A breadboard of a three-layer printed reflectarray for dual polarization with a different coverage in each polarization has been designed, manufactured, and tested. The reflectarray consists of three layers of rectangular patch arrays separated by a honeycomb and backed by a ground plane. The beam shaping for each polarization is achieved by adjusting the phase of the reflection coefficient at each reflective element independently for each linear polarization. The phase shift for each polarization is controlled by varying either the x or y patch dimensions. The dimensions of the rectangular patches are optimized to achieve the required phase shift for each beam at central and extreme frequencies in the working band. The reflectarray has been designed to produce a contoured beam for a European coverage in H-polarization in a 10% bandwidth, and a pencil beam to illuminate the East Coast in North America in V-polarization. The measured radiation patterns show that gain requirements are practically fulfilled in a 10% bandwidth for both coverages, and the electrical performances of the breadboard are close to those of a classical dual gridded reflector

A 1.2-meter reflectarray antenna has been designed to accomplish the requirements of a Direct Broadcast Satellite (DBS) mission, which provides a South America transmit-receive coverage in Ku band. The antenna has been designed by applying first of all a pattern synthesis technique to obtain the required phase distribution on the reflectarray at several frequencies in transmit and receive bands. Then, the patch dimensions have been optimized in a configuration made of three stacked layers of varying-sized patches in order to provide the required phase distribution at transmit and receive frequencies. An antenna demonstrator has been manufactured and tested. The measured patterns are in good agreement with the simulations and they are close to fulfill the coverage requirements in both transmit and receive bands.

A two-layer reflectarray is proposed as a central station antenna for a local multipoint distribution system (LMDS) in the 24.5-26.5 GHz band. The antenna produces three independent beams in an alternate linear polarization that are shaped both in azimuth (sectored) and in elevation (squared cosecant). The design process is divided into several stages. First, the positions of the three feeds are established as well as the antenna geometry to produce the three beams in the required directions. Second, the phase distribution on the reflectarray surface, which produces the required beam shaping, is synthesized. Third, the sizes of the printed stacked patches are adjusted so that the phase-shift introduced by them matches the synthesized phase distribution. Finally, the radiation patterns are computed for the central and lateral beams, showing a shaping close to the requirements. A breadboard has been manufactured and measured in an anechoic chamber, showing a good behavior, which validates the designing methodology.

A machine learning technique is applied to the design and optimization of reflectarray antennas to considerably accelerate computing time without compromising accuracy. In particular, Support Vector Machines (SVMs), automatic learning structures that are able to deal with regression problems, are employed to obtain surrogate models of the reflectarray element to substitute the full-wave analysis tool for the characterization of the unit cell in the design and optimization algorithms. The analysis, design and optimization of a very large reflectarray antenna for Direct Broadcast Satellite applications are accelerated up to three orders of magnitude. This is here demonstrated with three examples: one showing the design of a reflectarray; and two for the crosspolar optimization, one with one coverage for each linear polarization (Europe and the Middle East) and another with a Middle East coverage working in dual-linear polarization. The accuracy of the proposed approach is validated by means of a comparison of the final designs with full-wave simulations based on local periodicity obtaining good agreement. The crosspolar dicrimination and crosspolar isolation are greatly improved using the SVMs while considerably reducing computing time.

In recent years, artificial intelligence (AI) techniques have been developed rapidly. With the help of big data, massive parallel computing, and optimization algorithms, machine learning (ML) and (more recently) deep learning (DL) strategies have been equipped with enhanced learning and generalization capabilities. Besides becoming an essential framework in image and speech signal processing, AI has also been widely applied to solve several electromagnetic (EM) problems with unprecedented computational efficiency, including inverse scattering (IS) and EM imaging. In this article, a review of the most recent progresses in the application of ML and DL for such problems is given. We humbly hope a brief summary could help us better understand the pros and cons of this research topic and foster future research in using AI to address paramount challenges in the field of EM vision.

A new approach to designing frequency selective surface (FSS)-based linear to circular polarization (LP-CP) converters is presented. It is based on the use of FSSs which exhibit dual diagonal symmetry, and a novel four-port equivalent circuit able to describe the electrical behavior of the cells for the two linear incident polarizations at the same time. The equivalent circuit allows the use of standardized branch line coupler theory to design LP-CP converters comprising a cascade of an arbitrary number of layers, whose synthesis includes the phase and makes it possible to achieve prescribed electrical conditions systematically. A full design procedure has been developed using the new approach and several designs in both transmission and reflection modes are presented and evaluated. It has been proven that single layer reflective converters exhibit large bandwidths as the two reflected field components are in quadrature independently of the frequency. One of these devices was designed and showed an axial ratio <;0.2 dB within the band from 21.5 to 28.5 GHz. The reflective LP-CP converters designed was also manufactured and tested, and the measurements were used to validate the design procedure.