ABSTRACT. Nonlinear modeling of GaN HEMTs is a challenging task especially when accurate predictions are desired under nonlinear dynamic operation. This is mainly due to the dispersion phenomena affecting state-of-the-art devices that must be accurately characterized and correctly accounted for during the development of their nonlinear model. The challenge becomes harder as frequency, power and circuit complexity increase as being demanded by the marketplace for the upcoming 5G-related applications. In this paper, nonlinear measurement techniques are discussed which allow the accurate characterization of microwave GaN HEMTs under actual operating conditions. The paper also discusses how to use these measurements in the extraction phase of nonlinear models highlighting the benefits of their adoption for modeling purposes.

ABSTRACT. The present investigation is aimed at investigating the temperature effects on the forward transmission coefficient (S21) for high electron-mobility transistors (HEMTs) based on various technologies. The analysis is carried out on six different devices under different bias conditions with the ambient temperatures varied from -40°C to 150°C. The GaN-based HEMTs show a significant reduction of the low-frequency magnitude of S21 when the temperature is increased, whereas an operating bias condition at which this quantity is temperature insensitive is found for the GaAs-based HEMTs.

ABSTRACT. Fin field-effect transistors (FinFETs) are a type of device that has received great attention in recent years, owing to their ability to scale down, low cost, and high efficiency for advanced chip technology. In order to contribute to the study of the FinFET architecture, a three-dimensional (3D) silicon-on-insulator (SOI) n-FinFET device is developed and calibrated to IBM’s measured model. This paper is focused on investigating the influence of the gate dielectric material on the device RF parameters, such as transconductance, gate capacitance, and cut-off frequency. The achieved results demonstrate the positive effect of using a high-k material as a single layer between the gate contact and the channel. The studied device with a titanium dioxide (TiO2) gate dielectric allows achieving superior gate controllability compared to the low-k dielectric materials. In particular, the reported analysis leads to achieving a transconductance of 37.6 µS and a cut-off frequency of 4.48 THz at VDS = 0.05 V with VGS = 0.24 V and VGS = 0.2 V, respectively, by considering TiO2 as the gate dielectric material.

ABSTRACT. The miniaturization of the SOI-MOSFET transistor reduces the gate electrostatic control and reliability of the integrated circuit (IC). The fabrication of the transistor was reaching 7 nm, but the scaling to 5 nm technology node was projected to occur in the coming future, as a result, the researchers expect to identify new solutions for different design issues. In this paper, the double-gate FinFET for 5 nm technology is investigated with different gate dielectric materials (SiO2, SnO2, ZrO2, Ta2O5, and TiO2) using 3D-TCAD-SILVACO tools. The results showed that using TiO2 (k = 85) as gate dielectric material produces better performance and excellent analog characteristics such as Vth, SS, Ioff, Ion/Ioff ratio, and DIBL. It discovered that the proposed device is the most compatible for improving device efficiency (reliability, lower power, manufacturing cost, and faster circuit) for the future of nanoscale devices due to the use of the high-k gate material TiO2, as well as a shorter gate (Lg= 5 nm). Furthermore, the maximum oscillator frequency (fmax) and cut-off frequency (ft) are both increased which makes the proposed device a delicate choice for RF applications and potential manufacturing process innovation.

ABSTRACT. This paper presents a balun less differential output LNA by accounting for various interconnect losses. The modified cascode with current reuse and body biasing in the 1st stage of the LNA to achieve dc power consumption PDC <13mW. On the other hand, the 2nd stage employs a differential output for improved noise (NF<6.0dB) and linearity (IIP3>2.0dBm) performances. Moreover, the optimized MOSFET transistors are used in the proposed design to achieve noise resistance ROPT close to 50 required for power matching, with the rejection of image signal @50 GHz by more than +60 dB without sacrificing the LNA performances. The ground shielded inductors used in the design are realized in the IHP 0.13m SiGe BiCMOS process, and optimized under Keysight ADS EM environment to realize high-quality factor Q >25. The proposed LNA design layout occupies a chip area of 0.52mm2. Theoretically, predicted S-parameters of the LNA were found to overestimate the simulated results due to the simplified equivalent circuit used. By accounting layout area, a new Figure of-Merit (FoM) involving key performances like S21, noise factor F and dc power consumption PDC was proposed and found to be significantly higher for the proposed LNA than reported work in the same technology. All the results presented in this paper are verified using the full EM simulation. However, their experimental verifications are not included due to the limitation of experimental testing facilities at the authors' Institutes

ABSTRACT. In this paper, the experimental verification of two linearization methods applied on a broadband two-way microstrip Doherty amplifier is performed. The laboratory set-ups are formed to generate the baseband nonlinear linearization signals of the second-order. After being tuned in magnitude and phase in the digital domain the linearization signals modulate the second harmonics of fundamental carrier. In the first method, adequately processed signals are then inserted at the input and output of the main Doherty amplifier transistor, whereas in the second method, they are injected at the outputs of the Doherty main and auxiliary amplifier transistors. The experimental results are obtained for 16QAM and 64QAM digitally modulated signals.

ABSTRACT. ANN model for one-dimensional DoA estimation of the signal received by two-element textile wearable antenna array is proposed in this paper. Multilayer perceptron network is used to create the MLP_DoA module which is capable to provide information of a radio gateway location in azimuthal plane as output when a spatial correlation matrix, found by receiving the radio gateway signal using two-element textile wearable antenna array, is on its input. The noise impact on the accuracy of the proposed ANN model is investigated. Comparison of the presented model with the root MUSIC algorithm in terms of accuracy and program execution time is also demonstrated.

ABSTRACT. Application of one new algorithm for construction of wide bandpass filter is introduced. It is verified with successful fabrication of structure in microstrip technology. Characteristics of the filter can be adjusted only with characteristic impedance of the short-circuit stubs and number of cells. All stubs are uniform and have equal characteristic impedances.

ABSTRACT. In last century World population has more than tripled, while the population of the great cities in some cases have risen more than tenfold. This population boom is presenting not only sustainability and housing problem, but a problem to successfully maintain law and order in those cities. Law enforcing institutions are traditionally relying on manpower and electro – optical devices (EODs) installed either on fixed positions or on helicopters. These approaches are showing a lot of shortcomings, such as lack of manpower, EOD’s weather susceptibility and high exploitation costs. In this paper, we are presenting an IoT based surveillance principle which is relying on Remotely Piloted Aircraft Systems (RPASs) equipped with Multi-Functional Radar Systems (MFRSs) as primary “ears” and “eyes”. Benefits of these principles are multiple: firstly, it significantly reduces manpower needed to achieve complete surveillance, since all RPASs can be controlled from a single command center. Secondly, influence of weather factors is significantly reduced since MFRSs are more resilient to weather conditions than EODs. Finally, exploitation costs of the system in a long run can be lower compared to usual helicopter based mobile surveillance units.

ABSTRACT. The rapid growth of information and communication technologies and increased number of internet connections during the last few years has led to the extensive use of various IoT applications. Nowadays, IoT applications are built using various embedded platforms with constrained processing and energy resources. Achieving real-time performance in such IoT systems, with more efficient usage or with limited energy resources, can be a serious challenge. In this paper a set of parameters, that control the operation of IoT end-devices, and in this way affecting overall IoT system consumption and its real-time behavior are introduced. To investigate and illustrate this potential adequate simulation environment is designed and developed. The simulation of the simplified data producer-consumer configuration in IoT edge tier, under push-based communication model, have confirmed that careful tuning of the identified set of parameters can lead to more efficient IoT end-device operation under the given real-time constrain.

ABSTRACT. Mobile network operators are faced with tremendous data growth in their networks. Complex and multilayered networks, high-quality signal demands, and strong competition are only some of the aspects that push operators to further optimize their networks. Also, introduction of 5G in their networks enabled mobile operators to offer broad spectrum of 5G services, especially IoT (Internet of Things) and M2M (Machine to Machine) services which further increase traffic volume and the amount of collected data. Traditional data management solutions currently used are not able to successfully respond to such huge amounts of data. Big data technologies represent a modern approach for coping with the enormous data quantities. In this paper, we propose a big data solution that can collect and process huge amounts of data to extract valuable information and help mobile operators to bring their networks to enhanced quality level and offer full IoT solutions to their customers.

ABSTRACT. Technological advancement is one of the main problems and causes of environment pollution. Industry, power plants, traffic, factories are the main pollutants in big cities. Therefore, it is necessary to use technology to decrease its bad influence. One of the ways is smart city sensor network with applications that are able to predict air quality for the next few days and warn the population in the case of the upcoming severe pollution. The prediction task in this paper is conducted by using the state-of-the-art algorithms such as CatBoost, with additional features that help improve quality of the model. Three models are examined, regular CatBoost, CatBoost with additional variable and the new approach where CatBoost model is boosted with another model. All the models are compared, while the best performance is achieved via new approach Boosted CatBoost.

ABSTRACT. Overlapping user grouping based on adjusted maximum correlation value (OUG-AMCV) has been proposed in Multi-User Multiple-Input Multiple-Output (MU-MIMO) systems with zero forcing precoding. It achieves higher throughput and much lower complexity when compared with capacity-based greedy zero-forcing schemes. In this paper we combine OUG-AMCV method with two-stage precoding technique and investigate its performance in more practical IoT oriented 5G MU-MIMO environment with imperfect 3D channel estimation. The proposed algorithm further increases system capacity and reduces computational complexity.

ABSTRACT. In this paper, we propose and simulate a layered architecture for Wireless Sensor Networks in Precision Agriculture in order to estimate power consumption of several light-weight energy-efficient cryptographic algorithms. Simulation was carried using the Atarraya simulator and simulation description, results and observations are presented.

ABSTRACT. The paper presents a conceptual model of a cyber-physical system (CPS) in the area of the road section of the Motorway 9th January (M9J) in Republic Srpska BiH. The properties, requirements, problems, entities, networks and technologies of the proposed CPS conceptual model are defined. The paper places special emphasis on networks and platform on which the proposed model relies: M:tel Banja Luka provider networks, Wireless Sensor Networks, Software platform.

ABSTRACT. Air pollution is rising problem, with its effects being especially severe in urban and industrial areas. The constant and local monitoring of air quality, and the appropriate presentation of the results to population demands deployment of large-scale IoT-based monitoring networks in which low-cost, low-quality sensors would be predominantly used. However, the inherent measurement errors could incur large AQI (Air Quality Index) calculation error. Also, appropriate presentation of air pollution demands that measurements of air pollutants’ concentrations are classified into Air Quality Classes, thus making the classification task for AQI of large interest. In this paper we analyzed a wide variety of Machine Learning (ML) and Deep Learning (DL) models in order to solve classification task for AQI, but under the assumption of the low-cost sensor deployment in the real-world application. The analysis results suggest that DL models designed, optimized and tested in this paper present a viable and most suitable solution under these demands.

ABSTRACT. Although the development and implementation of IoT systems is still very popular area in education, there is the perception that regarding the young engineers, still not enough attention is given to security of IoT systems, especially in practical manner. This is especially important for the utilization of IoT in mission critical applications that demand high level of security. Accordingly, in this paper an IoT education platform for testing security vulnerabilities of IoT systems is presented. During the professional internship, students will get familiar, on a practical way, with the ease of ARP spoofing attack, using proposed education platform. Also, simultaneously students will learn defense mechanisms.

ABSTRACT. In this paper, we investigate security-reliability tradeoff on physical layer in Internet of Things (IoT) network, when an eavesdropper tries to intercept secure transmission. The best-node scheduling scheme is employed as an auxiliary tool to decrease probability of intercept. The main as well as wiretap channels are assumed to be subjected to composite generalized-K fading. Asymptotic expression of security versus reliability, valid in the range of medium to high average main-to-eavesdropper’s signal ratio (MER) values, is derived. The impact of different system parameters such as IoT network dimension, fading/shadowing conditions, path loss and average MER on intercept probability versus outage probability is observed, and then accompanying numerical results are shown.

ABSTRACT. This paper presents a promising simulation approach for evaluating the performance of arbitrarily deformed flexible electronic components. Arbitrary deformations of electronic components are designed using a computer graphic method for three-dimensional object manipulation. The electromagnetic simulations of these objects are primarily enabled by bespoke tetrahedral discretization that allows for smooth boundaries and thus eliminates discretization noise. The method is used to evaluate the impact of twisting and crumpling on the performance of flexible RF antennas.

ABSTRACT. We investigate a compressive sampling (CS) stepped frequency ground penetrating radar for detection of underground objects, which uses Bayesian estimation and a probabilistic model for the target support. Due to the underground targets being sparse, the B-scan is a sparse image. Using the CS principle, the stepped frequency radar is implemented using a subset of random frequencies at each antenna position. For image reconstruction we use Markov Chain and Markov Random Field models for the target support in the B-scan, where we also estimate the model parameters using the Expectation Maximization algorithm. The approach is tested using Web radar data obtained by measuring the signal responses scattered off land mine targets in a laboratory experimental setup. Our approach results in improved performance compared to the standard denoising algorithm for image reconstruction.

ABSTRACT. This paper proposes a coded terahertz imaging method that combines with a traditional compression sensing theory. The coded imaging of the target object can be calibrated with the measured collimated spot to improve the image quality significantly. Subsequently, corresponding intensity-only measurements are collected with a CMOS detector for different masks, and a reconstruction algorithm is applied to recover 16 × 16 pixel images.

ABSTRACT. This paper presents a hybrid boundary element method analysis of single and coupled V-shaped microshield lines. The numerical results show quasi-static values of the characteristic impedance and the effective relative permittivity of those lines. The obtained results are presented graphically and in tables. The numerical accuracy of applied method is verified through the results obtained using a software simulation.

ABSTRACT. In this paper, an ANN-EM neural model of dual band square patch antenna with a floating rectangular slot is presented. For the desired resonant frequencies, this ANN-EM model allows to speed up the process of estimating the physical parameters of the antenna with a good match between the antenna and the power line. The ANN EM model makes it possible to obtain antenna parameters for resonant frequencies that belong to the frequency range 700-3000 MHz.

ABSTRACT. The investigation of the influence of the metal frame spectacles on the electric field and Specific Absorption Rate distribution within the user’s eyes exposed to the mobile phone electromagnetic field is presented in the paper. Numerical analysis of electric field distribution and Specific Absorption Rate has been performed when user's eye is exposed to radiation of the mobile phone operating at 2.45 GHz. Numerical solutions are obtained solving the equations of electromagnetic wave propagation. Accordingly, the obtained results are given for biological tissues within user’s eyes. The 3D models of human head and human eye and spectacles have been developed. The results obtained for the models with and without glasses have been compared in order to determine the effects of the presence of described glasses on distributions of values of interest.