ABSTRACT. Machine learning and data analytics continue to expand the fourth industrial revolution and affect many aspects of our lives. The talk will explore hardware accelerator architectures for deep neural networks (DNNs). I will present a brief review of history of neural networks. I will talk about our recent work on Perm-DNN based on permuted-diagonal interconnections in deep convolutional neural networks and how structured sparsity can reduce energy consumption associated with memory access in these systems (MICRO-2018). I will then talk about reducing latency and memory access in accelerator architectures for training DNNs by gradient interleaving using systolic arrays (ISCAS-2020). Then I will present our recent work on LayerPipe, an approach for training deep neural networks that leads to simultaneous intra-layer and inter-layer pipelining (ICCAD-2021). This approach can increase processor utilization efficiency and increase speed of training without increasing communication costs.

ABSTRACT. Similarity has a close relationship with learning, whether it be human or machine. It was recently proved that all machine learning that uses gradient descent, and that includes deep learning, is essentially a kernel machine, where the kernel function measures the similarity between data points. There are many applications possible that rely on similarity. Machine translation generates sentences in a different language, which are similar to the sentences in the original language. Tanscoders do the same for programming languages. We recently published work that uses deep learning to convert music in one melodic framework to similar music in another melodic framework. Another recent publication that we authored describes a music tutor project that takes as input, an amateur vocal rendition, identifies its melodic framework, also using deep learning, and plays back a similar vocal snippet from a master. The talk will expound on these and similar applications by first introducing the notion of latent feature space in machine and deep learning and the unsupervised, self-supervised, and supervised learning techniques that leverage the latent feature space and the notion of similarity.

ABSTRACT. As AI and other advanced technologies become ubiquitous in their influence and impact, touching nearly every aspect of life, we have increasingly seen the need to more consciously align powerful new technologies with core human values integrating consideration of societal and ethical implications of new technologies into the earliest stages of their development. Asking, for example, of every new technology and tool: Who will benefit? What are the potential ecological and social costs? Will the new technology amplify or diminish human accomplishments in the realms of justice, democracy, and personal privacy? At the same time, we see an opportunity for advanced technologies to help solve a host of stubborn political, economic, and social issues that trouble today’s world by integrating technology with a humanistic analysis of complex civilizational issues among them climate change, the future of work, and poverty issues that will yield only to collaborative problem-solving.

ABSTRACT. Diamond has the highest thermal conductivity at room temperature. With the existing thermal limitations observed in GaN-based devices, it is being evaluated as an alternative solution. This paper presents the electrical and thermal data collected at the wafer scale and illustrates the enhancement in these properties realized by integrating a Diamond substrate. The performance comparison of AlGaN/GaN high-electron-mobility transistors (HEMTs) on diamond and SiC substrates is also examined. However, using diamond substrates has limitations owing to availability, sample size, and coefficient of thermal expansion mismatch. Therefore, a novel approach, termed "diamond-before-gate," is shown to enhance the deposition process's thermal budget and facilitate large-area diamond without degrading the gate metal. Nanocrystalline diamond (NCD) thin films are used as heat-spreading capping layers on AlGaN/GaN HEMT devices. Observations show that the NCD-capped HEMTs exhibit almost 20% lower device temperature. NCD-capped HEMTs exhibited enhanced carrier density, hall mobility, sheet resistance, and threshold voltage and reduced contact resistance, on-state resistance, transconductance, and breakdown voltage. Additionally, the effects of using a p+ Boron doped NCD also stated.

ABSTRACT. The point of the undertaking is to limit the line at the energy meter charging counters and to confine the use of the energy meter naturally, in the event that the bill isn't paid. The undertaking additionally targets proposing a framework that will lessen the deficiency of force and income because of force robberies and other criminal operations. The working framework embraces an absolutely new idea of Prepaid Energy Counter. The GSM innovation is utilized so the buyer would get messages about the utilization of force (in watts) and on the off chance that it arrives at the base sum, it would naturally make the purchaser aware of re-energize. This innovation holds really great for all power dissemination organizations, private networks, IT parks, and self-containing lodging projects. There are clear outcomes from numerous nations, where the prepaid framework has decreased the income misfortune by an enormous sum. A GSM-based Energy Recharge Interface contains a pre-loaded card comparable to a versatile SIM card. This presents the plan and demonstration of a GSM-based Energy Recharge System for paid ahead of time Metering. The current arrangement of energy charging in India is mistake inclined and furthermore time and work consuming. Mistakes get presented at each phase of energy charging like blunders with electro-mechanical meters, human mistakes,handling mistakes. The point of the undertaking is to limit the blunder by presenting another arrangement of Prepaid Energy Metering utilizing GSM.

ABSTRACT. The performance evaluation of nanoscale transistors is a complex phenomenon challenging the fabrication of various components of the integrated chips. In the present study, digital parameters of the graphene based nanoribbon field-effect transistor (GNR FET) are investigated along with the gate stacked GNR FET, i.e., GS-GNR FET with gate stacking of lanthanum aluminate and silicon dioxide. The proposed device is simulated in Quantum ATK SE using the Slater-Koster method without SCF iteration followed by the non-equilibrium Green's function (NEGF). The gate stacking improves the device's performance by reducing the off-state current (Ioff) and improving the device's switching ratio (Ion/Ioff). Further, the subthreshold swing (SS) is reduced by 12% in the GS-GNR FET over GNR FET, portraying improved channel control and switching with less leakage current. The results show that the GS-GNR FET is an excellent replacement for the GNR FET in designing IC components in the electronics industry.

ABSTRACT. These days, data hacking is one of the most pressing concerns in the realm of cloud computing. While there are numerous effective solutions to this problem, data encryption stands out as the most practical one. Encryption is the process of transforming data into an indecipherable code that can be read only by those who have been granted access to the matching decryption key. This work describes a straightforward solution to the durability problem encountered by secure resistive main memory that uses far less power and space than existing approaches. This approach makes use of the random qualities of the AES-encrypted data in conjunction with a rotating shift operation to achieve its goals. The Random Shifter is a straightforward hardware implementation of a very effective energy-saving technique. Its size is much lower compared to other previously presented approaches. Together with other forms of error correction, the Random Shifter technique is employed to keep data in memory safe. When compared to other methods, Rand Shifter has a lower power need. This new pipelining technology reduces power consumption and processing time in comparison to AES without the optimization. This article begins with an overview of the topic at hand before moving on to define the problem, explore possible solutions, and conclude with some findings from using a pipeline approach to AES. These days, cloud users worry more about having their data hacked than they did even a few years ago. The data is encrypted using the symmetric cryptographic methods Data Encryption Standard (DES) and Advanced Encryption Standard (AES). It is difficult to design a hardware architecture that uses a combination of AES and DES (Hybrid) efficiently, especially when working with limited hardware resources. The hybrid algorithm performs many changes on the input data to produce a safer dataset. Memory stuck-at-faults are caused by endurance problems in the primary memories that hold the encrypted data. This article describes a straightforward, low-power, low-space solution to the problem of memory weariness in safe resistive main memories. A random shifter methodology with an error correcting mechanism for the memory was used in the suggested work. Fpga Arrays are often used for the hardware of cryptographic algorithms because of their programmability. Implemented on the VIRTEX FPGA, the suggested random shifter technique is efficient and well-suited for hardware-critical applications involving AES and DES (Hybrid) data. Model sim & Xilinx 14.5 version are used in the development of this paper.

ABSTRACT. A VLSI implementation of the Reordered Normal Basis (RNB) finite field multiplier is implemented in this paper. This multiplier uses RNB, which is the type-II Optimal Normal Basis (ONB), to perform multiplication. The Critical Path Delay (CPD) is influenced by the XOR-AND-XOR (XAX) module of this multiplier. This block is designed in various logic styles, including, Static CMOS logic, Pseudo NMOS logic, Domino logic, Domino keeper logic, and NP Domino logic. The Tanner EDA 45nm tool is used to design the multiplier using the Full-custom design. The major goal is to determine the optimum logic style that meets the VLSI optimisation requirements like the area, multiplication delay, CPD, power dissipation, Area-Delay Product (APD), and Power-Delay Product (PDP). When compared to other logic styles, the delay and area of the multiplier employing NP Domino logic are lower, whereas the power dissipation is similar to other domino logic styles. Also, the architectures of Serial-In Serial-Out (SISO), Parallel-In Serial-Out (PISO), Serial-In Parallel-Out (SIPO), and Parallel-In Parallel-Out (PIPO) multipliers were implemented to analyse their efficiencies.

ABSTRACT. As the technology is increasing day by day, customers are expecting high performance and low power devices. Though many researchers are working under sub threshold operation still there is a demand for low power circuits working under sub threshold or near threshold by reducing leakage power. There triggers a need of low power high computational speed and performance. Therefore there is a demand of circuit design that operates under the threshold region. Though low power circuit designs are attractive, the limiting factor involved here is leakage power when circuit is operating under sub-threshold operation. However, this logic is more sensitive to temperature, process variations and the supply voltage as current is exponentially related voltage in sub- threshold region. i.e. (Vgs<Vth). The novel design techniques proposed here are Source Coupled Logic, GDI SRAM to evaluate the functionality, performance of SRAM Design. Some of the applications like pacemakers and wrist watches to evaluate the health conditions in specific to bio-medical applications.

ABSTRACT. Rounding is mostly performed on floating point numbers. Rounding can be made simpler by using HUB format. The HUB based floating point adder is termed as Architecture A. But some bias may be introduced by using the HUB format. Bias occurs while rounding post truncation, when all the bits to be discarded are zeros. This scenario is called a tie condition. A tie condition occurs while rounding in floating-point related operations. The architecture to partially eliminate the bias is proposed in this paper by making slight modifications to the existing HUB floating point adder. The proposed architecture is termed as Architecture A+. FPGA implementation of three architectures - 64-bit conventional adder for floating point numbers, Architecture A and, Architecture A+ have been done. FPGA implementation results of partial unbiased HUB adder for floating point numbers is compared with 64-bit conventional adder for floating point numbers. The results show that the delay has decreased from 0.197ns to 0.182ns (8.24%), power has reduced from 0.94W to 0.86W (9.3%). Along with these the number of LUT’s used have decreased from 2894 to 2315 (25%). Hence, the partial unbiased HUB adder is better than conventional floating point adder.

ABSTRACT. The classification of fetus expressions is done to recognized mood of the fetus. The emotions of unborn baby facial actions is a progressive process related to brain growth. From the 4D scans of unborn baby, showed that they develop expression at 24 weeks by making very simple one-dimensional emotion, such as moving their lips to complex emotions such as “pain” expressions. We created the landmark points of all images and then it is possible to compare the fetal images with landmarks. We have data of fetal images of various expressions, we created landmark points of face considering eyes, nose and lips corner points by using matlab 15.We propose dark channel prior algorithm for fetal image enhancement.We developed a multi-layered feature extraction system using Local Feature Detectors and Descriptors like harris key-points, Speed Up Robust Features (SURF) and compaired result by Gabor features extraction method. Artificial neural network is used to accurately classify the mood of the fetus. Also the relation between fetus age and expressions are studied, which can be used to give another index of the health indicator of a baby in womb to the doctors.

ABSTRACT. The main goal of this paper is to design LPI radar waveforms and compare their performance. Reduced side lobes, low average transmitted power, low radar noise and temperature, and, most crucially, wide bandwidth are all properties of the LPI radar waveform. The long pulse bandwidth can be expanded by modulating the signal using frequency or phase. We apply a pulse compression filter at the receiver to compress the signal. We apply pulse compression techniques such as linear frequency modulation and polyphase modulation in this paper. The peak side ratio and integrated side lobe ratio are calculated using the autocorrelation and ambiguity functions of these waveforms. In this study, we examine the performance of LPI waveforms using the peak side lobe ratio of the ambiguity function and the integrated side lobe ratio of the autocorrelation function.

ABSTRACT. As a result of technology advancements in the signal processing industry, radar performance has improved during the previous few decades. The RCS of an object is a measurement of how much energy it reflects back to a radar. The radar cross section (RCS) of simple and sophisticated objects is investigated because it is critical for identifying targets and improving or decreasing their radar visibility over a wide range of frequency ranges. To determine the RCS of any object, first learn how the RCS parameters change as the aspect angle changes. We simulate the RCS to represent the required objectives under various situations such as angle of view and frequency range in this project. This method is developed to assess radar overall performance. After then, the radar target's data is simulated. Swerling models were used in this project. First, we create and test the simulation's ideology. In order to potentially , successfully implement these models on a radar system, they are simulated in MATLAB and Xilinx. In MATLAB, Radar Cross Section signatures are detected for further improvement in order to find accurate RCS. The objective of this research is to provide an overview of the conceptual foundation and technical approach to RCS prognosis for readers.

ABSTRACT. To Obtain a low complexity nonrecursive fractional Hilbert transformer with a narrow transition band and small ripples explored. The method relies on merging Frequency Response Masking (FRM) and, Frequency transformation (FT) approaches. The frequency response masking approach is a simple multiplierless subfilter that can be obtained by rounding off. On the other hand, the Frequency transformation approach has the benefits of using an identical subfilter repeatedly as the overall architecture. An efficient method is shown by utilizing examples that the merging FRM-FT can achieve reduced computational complexity than earlier methods. The FRM-FT can be of a relatively small length of sample filter because of its ripple size. The final design is less complex compared to the direct approach.

ABSTRACT. COVID – 19 had put the whole world at standstill, however, nowadays its effect had been gradually slowing down. It’s been around four years completed since it was first observed. Until now many people had been already vaccinated and people had developed immunity against the COVID-19. However, there had been still variants of the COVID-19 which are getting mutated and spread. It is necessary to diagnose patients with the COVID -19 as early as possible. Different methods are available for the prediction among them chest x-ray is one. From the study, it is found that chest X-ray had distinct features as compared to the other, so deep learning-based methods were used to identify that. But those methods take time to train the model, which can be very critical if new variants start to spread again. And its detection should be made as quick as possible. For that reason, in this study, a COVID 19 chest X-ray classification is performed using the Long short-term memory (LSTM) neural network and Convolution Neural Network (CNN) model which is based on the Graph with Fast Localization spectral filters (GFL). The GFL model can be trained in couple of iterations which is necessary. Due to this the prediction can be done as early as possible but the testing accuracy of GFL model is low as compared with the LSTM model.

ABSTRACT. Brain injury is the greatest cause of death worldwide, according to a WHO (World Health Organization) research. Deep Learning is one of the most often used algorithms for segmentation of the brain, and also it can be used in conjunction with MRI, CT, PTE, and other modalities. Many anatomical properties, such as shape, size, and also the structure of the labels of distinct organs of humans, are not relevant to the suggested framework. When compared to state-of-the-art procedures, the developed frame may be studied using various functions such as classification and enhancement of image, and it also helps to improve the level of prediction. We should have a basic understanding of the organs before beginning an in-depth inspection. Using CNN (Convolution Neural Network) which is also known as ACNN (Anatomically Constrained Neural Networks), the algorithm that is proposed leverages semantic segmentation notions. Throughout brain imaging, this approach will develop and partition parts of the organ. To classify portions of an organ, various existing algorithms (such as NN, SVM, and KNN) are utilized, however, these methods fail if the input image is noisy. The method we proposed will segment the organ from brain images with the highest accuracy (CDR: Correct Detection Rate) and the highest indication of similarity. Through objective analysis and purpose, we have shown that our approach is more effective than existing strategies.

ABSTRACT. By identifying the landslide-prone locations in a chosen research area, Landslide Susceptibility Mapping (LSM) assists in reducing the danger of landslides. This article presents the LSM prepared for the state of Goa using Weight of Evidence statistical method. In the study area, 78 landslides were found and randomly divided into training landslides (80%) and validation landslides (20%). Ten landslide conditioning factors- slope, elevation, total curvature, plan curvature, profile curvature, annual rainfall, Stream Power Index (SPI), Topographic Wetness Index(TWI), distance to road and aspect have been used in preparing the LSM. The final map of landslide susceptibility has been classed into susceptibility classes of very low, low, moderate, high, and very high. It is found that, the eastern and southern portion of Goa includes areas with high probability of landslide occurrence. Area under the ROC curve has been used to validate this susceptibility map. This model’s validation outcome revealed testing accuracy of 71%. The finding of this study helps in identifying the landslide prone locations of the study area.

ABSTRACT. Low probability of intercept (LPI) radars are one of the most significant radars in radar technology because of their functionality. It has certain features such as low power, wide bandwidth, etc. To accumulate this, the LPI radars use a particular form of changed wave bureaucracy, which might be tough to intercept. The precept obligations of intercept receivers are to discover, estimate, and classify LPI signs even in the presence of excessive noise. Additionally, the estimation of the modulation parameters additionally offers statistics approximately the danger to radar so that essential counter measures may be initiated towards the enemy radar. In this mission, we initially describe the simulation/contour plots of poly time code indicators (T1, T2, T3, T4) and then the main intention of the task, i.e., the class of poly time codes are effectively labelled. All the four types of Poly time codes are classified using python in anaconda navigator and jupyter notebook. The algorithm used here involves image pre-processing, segmentation, and feature extraction, whose output is fed to an XGBOOST classifier used to classify polytime code images into T1, T2, T3, and T4. In this proposed method, the outcome of testing is very high, which is 89% (accuracy) and the AUC score is 92%, which is an appreciable value.

ABSTRACT. Abstract— The need for safe large data storage services is at an all-time high and confidentiality is a fundamental need of any service. Consideration must also be given to service customer anonymity, one of the most important privacy considerations. As a result, the service should offer realistic and fine-grained encrypted data sharing, which allows a data owner to share a cipher text of data with others under certain situations. In order to accomplish the aforesaid characteristics, our system offers a novel privacy-preserving cipher text multi-sharing technique. In this way, proxy re-encryption and anonymity are combined to allow many receivers to safely and conditionally receive a cipher text while maintaining the confidentiality of the underlying message and the identities of the senders and recipients. In this paper, a logical cloud security scheme is introduced called Modified Data Cipher Policies (MDCP), in which it is a new primitive also protects against known cipher text attacks, as demonstrated by the system.

ABSTRACT. The recent arrival of Smart Grid has influenced the creation of Big Data in the field of energy consumption. Smart Grid provides with an efficient and reliable end to end two-way delivery system. The electricity consumption for an individual user is available at real time. Also, enables real-time monitoring and controlling of power system from utilities perspective, which helps them reducing the power losses. So, it is going to replace the old methodologies of sharing electricity, to fulfill the exponential needs of electricity in terms of flexibility, reliability and quality, etc. The aim of the study is to find the possible number of appliances used by an individual user and accordingly find the star rating of the individual appliance. Further will be informing user to upgrade specific appliance with a lower star rating to a higher star rating. This would eventually help appliance development companies with respect to targeted advertisement. Also, the cost to the energy consumed would be reduced. So, it is going to help users to save their money on energy consumption and would also help to save energy. This would eventually lead to a healthy environment for the conservation of energy.

ABSTRACT. Neuroscience studies mental health and illness extensively, from both clinical and non-clinical viewpoints. Psychiatric diseases, such as schizophrenia, depression, and autism spectrum disorders, are associated with atypical facial expressions. Emotional expressions and the detection of mental illness can be significant in diagnosing and treating mental health issues. An early diagnosis of an illness can be made using OpenCV facial recognition and CNN feature extraction and classification. With the help of early diagnosis, there would be a lot of benefits. Deep learning algorithms in neural networks like CNN, and RNN plays a huge role in such people's life by diagnosing their illnesses.

ABSTRACT. The quick advancement of cloud computing has brought in an increment in the number of businesses and people who are using the public cloud to store and exchange data. In order to maintain the privacy of the data stored, the owner of the data encrypts the data. This enables the data to be accessible to only specific users who can decrypt it and are authorized to do so. A severe difficulty occurs when this encrypted data has to be shared in a vehicular ad-hoc type of network with people. To address this problem , we have introduced a hybrid protocol which includes Identity based encryption which uses public key cryptography and for broadcast using LCM which uses Symmetric key cryptography. We have designed a more efficient and lightweight protocol based on Bilinear maps and also proved its correctness.

ABSTRACT. The growth of the online social networking (OSN) platform has been huge in recent decades. Apart from OSN platforms, online dating websites are gaining attention day by day. People can now meet their potential life partners more easily due to the rise of online dating platforms. However, it has made it easier for fraudsters to take advantage of this growing market. Online social networking fraud ranges from spamming to human targeted fraud, in which the fraudsters build fake relationships with the victims and demand money. This paper provides an extensive review of online dating fraud methodology, global impact on users, different dating site features, and recent statistics about the rise of online dating users and online dating fraud. In addition to this, we have analyzed fraud prevention methodologies, taxonomies of scam detection, and scammer detection techniques using machine learning. Finally, this survey discusses open research questions, challenges, and important online dating safety guidelines.

ABSTRACT. Social media is increasingly becoming a window to the user’s personality. Hiring the right candidate is a formidable task for any organization and particularly in the highly competitive software industry. This paper presents a machine learning and natural language processing based system to leverage social media to assess job applicants for their suitability for a given job. We use LinkedIn profiles to assess the technical suitability and combine Twitter posts with them to assess emotional intelligence of the applicant. The system thus provides an indication of both the technical and soft skills perspective of the job applicants. The system can be used by both the prospective employers and employees. Employers can use it to shortlist job applicants and the prospective employees can use it to evaluate their chances, retrospect, and take any corrective action. The results from the created system are encouraging.

ABSTRACT. E-commerce applications allow merchants to sell the good and consumer to purchase the variety of products. Electronic transactions (e-transaction) used in e-commerce sites currently send the encrypted bank account information in text format using Secure Electronic Transaction (SET). The existing payment transaction contains encrypted information of the bank card information, the security code information, and the order information in the form of text. Quick Response code (QR code) is an image which hides the information in a 2-dimensional (2D) barcode like in steganography technique. This information can be about the product, website, payment, tracker id that links to an application. The static QR code contains only one fixed data which can be hacked in cyber-attacks. Dynamic QR code contains the unique code for the transaction created on the order creation. This paper proposes a secured two-layer mechanism for e-transactions. It proposes to use dynamic QR code as a payload for the payment transaction. The first layer is to enclose the payment information in dynamic QR code. The dynamic QR code is created with the bank information, user information, secure code and order information which is unique for an order. Second layer of security is encryption of this QR code using Secure Electronic Transaction (SET) transactions to secure the payment transaction. The receiver decrypts using the unique public key the payment information which is a 2-D QR code. This image data is decrypted to retrieve the bank information, code, user information and order information. Dynamic unique code per order which would enhance the security to reduce the vulnerability to the cyber-attacks. The two-layer secure payment transactions enhance the security two times by using dynamic QR code image as payment transaction.

ABSTRACT. In this paper, both small cell and Wi-Fi Access point conjointly deliver data to the multiple users in indoor environment. Further, to deliver the augmented data rates, small cell operates in both licensed and unlicensed bands. As a result, small cell generates interference to Wi-Fi Access Point users in unlicensed bands and to macrocell users in licensed bands. However, this produced interference is to be restricted. Moreover, total data rate rendered from both the small cell and Wi-Fi Access point is intended to be above the required limit. Thus, to comply with all these requirements and also intending to optimize the total imparted data rate, optimal power allocation is done to the Resource Blocks (or channels). While solving this optimization problem, the novelty of this paper consists of identifying the initial lagrange multipliers of the ‘Iterative Water Filling Algorithm’ (instead of considering the random values, as was carried out precedently). Finally, simulation results reveal the optimized results.

ABSTRACT. Massive multiple-input multiple-output or massive MIMO, is an expansion of MIMO that brings several antennas at the transmitter and receiver to enhance throughput and spectrum efficiency. Because of its potential to increase the volume of antennas, this technique has become an important part of wireless standards. Nevertheless, as Massive MIMO works with large number of antennas it requires progressive detecting techniques. Many linear conventional detecting techniques like Zero Forcing and MMSE techniques gives optimal BER performance. But both techniques gives rise to high computational complexity because of exact matrix inversion operations. Many iterative inversion methods like Nuemann Series, Newton Iteration method and Richardson methods were introduced to reduce that complexity. Paper proposes a fast convergent Richardson method which gives the near best BER performance as that of MMSE in less number of iterations.

ABSTRACT. The sixth-generation (6G) of mobile cellular networks is supposed to integrate the competence to preserve new and anonymous services. In order to improve its competence by 10–100 times compared with 5G, 6G should also be quick and open to adapt to the ever-changing services. In 6G, Device to Device communication (D2D) gives the high data rate services and thus improves total system throughput. But, the energy efficiency during the D2D communication was not improved with less latency in previous methods. This paper aims to improve resource aware D2D communication in 6G system using Deep Neural Regressive Tangent Transfer Classifier (DNRTTC) model. DNRTTC model includes different types of layers such as input layer, hidden layer and output layer for achieving resource aware D2D communication. The input layer obtains the number of mobile devices as input and it is fed into the hidden layers. In hidden layer one, energy, received signal strength and connection speed of each mobile device is computed. The results of hidden layer one is transferred to the hidden layer two where the regression analysis is employed to analyze the mobile device with their threshold. Then, these results are given to the output layer. In that layer, tangent sigmoid transfer activation function is employed to detect the resource efficient mobile devices. From that, the higher energy, received signal strength and connection speed of mobile devices are chosen for better communication in 6G. The device to device communication through the selected devices enhances the data delivery rate with less minimal latency. The simulation results verify the superior performance of the proposed model in 6G network compared with some benchmarks methods in terms of energy efficiency, data delivery rate, data loss rate, latency and throughput.

ABSTRACT. X-ray machine systems are used to acquire X-ray images of internal human body parts. A C-arm is a type of X-ray machine system used to produce X-ray images of desired body parts. The creation of X-rays is essential to the development of X-ray images. An X-ray generator built within the C-arm system produces the X-rays. The Master Controller is the brain of C-arm systems, which monitors and controls all the components of the C-arm system. Communication between the Master Controller and the X-ray generator is required in order to regulate the various functionalities of the X-ray generator. Using an actual X-ray generator to test the C-arm system's performance can be expensive and time-consuming. Thus, it is necessary to create a simulator that may take the role of an X-ray generator. In this research, a novel method for creating an X-ray generator simulator that uses an Arduino board in place of the X-ray generator is discussed. The Arduino board is programmed to perform like an X-ray generator, and communication between the Arduino board and Master Controller is obtained to simulate the functionalities of an X-ray generator.

ABSTRACT. Vehicular Ad Hoc network-VANET plays a crucial role in daily vehicular users and ensures safety for drivers and passengers. But there are some issues before VANET can be utilized. One of those issues is spectrum scarcity i.e., some vehicular users have insufficient spectrum which is used for their communication. In the United States of America, the Federal Communications Commission (FCC) allocated the 75 MHz in the 5.850-5.925 GHz band covering 7 channels with 10 MHz each and 5 MHz reserved as a guard band. So, to overcome the above issue we have Dynamic spectrum access (DSA) via Cognitive radio technology. CR main function is to find the spectrum holes in licensed frequency channels so that unlicensed spectrum users can utilize the unused spectrum. But due to their high speed and topology there are some challenges for spectrum sensing which are as follows .At first to develop effective spectrum sensing solution for vehicular networks we present Existing data for complete classification of CVN spectrum sensing techniques and next we talk about Results of the automotive environment like speed, traffic density fading on spectrum sensing ,at last we proceed to a set of prerequisite for CVN spectrum techniques that gives specific cognitive vehicular networks environment characteristics.

ABSTRACT. Abstract— Wireless Sensor Networks (WSN’s) consist of sensor nodes spread over a wide geographical region and are equipped with batteries for power supply. Sensor nodes are deployed to capture the information in the region of significance. Low sensing power may drain the battery quickly, reducing the network lifespan. Improper placement of sensors leads to low sensing power. To enhance the coverage rate of the sensor network, we proposed a solution based on a metaheuristics algorithm called the Harris Hawk Algorithm (HHO). HHO is executed at the base station to determine the location of sensor nodes. HHO is then executed at the sensor nodes to determine the coverage rate of the network based on the probability sensing model. The proposed approach is evaluated for coverage efficiency, sensor costs, and stability. The results validate the applicability of HHO for solving the WSN coverage problem. Simulation results substantiate that HHO outperforms other techniques in achieving a better coverage rate.

ABSTRACT. Cryptocurrency is a kind of virtual currency that came into existence with the recent advancement of technology in finance. It is used to complete transactions in a secure way by using the techniques of cryptography. This virtual currency is created with the help of block chain technology. In many countries, the transactions using cryptocurrency are not legalized by the banks. Some of the most popular cryptocurrencies are Bitcoin, Dogecoin, Litecoin, etc. The value of each cryptocurrency keeps varying from time to time. In this project, we build a data analytics model of the various cryptocurrency and also a machine learning model using the LSTM(Long Short Term Memory) algorithm to forecast the value of a certain cryptocurrency on a particular day. We make use of a web application called Yahoo Finance(yfinance) which has all the details of the live stock market and this acts as a source of dataset for our project. We use various python packages such as numpy, pandas, tensorflow, seaborn,matplotlib,etc for building a model. The LSTM algorithm makes use of RNN(Recurrent Neural Network) which is powerful to model data sequencer because it has an internal memory state to store the past seen data. We implement various layers of the LSTM to determine the price of a cryptocurrency on a particular day.

ABSTRACT. Taking ineffective and ineffective medicine can have serious consequences for our health. A incorrect or overdose of medicine can have serious consequences for our health, with long-term or short-term consequences. Human tendency involves taking medications based on marketing or knowing that someone we know has taken them. It's completely erroneous, so in order to circumvent this we have designed cutting edge technology and science model named “Automatic Voice Command Drug Dispenser”(VCMD). This proposed idea of an automatic voice command medicine dispenser is a technical and pharmacal device which is designed to dispense the medicines when the commands are given in the form of voice and the medicines are collected by the patients. There are numerous challenges which may influence the individuals when VCMD was not in utilization. The project VCMD can control the medicine dispensing, checking heart beat and temperature sensing to the patient etc. Here we using Arduino mega to can control all these operations. VCMD uses so many ways like military, old age homes. To continuously check the VCMD status, the Arduino Mega board to collect data from sensors, then it checks the data it displays on LCD. This research paper aims to replace manual maintenance of medicine dispensing with an automated system by using Arduino.

ABSTRACT. Remotely operated vehicles (ROVs) are underwater robots that are commanded from the surface by a person. This robot is made up of embedded systems and mechanical systems, and it is connected to the outside world via a series of load-bearing umbilical cables that hold power, data, and communication cables. For the ROV, data transfer as well as control signals are required. The navigation team was made up of working on the ROV's navigation system. It has a video camera and a ballast system for balance, movement propulsion system pumps, and IMU to locate the position of a vehicle microcontrollers to compute, process, and offer the navigation system, sensor for feedback data back to the lighting and components controlling and coordinating all of this is a challenge. A manipulator, subsea equipment, and devices to assess clarity, temperature, and depth are all included in the proposed task. The purpose of the proposed effort is to create such a ROV to execute a given mission involving several persons. In addition, a novel form of ROV attitude control is introduced, which makes use of floats to modify the ROV's centre of buoyancy. This is an interdisciplinary endeavour in which the members are from many fields. Together with the mechanical department, the electronics navigation team worked on this.

ABSTRACT. According to some estimates, 50 percent of accidents are caused [1] by poor road conditions and the failure to wear a helmet. A smart interactive robotic helmet with features such as a traffic adaptive music player, wireless bike authentication, and road hazard warning playing can be offered to prevent accidents and encourage people to wear helmets. When a traffic hazard is coming, the helmet will inform the rider, communicate with him if he is not wearing it, and perform wireless bike authentication to avoid theft. Because two-wheelers are used more frequently in India than four-wheelers, it necessitates chevalier care in terms of safety. In today's culture, safety, like security, is extremely vital. The goal of this anti-theft mechanism system is to create a vehicle safety and security system that is implanted integrating and altering existing components. This system is primarily comprised of three modules: gas sensing, obstacle detection, and anti-theft warning system, all of which are connected to an ATmega16 microprocessor. This paper presents an overview of a smart helmet having various security and safety-related features.

ABSTRACT. The impact of Covid around the globe is imperative all around the world. Though the pandemic had glaring effects on the market, sectors have been showing a reviving trend post-covid. This is possible with the right mapping and affective forecasting. The majorly hit sector was the hotel industry, and situations showed a positive surge with potential sales incrementing post covid. Hospitality industry in particular is much prone to act post covid after easing of travel restrictions and social distancing. Sustainability seems to be the difficult arena with many hotels having low profile financial conditions. The aim of the research is to understand the potential indicators in hotels in India which are expected to boost the sales and sustain the sector. The authors have used a combination of literature review to identify the potential indicators through past studies. Further the qualitative opinion polls, focus group and interviews were conducted to evaluate the indicators and their comparative ranking in the industry. Data were gathered through hotel employees representing various domains like Front office, F & B, Marketing, and promotions department of the premium leisure hotel properties. Further the analysis has been employed for classifying the potential indicators with their relative impact, examining their cause-and-effect phenomena. The study has been validated by expert opinion poll supporting the hypothesis tested. Mapping such inclusive indicators; the hotel industry may be well prepared for the coming future. The most significant indicators have been identified and the mutual interdependence between the potential indicators have been seen through the analysis. will be highly beneficial for the hotel industry to have strong strategic preparation post covid. It also shows that estimating the critical elements at the right time can lead to vowing revenues, saving the industry. The study has practical application for promoting the hotel industry.

ABSTRACT. Now-a-days the concept and the use of Internet Of Things is gaining huge popularity with increase of smart cities. To increase the productivity and reliability of urban infrastructure consistent development is being made in the field of IoT. The population in the smart cities is huge and most of the people living in these smart cities own their vehicle. Due to the limited parking facilities problems such as traffic congestion is being continued in these smart cities. Due to this people waste their time in finding the parking slots. Also while parking the vehicle in multi complex areas people will be charged to park their vehicle. During their exit they should pay the amount charged for parking their vehicle and here with the use of physical money the payment process gets delayed and hence it leads to the traffic congestion. In this paper, an IoT based smart parking system with E-ticketing was proposed. Here, In this parking system we are using Arduino UNO as the processing unit and RFID cards to identify each vehicle individually and deduct the charge for the parking before they enter into parking area. Only if there is sufficient amount in the account of that particular vehicle owner, it will be deducted and a message will be sent to their mobile phone and the gate will open to park their vehicle. Also the slots that are available for parking will be shown on the display so that the user can directly head towards that slot without wasting much time. By this we can minimize the time that is being wasted by the user in finding a vacant parking slot to park the vehicle.

ABSTRACT. The project's primary goal is to ensure kid protection monitoring. Pre-defined navigation is one of the primary challenges in the robotic industry. Many technologies have been developed to overcome these problems. In this project a night vision camera is equipped with the robot and the robot moves in the pre-defined path captures the image and video of child activity, and transmits the data to the main system. Robot is also equipped with the PIR sensor to monitor any other human face detection, and sound sensor for cry detection. sound sensor detects the crying of the child and gives alert to the parent.

ABSTRACT. This article compares MEMS capacitive accelerometer springs. This study designs MEMS single-axis accelerometers with alternative suspension systems. Structure analysis for 2kHz resonance. Comparing these devices shows that certain criteria must be compromised to satisfy requirements. Higher displacement sensitivity reduces mechanical stability, and vice versa. COMSOL Multiphysics runs the simulations.

ABSTRACT. The logistics sector may be made or broken by two factors: time and location. The performance of the logistics sector is determined by how quickly things can be delivered to a certain location or client. Technology, integration, and globalization control the logistics business. Second, we build a mobile elevator-conveyor system that is mounted between two movable frames. CASE STUDY: We will look at the logistics business, which involves moving items from one location to another. Ladder logic is built here for obtaining things from one target shelf. SYSTEM PROPOSED: The primary goal of this system is to automate the whole warehouse/logistics business, which may be accomplished with the help of PLC and SCADA. The commodities are transported from the main conveyor to the elevator, which transports them to the destination shelf. The items are stored on their intended shelf thanks to conveyor transfer. All of these functions are handled by the PLC. The contents from that shelf are now being moved to the elevator conveyor. The elevator then returns to the main conveyor, which transports the products to the appropriate vehicle through the main conveyor. A moveable elevator-conveyor arrangement controlled by the PLC does the task of storing and retrieving things from various locations automatically. Information, transportation, inventory, warehousing, material handling, packaging, and, in certain cases, security are all integrated into logistics. Following that, the elevator and shelf conveyors begin to run. CURRENT SYSTEM: In any business, there are several techniques for managing goods and materials. Plant simulation software can now model, analyze, visualize, and optimize the complexity of production logistics, but it is always evolving.

ABSTRACT. In order to improve healthcare management, it is important to track health outcomes. For medical researchers, machine learning techniques have become a popular means of making accurate predictions. Machine learning techniques can identify trends in large amounts of dataset, and capable of predicting cancer accurately. In the current medical research, medical imaging plays a significant role in thorough examination and diagnosis of the entire human body. Medical professionals rely entirely on computed tomography results acquired from the image sensors. Lungs are the most primary organ of the human respiratory system. Medical professionals face a challenging task in accurately predicting lung cancer. Detecting the lung cancer aids in determining the appropriate treatment, which increases the chances of survival for the lung cancer patient. In this project the lung cancer dataset is taken as input, the cancerous images are divided into five clusters based on the Features obtained using K-Means clustering. Now, detection of cancer cells can be done by using transfer learning. In this project Densenet121 model in transfer learning is used to identify whether the cancer cells are present or not. If the CT scan image is cancerous then the treatment associated with the type is suggested to the user. This would help the physician to guide the patient on whether to take surgery or to take other kind of treatments. This is also used for Insurance companies. In this project, Image processing, transfer learning, and K-means clustering are used to cluster the cancer images based on features and to identify whether the image is cancerous or not and to suggest treatment.

ABSTRACT. Pavement Health Monitoring has become a critical component of pavement maintenance and rehabilitation. To account for this, IoT (Internet of Things) sensors and low-cost connectivity have made it possible to remotely monitor and process data from far areas. Data from sensors can be read by IoT devices and transmitted with great accuracy. The Structural Evaluation of Flexible Pavements is proposed in this study using a Raspberry PI 4 model B, Piezoelectric (PZT) Sensors, DAC (Digital to Analog Converter), an ADC (Analog to Digital Converter) Python is used to run the algorithm on the Raspberry Pi. The Transducer sends a sinusoidal (Digital) wave, which is converted to an Analog by the DAC. The ADC converts the receiver's response signal. PZT sensors extract responses from the Pavement Structure, and a method is proposed to determine pavement strength. Initially, Responses are collected first from the traditional methods such as LWD (Light Weight Deflectometer), BBD, (Benkelman Beam Deflection) MR, (Resilient Modulus Test) British pendulum Tester, Roughness (Merlin) later responses are collected from developed IoT Module and correlated using soft computing techniques.

ABSTRACT. Over the years the number of road accidents increases dramatically with injuries. Currently, there are several safety measures in place to ensure driver safety such as awareness programs, penalties as well as celebrity campaigns which have a significant impact on reducing this number. Many accidents occur as a result of alcohol abuse and misalignment. Another major cause of road accidents is drowsiness caused by the driver while driving for long hours. A lot of research has been done on monitoring the condition of drivers in cars. By using measures, driving conditions can be monitored. In the present case, an independent vehicle control system was proposed to control the speed of the vehicle in the event of a defect in the driver's body parameters.

ABSTRACT. The progress of the nation ultimately depends upon the transport system it has. Many means of transport have been adopted but the roadway network plays a crucial role in all. The widespread of road connectivity to various cities and towns created a tough task for researchers in developing the road system. In this regard, effective experimental studies are in progress with respect to flexible and cement concrete pavements. This research paper holds a good promising attempt on cement concrete pavement by assessing its condition with the help of wide sensor network. An integrated platform of IoT for strength assessment of cement concrete samples is developed by using Wi-Fi module, Raspberry pi, ADC &DAC converters in also with (piezoelectric) PZT sensors. Testing of concrete specimens are strongly bonded with PZT transducers which acts as an actuator &sensor respectively. Sinusoidal wave is developed through an actuator simultaneously responses are recorded with the help of sensors which can be converted to digital form with the help of ADC medium. The entire collected data is analyzed by using MATLAB software.

ABSTRACT. Liquid filling is an essential feature in many industries such as pharmaceuticals, chemical, food and beverage, paint etc. The purpose of the paper is to design, implement and monitor a completely automated low-cost system for filling liquids from a main tank into bottles of different sizes. The size of the bottle is identified with the help of a vertical arrangement of three proximity sensors. The filling of the liquid is done with solenoid valve and timers are used to control the f illing time to attain maximum accuracy. The entire process is automated with PLC and no operator is needed during the functioning of the system.The sensors and solenoid valve are periodically tested automatically so that the efficiency of the system is maintained and no error occurs while filling. The control, monitoring and data acquisition is done with the help of SCADA. This enables the starting and stopping of the system through a computer system (local or remote). SCADA helps to transfer the data from the controller to another database for documentation purposes. Dynamic visualization is also made possible with SCADA. Since the entire system process is connected to SCADA via communication hardware, all the features done by the local SCADAsystem can also be accessed from a remote location using a remote desktop software with free of cost. When compared with other existing models, the proposed system has the additional advantages such as low-cost automatic size identification system based on height, dynamic data updating and monitoring, low-cost remote access and automatic error detection.

ABSTRACT. The automotive industry is rapidly growing every year, and we can see the automotive industry leaves more room for development and innovation. Nowadays, automation is needed to reduce human work. The in-car climate directly affects the driver, fatigue-free driving, and driving safety. To maintain the ambient temperature for the driver in the car, cars are fitted with an HVAC system controlled by the climate control unit (CCU). This paper will see test automation for CCU Panel using CAPL (Communication Access Programming Language). This paper gives knowledge about CAN protocol, HVAC (Heat Ventilation and Air Conditioning), CAPL scripting for panel logic module, and testing methods. The understanding of the system and its requirement for the development of CAPL test script for the automation in testing. For proceeding next, the CANoe setup for testing of CCU panel was designed. Previously manual testing is performed to test the ECU with the help of test cases, but it consumes lots of time. So here, we introduced a test automation script for our module. Using CAPL scripting, we automated the test cases, debugged them, and got the test report of pass and fail test cases. Here we can observe that manual testing takes almost 10 to 12 hr. to perform all test cases, but because of automation, it is done within 6.16 minutes

ABSTRACT. Internet of Things (IoT) enables many things like car, dustbin, and gas cylinder to be connected, communicated, and controlled. Furthermore, advances in Internet of Things makes home automation smarter. In the contemporary connected world it is highly required and comfortable to get a timely information which is priceless as it gets things accomplished smoothly and avoids many difficulties. Internet of Things, being the state of the art technology, features many capabilities such as facilitating a timely information. Many low power wireless devices and communication technologies are making Internet of Things based solutions available at a large scale. Internet of Things featured devices like ESP8266 and ESP32 become more popular to be end devices for the IoT network. The wireless capability of ESP8266 can be exploited to use it as a smart device to get data/information that can be utilized for applications such as a home automation. In this paper, an ESP8266 based weather forecaster prototype is implemented which communicates with an Open Weather Map web API to bring weather forecast information to assist people at home. In addition, the firmware for ESP8266 is developed using a platform called ‘Mongoose OS’ which efficiently provides lightweight OS/firmware for resource constrained devices such as ESP8266. The presented weather forecaster system is amenable with home environment.