Electrical and Computer Engineering
Sub-communities within this community
Collections in this community
Recent Submissions
-
COVID-19 detection via cough sounds using a hybrid MFCC-DWT based features mining and machine learningIn early 2020 the World Health association (WHO) declared the rapid transmission of the modern coronavirus (COVID-19) a pandemic, the current pandemic associated with the modern coronavirus since then researchers around the world have been working to aid in the diagnosis. Cough is an important symptom in many diseases and at times is the only major symptom to diagnose some ailments and the most frequent reason for seeking medical advice. Optimized care goes inevitably through an adapted recording of this symptom and automatic processing. Some of the coughs are different from each other which could define negative or positive for COVID detection. COVID-19 detection via cough sounds digital storage devices and sound sensors make it portable and accurate to record cough sounds, it would make a change in the computer technology and the availability of portable digital sound recording devices. In this context, this work focuses on the study, design, and development of an effective approach for COVID-19 detection via cough sounds-based features mining and machine learning. The objective is to achieve an effective solution with an improved level of precision compared to the existing counterparts. It can be done by smartly employing the hybrid features extraction and the robust classification techniques. The incoming audio segment will be enhanced by applying the appropriate preprocessing. The Mel-Frequency Cepstral Coefficients (MFCCs) and the discrete wavelet transform (DWT) will be extracted from the enhanced audio segment. Later appropriate lead based robust classifiers will be utilized to lead these extracted features with the reference database. The comparison outcomes will be used to make the classification decision. The classification decision will be transformed into a systematic signal wave sign. The system functionality will be tested with the help of a proposed system. The primary results and findings will be presented and discussed. The proposed approach has a potential to be helpful for the medical field to become a restful application for the majority of the people
-
Development and Implementation of pipeline Convolutional Coding using FPGAChannel coding is essential for ensuring reliable data transmission in challenging wireless communications. Improving spectrum efficiency involves leveraging efficient forward error correction (FEC) methods. Viterbi decoding plays a critical role in Convolutional channel coding for accurate error detection and correction, particularly in LTE and Satellite communication systems. This article discusses the simulation and FPGA implementation of a newly proposed non-systematic Convolutional system featuring a block interleaver and 64-QAM Mapping under AWGN and Rayleigh channel conditions. The system adopts a Convolutional coding rate of 1/3 and a constraint length of 7, utilizing a Trellis diagram for encoding and the Viterbi algorithm for decoding with hard decision decoding. Additionally, a pipeline coding approach is employed. Simulations are conducted using MATLAB-R2023b, and the implementation is executed on Virtex 6 (XC6VLX240T) FPGA using Xilinx 14.7. The study reveals that the pipeline technique demands more FPGA resources compared to traditional methods while still utilizing a small resource block from Virtex 6, with 3% and 9% usage of slice registers and LUTs, respectively. Moreover, the system's timing is reduced from 24 to 14 clock cycles, enhancing the efficiency of entirely LUT-FF pairs from 55% to 63%.
-
Optimizing Thermo-Electric Generation with Quasi-Z-Source Converter in Energy Harvesting for Wearable DevicesRecent advances in semiconductors industry and microelectronics have created new opportunities for integrating various technologies in energy harvesting projects. These advancements have simplified the process of capturing energy from different sources, such as thermal energy, and converting it into electrical power. Thermo-Electric Generators (TEGs) are commonly used in many applications such as wearable devices. However Thermo-Electric Generators produce very low output voltages, which is insufficient for most low-power electronic circuits to operate. Therefore, a boost converter is essential to step up the voltage output from the TEG to a usable level. In this paper, a DC-DC converter with a high performance in terms of voltage gain is implemented. The proposed quasi-Z-source (q-ZSC) DC-DC converter not only offers high voltage gain but also several significant advantages. The proposed q-ZSC is capable of both boosting and regulating the voltage output efficiently, even under variable input conditions, making it highly suitable for low-power energy harvesting systems such as Thermo-Electric Generators . The design is simulated using MATLAB Simulink software, focusing on integrated circuits, with a strong emphasis on optimizing the circuit’s efficiency. This work underscores critical factors in improving energy harvesting circuits, enhancing their performance and effectiveness in low-power applications for wearable technology.
-
Braille code classifications tool based on computer vision for visual impairedBlind and visually impaired people (VIP) face many challenges in writing as they usually use traditional tools such as Slate and Stylus or expensive typewriters as Perkins Brailler, often causing accessibility and affordability issues. This article introduces a novel portable, cost-effective device that helps VIP how to write by utilizing a deep-learning model to detect a Braille cell. Using deep learning instead of electrical circuits can reduce costs and enable a mobile app to act as a virtual teacher for blind users. The app could suggest sentences for the user to write and check their work, providing an independent learning platform. This feature is difficult to implement when using electronic circuits. A portable device generates Braille character cells using light- emitting diode (LED) arrays instead of Braille holes. A smartphone camera captures the image, which is then processed by a deep learning model to detect the Braille and convert it to English text. This article provides a new dataset for custom-Braille character cells. Moreover, applying a transfer learning technique on the mobile network version 2 (MobileNetv2) model offers a basis for the development of a comprehensive mobile application. The accuracy based on the model reached 97%.
-
A Novel PAPR in OTFS Systems through ISSA-PTS and ISPR TechniquesIn modern communication systems, managing the Peak-to-Average Power Ratio (PAPR) remains a critical challenge, particularly in Orthogonal Time-Frequency Space (OTFS) systems. PAPR significantly influences the efficiency and performance of these systems, with high PAPR leading to signal distortion and reduced system reliability. This is especially problematic in high-speed mobility environments, such as high-speed railway systems, where OTFS modulation is favored for its robustness in dispersive channels. However, the high PAPR in OTFS systems can compromise the performance of power amplifiers, increasing signal distortion and Bit Error Rate (BER). This study introduces two novel PAPR reduction techniques: a hybrid approach combining the Improved Salp Swarm Algorithm (ISSA) with Partial Transmit Sequence (PTS) and an innovative Iterative Sub-block Phase Rotation (ISPR) method. The hybrid ISSA-PTS approach optimizes phase rotations to reduce PAPR, while the ISPR technique further enhances PAPR reduction by iteratively rotating sub-block phases without compromising BER. Simulation results demonstrate that both proposed techniques significantly reduce PAPR and improve BER performance, making them promising solutions for enhancing the reliability and efficiency of OTFS systems in high-mobility communication environments.
-
Comparative study of DCT-and DHT-based OFDM systems over doubly dispersive fading channelsIn high-mobility operating scenarios orthogonal frequency division multiplexing (OFDM) system lacks its optimality, intercarrier interference (ICI) occurs and the Doppler shifts deteriorate the orthogonality of the subcarriers. However, this problem can be overcome by utilizing complicated equalizers at the receiver. Discrete cosine transform (DCT) and discrete Hartley transform (DHT) have been used instead of discrete Fourier transform (DFT) in the standard OFDM system. The performance results achieved enhancement over dispersive selective channels that cannot be accomplished with a standard OFDM system, even with utilizing complex equalizers. In this paper, the performance of the DCT-based OFDM system and DHT-based OFDM system is compared and analyzed with respect to DFT-based OFDM system performance over doubly dispersive fading channel at various Doppler shifts, the block diagrams of the proposed systems are provided to simplify the theoretical analysis by making it easier to follow. Simulation results emphasized that the performance of DCT and DHT-based OFDM systems under doubly dispersive fading conditions scenario outperforms DFT-based OFDM system of order 3 dB energy per bit to noise power spectral density ratio (Eb/N0).
-
Spline Global MPPT with PID Controller Based on Levy Invasive Weed Technique for Renewable Energy Resources of Smart HomesRenewable energy systems, particularly photovoltaic (PV) systems, have been played important role in the reducing carbon emissions. A primary concern in the field of photovoltaics (PV) based on the design of the maximum power point tracking (MPPT) is the capacity to accurately monitor power across many parameters in addition to ascertain the power production of solar cells or wind turbines and adjust the load to maximise power efficiency under varying weather conditions. On another hands, the hybrid smart system uses a wind catcher to reduce the amount of energy consumed by buildings from the grid, which is a historically significant architectural component for cross ventilation and passive cooling. This paper presents an updating of model that proposes improvements to the regulation of Spline MPPT with tuning PID controller by using Levy Invasive Weed optimization (LIWO)technique. A rapid, accurate, and straightforward approach for determining the (MPPT) of PV systems under consistent irradiation and partial shade, as well as wind turbines, is presented using the Spline- MPPT technique. The model of this paper employs LIWO in conjunction with a PID controller to improve the selection of PID gains. Moreover, is applied to generate the optimal values of duty cycle of the model. The comprehensive system design depicted has been simulated using MATLAB Simulink to verify the functionality of the system. The model has attained an accuracy of 93%. The outcomes of this model contribute greatly to our understanding of the suitability and efficacy of both AI and traditional MPPT controllers. This modeling will be beneficial to the renewable energy industry.
-
Alcoholism Identification by Processing the EEG Signals Using Oscillatory Modes Decomposition and Machine LearningExcessive alcoholism has the potential to disturb the nervous system. A timely identification and prevention can cure it. This work proposes a novel approach to detect alcoholism by processing the electroencephalogram (EEG) signals. It is based on the signal processing and machine learning algorithm. We have used the oscillatory mode decompositions to decompose the EEG signals in modes. First six modes are considered for second order difference plots (SODPs). The central tendency measure, area, and mean are chosen as the three features from each intended SODP. Experiments are carried out by taking into account features collected from three different length time windows in order to establish an appropriate EEG signal segment length for the intended application. For classification, different machine learning algorithms are used.
-
Advances and development of wind–solar hybrid renewable energy technologies for energy transition and sustainable future in IndiaWhile solar power projects are built on a continuous ground, wind power projects require scattered land, raising transmission costs and increasing the risk of land-related complications. Wind–solar hybrid (WSH) projects have been proposed to address these issues and accelerate installation. WSH power projects will create a well-defined area with sufficient infrastructure, including evacuation facilities, where the project’s risks can be reduced. The extensive coastline of India is endowed with high wind flow speed and plentiful solar power resources, creating an ideal environment for WSH projects to prosper while simultaneously improving grid stability and reliability. WSH plants guarantee higher transmission efficiency and cost-effectiveness than their stand-alone counterparts. As of 30.11.2021, 3.75 GW of WSH projects have been granted, with 0.148 GW of operational capacity and 1.7 GW of WSH projects in various bidding phases. In this paper, we discussed state-wise WSH potential, the key players in the WSH project, the National WSH, and the State WSH policy and amendments. Also, the WSH project’s physical progress and commercial details are covered. A feasibility study of the WSH plant is performed, and the primary design strategy for deploying WSH power facilities in India is discussed. It covers every step of this process, from design technique to choosing and evaluating potential locations for such hybrid projects, optimally placing wind turbines and solar panels, overall capacity mix for hybrid plants, and ultimately power evacuation optimization. Additionally, a brief study of the savings from these hybrid plants and the environmental, social, and governance standards which are necessary to implement these projects are provided. The potential challenges connected with WSH technologies are examined in depth, and potential solutions and mitigations for the challenges are provided. Designing a WSH for small-scale irrigation is provided along with the size and choice of wind and solar systems. Degradation of PV systems and carbon savings are included, along with some policy measures to boost the proportion of WSH in the entire power mix. In India, the development of large-scale WSH projects is still in its early stages, and more research is required to explore technical, commercial, and policy elements that influence project design. The policy suggestions for improvement of the WSH project are provided. The WSH project developers, potential investors, stakeholders, innovators, policymakers, manufacturers, designers, and researchers will benefit from the recommendations based on the review’s findings.
-
Advances in electric vehicles for a self-reliant energy ecosystem and powering a sustainable future in IndiaElectric vehicles (EVs) are essential for solving various mobility, environmental sustainability, and energy security issues. They help reduce greenhouse gas emissions, improve air quality, and promote economic development by creating jobs and developing novel innovations. Despite the numerous benefits of EVs, several difficulties persist, including range anxiety, limited infrastructure for charging, and high starting prices. However, continued breakthroughs in battery innovation, charging infrastructure development, and supporting government regulations steadily reduce these obstacles and increase global EV adoption. Significant EV adoption will necessitate ongoing coordination among governments, stakeholders in the industry, and communities to meet infrastructure requirements, incentivize consumers, and encourage sustainable mobility behaviors. The full potential of EVs can be realized to build a more environmentally friendly, healthier, and more sustainable future for future generations by encouraging innovation, investing in facilities, and raising the public's consciousness. This research aims to analyze current technological advancements in EVs, problems, opportunities, and potential solutions that can serve as the foundation for effective strategies and assist policymakers in formulating plans for target adaptation and achievement in India.
-
Comparative Analysis of Deep Learning Architectures for Blood Cancer ClassificationAn increased growth in the blood cancer necessitates the development of efficient, cost effective, timely, and accurate diagnosis. Traditional diagnosis methods are often invasive, expensive and time-consuming. The rapid artificial intelligence (AI) assistive advancement in the digital healthcare permits the realization of effective solutions in this framework. Specifically, the deep learning (DL), seems promising in an automated diagnosis. However, still a critical gap needs to be covered by understanding that which DL architecture performs better for the blood cancer detection. To address this crucial need, this paper presents a comprehensive comparative analysis of the key DL methods, used in an automated categorization of the blood cancer. The considered DL architectures are the MobileNetV2, DenseNet121, VGG16, ResNet50, and Inception V3.The applicability is tested using two blood cancer datasets namely the Acute Lymphoblastic Leukemia (ALL) dataset and American Society of Hematology (ASH) dataset. Each model is meticulously trained and evaluated on the ALL dataset for binary classification and the ASH image bank for multi-class classification. The categorization performance is evaluated based on accuracy, precision, recall, F1 score, and latency. Results have shown an out performance of the MobileNetV2 compared to the counter DL architectures with a mean accuracy of 91.26%, precision of 92.94%, recall of 91.27%, F1 score of 90.58% and latency of 104.16 mins for ALL dataset and 88.11 accuracy, 90.23% precision, recall of88.11 %, 87.98% F1 score and latency of 11.16 mins for ASH dataset.
-
ECG-based emotion recognition using CWT and deep learningEmotion recognition can enhance the Human-Machine Interaction (HMI) in several aspects such as an enriched personalization with intention-based adaptation and responsiveness. In this context, several valuable studies have been conducted by exploring the physiological signals. This chapter discusses the significance of assessing the Autonomic Nervous System (ANS) through physiological indicators like electrocardiogram (ECG), Galvanic Skin Response (GSR), Blood Pressure (BP), and respiration rates, with particular emphasis on ECG and GSR due to their insights into various pathological and psychophysiological conditions. While ECG provides detailed heart electrical activity information, GSR reflects ANS activity through sweat gland function. The simplicity, effectiveness, affordability, and noninvasiveness of these measures make them preferable, although automatic interpretation is crucial for accurately identifying patterns associated with specific mental and physiological states. This chapter aims to improve healthcare applications and human-computer interaction by investigating the possibilities of emotion recognition using AI algorithms applied to ECG and GSR signals. Machine learning and deep learning algorithms evaluate ECG and GSR data to categorize emotions. These techniques have shown promise in various fields, including affective computing, mental health assessment, and human-computer interaction. To validate their effectiveness in differentiating emotions for multiple applications, the chapter shows how to create an effective ecosystem for real-time emotion recognition from ECG and GSR signals using a blend of wavelet transform, convolutional neural networks, and transfer learning.
-
Surface EMG-based gesture recognition using wavelet transform and ensemble learningThere are several uses for surface electromyography (sEMG), but one crucial application is in human-machine interaction (HMI). HMI systems can benefit from the integration of the sEMG to provide more flexible and responsive user interfaces. By adjusting their muscular activity, users can operate machines, computers, virtual reality platforms, and other electronic devices, providing an alternative to conventional input methods. The hands are essential for gripping and working with various objects. Human activity is impacted when even one hand is lost. For the subjects who lost their hands, a prosthetic hand is an enticing remedy in this regard. When designing prosthetic hands for industrial and assistive uses, the sEMG is a crucial component. By combining many classifier models in a weighted manner, the ensemble classifiers outperform other methods. Therefore utilizing sEMG signals that were captured during the grabbing actions with different objects for each of the six hand motions, the viability of the bagging and boosting ensemble classifiers is evaluated for the fundamental hand movement recognition in this research. There are three stages in the suggested procedure. Denoising is done using the Multiscale Principal Component Analysis (MSPCA) in the first stage. The second stage involves extracting features from the sEMG signals using a novel feature extraction technique called the Tunable Q-factor wavelet transform (TQWT), after which the statistical values of the TQWT subbands are mined to attain a dimension reduction. The final stage involves feeding the acquired feature set into an ensemble classifier to identify the desired hand movements. Different performance indicators are used to compare the Random Subspace and Rotation Forest algorithm-based ensemble classifiers’ performances. A 98.9% classification accuracy is obtained by using the TQWT-derived features in conjunction with the Rotation Forest plus SVM/Random Forest/REP Tree/LDA Tree. As a result, the findings indicate that the suggested approach is a strong contender for the realization of modern HMI systems.
-
Boosting Regression Assistive Predictive Maintenance of the Aircraft Engine with Random-Sampling Based Class BalancingThis study presents the development of a data-driven predictive maintenance model in the context of industry 4.0. The solution is based on a novel hybridization of Remaining Useful Life (RUL) generation, Min-Max normalization, random-sampling based class balancing, and XGBoost regressor. The applicability is tested using the NASA’s C-MAPSS dataset, which contains aircraft engine simulation data. The objective is to develop an effective and Artificial Intelligence (AI) assistive automated aircraft engine’s RUL predictor. It can maximize the benefits of predictive maintenance. The rules based RUL generation provides a ground truth for evaluating the performance of intended regressors. The Min-Max normalization linearly transforms the intended dataset and scales the multi subject’s data in a common range. The imbalance presentation among intended classes can lead towards a biasness in findings. This issue is intelligently resolved using the uniformly distributed random sub-sampling. Onward, the performance of robust machine learning and ensemble learning algorithms is compared for predicting the RUL of the considered aircraft engine by processing the balanced dataset. The results have shown that the XGBoost regressor, uses an ensemble of decision trees, outperforms other considered models. The root mean square error (RMSE) and mean absolute error (MAE) indicators will be used to evaluate the prediction performances. The devised method secures the RMSE value of 12.88%. It confirms a similar or better performance compared to the state-of-the-art counterparts.
-
EEG-based emotion recognition using AR burg and ensemble machine learning modelsEmotion recognition plays a crucial role in human-computer interaction, affective computing, and mental health assessment. In recent years, electroencephalography (EEG) has emerged as a promising modality for detecting and interpreting human emotions. The popularity of processing data using machine learning is becoming popular day by day. Because of a complex nature of the EEG signals and presence of artifacts and noise, the automated recognition is usually limited to a small number of emotion classes. This chapter proposes a novel approach for EEG-based emotion recognition using autoregressive (AR) Burg modeling combined with ensemble machine learning models. Relevant features are extracted from EEG signals using the AR Burg approach, which captures the spectral properties and temporal dynamics linked to various emotional states. These features are subsequently fed into ensemble machine learning models to characterize emotions effectively. The suggested method improves emotion recognition performance by utilizing the advantages of feature extraction and classification approaches. The proposed method's efficacy in properly recognizing emotional states from EEG signals is demonstrated by experimental results, underscoring its potential applications in affective computing, mental health monitoring, and human-computer interaction.
-
Speech-driven human-machine interaction using Mel-frequency Cepstral coefficients with machine learning and Cymatics DisplayWhen people engage with machines, gadgets, or programs through spoken language, this is known as speech-driven human-machine interaction (MHI). Speech recognition technology is used in this interaction to interpret words in commands that the computer can understand and process. An innovative method for accomplishing automatic speech command recognition is presented in this chapter. The concept is to blend efficient analysis and speech processing techniques. An effective technique for autonomously isolated speech-based message recognition is proposed in this context. The input voice segments are improved for postprocessing when the appropriate preemphasis filtering, noise thresholding, and zero alignment procedures are used. The Mel-Frequency Cepstral coefficients (MFCCs), Delta, and Delta–Delta coefficients are extracted from the improved speech segment. The machine learning algorithms are then used to process these features that have been retrieved to classify the intended isolated speech commands automatically. As a case study, the science of Cymatics is applied to convert classification decisions into systematic signs. The system's functionality is examined using an experimental setting, and the findings are reported. It was possible to attain an average isolated speech recognition accuracy, for the intended dataset, of 98.9%. The suggested methodology has potential uses in the visual arts, in noisy and industrial settings, in integrating individuals with hearing impairments, and in education.
-
EEG-based stress identification using oscillatory mode decomposition and artificial neural networkThe stress identification offers an understanding of the user's mental state. Therefore it has the potential to greatly improve the Human-Machine Interaction (HMI). In the recent era of industrial, assistive, and healthcare applications, a new trend is to use the “Artificial Intelligence” (AI) powered multimodal signal processing to detect the stress level. This work presents an original approach for classification of stress and nonstress subjects. We have used the multichannel Electroencephalogram (EEG) signals for this categorization. Intrinsic mode function (IMF) is obtained from these EEG signals using the “Empirical Mode Decomposition” (EMD). The “Variational Mode Decomposition” (VMD) is also applied to get the modes from EEG signals. For the selected IMFs and modes, the “Second-Order Difference Plots” (SODPs) are traced. The shape of these SODPs is used to distinguish the stress and without stress categories. The feature space is derived from first 7 IMFs and modes. This includes areas, “Central Tendency Measures” (CTMs), and means of SODPs. A publicly available EEG signals dataset is used to test the applicability of the work. We have used the machine learning algorithms such as the “Support Vector Machine” (SVM), “Multilayer Perceptron Neural Network” (MLPNN), and Boosting with “Random Forest” (RF) for an automated categorization. A detailed performance analysis is conducted for individual channels, subset of channels, and lobe-wise. The highest attained accuracy at subset level is 99.89%, channel-wise is 98.89%, and lobe-wise is 99.99%.
-
Development and Evolution of Hybrid Microgrids in the Context of Contemporary ApplicationsThis chapter provides a valuable insight to the development and evolution of hybrid microgrids, their applications, energy demand analysis, energy sources, mathematical modeling, objective parameters, and case studies. It begins with the definition of hybrid microgrids and highlights their importance in various sectors. Onward, it discusses the key microgrid applications. The analysis of energy demand and the concerned factors in the considered applications are examined. Additionally, the major energy sources are explored such as the photovoltaic systems, wind mills, diesel generators, fuel cells, electrolyzes, hydrogen tanks, and battery storage systems in hybrid microgrid configurations are emphasized. Afterwards, the objective parameters such as net present cost, levelized cost of electricity, and greenhouse gas emissions reduction are introduced and used as key metrics for evaluating the performance and environmental impact of hybrid microgrids. Moreover, three case studies are presented, examining the cost analysis and greenhouse gas emissions of a cement industry, conducting sensitivity analysis and assessing environmental impacts in the case of the city of Gwadar, and exploring the implementation of hybrid microgrids in a university campus setting.
-
An Overview of Artificial Intelligence Driven Li-Ion Battery State EstimationThe omnipresence of batteries in nowadays appliances such as portable devices, electrical vehicles, hybrid microgrids, etc. gives them a special priority in energy storage. In addition, the high cost of battery manufacturing makes the maintenance and the monitoring of their states critical. The state estimation is a complicated process that may include many parameters to take into consideration. The effective solutions can lead towards an efficient maintenance of the intended battery packs and can render a longer life of cells in that pack. This chapter focuses on the critical process of battery state estimation and the role of artificial The omnipresence of batteries in nowadays appliances such as portable devices, electrical vehicles, hybrid microgrids, etc. gives them a special priority in energy storage. In addition, the high cost of battery manufacturing makes the maintenance and the monitoring of their states critical. The state estimation is a complicated process that may include many parameters to take into consideration. The effective solutions can lead towards an efficient maintenance of the intended battery packs and can render a longer life of cells in that pack. This chapter focuses on the critical process of battery state estimation and the role of artificial intelligence in battery state estimation.