Proceedings of First International Conference on Computational Electronics for Wireless Communications

A novel tapered and slot loaded compact circular microstrip patch antenna (TSCCMPA) with DGS for ultra-wideband (UWB) applications is proposed. For designing of UWB TSCCMPA, a move toward optimizing the physical stricture, two vigorous techniques, namely, artificial neural network (ANN) and adaptive neuro-fuzzy inference system (ANFIS) are used. Commercially available Ansoft HFSS v.13 is utilized to extract the 144 datasets of TSCCMPAs having different sensitive parameters related to antenna dimensions and substrate materials. Apart from 144 simulated datasets, 129 were used for training and remaining 15 were utilized for testing of models. The average percentage errors (APE) for resonant frequencies of return loss are calculated regarding performance of tested datasets of ANN and ANFIS model. The APE in ANN and ANFIS model found in testing of resonant frequencies f1 and f2 are 0.8731%, 0.0699%, and 0.7698%, 0.0607%, respectively. The very low APE indicates that the models can be successfully applied for optimization of physical parameter of UWB TSCCMPA for computer-aided design (CAD) applications. The trained models can be used effectively for antenna parameter estimation, instead of running HFSS repetitively or other optimization techniques, which consume more time. Successful implementation of TSCCMPA shows broader impedance bandwidth of 2.6–53.5 GHz as 180% at 26.00 GHz of center frequency. The average gain of proposed antenna is found as 4.14 dBi. Due to very wideband the anticipated antenna can be effectively used in many applications.

A compactness MIMO ultra-wideband MIMO radiator is presented in this paper. A novel technique of circular split ring resonator (SRR) is inserted in a circular patch, loaded stubs are added at the bottom of the patch and DGS placed in the ground plane. By considering this novel structure miniaturized antenna structure greatly improves the isolation between the patches which is obtained at 3.5 GHz. Due to the use of monopole ground plane structure, bandwidth impedance of the design is broadened. A very great broadband impedance bandwidth is achieved as a result from 3.0 to 10.0 GHz of the proposed structure. Moreover, the MIMO structure obtains the stable radiation patterns, low ECC around less than 0.02, and an efficiency around 93%. The current design has electrically compact size of 35 × 53 × 1.6 mm3. The proposed system with a configuration of low profile especially is a better candidate for wireless performance communication of MIMO system.

A comparative analysis of dipole and bowtie antenna with and without tapered balun feeding technique operating at 2.4 GHz is presented in this paper. The glass epoxy FR-4 substrate of thickness of 1.6 mm and dielectric constant of 4.4 is used in the proposed antennas. For S11 <−10 dB, the bandwidths obtained for dipole antenna with and without tapered balun are 24% and 16%, respectively. For S11 <−10 dB, the bandwidths obtained for bowtie antenna with and without tapered balun are 44% and 20%, respectively. The proposed antennas achieve more than 2 dB gain for the entire bandwidth. The antenna parameters such as radiation pattern, return loss (reflection coefficient S11 <−10 dB), bandwidth, gain, and Smith chart have been simulated and analyzed using Ansys HFSS. The proposed antennas find application in Industrial, Scientific, and Medical (ISM band-2.4 GHz) system.

This paper presents a gain enhancement of dual-layer patch antenna for WLAN applications. With help of conventional rectangular patch and modified structure, the dual-layer patch antenna achieves the required frequency of 5.4 GHz. For dual layer, air was used for second layer with height of 1.6 mm mounted on the first FR-4 substrate layer with height of 1.6 mm with dielectric constant of 4.3. The total height is 3.2 mm and radiating element is mounted on top layer with thickness of 0.035 mm. The 50 Ω impedance transmission line is fed with radiating element. Single layer −10 dB return loss is −34 dB which is simulated, fabricated, and measured. The VSWR value is <2 and corresponding gain and directivity are 4.12 dB and 5.35 dB, respectively, and bandwidth is achieved from 5.35 to 5.43 GHz. For dual layer −10 dB return loss is −29 dB, VSWR is <2, bandwidth from 5.38 to 5.54 GHz, gain and directivity are 7.57 dB and 7.51 dB, respectively, which is simulated with the help of 3D CSTMW software. The measurement result of single layer achieved is 5.38 GHz and return loss is −34 dB which is 20 MHz shifting because of indoor measurement. By observation dual-layer micro-strip patch antenna gives good result compared to the single-layer antenna. In this work, overall gain, directivity, and bandwidth are compared with single-layer and dual-layer patch antenna. The proposed antenna S-parameter measurement using keysight E5053A (100 kHz–14 GHz) vector network analyzer and then prototype single-layer antenna was fabricated. The dual-layer patch antenna gain and bandwidth is improved. This result proved better performance which is more suitable for WLAN applications.

In this paper, a novel microstrip patch antenna (MPA) which is having dual-band characteristics is presented. The proposed MPA, with resonating frequencies of 2.43 GHz and 5.872 GHz, has applications in S-band and C-band (in license-free band, i.e., 2.4–2.4835 GHz and 5.825–5.875 GHz). It is designed on a substrate FR4 (εr = 4.3 and tanδ = 0.04). In this, line feeding scheme is used to supply current to the antenna so as to excite it. The results of actual fabricated antenna are close to that of simulation one. The proposed antenna having dual-band resonance and comparatively better and high gain performance can be easily mounted into systems for wireless communications such as radar and satellite communications.

The tag antenna exhibiting operation in European and North American regions covering major UHF RFID bands resonating at 866 MHz and 915 MHz, respectively, has been designed in this paper. The tag antenna operating in single UHF RFID region is converted to operate in dual UHF RFID region band tag antenna by modifying its geometry and optimizing the final geometry to obtain resonance at the required resonant frequencies. The tag antenna proposed in this paper comprises a meandered line element with extended lower stub to obtain additional band at European Region. The designed tag employs Alien Higgs-4 RFID chip having capacitive reactance. The designed tag utilizes inductive spiral loop to obtain conjugate impedance to match the capacitive RFID IC. Further, the designed modified tag antenna is simulated and its performance is analyzed based on different parameters such as its resistance, reactance, radiation efficiency, realized gain, etc. Also, it has been seen that the designed dual band antenna shows bidirectional and omnidirectional radiation pattern in E-plane and H-plane, respectively.

The rapid increase in the deployment of Wireless Local Area Network (WLAN) particularly over the existing infrastructure is growing tremendously. Such an infrastructure-based WLAN, widely known as Wireless Fidelity (WiFi) network, needs to support voice service because, in general, it contributes a large portion of the traffic supporting personal communication. Voice over Internet Protocol (VoIP) over WiFi commonly known as VoWiFi can be used for providing voice services. In order to provide quality of service (QoS) to such voice calls, it is essential to establish a call admission control (CAC) policy. This policy requires VoWiFi cell capacity. In this paper, we have estimated the number of VoWiFi calls possible using the IEEE 802.11ac standard Access Point (AP). To estimate the cell capacity, we have used DCF Inter-frame Spacing (DIFS) to sense channel status before sending data and Short Inter-frame Spacing (SIFS) is used for acknowledgement, Request To Send (RTS), and Clear To Send (CTS) frames. We have estimated the VoWiFi cell capacity using different voice codecs like G.729 and G.723.1. We have used packet retransmission technique to avoid packet loss and Arbitration Inter-frame Space (AIFS) to enhance the VoWiFi cell capacity.

Beamforming with MIMO (Multiple-Input Multiple-Output) system is only solution to maximize high data rate and extended cell coverage with satisfying quality of Service (QoS) for fifth generation (5G) cellular networks. In hybrid millimeter-wave (mmWave) massive MIMO systems, estimation of information about Channel State is difficult due to the large channel and small number of RF chains. The present paper accomplishes millimeter-wave-based massive MIMO for hybrid beamforming based on Sparse Estimation. The proposed iterative hybrid algorithms accomplish the low rank and beamforming sparsity properties in massive MIMO to gain full data recovery with minimal error for small time duration. The proposed work model is an mmWave-based beamforming system with imperfect channel state information (CSI) to minimize channel estimation errors. Experimental section highlights the efficiency of our proposed method over traditional methods for the sake of simulation with accounting its improved temporal efficiency, fast convergence, and tolerance to abnormality channel information.

In this paper, a review on some of the techniques of designing the FIR filter efficiently has been presented. Most of the Digital Signal Processing (DSP) devices use digital FIR filter due to its various advantages over IIR filter. In designing these filters, there are many performance parameters like area, power consumption, speed and complexity, which put challenges in front of a designer. In order to meet the desired performance, many techniques have been proposed. This paper is mainly focused on some of these techniques.

The key purpose of this paper is to implement an amplifier that retains less power and large gain which is appropriate for precise biomedical applications. The circuit design of a two-stage opamp is executed with the help of LTspice tools employing 180 nm technology files. Numerous analog systems, for example, filters, integrators, data converters like ADCs, and summing circuits are carried out with operational amplifiers as it is the most necessary building block for the above-mentioned circuits. It is essential to design an effective amplifier that enhances the performance of analog circuits. Present work largely focused on executing a two-stage opamp with maximum gain. Further, utilizing this two-stage opamp, an instrumentation amplifier is realized. Several performance parameters, for example, voltage gain, AC gain, slew-rate, average power consumption, and bandwidth are calculated and observed.

A new resistor-less current differencing buffered amplifier (CDBA)-based voltage-mode (VM) Multifunction filter is proposed with single CDBA, three NMOS transistors and two capacitors. The presented filter circuit implements all standard filter responses without altering the circuit properties. The study of non-idealities and parasitic effects of the CDBA and their effect on transfer function is also carried out. This circuit also proves to give adequate response even in the presence of non-idealities and process parameter variations. The presented circuit is having low (≤0.5) sensitivity. The proposed circuit feasibility has been justified by using PSPICE 0.18 µm CMOS technology with supply voltage ±0.6 V and power dissipation of 171 µW.

The way energy can be captured is by using photovoltaics. It will enable us to use the natural resources that are available around us rather than being dependent on fossil fuel and oils. We focus on the renewable source that is the light. A solar-powered house and the requirements to make the same has been familiarized and understood to design AC- and DC-based solar systems and compare the same. Materials used to create are silicon based, and now there are many emerging materials that can be used to capture light. The material that is effective and probably even more than silicon is called perovskite. Then, we look at the technologies that are making significant changes in the photovoltaics sector. Such technologies are 3D Printed Solar Panels that will help in cutting cost and create solar panels faster and 3D Printed Solar Trees that will mimic trees and help capture energy from all around its surroundings.

Communication protocols are useful for transfer of data between devices. Even though we have many protocols, SPI is one of the most important bus protocol. SPI is used to connect microprocessor to the peripheral devices. It works on the principle of FIFO (Ring topology). Normally, Flipflops were used to synchronize the data between the Master and Slave but it consumes more power. In this paper, SPI is implemented using Pulsed Latches to reduce the power consumption. Pulsed latches consumes less power, and area occupied by the pulsed latches is less compared to Flipflops. The design is simulated using Verilog. The Integrated Development Environment (IDE) used is Xilinx-ISE, and ISE Simulator (ISIM) is used to verify the waveforms.

In this paper, a hybrid 1-bit Full Adder (FA) using Pass Transistors, Transmission Gates, and conventional CMOS logic is designed and also extended for 4-bit and 64-bit. Less power, delay, and PDP are the foremost features of the proposed approach. Circuit parameters have been analyzed using Cadence Virtuoso 45 nm technology. Parameters such as power, delay, PDP, EDP were compared to extant designs. With the 1 V supply, the average power consumption (0.1379 µW) and delay (66.6878 ps) were found to be extremely low as compared with existing designs. For 4-bit, the power consumption is 1.95 µW with 285.6 ps and PDP 0.56 fJ with 0.8 V supply. To examine its scalability, the 64-bit ripple carry adder is designed. In this work, a low-power 4:2 Compressor is presented. The average power consumption of compressor is 542.7 nW, delay is 116.5 ps, PDP is 63.21 aJ, and EDP is 7.364 × 10−27 J with 1 V supply. Meticulous Monte Carlo simulation is performed in CADENCE SPECTRE at 45 nm CMOS technology. Simulation results conclude that proposed design signifies remarkable results in power consumption and delay which enunciated for low PDP. The proposed full adder cells attain a 44% power improvement metric in comparison with reported designs of full adder. In the field of digital technology specifically, the image and signal processing predicated applications that necessitate high speed, less power, and lessen delay, a proposed adder is suited, as adder unit, it is one of the primary blocks of FIR Filter.

This paper presents a novel controlled positive feedback RFC OTA. An extra current source is used in feedback to control and enhance the transconductance of RFC OTA. A flipped voltage follower is used to provide the biasing to the input differential pair, which performs the class AB operation. The simulation has been done using Mentor Graphics Eldo with the CMOS technology of 0.18 µm. The capacitive load of 15 pF and supply voltage of ±0.5 V has been used for the simulation. The proposed amplifier results in an improved gain, bandwidth, and phase margin of 73.29 dB, 4.37 MHz, and 69.8°, respectively. The power dissipated by the proposed OTA is 55.50 µW. The figures of merits validate the results of the proposed amplifier. The effects of PVT variations on the proposed amplifier are negligible, which verifies the stability of the circuit.

This paper carries a performance evaluation of delay elements based on propagation delay (tp), power (PWR), Power-delay product (PDP), and Energy-delay product (EDP). The research article analyses a MOSFET-based CMOS delay element (M-CMOS DE) and a MOSFET-based MCML delay element (M-MCML DE). Simulation results establish the superior performance of M-MCML DE in terms of tp, PWR, PDP and EDP. M-MCML DE exhibit improvement in tp (58.98×), PWR (1.66 K×), PDP (98.07 K×), and EDP (5781.99 K×). Simulation outcomes validate that the MCML based design is a competent candidate to replace the traditional CMOS-based design. As an extension of work, M-MCML DE is implemented using an emerging device––FinFET. The proposed FinFET-based MCML delay element (FinFET MCML DE) emerges as ultra-fast and low power design of delay element. It offers improvement in tp (1.1×), PWR (34.28×), PDP (37.71×), and EDP (41.29×). In this research article, we restrict our attention to design metrics such as tp, PWR, PDP, and EDP. This monograph will rouse more research activities in the area of low power and robust design of digital circuits.

The MOSFET has a wide range of usages like switching, amplification, sensing, and many more required in the modern world for industrial, biomedical, and environmental purposes. In this work, a DG-MOSFET (Double Gate Metal Oxide Semiconductor Field Effect Transistor) with optimized channel width is proposed. The DG-MOSFET is designed separately with Si and Ge wafers. The device is constructed on the silicon wafer with three different regions, namely source, channel and drain. The double gate concept introduced in the device increases the control over the channel region. Further, the simulation is done using Visual TCAD 2D analysis. This 20 nm gate length device shows very good electrical characteristics with Ion/Ioff ratio (~1012) and almost negligible DIBL. The performance of this device indicates that it is useful for amplification applications.

With growing health awareness and the increasing cost of medical care, there is an impetus to new and advanced technologies for disease prevention and early diagnosis and treatment. The weakest link exposed by the COVID-19 pandemic in India is health care. Investment in critical health Infrastructure aided by modern technology is the need of the hour. So, this project aims to develop a comprehensive healthcare monitoring system by blending IoT and VLSI. It can monitor a patient’s basic health signs as well as the room condition where the patients are now in real time. We use Nexys4 Artix7 as a processor. In this system, six sensors are used to capture the data from the hospital environment named heartbeat sensor, body, and room temperature sensor, fall detection sensor, blood pressure sensor, air quality monitoring sensors, ECG sensor. The condition of the patients is conveyed via the ThingSpeak website and telephonic calls/SMS to relatives, medical staff.

In the generation and distribution of electricity in the power grid, there is a chance for the occurrence of non-technical loss while transmitting by various means. One of the most concern needed loss varieties is electricity theft. The theft occurrence may cause significant loss and harm to the power grid and also to the economy by leading to unprofitable accounts for the power supply companies. Regular inspection on irregular consumption of power is inefficient and very time consuming. Utilizing machine learning in this theft detection system helps to prevent huge losses. In this paper, various machine learning models are employed to state the better performing model for the given data. By employing the various techniques of machine learning, an effective model for theft detection can be obtained and the problem associated with non-technical loss especially theft detection can be monitored and controlled.

Blockchain Technology (BT) enables decentralized, distributed, encrypted, and immutable logging of digital transactions. Blockchain features a decentralized database and public registry which helps transactions across a peer-to-peer network secured through cryptography. The aim of this paper is to how BT could be useful in enhancing the business performances of an organization based on research in the past decade. The study will be based on theoretical aspects of blockchain implementations, their application in various processes of an organization. This study focused on the detailed description of blockchain technology and its various business application, showing its impact on the different business models. The study highlights various barriers, motivators, and tools for BT and its impact on different parts of an organization.

Industry 4.0 is one of the rapidly growing technologies of present-day industries. The development of newer technologies such as cloud computing, internet of things and cyber-physical systems has brought a new revolution to the manufacturing industry. Interoperability among devices is one of the key goals of Industry 4.0. Modbus, TCP (Transmission Control Protocol), UDP (User Datagram Protocol) are common protocols that are universally accepted by all manufacturers of Industrial Controllers. These protocols help in standardizing the data communication between different vendor controllers through the server–client architecture. In recent days, Open Platform Communications (OPC) have proven to be an effective technology to support interoperability and heterogeneity in control and automation applications. In this paper, a case study of common server–client communication protocols present in the industry through the implementation in industrial controllers has been presented. This research looks at the features of these communication protocols and their usage in the existing systems. OPC-UA communication has been compared with existing protocols such as Modbus, UDP, and TCP and its features to support Industry 4.0 are shown with comparative analysis.

The target tracking using the passive multi-static radar system produces various detections via distinct signal propagation paths. Trackers solve the uncertainties that arise from the measurement path as well as the measurement origin. The existing multi-target tracking algorithms suffer from high computational loads, because they require the entire probable joint measurement-to-track assignments. This paper proposes to develop a comparative analysis on diverse heuristic algorithms for implementing the optimized JDPA model for tracking multiple targets using multi-static passive radar system in the presence of clutter. Here, the nature-inspired algorithms like Particle Swarm Optimization (PSO) and Grey Wolf Optimization (GWO) are used for analyzing the optimized JDPA model for tracking multiple targets in multi-static passive scenario. This paper further, aims to tune the position and velocity of the tracker towards the target using two heuristic algorithms, and intends to analyze the effect of those algorithms on improving the performance of multiple target tracking. The key objective of the proposed model is to minimize the Mean Absolute Error (MAE) between the estimated trajectory of the track and the true target state.

Flow Management is a key concern to ensure the Quality of Service in Software-Defined Network (SDN). This paper introduces an analytical model using M/M/c/K queueing where the arrival of flows is governed by Poisson process forming a single queue and there are ‘c’ number of controllers whose service times are exponential having a finite buffer ‘K’. The loss due to Reneging and Retention of flows in a Multi-Controller SDN is initially evaluated. A Cost model has been formulated to estimate the effect of Retention of Reneged flows on Total Expected Cost per unit time. The results are simulated which validates that the Total Expected Cost per unit time is a function of service rate of the controller and M/M/c/K queueing model have significant influence on the performance of a Multi-Controller SDN.

The current day scenario is all about the growing pandemic and a lot of possibilities arise regarding the resurgence of the pandemic in second and third waves which demands a good prediction model. The core idea of the project is around this. The aim of this project is to develop a prediction model which can provide an estimate of the effected population if second and third wave of resurgence occurs. The prediction model is based on the key behavior of the pandemic, which is a parabolic behavior. Considering this behavior, Numerical Methods have been used to find solution of Parabolic Differential Equations to arrive at an estimate simulated in the MATLAB tool. These can be of great use and provide accurate results if appropriate input parameters are provided. The IoT-based mechanism captures the outputs of the algorithm and provides a physical indicator through a dedicated device or mobile of the user.

The main motivation of this work is from the extensive observation in the way tasks are implemented within hardware subsystems to achieve the desired functionality. In most of the scenarios for the completion of a particular computation a predetermined sequence is implemented. This in terminology of subsystem level implementation is called as hardware internal sequences. A lot of the steps which happen within the whole hardware sequence are redundant in nature. The target of optimization in this work is redundancy occurring in hardware sequences. To make the optimization methodology more compatible to the current trend of AI development, a nature-inspired methodology is proposed. The algorithm used here is called as collapse and evolve approach. The focus optimization parameters are latency and power consumed when the whole hardware sequence gets implemented. Nature or the process of evolution of nature is some hundreds of billion years old when compared to human evolution which is some tens of billion years old suggesting natural evolution to be a more advanced flow. So, focus is put on the natural evolution process to extract an efficient technique which can achieve our target optimization within the hardware sequences. The manner in which initial set of species survived and evolved frames the base of the hardware computational algorithm. The work suggested falls in the unison of a very interesting area of AI to hardware and nature-inspired computing which allows the outcomes of the work to be utilized in far more diversified areas. As a part of the study, the proposed algorithm is implemented on the power processing block of digital system.

The post-COVID-19 era redefines safety and security for all of us. With increasing population, higher rate of unemployment and faster pace of life, we need a solution that can ensure safety as well as peace of mind for the family and their home. The proposed solution Home Shanti provides an integrated and holistic home management system which autonomously and remotely monitors the home without physical presence. It makes use of edge computing mechanisms to take on-board decisions via interaction with the sensors in the IoT environment. Smart home security is offered with a network of specific sensors. In addition, safety of the residents is taken care of in various regions of the house. The system is in constant touch with the owner through cloud, remote monitoring and real-time decision making. Hence, Home Shanti is the perfect solution for the safety and security of urban and rural home environments and its family, truly a ‘shanti’ peace of mind for the homeowner and his/her family.

As years pass by, the world is turning toward clean, renewable energy solutions and environmentally friendly practices. As a result, the use of Electric Vehicles (EVs) is being encouraged for sustainable transportation and they are also considered to be portable energy storing devices. It is highly expected of these EVs to make a major impact in the future grid stability which becomes a major concern when adopting renewable generation due to the intermittency of the sources. The concept of V2G (vehicle to grid) refers to the bidirectional integration between plug-in EVs and the grid which also allows the vehicle to contribute to reactive power support and active power regulation. Adding to that, filtering out harmonics, prediction of renewables’ behavior and amending grid operations like stability, volatility, efficiency, etc., are also some features of V2G. This paper covers the overview of design, grid impact, and cost benefit of the V2G technology.

A Wireless Sensor Network (WSN) consists of tiny sensor nodes possessing low power, storage capacity, computing capabilities, etc., that are spread in the sensing area to collect data. These sensor nodes can be grouped together to form clusters to ease communication and prolong network life. The coordinator node called Cluster Head (CH) plays the major role by collecting data from sensor nodes and sending it to the BS. The CH is either rotated or new CH is selected to use the network energy optimally. The CH rotation can be energy-based or time-based. The energy-based mechanism uses the remaining energy of the current CH whereas, the time-based mechanism uses the number of rounds after which CH needs to be changed. The paper proposes a dynamic time-driven CH rotation approach using fuzzy logic. The results demonstrate that the approach is more efficient.

Neural Style Transfer is a very good example to show the capability of Convolutional Neural Network (CNN) to generated high-quality images. In style transfer algorithms two input images are present one is a content image and the other is a style image and it generates output in such a way that the output is a mix of both content and style image. There are different methods and approaches that can be used to implement the style transfer. This paper will focus briefly on neural networks and their architecture used in the approaches and briefly describe the important loss functions and the implementation of neural style transfer. This review paper will explain the four major approaches or techniques that have been used to implement neural style transfer, these approaches will also tell how they have improved on the original method put forward by Gatys et al. 2015 to produce good quality results and then find the advantages and disadvantages by comparison of each method that is explained. This paper aims to provide knowledge of how different approaches work.

Software-Defined Networking (SDN) is a traditionally used method in directly connected networks, and very recently onwards it was also used in some wireless networks. SDN’s concept is to make the network more robust by leveraging the advantage of network interface re-programmability during initialization. In SDN-based networks, the control plane is transitioned to a level higher from the infrastructure to provide re-configuration. By having some input from nodes in the network, the SDN controller wants to change the configuration parameters. In Mobile Ad-hoc Network, the feedback may include insight into the quality of the connection and resources available, such as battery capacity and node position (hop count away from the destination). We introduce an anonymous authentication scheme to improve privacy inside SDN-based MANET along with which we proposed traffic monitoring system to avoid congestion and share the load from one controller to another controller. As our main contribution, a compact, stable authentication method with proper forward secrecy is developed, and what we consider to be a more interesting aspect is that it is possible to achieve user anonymity, perfect forward secrecy, and resistance to de-synchronization attacks at the same time. Our proposed work is implemented using ns-3 simulator and also proved that it was efficient by comparing with existing models.

The pandemic of 2020 has affected the every minor and major aspect of human life all over the world. The whole world went into lockdown, where socializing with other people was banned. Almost everything was being done online, whether it was the office work or classes. Schools and Colleges started conducting classes online through various video meeting apps, but this came with a lot of issues like students were losing interest in class, teachers were unable to see whether the students are paying attention in the class or not, etc. In this paper, a drowsy alarm system is proposed to tackle these issues which alerts the student as well as the teacher when a student is dosing off or losing interest in the class. MediaPipe FaceMesh, a lightweight Machine Learning (ML) face geometry solution, is used to create this system. With the help of Tensorflow.js, an ML library in Javascript, the system can be deployed in the web browser of any device whether it is a mobile or a computer. The system tracks the movement of eyes and mouth and alerts the person by starting the alarm if a closed eye or a yawn is detected. The effectiveness of the system when it runs on the WebGL, WebAssembly (WASM), and on computer’s CPU (Central Processing Unit) is compared and the result has been shown.

Facial Recognition is a technique that uses the face and its features to identify/verify a person. It is a Biometric Application, which is playing an integral role in today’s world in a wide variety of areas like Criminal Identification, Visitor Verification, and many other Real Time Identification systems. In this paper, we present a system that does facial matching for deformed faces. The proposed method combines the extraction of high-level feature representations using Deep Learning and the calculation of facial ratios using Image Processing to generate a robust model that can be used to identify/verify deformed faces. Whereas traditional face recognition systems show poor results when there is a face deformation. We use a function that tells us how similar or how different the two input images are. So, for this Siamese network is used. This network takes in an image as input and gives its feature vector. Then, we find the distance between the computed feature vectors of the two images which will give us a similarity score. Next, we use an Image Processing technique to calculate the facial ratios from macro features like eyes, lips, etc. These facial ratios are calculated based on facial landmarks spread across the face. We use multiple facial ratios, so even if deformities occur in a part of the face it will not have much effect on the system rendering it immune to facial changes. Hence, this paper tries to close the gap in the previously devised facial matching methods which were not designed for deformed faces.

In digital communication, transmitting data error-free is a major concern. The errors in data while transmitting results in wrong information at the receiving end. To protect the information in memories and register in digital circuits, error correction code plays an important role. Hamming code is one of such error-correcting codes used in digital transmission which can correct single-bit errors. Both even and odd parity checks can be used in the Hamming code. Here, even parity check method is used in implementing Hamming code encoding and decoding circuits. In Novel Hamming code, the parity bits are appended at the last to the data. Novel Hamming code requires less number of combinational gates than regular Hamming code. The pad-to-pad time delay is reduced in Novel Hamming encoding and decoding circuits as combinational gates are reduced while calculating parity bits. Here, both Hamming code and Novel Hamming code circuits are implemented in the Xilinx ISE Platform. A special redundant bit is used to spot the double-bit error.

The Steered Response Power with Phase Transform (SRP-PHAT) is a widely preferred technique for acoustic source localization due to its robustness in a harsh environment. However, it necessitates exploring the search region on a computationally intensive grid to maintain its robustness, making it unsuitable for real-time applications. Modified SRP-PHAT introduced an effective strategy that considers the influential cubic volume around the grid location to compute the SRP functional. Consequently, the method allows examining the search region on a coarser spatial grid. However, the resolution of the technique is marginalized by the selected spatial grid step size. Thus, the Modified SRP-PHAT was further improved by employing an iterative grid decomposition strategy. However, in challenging environments, the performance deteriorates considerably with the iterative approach. Thus, the different implementations of the iterative SRP-PHAT techniques have been proposed in the literature. This paper presents a performance comparison of the Iterative Modified SRP-PHAT, Iterative Volumetric Reduction SRP-PHAT, Iterative Modified SRP-PHAT with adaptive search space, and conventional SRP-PHAT methods. These methods have been extensively tested on the simulated environment and SMARD database to confirm their validity and reliability. The results exhibit that the Iterative Modified SRP-PHAT with adaptive search space localizes the source with the least root mean square error in the practical reverberant environment. Furthermore, this localization accuracy is very close to densely computed SRP-PHAT with very small functional evaluations.

Software testing has always been a crucial job in accomplishing and assessing the quality standards of a software product. Software testing is done to confirm the developed software product does what it is expected to do. However, testing is expensive in terms of time, effort, and is quite complicated. Studies report that software testing alone is responsible for almost half of the total budget incurred in software development. Additionally, manual testing is more prone to bugs and creating accurate and reliable software is an open issue. Specialists and experts have been exploring more effective and successful automation techniques for testing to deal with this issue. This paper is an endeavor to review the cutting edge of how machine learning and artificial intelligence have been figured out to automate and streamline software testing processes. It also provides an insight mapping of the research into these fields. Furthermore, a practical study on testing web applications is performed using selenium.

Nowadays, there are several types of ways to post your idea through social media or any kind of other funding platforms like Kickstarter, Indiegogo, Wishberry, RocketHub, and FundRazr. All these platforms raise funds for your Ideas and projects. However, these platforms do not evaluate or analyze your ideas. The analysis in our invention will provide the success percentage of the idea or project. Success percentage is the main factor to get your idea ready for funding. As mentioned in the terms and conditions in the existing applications, all of these charge for using their services. Some have a different charge for using the service without caring if your idea/project works or not, while some charge from the money that has been raised. These services do not guarantee your return on investments. This invention will provide a platform where anybody can share their thoughts and ideas. An AI algorithm will then assess these ideas and calculate how relevant the idea is as per the latest trends and requirements. The next algorithm in our application will also try and predict the chances of the project succeeding. It will also include a user rating feature where other users can rate the ideas and give opinions about them. Furthermore, our product will introduce a concept where the algorithm will give the idea a score based on the ratings and the results achieved by the other algorithms. Ideas will be ranked based on this score. In addition to this, our application will let users collaborate with other developers given that they have relevant skills. This will solve the first two shortcomings. For instance, you have an idea that you know is in demand and also a team capable of turning that idea into a reality. But how do you scale your product? This is where our final feature will come into play. Like normal users, potential investors can also go through the ideas and rate them. They can also choose to invest if there is an Idea/Product that captures their attention. Lastly, many features such as user sponsorship, product requests, and various other exciting features are being established to improve the experience of both the Users and the Investors.

Internet of Things (IoT) is a dimension which permits the things to be detected, sensed, and to be accessed from remote locations. Pre-existing network architecture provides an opportunity for the combination of advanced monitoring system. This provides more accurate, efficient, and economical results. IoT is playing an active role in various fields of our life such as to assists chronic patients, smart transportation, smart grid, smart city, etc. In this research paper, we have explained the application of IoT in home automation and its role in sensing the environmental factors for improving energy efficiency and comfort of life. We have also discussed the various examples of IoT in home applications. Along with that a model has been discussed based on solar energy using IoT home automation. Primary objective of the research is focused on striking a balance between efficient power utilization and generation of power. Sensors and actuators are combined with the internet and are discussed here.

Wireless sensor networks comprises of sensor nodes that consists of non-rechargeable batteries. The main constraint in WSNs is to develop energy-efficient protocols to save energy of sensor nodes to improve the existence of the network. Various energy-efficient clustering protocols have been developed as one of the solutions to energy efficiency. We propose and evaluate a new clustering protocol using game-theoretic approach based on probability for heterogeneous wireless sensor networks using two level nodes, i.e., standard and advanced nodes. In proposed protocol, cluster heads are elected using game theory in heterogeneous environment using various factors such as probability, initial energy, and residual energy. Finally, simulation experiments show that proposed protocol achieves more energy efficiency and enhances stability period, than other existing clustering protocols such as D-DEEC, SEP, LEACH in heterogeneous WSNs.

Outlier detection is a key component for the proper functioning of a wireless sensor network. It helps in identifying malicious events taking place in the field and faulty sensors. This paper presents a distributed outlier detection technique that separates outlying cases from normal data locally at sensors and sends only the summarized information of data to sink nodes to minimize the communication overhead in the resource-constraint wireless sensor network. The proposed solution employs variance of normal data as a measure to identify outliers. The performance of the proposed method has been evaluated on real-world datasets collected by WSN deployment projects. Experimental results prove the effectiveness of the proposed method in terms of higher outlier detection rate, lower false alarm rate and lower communication overhead as compared to the simulated centralized approach.

In this paper, we have propounded a neoteric procedure for the super-resolution of an image using a single image. An image of low resolution is given as input which is upscaled to an image while preserving the information that is stored in textural and structural details of an image. The image provided as input which is of low resolution is segregated into two sections, namely textured and non-textured according to the features of the image. Rational fractal interpolation is employed in the section of the image considered as textured and rational interpolation is employed in the remaining image which is considered to be non-textured. Thereafter, pixel mapping is performed. The result obtained from interpolation is found to contain Gaussian noise. To subdue the effect of this noise, an adaptive Wiener filter is applied. Finally, an image of high resolution is obtained. Profound simulations and assessments demonstrate that competitive performance is achieved by our algorithm. The mean square error reduces approximately up to $$5\%$$ 5 % , whereas the structural similarity index improves marginally.

Multipliers can generate a significant performance in designing filters, MAC units, memories, and image processing applications. Several multipliers have been designed and implemented for various applications. However, the main issue was the latency which is increasing as the number of bits increasing. Multipliers with compressors help in the reduction of latency of partial product which in turn helps in reducing the summation of adders in the final summation stage. In the proposed design of 8 * 8 Wallace tree multiplier, 16 partial products have been obtained by multiplication of operands and multiplicand. In the proposed Wallace tree multiplier design with 8–4 compressor, multiplication operation has been obtained in less number of steps as compared to conventional multiplier without compressors in which the number of stages is reduced from 16-12-8-6-4-3-2. The proposed 8 × 8 Wallace tree multiplier with an 8-4 compressor requires 24.7 ns for completion of the multiplication operation. It thus achieved a 43% reduction in delay as compared to the conventional Wallace tree multiplier (WTM). The simulation result also proves that the proposed design has achieved a 27% reduction in total power dissipation. Thus, the proposed Wallace tree multiplier (WTM) exhibits better interpretation in terms of latency as well as power dissipation, and hence, it can be used for high-performance image processing implementation. The proposed Wallace tree multiplier (WTM) is simulated using ModelSim Student Edition tool and synthesized using Xilinx ISE 14.7 navigator tool.

In this paper, the discussion is on non-contact vital signs monitoring using a frequency-modulated continuous-wave (FMCW) radar. Here, we analyze the phase of the intermediate frequency (IF) signal in detail. Any human or clutters exposed to the FMCW radar signal can reflect the chirp. The reflected chirp goes to the mixer to produce an IF signal. This paper deals with extracting vital signs such as breathing rate and heart rate from a subject who sits in front of the radar irrespective of the chest orientation toward the radar. The phase of the IF signal is processed to obtain the spectrum in the desired range-bin. The spectrum is further subjected to filtering and thresholding as will be detailed. Random body movements are detected, and those samples are discarded for the calculation of heart rate.

The most important sources of information about the products or services are online reviews. People trust online review comments while purchasing electronics items, hotel booking, college admission, movies, etc. Sometimes it has been intentionally written by the fake reviewers for monetary gain, business rivalry, etc. Many times, these messages were found fake after the judgments. The fake reviews transmission has a significant social and economic impact on society. Hence, an accurate detection mechanism must be there to identify fake reviews. In this paper, the opinion spam detection mechanism is proposed using sentiment analysis (SA) for content-based applications. In this technique, the sentiment score of the sentences is computed. It is detected as fake or not fake, depending on the sentiment score of reviews. The work also proposed a Long Short-Term Memory (LSTM) based deep learning approach to identify the topic of fake reviews. Combining these two approaches provides a more accurate opinion spam detection rate compared to other existing models. On the benchmark “Deceptive Opinion Spam Corpus v1.4” dataset is used. Our model’s accuracy is 92.46% with 9.23% of the false acceptance rate and 5.50% of the false rejection rate.

The images in hazy environment are not clearly visible due to atmospheric light scattering. Hence, image dehazing is required to reduce the haze effect. In this paper, a colour ellipsoid prior-based model is utilized for image dehazing. After dehazing, the output appears dark and with less contrast. These images experience with low brightness and may downgrade the performance for computer vision systems. The LIME (low-light image enhancement) is implemented in our proposed work, to avoid the over-dark and less-contrast outcome. The image enhancement scheme is implemented for the dehazed image using LIME. For a dehazed image, the Illumination Map (IM) is computed and it is further refined by using structure prior. Through the refined IM, the dehazed image can be enhanced effectively. For the proposed method, the experimental results show better qualitative and quantitative results when compared to the state-of-the-art methods.

Cloud computing provides many facilities such as storage, computing, analytics, scalability, etc. But, it also suffers from issues like traffic congestion, high latency and increased communication cost. Fog computing was introduced by Cisco Systems as an alternative to cloud computing. It acts as a mid-layer between user devices and cloud level thereby helps in managing the cons of the latter. However, the introduction of fog layer casts doubt on the security of data travelling through it. Security, being multidimensional in nature, can be handled in an efficient manner, if addressed through its dependent factors and sub-factors in a hierarchical manner. Hence, a hierarchical structure representing corresponding factors and sub-factors related to fog computing security is presented and the same have been prioritized deploying Fuzzy Analytical Hierarchy Process (FAHP). The obtained solutions have been compared to those obtained through classical AHP and are observed as coordinated in nature.

In spite of successful and remarkable advancement made in current studies on local binary methods, it requires groundbreaking research into theoretical perspectives as well as required more efficient approaches on algorithm to concern about the real-world problems occurring in natural images. The research indicated a precise indication of the target for texture characteristics and allowed the application to expand the LBP method into practical engineering, while optimizing theoretical analyses. The LBP method is relatively straightforward and uncomplicated for texture analysis, but also has an invariance of rotation, grayscale invariance, and other important benefits. In terms of precision of texture analysis based on LBP having multiple applications, the conventional local binary model (LBP) approach was to calculate the value of each pixel of an image and assign the model's binary value with their specified based on their relationship to neighboring pixels. In consideration of situation on real time images that are having efficient operator on texture analysis for images, this paper covers the various strains of local binary patterns with their area of application, benefit, and their disadvantages for future shortcomings.

For getting the improvement over the recognition accuracy of face recognition system, a method based on auto-luminance is proposed. Firstly, the face is detected from the query via input images, real-time web cam using the Viola–Jones face detection algorithm, then detected face is pre-processed with auto-luminance and histogram equalization method. Then, we use HOG feature descriptor measuring gradient amplitude and the orientation of the gradients. These features with the description are then classified with the help of SVM machine learning classifier to discriminate the different classes. By using this approach of auto-luminance with histogram equalization and HOG feature extraction when performed at input images, live web cam and the well-known ORL cropped face database, it is cogitated that this algorithm has a higher face recognition rate than the conventional one when applied pre-processing on them.

Today, security is a major issue in the cyberworld. The data generated by the devices are more susceptible to threats and attacks than being used. The emergence of cloud computing and IoT changed the whole concept of handling and interacting with data. The framework that we proposed for the security of data, which is being generated by IoT devices is divided into multiple phases. The first starting phase is mutual authentication. This phase guarantees the communicating parties are genuine. The paper presents the novel scheme for mutual authentication between two parties that are IoT device and Fog system. In the paper, we discuss and figure out how the IoT device and the Fog system will interact with each other and authenticate each other.

The neighborhood search algorithms recently emerged as a promising low-complexity detection algorithm for massive multiple-input, multiple-output (MIMO) wireless systems have fascinated recent research attention. They iteratively perform searches in a constrained Maximum-likelihood (ML) space for the solution vector. In addition, they execute with a complex matrix inversion an immense number of computations, resulting in higher complexity within a massive MIMO configuration. Motivational by these, we focus on unconstrained minimization problems to quickly decline in an ML cost function and devise for updating rules to improve performance. Using this robust approach and the reduction of the ML cost function concerning the descent direction, we propose a computationally efficient Conjugate Gradient-based local neighborhood search (CGLS) detection algorithm. The proposed CGLS detection algorithm is employed to give the most reduction in the ML cost metric by determining in a decent direction update within an unconstrained neighborhood. It gains acceleration for convergence with few iterations. Simulation results show that the proposed CGLS detector outperforms the counterpart detector.

Likelihood Ascent Search (LAS) algorithm is a low-complexity detection algorithm that emerges as an optimistic in massive multiple-input multiple-output (MIMO) systems. In real-time scenarios, channel estimation is challenging in wireless communication systems. This paper proposes a lattice reduction-based LAS (LR-LAS) detector for efficient detection under the assumption of imperfect channel state information. Our proposed LR-LAS detector takes computing the maximum likelihood (ML) cost in the ubiquity of a medium-to-strong estimation error. Our approach presents a novel training-based channel estimation and detection framework. Simulation results on the performance of the Bit Error Rate (BER) of LR-LAS are compared by drawing useful insights for massive antennas.

This article represents an improvement in the efficiency of Deep Neural Networks (DNN) employed in OFDM systems for estimation of channel, detection of symbols by using Hyperparameter tuning. Deep learning is increasing its popularity in each field and it’s a set of Artificial Intelligence (AI). Deep Neural Network (DNN) model is completely different from traditional Orthogonal Frequency Division Multiplexing (OFDM) receiver in such a way that OFDM receiver has to estimate the channel state information separately then it is able to detect the transmitted signal, but DNN model will be trained offline by simulated data over various channel statistics then directly employed in online for detection of the transmitted signal directly. Deep learning gives a performance better than traditional methods, i.e., Minimum Mean Square Error (MMSE) of channel estimation and symbol detection in wireless communications. First, we have analyzed the performance of traditional methods, i.e., MMSE over Winner II channel without deep learning by observing BER with cyclic prefix and without Cyclic Prefix (CP) and with variation in number of pilot carriers. Thereafter we will deploy hyperparameter tuning with proposed DNN model in OFDM for estimation of channel and detection of symbols in the winner II channel mainly for the Urban power delay profile using QPSK modulation. Here, we have claimed that by reducing the no of epochs from 20,000 to 10,000, training period will be reduced from 52 to 26 h, i.e., 50% while maintaining the almost same BER performance.

mmWave communication system encounters the higher path loss than the microwave communication system due to the high-frequency band. To defeat the path loss problem massive number of antennas with low wavelength are deployed at the transmitter and receiver side. To transmit multiple data streams and to get better spectral efficiency and capacity precoding is required. Developing the hybrid Precoding is economically high and consumes more power and it is divided into analog and digital parts. Due to the presence of large antennas and analog part in the hybrid precoding, mmWave massive MIMO requires some special algorithms to do the channel estimation and precoding. To construct the sparse precoding and combining problems in mmWave massive MIMO, channel is considered as -sparse channel. In this paper sparse precoding is designed based on the orthogonal basis pursuit algorithm for mmWave massive MIMO by using dictionary which consists of array response vectors of angle grid points.

A consolidated nomenclature for LTE wireless networks is proposed in which wavelength division multiplexing PONs are arranged in a ring topology. The setup is completely dynamic so that various base stations in the particular network may communicate over a dedicated λLAN channel. The irregularity in traffic of downstream channels may be monitored in this architecture. Traffic balancing may be achieved in this proposed work.

Considering Wireless Sensor Networks (WSNs) usage in sensitive applications, providing secure communication between the sensor nodes is of utmost importance. The key pre-distribution technique allows the sensor nodes to encrypt the messages employing the secret key to uphold the network security. Having limited computational powers and storage capacity are the constraints of sensor nodes. In this work, Combinatorial Design (CD) is employed to propose a deterministic scheme for key pre-distribution in WSNs wherein keyrings are generated from a given keypool. The network region is divided into many same-sized cells with regular sensor nodes and cell leaders deployed in each cell. The cell leaders possess higher resource and computational capabilities than the regular sensor nodes and thus are used for communication between cells. Whenever the regular sensor nodes need to establish communication links with other regular sensor nodes in the same cell, they can do so directly using the common secret key. The key pre-distribution scheme proposed for cell leaders is highly scalable. A detailed study of the scalability, the resiliency of the proposed scheme is also presented. The resiliency accomplished is comparable to other existing schemes. Still, at the same time, the given scheme provides full connectivity, high scalability without a significant increase in the storage overhead of the sensor nodes.

Millimeter-wave (30–300 GHz) communication entails a large number of transmitting and receiving antennas to accomplish high beamforming gains, so as to minimize the high path loss. Digital precoding/combining suffers from complex hardware due to the requirement of an enormous number of antenna arrays, and analog precoding/combining suffers from a poor performance rate. Therefore, hybrid precoding is a trade-off between analog and digital precoding/combining, when the number of users increases. This paper carries Kalman-based analog/digital, i.e. hybrid precoding to increase spectral efficiency in the case of multiuser system. Simulation results show the improved average achievable rate in millimeter-wave multiuser system.

In Wireless Sensor Networks (WSNs), issues including energy management, topology management, bandwidth estimation, packet loss calculation, etc. are dealt. Only a few works have paid attention to mitigating congestion while estimating delay and bandwidth for transmitting data packets in WSNs. Several works on queue management and congestion control have incorporated Soft Computing (SC) techniques to solve some of the problems in WSNs. In this paper, an Active Queue Management (AQM) model to estimate congestion using Random Early Detection (RED) and mitigate congestion is proposed. It is found that the results obtained outperform the existing methods in terms of delay between intermediate nodes, end-to-end delay, packet loss ratio, packet loss probability, queue size and energy consumption.

In this paper, a refractive index-based ethanol sensor is proposed. The hollow core Photonic Crystal Fiber (PCF) with hybrid cladding structure is used. Ethanol is taken as the chemical analytes for sensing at terahertz frequency. TOPAS is used as a fiber material in photonic crystal fiber with asymmetrical rectangular hollow core. The outer layer of the PCF is considered as Perfectly Matched Layer (PML) for boundary conditions. Finite Element Analysis (FEA) is used for mathematical analysis with COMSOL Multiphysics software. As results, the investigated parameters are sensitivity, confinement loss and effective area. The proposed PCF-based refractive index sensor shows the high relative sensitivity of 97.28%, low confinement loss (CL) of 6.27×10-13(dB/m) and high effective area of 14.4µm2 at terahertz frequency (0.8–1.6 THz). So, the proposed refractive index-based sensor can be used for various applications like marine industries, healthcare and food industry.

Wireless rechargeable sensor network (WRSN) is gaining a significant place in the current research field due to its wide range of applications including in the field of Internet of Things (IoT). The sensors in WRSN will soon deplete their energy due to their limited battery capacity. The wireless charging (WC) is one of the prominent methods for enhancing the sensors lifetime in wireless rechargeable sensor networks. In WC, with the help of mobile charger (MC), we can steadily provide energy to sensors using electromagnetic signals. However, the promising issue in this area is scheduling MC efficiently for charging the sensors so that the network is up to its maximum time. This article contributes to addressing the above-mentioned problem, by proposing an efficient way of scheduling the MC with the help of sensor’s Energy Consumption Rate Prediction. From our experimental results, it can be observed that the proposed method has a given good survival rate of sensors compared with the existing methods like FCFS and NJN without preemption.