Localized surface plasmons (LSPs) are localized oscillations of free electrons in metal nanoparticles at optical frequencies. Confined mode profiles and near‐field enhancements make LSPs ultrasensitive to the dielectric environment, making them good candidates as sensors. The concept and applications have been generalized to spoof LSPs in microwave and terahertz frequencies, via plasmonic metamaterials composed of subwavelength corrugations. Herein, the basic physics, sensing prototypes, detection schemes, and state‐of‐the‐art progress are broadly reviewed from optical LSP sensing to microwave spoof LSP sensing. While optical LSPs exhibit localized sensitivity enhancement with high attenuation, spoof LSPs in microwave and terahertz frequencies combine the characteristics of deep‐subwavelength confinement and sensitivity enhancement of optical LSPs with low loss, high quality factor, multipole modes, and on‐chip detection of optical microcavities. Meanwhile, advances in printed circuits, integrated circuits, wireless communications, wearable devices, and Internet of things have endowed microwave sensing with a solid technical foundation and promising prospects. Applications in liquid sensing, gas sensing, and wearable sensing are demonstrated. Discussions are extended to electromagnetic sensing throughout the wave spectra, with concerns about key supporting technologies. The prospect of microwave sensing is emphatically investigated, specifically on leveraging the advantages of plasmonic enhancement.

Driven by the miniaturization of integrated electronics, research on spoof plasmonic circuits has recently aroused widespread interest. On the other hand, nonreciprocal devices, such as isolators and circulators, are key components of integrated electronic systems. However, bulky magnets required to realize isolation and circulation prevent the application of traditional nonreciprocal technologies to integrated systems. Here, parametric amplification is explored to achieve magnetic-free plasmonic isolation, and an ultrathin reconfigurable spoof plasmonic isolator is realized experimentally. In this isolation system, the forward signal amplified by a spoof plasmonic parametric amplifier is coupled to a second linear plasmonic waveguide via a spoof localized surface plasmon resonator, whereas the transmission from the inverse direction is prohibited, giving rise to a measured isolation ratio of up to 20 dB. By tuning the nonlinear phase-matching condition through external bias voltage, multifrequency isolation of spoof surface plasmon polariton (SSPP) signals is also realized experimentally. This work demonstrates the possibility of producing miniaturized and low-cost nonreciprocal SSPP devices, holding great promise for applications in nonmagnetic information processing and radar detection.

Recently, some new forms of arti cial intelligence computing hardware and chips have been presented. However, most of them have di culties to simultaneously achieve advantages of light-speed computing, programmable weight matrix, and programmable nonlinear activation functions. Here, we propose a programmable surface plasmonic neural network (SPNN) with programmable weights and activation functions based on a spoof surface plasmon polariton (SSPP) platform, which can perform intelligent functions and sense electromagnetic (EM) waves at nearly light speed. We demonstrate a parallel coupling SSPP structure loaded with varactors to introduce four paths with tunable transmitting parameters. On this multi-port architecture, we further establish a real-time control and feedback method to enable arbitrarily designable activation functions under a detecting feedback loop. Experimental results show that a four-in and four-out fully-connected super-neuron can ful ll independently adjustable weights and programmable activation functions, where each input can be sensed for arbitrarily programming. To comprehensively show the above capabilities, we design and demonstrate experimentally a wireless communication system based on the SPNN for image decoding and recovery.We further illustrate a partially connected SPNN using the super-neurons with a high prediction accuracy. The proposed concept paves a new way for arti cial intelligence devices, stimulating the fascinating elds like large-scale EM computing and communication systems in the future.

Magnetic‐Free Plasmonic Isolation A reconfigurable non‐magnetic isolator based on parametric amplification is demonstrated experimentally on an ultrathin planar spoof plasmonic platform in article number 2100578. Xinxin Gao, Jingjing Zhang, and colleagues envision this work can establish a foundation for developing magnetic‐free non‐reciprocal devices suitable for on‐chip integration with microwave circuits, with high potential to miniaturize existing communication and radar systems.

We propose an ultrathin reconfigurable Mach–Zehnder interferometer (MZI) for realizing dynamic frequency and amplitude modulations of spoof surface plasmon (SSP) signal. Active varactor diodes are integrated in the SSP unit cells on one of the MZI arms to introduce asymmetry to the MZI structure, which can control the interference patterns by varying bias voltages applied on the varactor diodes. We show that the spectral positions of multiple sharp interference dips are very sensitive to the change of diode capacitance, thereby allowing for good frequency modulation. We also demonstrate continuous amplitude modulation by tuning the varactor diodes at multiple selected frequencies. To verify the reconfigurable feature of the proposed SSP MZI, the frequency shift keying (FSK) and amplitude modulations have been experimentally demonstrated on the same structure. The modulation depth of the amplitude modulation can be further improved by designing geometrical parameters of the SSP structure, reaching a significant amplitude change from 0.88 to 0.05 in experiments.