firstly acquired and encoded via a depth camera integrated with modified YOLOv5s target detection algorithm, then transmitted by software defined radio modules with long term evolution at 5 GHz, and finally reproduced in the form of holographic images by spin-decoupled programmable coding metasurfaces at 12 GHz after decoding through modified Gerchberg–Saxton algorithm. Experiments are conducted to demonstrate the brand-new concept of optical information conversion to electromagnetic one via above intelligent scheme. Our strategy may open a novel avenue toward applications of near-field communication based on adaptive variation of electric field patterns (i.e. holographic images).
1 Introduction
With rapid development of holography, holographic communication has been expected to be used in development of adaptive electromagnetic (EM) functions. It is an intelligent application scheme for information acquisition, encoding, transmission, decoding, and reproduction [1], [2]. The system realized based on above strategy can automatically generate holographic images containing specific information. Furthermore, additional techniques like beamforming can be realized by aforesaid scheme, which shows advantages of good developability and intelligence. However, related application scenarios are complex and manifest weak confidentiality. Moreover, the hardware system based on array antenna is inefficient due to their complexity, high cost, and large volume. Therefore, more flexible hardware architecture, higher-efficiency information acquisition, and more advanced theory of algorithms are urgently needed to address above issues.
Metasurfaces, as a two-dimensional version of metamaterial [3], [4], [5], afford many degrees of freedom to arbitrarily manipulate EM wavefront in terms of amplitude, phase and polarization, achieving fascinating practical applications such as diffusive scattering [6], [7], [8], [9], [10], [11], ultra-thin cloaks [12], [13], [14], [15], [16], [17], beam deflection [18], [19], [20], [21], [22], orbital angular momentum generation [23], [24], [25], [26], etc. However, reconfigurable properties cannot be realized by traditional passive metasurfaces once geometrical structures are fixed [27], [28]. More recently, active digital coding metasurfaces were proposed to dynamically control a variety of functions by integrating active components. For example, anomalous deflections and RCS reduction were realized by adopting 1-bit coding sequences (“0” and “1” states) in a single meta-atom, where “0” and “1” represent two phase states of 0° and 180°, respectively [29], [30]. Furthermore, by incorporating external control mechanisms like field-programmable gate arrays (FPGA), the digital coding metasurface concept has been extended to its programmable counterpart which allows dynamic EM manipulation with greater intelligence and intricacy [31], [32], [33], [34], [35]. Moreover, such metasurfaces with dynamically tunable properties of EM waves were also referred to reconfigurable intelligent surfaces in communication research, which serve as essential technology for wireless channel optimization [36], [37], [38].
To further improve intelligent level of metasurfaces, programmable metasurfaces have been applied to some self-adaptive applications. For example, a tracking communication system based on computer vision is proposed to improve communication quality of moving targets [39]. Moreover, a neuro-metasurface system based on supervised-evolving learning algorithms is proposed to achieve adaptive focusing [40]. These works provide creative ideas for designing programmable metasurfaces by combining advanced sensors with intelligent algorithms [41], [42], focusing on realization of adaptive functions. However, it is also important for an intelligent system to reproduce and reuse EM information. Therefore, a new strategy of metasurface-based holographic communication is highly desirable to develop adaptive EM functions, but unfortunately remains elusive so far.
Here, we propose for the first time a new strategy to achieve holographic communication by synthesizing intelligent algorithms, long term evolution (LTE) and programmable coding metasurface (PCM). Such a technique provides an intelligent solution for developing adaptive functions to convert optical information into EM one by applying accurate and real-time control. Benefiting from two independent direct current (DC) biased voltage groups supplied to two integrated PIN diodes, a spin-decoupled PCM was designed to achieve independent control of the targeted reflection field in two orthogonal circular polarization (CP) channel by a self-made microcontroller unit (MCU) development board. Moreover, depth camera and software defined radio (SDR) modules were employed to acquire and transmit target information. Therein, modified You Only Look Once version 5 small (YOLOv5s) target detection algorithm and Gerchberg–Saxton (GS) algorithm were applied for encoding and decoding. For verification, experiments were carried out, showing that our designed holographic communication system is capable of reproducing self-adaptive holographic images in real time based on particular environment. More importantly, our work offers a solid platform, expanding metasurface applications in adaptive function development.
2 Results and discussion
2.1 Application scenario and schematic architecture
Here, we conceive an application scenario of our designed holographic communication system, which is composed of spin-decoupled PCM, SDR modules, and a depth camera. Figure 1 illustrates schematic diagram of corresponding application scenario. Target information in the form of images is firstly acquired by a depth camera (Intel RealSense D435i) under complicated environment, then scaled to [608, 608, 3] before encoding, and finally fed to YOLOv5s target detection algorithm for prediction of classification results. The preliminary encoding procedure takes around 0.07 s to identify targets. Therein, each image is firstly collected at a rate of 30 FPS (frames per second), and then processed in the neural network within about 0.03 s. Furthermore, aforementioned real-time procedure makes it possible to recognize moving objects well. To further complete encoding, we map classification results to preset gray-scale image patterns by optimizing the algorithm to not only detect targets but also extract desired encoded information. Here, we use cell phone, mouse, keyboard and scissors as target examples in experiments. With the help of SDR modules ADALM-PLUTO and LTE toolboxes in MATLAB commercial software, baseband reference measurement channel waveform carrying information will be modulated to hardware memory for continuous transmission in free space.
Application scenario and scheme architecture of our designed holographic communication system. Target information of selected objects is firstly automatically acquired and encoded by depth camera and modified YOLOv5s target detection algorithm, then transmitted and decoded by SDR modules and modified GS algorithm, and finally transformed into coding sequences required for spin-decoupled PCM to reproduce particular holographic image. Our proposed holographic communication scheme with polarization and phase reconfigurable feature is expected to develop adaptive EM functions, which can be further used as an NFC terminal.
After the received information is demodulated, modified GS algorithm is used to decode it into a coding matrix rather than original data for information transformation with good confidentiality. Then, decoded results are transformed into coding sequences for applying to spin-decoupled PCM which can realize phase modulation of EM wave in left-handed circular polarization (LCP) and right-handed circular polarization (RCP) channels by controlling left and right PIN diode states at different DC voltages. As can be seen from the equivalent circuits of PIN diodes at two states shown in middle part of right panel of Figure 1, the values of C, L, and R are 0.03 pF, 0.03 nH, and 7.8 Ω, respectively. Finally, holographic images are adaptively reproduced in particular CP channels according to variational target information. As a result, optical information conversion into EM one can be realized progressively.
Considering subsequent application of EM information, our proposed strategy for holographic communication is expected to develop adaptive EM functions, which can be further used as a near-field communication (NFC) terminal to establish point-to-point communication with other NFC terminals by continuously reproducing versatile EM fields in near field sensing region, realizing exciting applications including access control, mobile identity identification, and anti-counterfeiting, etc.
2.2 Encoding and decoding
As mentioned previously, depth cameras adopted in our work can realize acquisition of three-dimensional coordinates and achieve versatile functions. Here, it was used to encode acquired target information for simple verification. More specifically, sampled images were processed to obtain accurate classification results and mapped gray-scale image patterns by using modified YOLOv5s target detection algorithm for its strong network with appropriate depth and width. Therein, new CSP-Darknet53 network with strong feature extraction ability is introduced to realize high-precision detection with high speed, as shown in Figure 2. By loading COCO pre-trained weight and adopting NVIDIA GeForce RTX 4080 GPU, the time used to train the network was reduced to 5 h. Moreover, adaptive image scaling and mosaic data enhancement were used for preprocessing to adaptively scale images to [608, 608, 3] and expand data sets from 300 to 2000 with low GPU storage occupation, solving the issue of low average accuracy induced by different image sizes and uneven distribution of small targets in large data sets. In this regard, detection efficiency and integrity of image data can be improved significantly.
Structure diagrams of modified YOLOv5s and GS algorithm. Acquired target information is firstly fed into modified YOLOv5s target detection algorithm for classification result prediction and gray-scale image pattern mapping, then fed into modified GS algorithm for coding matrix evaluation, and finally transformed into coding sequences which are sent to MCU development board for real-time reproducing of automatic images.
To obtain feature maps with different dimensions [76, 76, 255], [38, 38, 255], and [19, 19, 255] for ultimate classification results, preprocessed images are firstly put into backbone (New CSP-Darknet53) network for feature extraction, then fed into neck (SPPF, New CSP-PAN) and head network for operation and prediction. Finally, a mapping relationship between classification results and preset gray-scale image patterns should be implemented to optimize YOLOv5s algorithm, more details are presented in Note S1 of Supporting Information.
Here, modified GS algorithm is applied for decoding based on Rayleigh–Sommerfeld diffraction theory to analyze required phase profiles, which can be expressed as [43]:
where U
h
(x
h
, y
h
) and U
0(x
0, y
0) stand for E-field distribution on holographic images plane and imaging plane, respectively. λ and k denote operating wavelength and wave number in free space, respectively.
To retrieve ultimate coding matrix, it is essential to discretize continuous Equation (1) and adopt binarization and iterative method. The detailed process is presented in Note S2 of Supporting Information. Finally, evaluated coding matrixes are transformed into form of sequences, and then sent to MCU development board for automatic reproduction of holographic images.
2.3 Information transmission scheme
The system based on multi-terminal collaboration can extend spatial dimension and enrich substantially application scenarios, fulfilling increasingly complex application requirements. Therefore, a stable scheme adopting two SDR modules ADALM-PLUTO is proposed here to achieve reliable information transmission between two terminals, see the hardware shown in Figure 3a and b. For experimental verification, cell phone targets were used as an experimental sample. Furthermore, LTE toolbox in MATLAB commercial software was applied to modulate preset gray-scale image patterns into transmitted radio frames, and then demodulate corresponding received data to original image patterns. Moreover, detailed procedure is illustrated below for great understanding.
SDR modules and related measured results. (a–b) SDR modules used for transmitting and receiving information. (c) Transmitted gray-scale image pattern. (d) Received gray-scale image pattern. (e) Received baseband LTE signal spectrum. (f) Constellation diagram of equalized physical downlink shared channel symbols.
By converting into binary data stream, gray-scale image patterns were packed into transport blocks of a downlink shared channel to generate baseband LTE signals, which were applied to an SDR module for continuous transmission at a center frequency of 5 GHz. After capturing abundant frames by another SDR module, offsets of captured signals were determined and corrected firstly. And then, processed information was synchronized to the start of LTE frames for demodulation to get resource grids. Finally, a channel estimation has been performed. Therein, downlink shared channel was analyzed to obtain raw data which were then recombined to form ultimate image patterns.
To reflect performance of our proposed information transmission scheme, transmitted and received image patterns are shown in Figure 3c and d. Besides, power spectral density of captured signal and constellation diagram of equalized physical downlink shared channel symbols are shown in Figure 3e and f. Therein, the spectrum shows that transmitted bandwidth of signal is approximately 9 MHz. Moreover, the constellation diagram shows that noise is randomly distributed around the ideal value in forms of cloud, indicating that the signal is slightly disturbed by a white noise. In addition, root mean square error of vector magnitude and bit error rate are 2.431 % and 0.001 % in measurement, demonstrating that the designed scheme is functioning well and image patterns have retained comparatively complete information throughout the transmission process.
2.4 Meta-atom design and EM property
To encrypt aforementioned information in different polarization channels for holographic communication applications, polarization-multiplexing metasurfaces should be used here. However, there are no related technology being used to achieve independent control of EM wave functions using AA (Aharonov–Anandan) phase used in our meta-atom design under two orthogonal linear polarization channels. Therefore, a CP metasurface was used in our work rather than a linear polarization counterpart. Here, AA phase is defined as the geometric phase accumulated in rotation path i.e. phase offsets in CP channels can be varied by changing length of meta-atom along rotation direction [44]. To construct a PCM working under two orthogonal CP incident waves based on AA phase, a triple-layer reflective element is designed with spin-decoupled phase control and high reflection coefficient, which provides stable reflection responses in orthogonal CP channels. As shown in Figure 4a, the designed programmable meta-atom is composed of four layers with diverse metallic patterns. Three F4B spacers (with a relative permittivity of 2.65 + 0.001i) separate the layers, thickness of which are h = 3 mm and h 1 = 0.2 mm, respectively. In addition, the top layer is an umbrella-type resonator, the middle layer is metallic ground and the bottom two are mutually orthogonal biasing lines. Then two PIN diodes (denoted as “1” and “2”) are loaded in two arcs of umbrella type, respectively. To simplify control circuit, two PIN diodes share same ground by a metallic via hole to connect negative poles of each diode, while positive poles are connected to different biasing lines by two metallic via holes without any connection with ground. As shown in Figure 4b, two PIN diodes are loaded in the position and vertical-symmetrical one of β = 45°, respectively, which is beneficial for generating 180° phase difference under two orthogonal CP channels between diodes states “ON” and “OFF”. In Figure 4c, two biasing lines with a gap of g = 0.1 mm are orthogonally printed in upper and lower layers, ensuring independent phase control of each element by applying DC bias voltages. Other parameters are given as follows: P = 10 mm, r = 4 mm, l = 3.4 mm, α = 135°, d 1 = 0.7 mm, w 1 = 1 mm, w 2 = 0.2 mm, and d 2 = 0.9 mm.
The schematic view of our designed meta-atom and its simulated co-polarized performances. (a) Perspective, (b) top, and (c) bottom view of the programmable meta-atom. Simulated co-polarized (d–f) reflection amplitudes and (e–g) reflection phases responses of the proposed meta-atom with four different PIN diode combination states under orthogonal CP wave incidence. (h) Simulated surface current distribution of the spin-decoupled meta-atom under orthogonal CP wave incidence.
To investigate reflection properties of our designed programmable meta-atom, we perform finite-difference-time-domain (FDTD) calculations in CST Microwave Studio. In Figure 4d and e, we give simulated reflection amplitude and phase under LCP wave excitation. Here, the overall working states of the meta-atom are denoted as “(diode 1, diode 2)” with each PIN diode 1, 2 ∈ [ON, OFF], i.e. only two possible states for each diode. It is obvious that reflection amplitude maintains over 0.9 and phase difference is 180° between states “ON” and “OFF” of PIN diode 1 at 12 GHz. Moreover, the calculated results indicate that diode 2 states are immune to LCP wave incidence, which guarantees independent phase control in LCP channel. Next, we explore EM response of meta-atom under RCP wave incidence when states of PIN diode 2 are switched under “ON” and “OFF”, see Figure 4f and g. Simulated results show that the programmable meta-atoms achieve a high reflection rate of more than 0.9. Meanwhile, phase difference of 180° is generated between states “ON” and “OFF” of diode 2 at 12 GHz. Similarly, there is little influence on EM properties when switching the states of diode 1 to “ON” or “OFF” state in RCP channel. Therefore, our proposed meta-atom indeed achieves 1-bit spin-decoupled phase control of CP wave. To further verify above independent control of two orthogonal CP channels, surface currents are simulated, as depicted in Figure 4h. Results show that the induced surface current flows along the left and right arc under LCP and RCP wave incidence, respectively.
2.5 Holographic communication system and experiments
In the following, we establish a system to experimentally validate the performance of our proposed holographic communication scheme. Therein, the spin-decoupled PCM fabricated based on printed circuit board (PCB) technology was composed of 21 × 21 meta-atoms, where 2 × 21 × 21 PIN diodes were embedded on metal surface, more details are shown in Note S3 of Supporting Information. Moreover, the DC bias voltages of each PIN diode were provided by a self-made MCU development board (detailed information is given in Note S4 of Supporting Information).
Experimental setups and PCM prototype details of the holographic communication system are presented in Figure 5a and b. Therein, three-dimensional coordinate system is provided and forward direction is defined as the orientation along −z. At the transmitter end, variational target information was firstly fed into modified YOLOv5s target detection algorithm for encoding to predict classification results and obtain mapped gray-scale image patterns, the prediction outcome during detection process is presented in Figure 5c. After encoding, LTE-based SDR modules connected with high-power horn antennas were employed for reliable directional transmission. At the receiver end, received information was decoded into coding matrix based on modified GS algorithm, visualized results of measurement were presented in Figure 5d. Moreover, corresponding simulated near-field electric field distributions of holographic images as shown in Figure 5e. And then, coding sequences transformed from matrix form were applied to PCM in the form of DC bias voltage by MCU development board, the detailed control process of DC bias voltage is presented in Note S5 of Supporting Information. Here, PCM was excited by a switchable CP horn from forward incidence, and intensity distribution of reflected wave was captured by a 6 mm-long monopole antenna placed at 120 mm away from x to y plane. Furthermore, the incident wave in free space was deducted from the total fields to obtain targeted reflection field. In all measurements, a CP horn and a monopole antenna were connected to two ports of an AV3672B vector network analyzer, respectively. To obtain local Ex and Ey field patterns, a monopole antenna was connected with a two-dimensional electronic step motor which could move automatically in a maximum area of 0.21 m × 0.21 m with a step resolution of 4.2 mm. Then, LCP and RCP components could be calculated by incorporating both measured information according to
Simulated and measured results of our designed holographic communication system and experiment setups. (a) Experimental setups at transmitter (left panel) and receiver (right panel) end. (b) Additional experimental setups at receiver end (left panel) and details on the PCM prototype (right panel). (c) Predicted classification results by modified YOLOv5s target detection algorithm. (d) Retrieved coding matrix by modified GS algorithm. (e) Simulated and (f) measured near-field electric field distributions of holographic images. Each column corresponds to the same target.
3 Conclusions
In this work, we propose a new strategy of holographic communication with self-adaptive capacity to reproduce versatile holographic images in real time based on particular environment. With depth cameras and SDR modules, variational target information is captured and transformed into coding matrix required for PCM based on modified YOLOv5s and GS algorithms. Moreover, a spin-decoupled PCM is designed to control targeted reflection field in two orthogonal CP channels by a self-made MCU development board. As a prototype, a holographic communication system is constructed to verify our design and experimental verifications have been carried out to prove that our proposed strategy paves a new way for continuous reproduction of versatile electric field patterns (i.e. holographic images), which can be further extended in application of NFC. In future, three-dimensional coordinates of targets acquired by depth cameras will be utilized to achieve broader application scenarios with abundant information. In addition, more advanced NFC terminals using our designed system can be designed for establishing point-to-point communication.
Supporting Information
Mapping relationship between classification results and preset gray-scale image patterns; Decoding using modified GS algorithm; Additional information for programmable coding metasurface; Design of self-made MCU development board; Detailed control process of DC bias voltage.
Funding source: National Natural Science Foundation of China
Award Identifier / Grant number: 62171459
Funding source: National Defense Foundation for Distinguished Young Scholars
Award Identifier / Grant number: 2022-JCJQ-ZQ-006
Funding source: Youth Talent Lifting Project of the China Association for Science and Technology
Award Identifier / Grant number: 2021-JCJQ-QT-018
-
Research funding: This work was supported by the National Natural Science Foundation of China (62171459); National Defense Foundation for Distinguished Young Scholars, grant number 2022-JCJQ-ZQ-006. This work was supported in part by the Youth Talent Lifting Project of the China Association for Science and Technology No. 2021-JCJQ-QT-018.
-
Author contributions: All authors have accepted responsibility for the entire content of this manuscript and approved its submission.
-
Conflict of interest: Authors state no conflicts of interest.
-
Data availability: Data sharing is not applicable to this article as no datasets were generated or analyzed during the current study.
References
[1] Y. Huang, Y. Zhu, X. Qiao, X. Su, S. Dustdar, and P. Zhang, “Toward holographic video communications: a promising AI-driven solution,” IEEE Commun. Mag., vol. 60, no. 11, pp. 82–88, 2022. https://doi.org/10.1109/mcom.001.220021.Search in Google Scholar
[2] A. E. Essaili, et al.., “Holographic communication in 5G networks,” Ericsson Technol. Rev., vol. 2022, no. 5, pp. 2–11, 2022. https://doi.org/10.23919/etr.2022.9904689.Search in Google Scholar
[3] H.-X. Xu, G. Hu, X. Kong, Y. Shao, P. Genevet, and C.-W. Qiu, “Super-reflector enabled by non-interleaved spin-momentum-multiplexed metasurface,” Light Sci. Appl., vol. 12, no. 1, p. 78, 2023. https://doi.org/10.1038/s41377-023-01118-1.Search in Google Scholar PubMed PubMed Central
[4] H.-T. Chen, A. J. Taylor, and N. Yu, “A review of metasurfaces: physics and applications,” Rep. Prog. Phys., vol. 79, no. 7, p. 076401, 2016. https://doi.org/10.1088/0034-4885/79/7/076401.Search in Google Scholar PubMed
[5] H.-H. Hsiao, C. H. Chu, and D. P. Tsai, “Fundamentals and applications of metasurfaces,” Small Methods, vol. 1, no. 4, p. 1600064, 2017. https://doi.org/10.1002/smtd.201600064.Search in Google Scholar
[6] H.-X. Xu, et al.., “Wavenumber-splitting metasurfaces achieve multichannel diffusive invisibility,” Adv. Opt. Mater., vol. 6, no. 10, p. 1800010, 2018. https://doi.org/10.1002/adom.201800010.Search in Google Scholar
[7] G. Wu, W. Yu, T. Lin, Y. Deng, and J. Liu, “Ultra-Wideband RCS reduction based on non-planar coding diffusive metasurface,” Materials, vol. 13, no. 21, p. 4773, 2020. https://doi.org/10.3390/ma13214773.Search in Google Scholar PubMed PubMed Central
[8] W. Ying-Jie, C. Ke, Z. Fang-Kun, T. Jiang-Xiao, F. Yu-Feng, and C. Ping, “Resonance phase and geometric phase integrated diffusion metasurface for broadband scattering control,” J. Phys. D Appl. Phys., vol. 54, no. 16, p. 165101, 2021. https://doi.org/10.1088/1361-6463/abda82.Search in Google Scholar
[9] K. Chen, et al.., “Geometric phase coded metasurface: from polarization dependent directive electromagnetic wave scattering to diffusion-like scattering,” Sci. Rep., vol. 6, no. 1, p. 35968, 2016. https://doi.org/10.1038/srep35968.Search in Google Scholar PubMed PubMed Central
[10] H.-X. Xu, et al.., “Deterministic approach to achieve broadband polarization-independent diffusive scatterings based on metasurfaces,” ACS Photonics, vol. 5, no. 5, pp. 1691–1702, 2018. https://doi.org/10.1021/acsphotonics.7b01036.Search in Google Scholar
[11] K. Chen, W. Guo, G. Ding, J. Zhao, T. Jiang, and Y. Feng, “Binary geometric phase metasurface for ultra-wideband microwave diffuse scatterings with optical transparency,” Opt. Express, vol. 28, no. 9, pp. 12638–12649, 2020. https://doi.org/10.1364/oe.392182.Search in Google Scholar
[12] P. Salami and L. Yousefi, “Wide-band polarisation-independent metasurface-based carpet cloak,” IET Microw., Antennas Propag., vol. 14, no. 15, pp. 1983–1989, 2020. https://doi.org/10.1049/iet-map.2020.0396.Search in Google Scholar
[13] Z. Zhen, et al.., “Realizing transmitted metasurface cloak by a tandem neural network,” Photon. Res., vol. 9, no. 5, pp. B229–B235, 2021. https://doi.org/10.1364/prj.418445.Search in Google Scholar
[14] H.-X. Xu, et al.., “Polarization-insensitive 3D conformal-skin metasurface cloak,” Light Sci. Appl., vol. 10, no. 1, p. 75, 2021. https://doi.org/10.1038/s41377-021-00507-8.Search in Google Scholar PubMed PubMed Central
[15] C. Wang, et al.., “Multi-frequency metasurface carpet cloaks,” Opt. Express, vol. 26, no. 11, pp. 14123–14131, 2018. https://doi.org/10.1364/oe.26.014123.Search in Google Scholar
[16] J. Yang, C. Huang, X. Wu, B. Sun, and X. Luo, “Dual-wavelength carpet cloak using ultrathin metasurface,” Adv. Opt. Mater., vol. 6, no. 14, p. 1800073, 2018. https://doi.org/10.1002/adom.201800073.Search in Google Scholar
[17] X. Wang, D. Mao, and Y. Li, “Broadband acoustic skin cloak based on spiral metasurfaces,” Sci. Rep., vol. 7, no. 1, p. 11604, 2017. https://doi.org/10.1038/s41598-017-11846-1.Search in Google Scholar PubMed PubMed Central
[18] F. Yu, et al.., “Reconfigurable metasurface with tunable and achromatic beam deflections,” Opt. Mater. Express, vol. 12, no. 1, pp. 49–58, 2022. https://doi.org/10.1364/ome.445151.Search in Google Scholar
[19] X. Yang, T. Chen, P. Lu, Y. Zou, W. Li, and Q. Liu, “Phase-shifted metasurface design for pseudo-nondiffractive beam deflection,” IET Microw., Antennas Propag., vol. 16, no. 4, pp. 240–247, 2022. https://doi.org/10.1049/mia2.12235.Search in Google Scholar
[20] C. Wang, et al.., “Heterogeneous Amplitude−Phase metasurface for distinct wavefront manipulation,” Adv. Photon. Res., vol. 2, no. 10, p. 2100102, 2021. https://doi.org/10.1002/adpr.202100102.Search in Google Scholar
[21] C. Wang, et al.., “Full-space spin-decoupled versatile wavefront manipulations using non-interleaved metasurface,” Adv. Opt. Mater., vol. 12, no. 15, pp. 3149–3158, 2023. https://doi.org/10.1515/nanoph-2023-0171.Search in Google Scholar
[22] M. R. M. Hashemi, S.-H. Yang, T. Wang, N. Sepúlveda, and M. Jarrahi, “Electronically-Controlled beam-steering through vanadium dioxide metasurfaces,” Sci. Rep., vol. 6, no. 1, p. 35439, 2016. https://doi.org/10.1038/srep35439.Search in Google Scholar PubMed PubMed Central
[23] F. Qin, L. Wan, L. Li, H. Zhang, G. Wei, and S. Gao, “A transmission metasurface for generating OAM beams,” IEEE Antenn. Wireless Propag., vol. 17, no. 10, pp. 1793–1796, 2018. https://doi.org/10.1109/lawp.2018.2867045.Search in Google Scholar
[24] Y. Li, et al.., “Orbital angular momentum multiplexing and demultiplexing by a single metasurface,” Adv. Opt. Mater., vol. 5, no. 2, p. 1600502, 2017. https://doi.org/10.1002/adom.201600502.Search in Google Scholar
[25] H. H. Lv, Q. L. Huang, X. J. Yi, J. Q. Hou, and X. W. Shi, “Low-profile transmitting metasurface using single dielectric substrate for OAM generation,” IEEE Antenn. Wireless Propag., vol. 19, no. 5, pp. 881–885, 2020. https://doi.org/10.1109/lawp.2020.2983400.Search in Google Scholar
[26] C. Ji, J. Song, C. Huang, X. Wu, and X. Luo, “Dual-band vortex beam generation with different OAM modes using single-layer metasurface,” Opt. Express, vol. 27, no. 1, pp. 34–44, 2019. https://doi.org/10.1364/oe.27.000034.Search in Google Scholar PubMed
[27] L. Li, H. Zhao, C. Liu, L. Li, and T. J. Cui, “Intelligent metasurfaces: control, communication and computing,” eLight, vol. 2, no. 1, p. 7, 2022. https://doi.org/10.1186/s43593-022-00013-3.Search in Google Scholar
[28] Q. Ma, et al.., “Information metasurfaces and intelligent metasurfaces,” Photonics Insights, vol. 1, no. 1, p. R01, 2022. https://doi.org/10.3788/pi.2022.r01.Search in Google Scholar
[29] M. Feng, et al.., “Two-dimensional coding phase gradient metasurface for RCS reduction,” J. Phys. D Appl. Phys., vol. 51, no. 37, p. 375103, 2018. https://doi.org/10.1088/1361-6463/aad5ad.Search in Google Scholar
[30] X. Han, et al.., “Multiple diffuse coding metasurface of independent polarization for RCS reduction,” IEEE Access, vol. 8, pp. 162313–162321, 2020, https://doi.org/10.1109/access.2020.3021650.Search in Google Scholar
[31] W. Tang, et al.., “Wireless communications with programmable metasurface: new paradigms, opportunities, and challenges on transceiver design,” IEEE Wireless Commun., vol. 27, no. 2, pp. 180–187, 2020. https://doi.org/10.1109/mwc.001.1900308.Search in Google Scholar
[32] C. Wang, H.-X. Xu, Y. Wang, G. Hu, H. Luo, and K. Wang, “Reconfigurable transmissive metasurface synergizing dynamic and geometric phase for versatile polarization and wavefront manipulations,” Mater. Des., vol. 225, p. 111445, 2023, https://doi.org/10.1016/j.matdes.2022.111445.Search in Google Scholar
[33] Y.-L. Sun, X.-G. Zhang, Q. Yu, W.-X. Jiang, and T.-J. Cui, “Infrared-controlled programmable metasurface,” Sci. Bull., vol. 65, no. 11, pp. 883–888, 2020. https://doi.org/10.1016/j.scib.2020.03.016.Search in Google Scholar PubMed
[34] L. Li, et al.., “Electromagnetic reprogrammable coding-metasurface holograms,” Nat. Commun., vol. 8, no. 1, p. 197, 2017. https://doi.org/10.1038/s41467-017-00164-9.Search in Google Scholar PubMed PubMed Central
[35] T. Shan, X. Pan, M. Li, S. Xu, and F. Yang, “Coding programmable metasurfaces based on deep learning techniques,” IEEE J. Emerg. Sel. Top. Circuits Syst., vol. 10, no. 1, pp. 114–125, 2020. https://doi.org/10.1109/jetcas.2020.2972764.Search in Google Scholar
[36] Y. Feng, Q. Hu, K. Qu, W. Yang, Y. Zheng, and K. Chen, “Reconfigurable intelligent surfaces: design, implementation, and practical demonstration,” Electromagn. Sci., vol. 1, no. 2, pp. 1–21, 2023. https://doi.org/10.23919/emsci.2022.0011.Search in Google Scholar
[37] Q. Hu, et al.., “An intelligent programmable omni-metasurface,” Laser Photon. Rev., vol. 16, no. 6, p. 2100718, 2022. https://doi.org/10.1002/lpor.202100718.Search in Google Scholar
[38] J. You, et al.., “Electromagnetic metamaterials: from classical to quantum,” Electromagn. Sci., vol. 1, no. 1, pp. 1–33, 2023. https://doi.org/10.23919/emsci.2022.0005.Search in Google Scholar
[39] W. Li, et al.., “Intelligent metasurface system for automatic tracking of moving targets and wireless communications based on computer vision,” Nat. Commun., vol. 14, no. 1, p. 989, 2023. https://doi.org/10.1038/s41467-023-36645-3.Search in Google Scholar PubMed PubMed Central
[40] H. Lu, et al.., “Eye accommodation-inspired neuro-metasurface focusing,” Nat. Commun., vol. 14, no. 1, p. 3301, 2023. https://doi.org/10.1038/s41467-023-39070-8.Search in Google Scholar PubMed PubMed Central
[41] S. V. Mahadevkar, et al.., “A review on machine learning styles in computer vision—techniques and future directions,” IEEE Access, vol. 10, pp. 107293–107329, 2022, https://doi.org/10.1109/access.2022.3209825.Search in Google Scholar
[42] M.-H. Guo, et al.., “Attention mechanisms in computer vision: a survey,” Comput. Vis. Media, vol. 8, no. 3, pp. 331–368, 2022. https://doi.org/10.1007/s41095-022-0271-y.Search in Google Scholar
[43] C. Wang, H.-X. Xu, T. Liu, and F. Zhang, “Hybrid-phase assisted amplitude and phase control using full-space metasurface,” Adv. Opt. Mater., p. 2302153, 2023, https://doi.org/10.1002/adom.202302153.Search in Google Scholar
[44] G. D. Bai, et al.., “Spin-symmetry breaking through metasurface geometric phases,” Phys. Rev. Appl., vol. 12, no. 4, p. 044042, 2019. https://doi.org/10.1103/physrevapplied.12.044042.Search in Google Scholar
[45] X. Ni, A. V. Kildishev, and V. M. Shalaev, “Metasurface holograms for visible light,” Nat. Commun., vol. 4, no. 1, p. 2807, 2013. https://doi.org/10.1038/ncomms3807.Search in Google Scholar
Supplementary Material
This article contains supplementary material (https://doi.org/10.1515/nanoph-2023-0925).
© 2024 the author(s), published by De Gruyter, Berlin/Boston
This work is licensed under the Creative Commons Attribution 4.0 International License.
