We establish an adaptive batch-normalized neural network, aiming to implement smart and rapid inverse design for graphene-based metamaterials with on-demand optical responses.

No abstract

VII. CONCLUSIONReflective diffractive gratings have numerous applications in optics and in electromagnetism, but their thickness can be a problem for applications at VHF frequencies or lower. We present a solution to this problem in the form of a thin diffractive panel that comprises resonant elements. A particular implementation was designed and tested for applications to prevent building perturbations to aircraft ILS. The screen measured only 1/34th of a wavelength in depth (1/15th of a wavelength with its protective weather screen). Measurements for an angle of incidence from 50 • to 55.5 • with respect to the plate normal displayed a 91% reduction in scattered field in dry conditions and a 90% reduction in simulated extreme weather conditions.The technology was developed with a particular application in mind; therefore, the main driver was whether it could be implemented simply and cheaply for applications in the VHF band. We did not pursue designs with thicknesses less than 1/34th of a wavelength as these were found to have a reduced bandwidth due to the reduced inductance. The tradeoff between thickness and bandwidth should be studied on a case-by-case basis for other applications. A video of the field trials is available on the Airbus website [12].Abstract-A new single-layer linear-to-circular polarizer is presented in this communication. A novel hybrid meander line and loop configuration is employed to transform linearly polarized field to circularly polarized field over a wide frequency band. Numerical simulations indicate that the proposed polarizer is robust under oblique and deflected illuminations. A prototype of the proposed polarizer is designed and fabricated. Measured results show that the band of axial ratio less than 3 dB ranges from 18 to 29 GHz with about 3-dB insertion loss.

Intelligent transportation systems (ITSs) of industrial systems have played an important role in Internet of things (IOT). The assistant calibration system (ACS) of vehicles is an emerging technology, which services the driver to drive the vehicle safely. To solve some existing problems in ACS such as frequency pairing, vehicle localization judgment, and driving in the curve road, two direction-of-arrival (DOA) estimation-based approaches are proposed to resolve these problems. However, the performance of most conventional DOA estimation algorithms is affected by the mutual coupling among the elements. The special structure of the mutual coupling matrix of the uniform linear array is applied to eliminate the effect of mutual coupling. Then, a novel on-grid DOA estimation algorithm based on compressive sensing (CS) strategies is proposed in the presence of unknown mutual coupling. In order to compensate the aperture loss of discarding information that the array receives, the array aperture is extended by the vectorization operator. In order to deal with the effect of grid mismatch, an offgrid DOA estimation algorithm based on sparse Bayesian learning (SBL) is proposed in this paper. The temporal correlation between the neighboring snapshot numbers is considered in the off-grid algorithm. The computer simulation verifies the effectiveness of the proposed algorithms.Index Terms-Compressive sensing (CS), direction-of-arrival (DOA) estimation, intelligent transportation systems (ITSs), sparse Bayesian learning (SBL).

Enhancing the light–matter interaction of graphene is an important issue for related photonic devices and applications. In view of its potential ultraviolet applications, we aim to achieve extremely high ultraviolet absorption in graphene without any nanostructure or microstructure patterning. By manipulating the polarization and angle of incident light, the ultraviolet power can be sufficiently coupled to the optical dissipation of graphene based on single-channel coherent perfect absorption in an optimized multilayered thin film structure. The ultraviolet absorbance ratios of single and four atomic graphene layers are enhanced up to 71.4% and 92.2%, respectively. Our research provides a simple and efficient scheme to trap ultraviolet light for developing promising photonic and optoelectronic devices based on graphene and potentially other 2D materials.