The present work proposes a novel design of a dual-band-printed antenna for operation at the millimeter-wave frequencies 28 and 38 GHz that are utilized for the modern and future generations of mobile communications. The antenna is composed of two radiating elements. The first element is the main patch that is fed through a microstrip line with inset feed, and the second element is a parasitic element that is fed through capacitive coupling with the main patch. The design parameters of the proposed antenna are optimized through a complete parametric study to give excellent impedance matching at 28 GHz over the band 27.7–28.3 GHz and at 38 GHz over the band 37.7–38.3 GHz. The surface current distributions at the two operational frequencies are investigated. The designed antenna is used to construct a four-port efficient multi–input–multi–output (MIMO) system. The MIMO system performance is investigated regarding the envelope correlation coefficient (ECC), diversity gain (DG), and the channel capacity loss (CCL) showing very good performance. The single-element antenna and the MIMO are fabricated and experimentally evaluated showing excellent impedance matching over the lower and higher frequency bands, which come in agreement with the simulation results. It is shown that the antenna produces maximum gain of 7.4 and 8.1 dBi at 28 and 38 GHz, respectively. The average radiation efficiencies of the proposed antenna are 88% and 88.8% over the lower and higher frequency bands, respectively. In addition, the coupling coefficients between the MIMO antenna systems are measured experimentally showing very low coupling values resulting in an efficient MIMO system that is suitable for future millimeter-wave (mm-wave) applications.

A novel design for Vivaldi antenna is introduced to operate in more than one frequency band in the millimeter-wave spectrum range with high gain. The proposed Vivaldi antenna is planar and printed on both sides of Rogers 3003C flexible substrate. A parasitic annular ring is added to each arm of the Vivaldi antenna to adjust the impedance matching and maximize the gain. The dimensions of the antenna are optimized using the CST microwave simulator to obtain the best impedance matching bandwidths and highest gain in a compact-size antenna. The 28-GHz band is then selected due to its importance to enhance the gain and improve the operating bandwidth for the 5G applications. A gain of 10 dBi and bandwidth of 6.1 GHz are obtained at 28 GHz. A multiple-input multiple-output (MIMO) antenna system is then designed for pattern diversity for the fifth-generation mobile applications. The performance of the MIMO system is evaluated regarding the envelope correlation coefficient and the diversity gain showing good results. The optimized antenna is fabricated and its return loss and radiation patterns are measured showing good consent between the simulations and the practical work. The MIMO antenna system is also fabricated, and its performance is measured regarding the reflection and coupling coefficients showing good agreement between the measurements and simulations.

A circularly-polarized (CP) printed antenna is proposed for millimeter-wave mobile communications. The antenna has a defected ground structure (DGS), and it operates at 28 GHz. The geometries of the antenna on the upper surface and the DGS are symmetric about a $$45^\circ$$ -diagonal to produce perfect circular polarization. The antenna is fed through a three-stage microstrip line that acts as an impedance transformer for perfect matching of the antenna impedance to $$50\,{\Omega\:}$$ at $$28\:\text{G}\text{H}\text{z}$$ . The development of the antenna design is described in detail, starting with a square patch that is subjected to progressive geometrical modifications in the design stages to reach the final design of the patch antenna. The simulation results show that the antenna has a reflection coefficient magnitude of less than $$-35\:\text{d}\text{B}$$ at an axial ratio of less than 0.5 dB at 28 GHz. The technique used to generate the radiated circular polarized waves is demonstrated by investigating the surface current on the printed patch at the resonant frequency. The primary objective of the proposed design is not to produce a high-gain antenna, but rather to produce a low-profile planar antenna to serve as a building block for CP antenna systems such as multiple-input-multiple-output (MIMO) antennas, beam steering, and beamforming antenna arrays operating at 28 GHz. The designed antenna is fabricated and subjected to experimental evaluation. The numerical simulation results and the practical measurements came in good consent from the impedance matching and axial ratio (AR) of the radiated circular polarized wave perspectives. The input impedance is matched to the feeding line over a 1-GHz band, from 27.5 GHz to 28.5 GHz. The radiated wave is circularly polarized with AR less than 3 dB over about 0.4 GHz band starting from 27.8 GHz to 28.2 GHz. The peak gain of the designed antenna is 7 dBi and the radiation efficiency exceeds 90% all over the band of impedance matching. The performance of the antenna is compared with those of other antenna designs presented in some recent publications.

A compact planar antenna is proposed to operate in five millimeter-wave frequency bands at 28, 51, 55, 59, and 63 GHz. The antenna is designed to produce circular polarization at 28 GHz and linear polarization at the other four frequencies. The 28 GHz frequency band is recommended for long range communications, whereas the other four frequencies are recommended for short range communications due to the atmospheric absorption in this range of the millimeter-wave frequencies and to provide more secure communications. The reasons behind the selection of the specific values of the higher frequency bands (51, 55, 59, and 63 GHz) are (i) to provide enough number of multiple bands at such unlicensed frequency range to be used as alternatives for different applications, (ii) to give about 4 GHz separation between the center frequencies of each band and, (iii) finally, to avoid the operation at 60 GHz, in particular, due to the peak atmospheric absorption of millimeter-wave encountered at this frequency. The antenna is constructed as a square patch surrounded by five parasitic elements that are capacitively coupled to the square patch so as to realize the location of the resonant frequencies of the higher frequency bands. The antenna is fabricated and its performance is experimentally evaluated. The proposed antenna is shown to operate efficiently over the five frequency bands 28, 51, 55, 59, and 63 GHz providing impedance matching bandwidths of 2.0, 0.9, 1.2, 1.1, and 1.6 GHz, respectively. The corresponding values of the maximum gain are 7.0, 6.8, 6.0, 6.0, and 8.0 dBi, respectively. Also, the corresponding radiation efficiencies are 90, 90, 91, 78, and 86%, respectively. The 3 dB-axial ratio bandwidth at 28 GHz is about 1.2 GHz.

In this article, a new technique for field effect transistor (FET) small-signal modelling using neural networks is proposed. This technique is based on the combination of the Mel-frequency cepstral coefficients (MFCCs) with different discrete transforms such as the discrete cosine transform (DCT), the discrete sine transform (DST) and the discrete wavelet transform (DWT) of the inputs to the neural networks. The input datasets to traditional neural systems for FET small-signal modelling are the scattering parameters and the corresponding frequencies in a certain band, and the output datasets are the circuit elements. In the proposed approach, these datasets are considered to form random signals. The MFCCs of the random signals are used to generate a small number of features characterising the signals. In addition, other vectors are calculated from the DCT, DST or DWT of the random signals and appended to the MFCC vectors calculated from the signals. The new feature vectors are used to train the neural networks. The objective of using these new vectors is to characterise the random input sequences with more features, to encourage robustness against measurement errors. There are two benefits to these approaches: (1) a reduction in the number of neural network inputs, and hence a faster convergence of the neural training algorithm and (2) robustness against measurement errors in the testing phase. Experimental results show that the techniques based on the discrete transforms are less sensitive to measurement errors than using the traditional and MFCC methods.