Emerging Technology Trends in Electronics, Communication and Networking

The power budget is a very stringent requirement for portable instruments. In this paper, static dynamic methods for ultra-low power and voltage are identified like gate driven, bulk driven (non-conventional method), and dynamic threshold methods which are applied on basic differential amplifier cell, and comparative analysis is performed for various parameters like power dissipation, gain bandwidth product, etc. The ultra-low voltage is selected for low power dissipation and simulations performed under TSMC 180 nm Technology node. The results show that the dynamic threshold method (gate driven, bulk driven) method is superior as compared to other methods in the aspect of all performance parameters. For selected technology nodes, total power dissipation is 4nW and maximum gain 40 dB with the acceptable value of the gain bandwidth product.

The design methodology of the overmoded circular waveguide tapers to connect the gyrotron (ø85mm) output power at 42 ± 0.2 GHz/200 kW/3secs with different diameters (ø63.5 and ø31.75 mm) of transmission line components has been carried out. Design parameter of taper is optimized using coupling theory in such a manner that it provides an appropriate match between input and output of transition with lower spurious modes conversion. There are linear and nonlinear applicable methods for designing internal taper profile. These methods are implemented, and the results of linear tapers are compared with the analogous expressions for raised cosine (nonlinear) tapers. The overall loss in circular linear tapers is less than 1% for the length of approximately 107λ. However, raised cosine tapers may provide lower insertion loss, but linear tapers with moderate values of diameter ratio may be attractive because of its simplicity in fabrication. Both the tapers are designed and simulated using computer simulation technology (CST) Microwave Studio software.

A convenient way to estimate and optimize the delay of VLSI digital circuits is the popular logical effort-based optimization. In this paper, we analyzed the effect of various circuit parameters such as logical effort (G), branching effort (B), electrical effort (H), and parasitic effort (P) on the delay of a given circuit for two different technology nodes, namely 180 and 16 nm. The analysis results show the variation of delay with a particular logical effort parameter. The variation between simulation delay and logical effort delay is indicated by a parameter τ’, which is compared with the τ which is the delay of an inverter driving an identical inverter with no parasitic for a chosen technology. The effectiveness of the logical effort-based optimization is explored. Further, the logical effort-based delay reduction, a super buffer-based delay reduction, and delay of an un-optimized circuit are also compared. The effect of technology on logical effort method for each parameter in the deep submicron sizes has also been investigated in this research work.

Floating point numbers provide more range as compared to the fixed point values. The multiplier is one of the main blocks of a processor. For improved performance, there is a need for fast and efficient floating point multipliers. The single electron transistor (SET) is a prominent advanced device structure for achieving high-end computing system. This paper describes the implementation of single precision floating point multiplier using SET (single electron transistor) for better performance. Design and simulation of floating point multiplier have been performed using Cadence Virtuoso. In this paper, we are comparing SET-based floating point multiplier with 16 nm CMOS and then power and delay had been compared. The main objective of this paper is to reduce power consumption and increase the speed or reduce the delay of execution. IEEE 754 standard has been used to represent floating point numbers. Here, floating point multiplier is implemented and verified using the Cadence Virtuoso tool. Thus, SET-based floating point multiplier provides better execution in lowering the power and increasing the speed.

The long-term reliability of metal oxide semiconductor (MOS) devices in space technology depends on the total ionizing dose (TID) effect. In MOS technology, power consuming, expensive, and bulky triple modular redundancy and shielding techniques are required to address radiation related issues. In this work, we simulate Semi-Conductor Laboratory (SCL) 180 nm silicon on insulator (SOI) and Bulk NMOS device for comparative study of TID effects in space technology applications. Both devices after simulation show 0.42 V and 0.62 V threshold voltage, respectively. Devices are irradiated for 15 s to achieve doses of 100 K Rad, 200 K Rad, 500 K Rad, 800 K Rad, 1 M Rad, respectively with different dose rates. Bulk 180 nm NMOS was found to be more radiation-sensitive than SOI devices. Dose rate (DR) effect of 35 µV on a Bulk device and 16 µV on SOI was observed. 267% on Bulk and 256% on SOI leakage current shift observed due to radiation. Devices show the dose rate sensitivity with varying leakage current from the range of 1.8 to 3nA/um. In both the devices, leakage current is generated because of interface charge trapped due to radiation and charge trapped. Post radiation major shift transconductance characteristics are observed.

In this paper, we have proposed the design and performance analysis of corrugated horn antenna for non-contact liquid level measurement applications. The performance of corrugated horn antenna is dependent on corrugation profile, depth and period of corrugations. It gives higher gain, wide bandwidth, low side lobe level and better isolation from cross-polarization component. One of the best merits of the profiled corrugated horn antenna compared to rectangular or circular horn antenna is in compact size. The proposed design is simulated using computer simulation technology (CST)-microwave studio software based on finite integration technique. The simulation results show that it gives very narrow beam width and better isolation value in operating band.

A compact (32 × 32 × 1.6 mm³) quad-element with penta-band key-shaped Multiple-Input-Multiple-Output (MIMO) antenna is proposed having a wide bandwidth for targeting Ku, K and Ka band applications and to be used for 5G millimeter-wave applications. Four antenna designs (Ant-1, Ant-2, Ant-3 and Ant-4) are methodically investigated in terms of simulated gain, return loss, radiation efficiency, radiation pattern, envelope correlation coefficient (ECC), diversity gain (DG) and total active reflection coefficient (TARC). The design parameters of Ant-4 have been optimized for desired performance parameters and operation. Simulated isolation < −17 dB, ECC <0.08, DG between 9.979–9.999 (dB), TARC <0 dB and radiation efficiency > 70% are obtained.

Here, a simple, planar monopole antenna with polarization diversity for quad band application using a single feed mechanism has been reported. The design antenna which is proposed here involves two L-shaped radiators with two asymmetric cross slots within it creating mutual coupling to attain dual circularly polarized (CP) bands. To generate quad impedance bandwidth (IBW), defective ground plane is utilized on the bottom portion on the substrate. The antenna which is proposed covered the quad simulated IBs that are from 3.631–4.167 GHz (535.6 MHz, f_rc1 = 3.9 GHz, 13.73%), 6.542–8.494 GHz (1951.8 MHz, f_rc2 = 7.52 GHz, 25.96%), 8.859–10.253 GHz (1134.3 MHz, f_rc3 = 9.56 GHz, 14.59%) and 10.8494–beyond 15 GHz (4150.6 MHz, f_rc4 = 12.92 GHz, 32.19%), respectively. The simulated dual ARBWs span over 3.865–4.069 GHz (203.7 MHz, f_cp1 = 3.97 GHz, 5.14%) within 1st IB and 8.888–9.785 GHz (897.3 MHz, f_cp2 = 9.34 GHz, 9.6%) within 3rd IB. The simulated peak gain between 2.2 and 6.01dBi span throughout IBW region makes the dual CP bands appropriate for some part of S- and X-band, particularly fixed-satellite service (FSS) and synthetic aperture radar (SAR) applications.

In recent times, reconfigurable antennas become essential components of communication and radar systems because of their capability to adapt to changing system requirements or environmental conditions. Reconfigurability is a factor that changes the characteristics of a particular antenna for obtaining additional functionalities for any system. These antennas are low profile and low priced resulting in a good performance. Moreover, one single antenna can deliver the same purpose of more than one traditional single functionality antenna. Mainly reconfigurable antennas possess the ability to change the properties of antenna such as polarization, pattern, and frequency dynamically. The proposed antenna design is desirable for emerging wireless communication systems. The design consists of a rectangular patch. L-shaped stubs are used between patch and ground using PIN diodes as switches. By changing the states of PIN diode switches, an antenna can be made reconfigurable. The substrate used is FR4 having 1.6 mm thickness and a 4.4 dielectric constant. This antenna is analyzed using HFSS 15.0 simulation software. The proposed structure can produce all three reconfiguration properties.

IoT applications have opened up new avenues of research in all domains of electronics engineering. Antenna designing is one of them. This paper presents the design of a wearable antenna designed for IoT applications like medical healthcare and home automation. The design covers the W-BAN applications. The antenna design uses combination of Minkowski fractal and CSRR metamaterial-based antenna design on flexible materials like Jeans and polyimide used as a substrate. The antenna gives a gain of 4.6031 dB at 3.9 GHz using Jeans as a substrate and its performance remains good under bent conditions also.

The satellite navigation system is the most important asset for any country in the world. Global navigation satellite system (GNSS) encompasses all global satellite positioning systems like global positioning system (GPS) from the USA, global navigation satellite system (GLONASS) from Russia, Galileo from Europe and BeiDou from China, regional Navigation systems like Quasi-Zenith Satellite System (QZSS) from Japan, and Indian Regional Navigation Satellite System (IRNSS) from India. The Indian Regional Navigation Satellite System (IRNSS) by Indian Space Research Organization (ISRO) consist of a group of seven satellites that cover the Indian boundary and 1500 km extended area near the border. The aim of this work is to develop graphical user interface (GUI) for IRNSS receiver that can use the navigation files in excel format, process it and provide position, velocity, time (PVT) information. It can also plot PVT data and display the coordinates on Google Maps. The GUI is developed with using Python 3.6 with Python Tkinter toolkit. Plot facility of the GUI takes navigation file as input and plots it into six kinds of information, i.e., ground track, position, velocity, altitude, clock bias, and intersystem bias. It also allows to save the plots at desirable locations in the computer. Google Maps facility provides the receiver location on Google Maps with the coordinates.

Visible light communication is a fast-growing technology that is being explored as a prominent option for future communication. It solves the problem of RF spectrum crunch with a license-free spectrum by reducing the cost of the communication system. A transmitter architecture including an external modulator is proposed in this article which can be employed in the indoor environment. The calculated BER at 1 Gbps data rate is \(5.6 \times 10^{-5}\).

In this paper, the focus is to obtain performance of a multiple input and single output (MISO) wireless communication system in a real-time scenarios of time division duplex (TDD) and frequency division duplex (FDD). We consider a MISO system equipped with 2N transmit antennas. In the total N pairs of antenna, there is a spatial correlation in each pair (two antennas) only. We assume quasi-static Rayleigh fading channels. We select one pair of antennas with the highest channel power gain. In case of FDD, we assume antenna selection and detection, both based on imperfect channel state information (CSI), whereas in case of TDD, detection is based on imperfect CSI, but antenna selection is to be done using perfect CSI. Finally, we use Alamouti transmit diversity. For this system, we present the performance in terms of bit error rate (BER) versus avg. signal-to-noise ratio (SNR) and outage probability versus avg. SNR using Monte Carlo simulations. We have seen the adverse effect of spatial correlation and imperfect CSI on the performance. We conclude that TDD outperforms the FDD due to higher antenna selection gain.

The Free Space Optics (FSO) is one of the best solutions for ever-increasing demand of higher bandwidth requirements. As in the FSO, the transmitted optical signal passes through the free space channel where the atmospheric attenuation due to fog and rainfall considerably decreases the performance of the system. In our analysis, link range of 1 km is considered in 64 QAM-OFDM system with a data rate of 10 Gbps, where the system is analyzed under different atmospheric conditions of rainfall and fog. Results show that the optical beam cannot prolong beyond 1 km as the atmospheric conditions worsens.

Free space optical communication (FSO) is featured as a solution to the massive demand for large bandwidth and last-mile connectivity. However, the link performance degrades due to atmospheric turbulence. Like RF wireless communication link, the diversity and combining techniques were found effective in FSO link too. This paper compares BER performance of various diversity and combining techniques. We have simulated maximum ratio transmission (MRT) and transmit aperture selection (TAS) technique as transmit diversity scheme. For the receive diversity techniques, we have used selection combining (SC), maximal ratio combining (MRC), and equal gain combining (EGC). We have also considered MIMO configuration and wavelength diversity. These all techniques are compared for various diversity orders and different strengths of atmospheric turbulence. We found that maximum ratio combining and maximum ratio transmission schemes have the best BER performance among all diversity techniques. However, the MRT technique requires additional resources to make CSI information available at the transmitter.

The objective of this research is to design an abstract framework that can be used to represent and visualize any kind of geographical phenomena. The framework aims to be generalized in nature so that it can be applicable to all different kinds of geographical phenomena. There are three different proposed abstractions, namely objects, events and processes. We also propose a generic system design framework for developing a big data geospatial analytics application. The system design proposed provides a brief overview of what a usual big data geospatial analytics system might look like. The design highlights a few important aspects of the system like data ingestion and data store selection.

A sector-shaped circularly polarized (CP) corporate feed inspired fractal defected ground structure (FDGS) antenna is proposed in this paper. Fine-tuning the ground plane's defected ground structure (DGS) dimensions to aid CP radiation with enhanced axial-ratio bandwidth. Comparative analysis of the fine-tuned fractal structure with the conventional patch is provided. The higher iterations of the FDGS help achieve good polarization purity with axial ratio enhanced bandwidth compared to traditional sector antenna. The proposed sector patch antenna illustrates an impedance bandwidth of around 110 MHz (1.75–1.86 GHz), while its 3 dB axial-ratio bandwidth is 40 MHz (1.78–1.82 GHz). The gain ranges from 3.39 to 3.75 dBi across the axial-ratio bandwidth.

A two-element multiple-input multiple-output (MIMO) antenna of size 80 × 40 mm with circular polarization diversity for sub-6 GHz application is proposed in this paper. The single antenna operating at 3.5 GHz frequency consists of a sidewalk brick-shaped patch antenna. An offset feed is employed to get circular polarization. The step monopole ground of length λ/4 is used to get broad bandwidth. The axial ratio bandwidth improves further when an inverted L-shaped strip and an L-shaped strip are added to the resonating edge of the patch. The impedance bandwidth of 51.6% is obtained from 3.06 to 5.19 GHz, while the 3-dB axial ratio bandwidth of 16% is from 3.32 to 3.91 GHz. The antenna exhibits a tilted directional radiation pattern with 2.4 dB gain and 90% efficiency. Two-element MIMO antenna with a circular polarization diversity arrangement is designed with a distance of approximately λ/4. The isolation obtained between two antennas is more than 14 dB. Envelope correlation coefficient (ECC) is well below 0.005 with a diversity gain value close to 10. The antenna has application in a 5G sub 6 GHz band and wireless communication.

Employees are a company’s most valuable assets. However, if they left their jobs suddenly, it might cost a company a lot of money. Consequently, companies nowadays are actively seeking tools and technologies that can help in accurate and early employee attrition prediction. The main focus here shall be on the visualization of the available employee data in order to gain intuitive insights on the correlation among attributes and top causes behind the attrition. In this paper, we present the comparison of four algorithms used to predict whether an employee will leave or not, based on various attributes like age, salary, experience, etc., along with intuitive visualizations and its importance. The models used for prediction are random forest, logistic regression, K-nearest neighbor, and Naïve Bayes classifier. Thus, the visual analysis of employee attrition problem and accurate prediction can allow HR managers to take precautionary actions to retain the employee within the company.

Plants are prone to climatic changes which is considered to be one of the crucial challenges in agriculture. As a major part of the agricultural industry depends on rainwater for irrigation, climatic changes such as lack of rainfall may lead to shortage in water supply which may cause leaf wilting. If the extent of leaf wilting is not identified in the early stages, it may adversely affect the total yield. Leaf wilting is widely used as a parameter to compute drought tolerance in plants. The breeders manually collect the data for the extent of leaf wilting which consumes a lot time and efforts. This paper proposes a model with ResNet152 as the base architecture using transfer learning approach. The developed model uses leaf images as the input data and predicts the extent of leaf wilting into five different classes, namely 0, 1, 2, 3, and 4; 0 being the least amount of wilting and 4 being the highest amount of wilting. The proposed model achieves an accuracy of 87.00% which is better than other existing models. Further, the model was tested against some unlabelled images taken from the Internet and from the field for classification.

Catering to the widespread COVID-19 pandemic, the authors aim to develop a system based on machine learning combined with the knowledge of medical science. Considering the prevailing situation, it becomes necessary to diagnose the COVID-19 at initial stages. The idea behind the described designed model is to identify the spread of infection in patients as fast as possible. The paper sketches two different approaches: K-fold cross-validation and deep network designer which are based on deep learning technology for the prediction of COVID-19 in the initial stages by using the chest X-rays. The performance evaluation of the cross-fold validation process is compared with the designed application in the deep network designer to find an effective and efficient methodology for classification which attained better accuracy.

Optical character recognition (OCR) involves the electronic transcription of handwritten, printed text from either scanned documents or any sort of images. This multiclass classification problem comprising of recognizing the various characters in a language and correctly classifying them has found its way in plenty of computer vision applications. Although much work has been done for the English language, there have been only a few explorations pertaining to OCR for the Devanagari script. This paper proposes a hybrid CNN-RNN model to classify characters using the Devanagari handwritten character dataset. The main objective is to design a model with higher accuracy than the CNN model reported in literature for the same purpose. The models are trained and evaluated using the same procedures. On evaluating the models, the hybrid CNN-RNN model has a testing accuracy of 98.71%, which is higher than the CNN model, having 97.71% testing accuracy. It also fares better than the standard neural network architectures-VGG16 and AlexNet which when taken without the pre-trained weights result in 97.62 and 98.20% testing accuracy respectively. Hence, this successfully demonstrates the attributes of RNN in improved feature extraction when used along with convolutional layers.

Education data mining (EDM) is an important application domain of data mining. It focuses on generating interesting, novel, and useful patterns in benefit of academia. Institute Graduation Rate prediction, student dropout prediction, student group formation, students’ performance prediction, and student failure analysis are some of the challenging problems to address in field of education data mining. For education domain, useful research is conducted with either online available datasets or local dataset available with the organization. Dataset preparation for generalized analytical use is a challenge for field of EDM. Much of the time of the researchers is invested in identification and preparation of proper dataset that can be used for development of novel techniques and algorithms. In this paper, a novel framework of EDM dataset preparation is presented. Illustratively, this framework is successfully applied and tested for Institute Graduation Rate prediction problem. It is vital for researchers to have data at hand first to carry out further research. This paper illustrates a framework by which, education domain researchers could prepare datasets from huge data repositories as per their application need.

Recently, convolutional neural networks (CNNs) have been explored to achieve exceptional performance on super-resolution (SR) in terms of distortion metrics. In these methods, pixel-based loss functions have been employed to optimize their networks leading to overly smooth blurry solutions. In contrast, a generative adversarial network (GAN) has the proficiency to bring out perceptually better SR solutions. In the case of larger upscaling factors, some degradations are still discovered in the SR observations that can be reduced by increasing the number of convolution layers. However, such approach tends to increase the number of trainable parameters and additionally provides a lot of burden on resources which leads it to be unavailable for many real-world tasks. Here, we propose an improved progressive approach for SISR using GAN (i.e. ImProSRGAN). The potency of the proposed model has been seen by conducting different experiments where we observe that the introduced ImProSRGAN model performs better than existing GAN-based SISR approaches even though picking up fewer training parameters.

Community detection is widely used research topic. Community detection refers to detect same type of community structure in given graph or network. Nowadays, community detection is used for many applications like fraud detection, recommendation system, segmentation, etc. In this paper, our objective is to find the label for the unlabelled node using random walk and then label propagation algorithm. In this method, labelled and unlabelled data is provided as input and then, we take random walk until we find the unlabelled node, after that label for unlabelled node is provided based on labelled node using label propagation algorithm. The output of this method will be label for unlabelled node by using this we can divide the network into community. Comparison of different algorithm for community detection is discussed in this paper.

In the past decade, a convolutional neural network that can work on image became most popular since there were many powerful GPUs and datasets available to work. Convolutional neural network advanced the performance in various image processing tasks. In single-image-based super-resolution, convolutional neural network predicated methods have obtained remarkable performance in terms of error quantification (i.e., PSNR and SSIM) than earlier traditional machine learning predicated methods. Even though achieving better error measurements, the super-resolution results look blurry in appearance because of the classical \(L_1\) or \(L_2\) loss functions that have been used in the training process. Recently, generative adversarial network has been used to obtain a super-resolve image with better perceptual quality as compared to a convolutional neural network. In this manuscript, the technique and performance of those generative adversarial networks for single-image super-resolution approaches are compared for upscaling factor \(\times \, 4\).