RSS 1.0Electrical and Computer Engineering
Sub-communities within this community
Collections in this community
Recent Submissions
-
Artificial Intelligence and Multimodal Signal Processing in Human-Machine InteractionArtificial Intelligence and Multimodal Signal Processing in Human-Machine Interaction presents an overview of an emerging field that is concerned with exploiting multiple modalities of communication in both Artificial Intelligence and Human-Machine Interaction. The book not only provides cross disciplinary research in the fields of multimodal signal acquisition and sensing, analysis, IoTs (Internet of Things), Artificial Intelligence, and system architectures, it also evaluates the role of Artificial Intelligence I in relation to the realization of contemporary Human Machine Interaction (HMI) systems. Readers are introduced to the multimodal signals and their role in the identification of the intended subjects, mental state and the realization of HMI systems are explored, and the applications of signal processing and machine/ensemble/deep learning for HMIs are assessed. A description of proposed methodologies is provided, and related works are also presented. This is a valuable resource for researchers, health professionals, postgraduate students, post doc researchers and faculty members in the fields of HMIs, Brain-Computer Interface (BCI), Prosthesis, Computer vision, and Mental state estimation, and all those who wish to broaden their knowledge in the allied field.
-
Enabling Trust in Automotive IoT: Lightweight Mutual Authentication Scheme for Electronic Connected Devices in Internet of ThingsThe emergence of smart and embedded devices and the adaptation of new technologies with the Internet of Things have entirely changed online shopping with a new ecommerce paradigm. However, while the accessibility of IoT services has many advantages, it also creates hazards to customerrelated data. Due to pervasive e-commerce services, anyone can intercept and be compromised, creating great security and privacy concerns. To address these security challenges and to provide lightweight authentication for all entities, we introduce a “Lightweight Mutual Authentication (LMA) Scheme for Connected Devices in IoT”. The proposed scheme uses automatic and mutual authentication for all entities, employing a distributed approach within a server-based architecture. It is lightweight because it provides a secure way of using e-services with fewer steps, and it is automotive because the entities automatically authenticate each other. The LMA is formally validated in TCL, and the experimental results show that it decreases computation cost by about 56%, increases throughput by about 33.3%, and communication cost remains the same as the average of the other three schemes. In evaluation, the results demonstrate that the presence of the LMA leads to a 20-millisecond increase in delay and a 2% decrease in PDR for 100 devices.
-
Appliances Load Pattern Reconstruction from Adaptive Delta-Driven Sampled Smart Meter DataIn recent days the interest in the usage of smart meters is raising. It is evident from the widespread use of smart meters in contemporary society. The stakeholders in the smart grid must profit from the gathering and processing of fine-grained metering data. Time invariance characterizes the classical data sampling method. As a result, a significant volume of unnecessary data is gathered, sent, and analyzed. A method of adaptive delta-driven sampling (ADDS) of the smart meter data is proposed. It compensates the aforementioned shortfall and can lead towards a significant real-time compression without losing pertinent information. Subsequently, the compressed form of data can be efficiently processed, analyzed, stored and transmitted. It promises a significant transmission and computational effectiveness with a diminished latency. It is shown that the devised form of compressed data can be effectively reconstructed using a low complexity reconstruction algorithm. The reconstruction error is measured in terms of the root mean square error (RMSE) and the mean absolute error (MAE). The applicability is tested using the power consumption patterns of coffee machines, computer stations, fridges and freezers. The proposed solution attains an overall compression gain of 1.84-times, 2.49-times, 7.55-times respectively for the coffee machine, computer stations, and fridges and freezers. Moreover, the obtained values of RMSE and MAE confirm an appropriate reconstruction using the devised method.
-
Electrooculogram Compression based on Wavelet Packet DecompositionThe electrooculogram (EOG) represents the electrical activity of the eye. Since most EOG-based applications need a large amount of data to be stored and transmitted, compression is required. A study on the use of discrete wavelet packet decomposition (DWPD) for compressing EOG signals is presented in this paper. The EOG signals are recorded using a commercial bioamplifier. The compression algorithm is implemented using the MATLAB software. The performance of the compression was evaluated using three parameters: compression ratio (CR), energy retention, and percent root-mean-square difference (PRD). The experimental results show that the DWPD allows for CRs of 82% while retaining almost 100% of the signal energy with a PRD of only 7%.
-
Enhanced the Hosting Capacity of a Photovoltaic Solar System Through the Utilization of a Model Predictive ControllerThe global expansion of solar-powered within distribution networks with Low Voltage (LV) is experiencing substantial expansion. Despite the various advantages offered by solar photovoltaic generation, surpassing the constraints on Hosting Capacity (HC) within these networks persist a significant technical problem in system operation, especially in relation to voltage operation. This research delves into the effectiveness of improving the Hosting Capacity (HC) of a photovoltaic (PV) system within an LV distribution system. It utilizes a Model Predictive Controller (MPC) to achieve this enhancement and contrasts its performance with reactive power control. The study examines scenarios encompassing both linear and non-linear loads to assess the impact of these control strategies on the PV system’s harmonic current in the LV distribution network. Through detailed analysis, the MPC controller demonstrates superior adaptability and responsiveness, maintaining stable active power at 95.5 kW before accommodating a 100% PV system penetration and experiencing a substantial increase to 192 kW. The hosting capacity, thereby, sees a notable 101.05% improvement under MPC control. Additionally, the study reveals that MPC optimizes reactive power utilization, resulting in a 17.9% reduction in reactive power and an 18.3% enhancement in bus voltage compared to reactive power control. Notably, MPC exhibits superior adaptability to both linear and non-linear loads, emphasizing its potential as an effective solution for optimizing the performance of PV systems within LV distribution grids. This research underscores the significance of advanced control strategies in facilitating the integration of renewable energy systems while ensuring grid stability and reliability.
-
Anwendung von Wavelet-Zerlegung und maschinellem Lernen für die sEMG-Signalbasierte GestenerkennungIn German Language: Amputierte auf der ganzen Welt haben begrenzten Zugang zu hochwertigen intelligenten Prothesen. Die korrekte Erkennung von Gesten ist eine der schwierigsten Aufgaben im Kontext der Entwicklung von auf Oberflächen-Elektromyographie (sEMG) basierenden Prothesen. Dieses Kapitel zeigt eine vergleichende Untersuchung mehrerer auf maschinellem Lernen basierender Algorithmen zur Identifizierung von Handgesten. Der erste Schritt im Prozess ist die Datenerfassung aus dem sEMG-Gerät, gefolgt von der Merkmalsextraktion. Anschließend werden zwei robuste maschinelle Lernalgorithmen auf den extrahierten Merkmalsatz angewendet, um ihre Vorhersagegenauigkeit zu vergleichen. Die mittlere Gaußsche Support Vector Machine (SVM) funktioniert unter allen Bedingungen besser als der K-nearest neighbor. Verschiedene Parameter werden für den Leistungsvergleich verwendet, darunter F1-Score, Genauigkeit, Präzision und Kappa-Index. Die vorgeschlagene Methode zur Erkennung von Handgesten, basierend auf sEMG, wird gründlich untersucht und die Ergebnisse haben eine vielversprechende Leistung gezeigt. In jedem Fall kann ein Fehlverhalten bei der Merkmalsextraktion die Erkennungsgenauigkeit verringern. Die tiefgreifenden Lernmethoden werden verwendet, um eine hohe Präzision zu erreichen. Daher berücksichtigt das vorgeschlagene Design alle Aspekte bei der Verarbeitung des sEMG-Signals. Das System sichert eine höchste Klassifizierungsgenauigkeit von 92,2 % für den Fall des Gaußschen SVM-Algorithmus.
-
"Fiction to Function" Shaping Renewable Energy Education with MATLAB and ChatGPT-Driven EnvironmentsAddressing the gap in effective and engaging renewable energy education, this study, 'Fiction to Function,' transitions imaginative frameworks into practical educational tools, focusing on solar and wind energy in line with Sustainable Development Goals (SDGs). The goal is to empower users with deep insights into renewable energy through the novel integration of MATLAB and ChatGPT-driven environments, enhancing the analysis and exploration of these energy sources. The study employs a structured approach using MATLAB App Designer to develop a user-friendly GUI, aligning instructional content with the SDGs and integrating ChatGPT via the OpenAI API for dynamic, interactive learning experiences. This integration provides a multifaceted platform that not only educates but also engages users in practical scenarios, bridging the gap between theoretical knowledge and real-world application. The results demonstrate an enhanced understanding of renewable energy concepts among users, showcasing the efficacy of combining MATLAB's analytical capabilities with ChatGPT's AI-driven guidance. Ultimately, the study highlights the transformative potential of innovative technology in sustainable energy education, setting a precedent for future educational tools in this domain.
-
Gehirn-Computer-Schnittstelle (BCI), basierend auf der EEGSignalzerlegung, Schmetterlingsoptimierung und maschinellem LernenIn German Language: Die Gehirn-Computer-Schnittstelle (BCI) ist eine Technologie, die Menschen mit Behinderungen hilft, Hilfsgeräte zu bedienen, indem sie neuromuskuläre Kanäle umgeht. Diese Studie zielt darauf ab, die Elektroenzephalographie (EEG) Signale zu verarbeiten und diese Signale dann durch Analyse und Kategorisierung mit Maschinenlernalgorithmen in Befehle zu übersetzen. Die Ergebnisse können weiterhin zur Steuerung eines Hilfsgeräts verwendet werden. Die Bedeutung dieses Projekts liegt in der Unterstützung von Menschen mit schweren motorischen Beeinträchtigungen, Lähmungen oder denen, die ihre Gliedmaßen verloren haben, um unabhängig und selbstbewusst zu sein, indem sie ihre Umgebung kontrollieren und ihnen alternative Kommunikationswege bieten. Die erworbenen EEG-Signale werden digital mit einem Tiefpass gefiltert und dezimiert. Anschließend wird die Wavelet-Zerlegung zur Signalanalyse verwendet. Die Merkmale werden aus den erhaltenen Unterbändern abgebaut. Die Dimension des extrahierten Merkmalsatzes wird durch Verwendung des Butterfly-Optimierungsalgorithmus reduziert. Der ausgewählte Merkmalsatz wird dann von den Klassifikatoren verarbeitet. Die Leistung des k-Nearest Neighbor, der Support Vector Machine und des Artificial Neural Network wird für die Kategorisierung von motorischen Imagery-Aufgaben durch Verarbeitung des ausgewählten Merkmalsatzes verglichen. Die vorgeschlagene Methode sichert eine höchste Genauigkeitsbewertung von 83,7 % für den Fall des k-Nearest Neighbor-Klassifikators.
-
Fortschritte in Der Nicht-Invasiven Biomedizinischen Signalverarbeitung Mit MLIn German Language: Dieses Buch stellt die modernen technologischen Fortschritte und Revolutionen im biomedizinischen Sektor vor. Fortschritte in der zeitgenössischen Sensorik, dem Internet der Dinge (IoT) und bei Maschinenlernalgorithmen und -architekturen haben neue Ansätze im mobilen Gesundheitswesen eingeführt. Eine kontinuierliche Beobachtung von Patienten mit kritischer Gesundheitssituation ist erforderlich. Sie ermöglicht die Überwachung ihres Gesundheitszustandes während alltäglicher Aktivitäten wie Sport, Gehen und Schlafen. Dank moderner IoT-Rahmenbedingungen und drahtloser biomedizinischer Implantate, wie Smartphones, Smartwatches und Gürtel, ist dies realisierbar. Solche Lösungen befinden sich derzeit in der Entwicklung und in Testphasen durch Gesundheits- und Regierungsinstitutionen, Forschungslabore und biomedizinische Unternehmen. Die biomedizinischen Signale wie Elektrokardiogramm (EKG), Elektroenzephalogramm (EEG), Elektromyographie (EMG), Phonokardiogramm (PCG), bei chronisch-obstruktiver Lungenkrankheit (COP) und Elektrookulographie (EoG), Photoplethysmographie (PPG), Positronenemissionstomographie (PET), Magnetresonanztomographie (MRI) und Computertomographie (CT) werden nicht-invasiv erfasst, gemessen und über die biomedizinischen Sensoren und Gadgets verarbeitet. Diese Signale und Bilder repräsentieren die Aktivitäten und Zustände des menschlichen kardiovaskulären, neuralen, visuellen und zerebralen Systems. Eine Mehrkanalerfassung dieser Signale und Bilder mit einer angemessenen Granularität ist für eine effektive Überwachung und Diagnose erforderlich. Sie erzeugt ein großes Datenvolumen, und seine Analyse ist manuell nicht machbar. Daher sind automatisierte Gesundheitssysteme in der Entwicklung. Diese Systeme basieren hauptsächlich auf der Erfassung und Sensorik von biomedizinischen Signalen und Bildern, Vorverarbeitung, Merkmalsextraktion und Klassifizierungsstufen. Die zeitgenössischen biomedizinischen Signal-Sensorik, Vorverarbeitung, Merkmalsextraktion und intelligente maschinelle und tiefgreifende Lernalgorithmen für die Klassifizierung werden beschrieben.
-
Energy-efficient architecture for high-performance FIR adaptive filter using hybridizing CSDTCSE-CRABRA based distributed arithmetic design: Noise removal application in IoT-based WSNAn energy-efficient architecture of high-performance FIR adaptive filter design using approximate distributed arithmetic (DA), which is integrated with canonic signed digit-based triangular common sub expression elimination (CSDTCSE) and carry-resist adder based Booth recorder adder (CRABRA) is proposed for noise removal in sensor nodes. Distributed arithmetic is coupled with two signed 32-bit, 16-bit radix-8 Booth algorithms and approximate computation under 2-bit adder to design FIR adaptive filter for decreasing partial products (PP) together with accumulation circuits. The truncation of LSB in the PP is presented to approximate the PP to reduce memory complexity and hardware overhead. An approximation recoding adder decreases the energy usage, area, and critical path. Approximate Wallace trees are applied to the PP accumulation to lessen the latency. The canonic signed digit-based triangular common sub-expressions elimination framework is proposed, which significantly reduces a count of logic operators and logic depth in implementing the FIR filter. The proposed algorithm is activated in Verilog coding and synthesized using Xilinx 14.5 ISE simulation software. The proposed design successfully reduces delay, area, and power by maintaining better accuracy with performance.
-
Prediction of Adsorption and Desorption Isotherms for Atmospheric Water HarvestingThis study improves the mathematical modeling for the potential of atmospheric water harvesting (AWH) using desiccant materials. This research is crucial in highlighting the sustainable water management strategies of AWH systems in converting atmospheric moisture into a vital water source. The primary focus of our research methodology was the analytical derivation of sorption isotherms, essential for the hygrothermal simulation of desiccant materials. This was accomplished through the application of two established models, namely, the Guggenheim-Anderson-de Boer (GAB) and Van Genuchten (VG). Experimental data on various anhydrous salts from existing literature have been used. An in-depth comparative analysis of these models reveals that the VG model aligns more closely with the experimental data, thus asserting its superiority in enhancing the selection and efficiency of desiccant materials in AWH systems. By confirming the VG model’s superiority in accurately modeling sorption isotherms, our research not only improves the model of AWH systems but also, importantly, contributes to the development of advanced water harvesting technologies.
-
Integrating Human-Centricity, Sustainability, and Resilience in Digital Twin Models for Industry 5.0 : A Multi-Objective Optimization ApproachThis paper presents the InduDesc framework, an innovative digital twin model within the CupCarbon software, designed for the advanced needs of Industry 5.0. It integrates human-centred ergonomics, sustainability and resilience into the Flexible Job Shop Scheduling Problem (FJSP), traditionally an NP-hard challenge. By minimising operating times and balancing machine utilisation with ergonomic and sustainability considerations, the framework provides a dynamic workload management tool based on real-time 'fatigue' metrics. Using a tabu search algorithm, InduDesc generates a Pareto frontier to help decision makers identify strategies that efficiently align with the integrated goals of Industry 5.0.
-
Physics‐based and data‐driven approaches for lifetime estimation under variable conditions: Application to organic light‐emitting diodesThe prognosis of organic light-emitting diodes (OLEDs) not only requires early detection of a bearing defect, but also the capability to predict their life data under all operational scenarios. The use of sophisticated machine learning (ML) algorithms is undoubtedly becoming an increasingly exciting research direction, as these algorithms can yield high predictive models with minimal domain expertise. The central question of this perspective is: how well can ML models advance our ability to forecast the lifetime of OLEDs compared to the physics-based models? In this paper, data-driven methods, feed-forward neural networks (FFNN), support vector machines (SVMs), k-nearest neighbors (KNNs), partial least squares regression (PLSR), and decision trees (DTs), are used to predict the lifetime and reliability of OLEDs through analyzing the lumen degradation data collected from the accelerated lifetime test. The final predicted results indicate that both the data-driven and our physics-based OLED lifetime models fit well the experimental data. The main drawback of the former method is that their efficacy is highly contingent on the quantity and quality of the operational dataset. Among all these methods, much more reliability information (time to failure) and the highest prediction accuracy can be achieved by FFNN.The prognosis of organic light-emitting diodes (OLEDs) not only requires early detection of a bearing defect, but also the capability to predict their life data under all operational scenarios. The use of sophisticated machine learning (ML) algorithms is undoubtedly becoming an increasingly exciting research direction, as these algorithms can yield high predictive models with minimal domain expertise. The central question of this perspective is: how well can ML models advance our ability to forecast the lifetime of OLEDs compared to the physics-based models? In this paper, data-driven methods, feed-forward neural networks (FFNN), support vector machines (SVMs), k-nearest neighbors (KNNs), partial least squares regression (PLSR), and decision trees (DTs), are used to predict the lifetime and reliability of OLEDs through analyzing the lumen degradation data collected from the accelerated lifetime test. The final predicted results indicate that both the data-driven and our physics-based OLED lifetime models fit well the experimental data. The main drawback of the former method is that their efficacy is highly contingent on the quantity and quality of the operational dataset. Among all these methods, much more reliability information (time to failure) and the highest prediction accuracy can be achieved by FFNN.
-
Battery Management System for Enhancing the Performance and Safety of Lithium-Ion BatteriesBattery packs integrated into the grid offer a promising solution for energy storage, but their efficient operation requires precise monitoring and control, which is achieved through Battery Management Systems (BMS). This paper proposes a temperature-dependent second-order RC equivalent circuit model to reflect the battery’s dynamic characteristics accurately. Then, a novel BMS design, incorporating Extended Kalman Filtering (EKF), a constant current-constant voltage (CCCV) charging, and a passive balancing algorithm to estimate the battery state of charge (SOC), balance voltage levels, and monitor thermal characteristics. The research also includes a comprehensive simulation study conducted in SIMULINK with the Simscape toolbox to assess the effectiveness of the analyzed BMS. The simulation results demonstrate the effictiveness of the proposed BMS design in monitoring the battery pack’s state, maintains cell balancing, estimates SOC, and keeps the temperature and current levels within safe limits. This model can help guide a more efficient and accurate BMS design for future studies.
-
Bayesian Optimization Algorithm for ConvLSTM-AE model To Forecast Solar IrradiationAs much as the development of smart grid systems, accurate and reliable solar irradiance is crucial. In smart grid systems, forecasting solar irradiance remains a vital solution to mitigate the best management of energy.This study presents a new approach using a hybrid deep learning model, Conv-LSTM-AE, enhanced by Bayesian algorithm optimization. In four seasons of the years, the model was evaluated based on the accuracy using different metrices errors, and time computing in training. The model exhibits exceptional adeptness as compared to other deep learning models, showing that the Conv-LSTM-AE with Bayesian optimization is a standout performer with remarkable accuracy. This development represents a substantial step towards utilizing state-of-the-art technology for the efficient use of solar power within intelligent grid systems.
-
Rechargeable Battery State Estimation Based on Adaptive-Rate Processing and Machine LearningThe generalization of the use of electronic systems and their integration in industrial systems and different aspects of modern life (internet of things, electric vehicles, robotics, smart grids), give rise to new challenges related to the storage and optimized management of energy. Lithium-on batteries perfectly meet this objective due to their good qualities such as high energy density, small installation size, low self-discharge and high supply capacity. However, their wide application requires further research on battery failure prediction and health management. Intelligent “battery management systems” (BMSs) employ the real-time estimation and control algorithms to improve the battery safety while enhancing its performance. Nevertheless, BMS are complex and require increased processing power which could lead to more power consumption. In this context, the present article provides a new approach for efficient prediction of the “Lithiumion” (Li-ion) battery cells capacities by analysing and exploiting the battery parameters, acquired by an event-driven module. It acquires the intended cells voltages, currents and temperature values during the charge-discharge cycles. The solution is based on the machine learning algorithms and event-based segmentation. The “National Aeronautics and Space Administration” (NASA) has provided a high-power Li-Ion cells dataset for the purpose of research and innovation. This dataset is used to test and evaluate the suggested approach. The evaluation of the overall performance of the chain has shown encouraging results of the proposed approach.
-
Model Predictive Control of Consensus-based Energy Management System for DC MicrogridThe increasing deployment and exploitation of distributed renewable energy source (DRES) units and battery energy storage systems (BESS) in DC microgrids lead to a promising research field currently. Individual DRES and BESS controllers can operate as grid-forming (GFM) or grid-feeding (GFE) units independently, depending on the microgrid operational requirements. In standalone mode, at least one controller should operate as a GFM unit. In grid-connected mode, all the controllers may operate as GFE units. This article proposes a consensus-based energy management system based upon Model Predictive Control (MPC) for DRES and BESS individual controllers to operate in both configurations (GFM or GFE). Energy management system determines the mode of power flow based on the amount of generated power, load power, solar irradiance, wind speed, rated power of every DG, and state of charge (SOC) of BESS. Based on selection of power flow mode, the role of DRES and BESS individual controllers to operate as GFM or GFE units, is decided. MPC hybrid cost function with auto-tuning weighing factors will enable DRES and BESS converters to switch between GFM and GFE. In this paper, a single hybrid cost function has been proposed for both GFM and GFE. The performance of the proposed energy management system has been validated on an EU low voltage benchmark DC microgrid by MATLAB/SIMULINK simulation and also compared with Proportional Integral (PI) & Sliding Mode Control (SMC) technique. It has been noted that as compared to PI & SMC, MPC technique exhibits settling time of less than 1μsec and 5% overshoot.
-
Design of receiver RF front end for mm-Wave 5G applicationsRecently, the research emphasis has shifted towards 5G due to its potential to accommodate the increasing demand for data traffic, extensive interconnectivity of devices, and the emergence of numerous novel applications. Given the inadequacy of speed provided by spectrum resources in the lower frequency bands of 5G, a novel sub-generation utilizing millimeter wave frequencies has been introduced. The new sub-generation is called mm-wave 5G and provides higher bandwidth for faster speed and higher capacity. RF system architecture, circuits, and antenna innovation will be required to provide the requisite speed and capacity. In this paper the design challenges and trade-offs in RF front-end circuits and receiver sub-systems are discussed. Moreover, the massive multiple-input multiple-output (MIMO) techniques are examined. In addition, design, and simulation of key components in 5G mm-wave receiver, including the design of linear phased array antenna receiver and an analog-to-digital converter (ADC) is presented. Sigma-Delta (ΣΔM) Converters are a type of ADC with a technique that involves oversampling and shaping quantization noise to achieve a higher resolution which make it a best choice to be used in the design of the mm wave 5G communication applications. The proposed ΣΔM uses multistage architecture to provide high-order noise shaping and high resolution. The simulation results shows that the designed Cascade 2-2 MASH ΣΔ achieve a signal to noise ratio of 110(resolution of 16 bits).
-
Accurate Classification of Cervical Cancer Based on Multi-layer Perceptron Hunger Games search Optimization techniqueCancer is distinguished by the presence of abnormal cellular proliferation and growth, both of which serve as signs and symptoms for this kind of illness. Computer vision, deep learning, and metaheuristics optimization techniques are increasingly important for solving complex medical Artificial Intelligence (AI) problems such as cancer detection. This paper introduces a new methodology for training the Multi-Layer Perceptron (MLP) using the optimization algorithm known as Hunger Games search Optimization technique (HGO) and apply this method to classify the cervical cancer. The main goal of this method is to reduce the error and enhance the classification rate of cervical cancer. The outcomes show that the MLP with HGO algorithm performed better than other algorithms in terms of classification efficacy and accuracy rate. Simulation outcomes indicate that the proposed strategy performs better than previously published research in terms of effectiveness for the classification optimization methods.
