This paper presents the design, simulation, and experimental validation of a PCB-based multi-turn square loop antenna on an FR-4 substrate for detecting artificial surface partial discharges (SPD). SPD is a specific type of partial discharge (PD) that occurs along insulation surfaces in high-voltage equipment and is often caused by contamination, moisture, or poor installation. Early detection is critical to avoid insulation failure and equipment damage. The proposed antenna is optimized with a self-resonant frequency of 40 MHz, within the VHF range where SPD signals typically radiate. After simulation, the antenna was fabricated using standard PCB processes and tested in a controlled SPD environment. Measurements confirmed successful SPD detection, with a 30 mV output observed at a 1 MΩ oscilloscope input. Compared to conventional sensors, the developed solution offers key advantages: compact size, lightweight construction, and low manufacturing cost—making it a practical and scalable alternative for PD monitoring in high-voltage applications.

Partial discharge (PD) classification is essential for the reliable monitoring and maintenance of high-voltage insulation systems. Traditional classifiers based solely on either Phase-Resolved PD (PRPD) images or raw time-domain current pulses are limited in accuracy, particularly under noise contamination, as they fail to jointly exploit spatial and temporal discharge features. To address these challenges, this paper proposes a multimodal Deep-Learning (DL) framework that fuses convolutional neural network (CNN) features extracted from PRPD maps with long short-term memory (LSTM) representations of synchronized High-Frequency Current Transformer (HFCT) waveforms. The fused architecture integrates spatial morphology with pulse dynamics through lightweight fully connected layers. A laboratory dataset of 840 labeled events from 15 kV cross-linked polyethylene (XLPE) cables-including corona, internal, surface, and noise signals-was used for training and validation (70/15/15 split, targeted augmentation, z-score normalization, five-fold cross-validation). The fused CNN-LSTM achieved 96.3% accuracy, 94.8% recall, and a 95.5% F1-score, outperforming CNN-only (92.4%), LSTM-only (90.1%), and parallel fusion (94.9%) baselines. The similar to 2.1M-parameter model was deployed and profiled on an NVIDIA Jetson TX2. Embedded inference on Jetson TX2 completes in approximate to 15 ms per sample (FP16, batch = 1), satisfying within-cycle operation for 50 Hz systems (20 ms per cycle); half-cycle (approximate to 10 ms) targets are feasible with INT8 quantization/pruning. By jointly modeling PRPD spatial patterns and temporal waveform characteristics, the proposed approach substantially enhances PD diagnostic reliability and robustness for intelligent condition monitoring of medium-voltage power assets.

The accurate and efficient simulation of biosensors is essential for applications in healthcare, environmental monitoring, and diagnostics. This study presents a co-simulation framework integrating COMSOL Multiphysics and Continuous DIscrete Simulation (CODIS+), enabling a synchronized and multi-domain simulation approach to enhance the accuracy and execution time estimation of biosensor systems. The proposed framework leverages COMSOL for high-fidelity multiphysics modeling of biosensor behavior and CODIS+ for real-time signal processing, incorporating a 1D Convolutional Neural Network (CNN) for advanced noise reduction. Furthermore, Worst-Case Execution Time (WCET) estimation is implemented to ensure predictable real-time performance, relying on profiling tools within SystemC and CODIS+. Unlike traditional standalone simulations, the proposed framework eliminates iterative feedback between control and physical modeling, optimizing computational efficiency while maintaining high detection accuracy. A high-fidelity COMSOL model is used as the reference for validation due to the absence of experimental data, ensuring a reliable benchmark for performance evaluation. The framework achieves a low Execution Time Error (ETE) of approximately 4%, validating the precision of execution time estimation and ensuring computational predictability. Performance evaluation is conducted using Root Mean Square Error (RMSE) and Signal-to-Noise Ratio (SNR) metrics. The proposed approach achieves a significant reduction in RMSE (from 7.8 to 2.1) and outperforms traditional noise reduction techniques in terms of SNR improvement, demonstrating its effectiveness in preserving biosensor signal integrity. These results confirm that integrating physics-based modeling with AI-driven noise filtering enhances both biosensor signal accuracy and real-time feasibility. The validation presented in this study is based solely on simulation and profiling results; hardware-level testing is planned for future work. The proposed co-simulation framework presents a scalable and reliable solution for optimizing biosensor design and real-time signal processing, ensuring its applicability in critical biomedical and environmental monitoring applications. It underscores the extensibility, modularity, and reusability of our integration approach, allowing other COMSOL models and CODIS+ functionalities to be easily incorporated and customized.

Impedance-based biosensing has emerged as a critical technology for high-sensitivity biomolecular detection, yet traditional approaches often rely on bulky, costly impedance analyzers, limiting their portability and usability in point-of-care applications. Addressing these limitations, this paper proposes an advanced biosensing system integrating a Silicon Nanowire Field-Effect Transistor (SiNW-FET) biosensor with a high-gain amplification circuit and a 1D Convolutional Neural Network (CNN) implemented on FPGA hardware. This attempt combines SiNW-FET biosensing technology with FPGA-implemented deep learning noise reduction, creating a compact system capable of real-time viral detection with minimal computational latency. The integration of a 1D CNN model on FPGA hardware for adaptive, non-linear noise filtering sets this design apart from conventional filtering approaches by achieving high accuracy and low power consumption in a portable format. This integration of SiNW-FET with FPGA-based CNN noise reduction offers a unique approach, as prior noise reduction techniques for biosensors typically rely on linear filtering or digital smoothing, which lack adaptive capabilities for complex, non-linear noise patterns. By introducing the 1D CNN on FPGA, this architecture enables real-time, high-fidelity noise reduction, preserving critical signal characteristics without compromising processing speed. Notably, the findings presented in this work are based exclusively on comprehensive simulations using COMSOL and MATLAB, as no physical prototypes or biomarker detection experiments were conducted. The SiNW-FET biosensor, functionalized with antibodies specific to viral antigens, detects impedance shifts caused by antibody–antigen interactions, providing a highly sensitive platform for viral detection. A high-gain folded-cascade amplifier enhances the Signal-to-Noise Ratio (SNR) to approximately 70 dB, verified through COMSOL and MATLAB simulations. Additionally, a 1D CNN model is employed for adaptive noise reduction, filtering out non-linear noise patterns and achieving an approximate 75% noise reduction across a broad frequency range. The CNN model, implemented on an Altera DE2 FPGA, enables high-throughput, low-latency signal processing, making the system viable for real-time applications. Performance evaluations confirmed the proposed system’s capability to enhance the SNR significantly while maintaining a compact and energy-efficient design suitable for portable diagnostics. This integrated architecture thus provides a powerful solution for high-precision, real-time viral detection, and continuous health monitoring, advancing the role of biosensors in accessible point-of-care diagnostics.

This study deals with the optimization of Fe(III) ion removal using activated carbon from olive stone waste using advanced machine learning models. The main objective is to evaluate and compare the performance of machine learning models, specifically multilayer perceptron artificial neural network (MLP‐ANN), general regression artificial neural network (GR‐ANN), radial basis function artificial neural network (RBF‐ANN), and particle swarm optimization artificial neural network (PSO‐ANN) in predicting Fe(III) removal efficiency. Experimental data on adsorption parameters were used to train and test the models. Techniques such as tuning hidden layer neurons, optimizing propagation values, and using a Taguchi approach PSO algorithm were applied to improve the models. For the MLP‐ANN model, the optimal configuration contains 13 neurons in the hidden layer. Concerning the parameters involved in the PSO‐ANN model, the coefficient C2 and the particle have the main effect on the reduction of the error. Their contributions are respectively 49% and 19%. The PSO‐ANN model showed superior performance with the highest regression coefficient (0.9997) and remarkable prediction accuracy, surpassing other models such as MLP‐ANN and GR‐ANN. This research suggests that innovative optimization techniques, particularly using PSO algorithms, significantly enhance the predictive capabilities of machine learning models in complex adsorption processes, contributing to more accurate Fe(III) removal models.

This paper delves into how social media, an increasingly pervasive global phenomenon, shapes the language nowadays. It zeroes in on emojis, a rapidly evolving means of communication that deviates from traditional verbal and nonverbal expressions, examining their impact on linguistic development. Emojis, initially conveying emotions, have shifted towards representing feelings rather than explicit meanings. Users now create entire messages using strings of emojis instead of sentences or phrases. The present study examines how people understand emoji-based messages, particularly through a survey conducted in Saudi Arabia. Additionally, an intelligent system leveraging machine learning is introduced to decode the meanings within emoji messages. The creation of EmojiString, a novel dataset, aids in better understanding these messages by utilizing advanced models like long short-term memory (LSTM) and MultiLayer Perceptron (MLP). The proposed model boasts an average accuracy of 82.22%, surpassing the existing methods. These results strongly support the idea that emojis serve as vital contextual cues in everyday communication. They are not just whimsical symbols, but meaningful elements that shape the interactions between persons. This research underscores the need to recognize emojis' nuanced roles in the evolution of modern language, marking a significant step forward in understanding their impact on how people communicate with one another.

This article delves into the application and efficacy of planar loop sensors for detecting partial discharge (PD) phenomena. The preference for loop sensors in PD detection stems from their simple design, non-destructive testing feature, and contactless characteristics. The paper explores both experimental and finite element modeling (FEM) approaches to evaluate the performance of distinct planar square spiral sensors in PD detection. The primary objective of the study is to compare inductive loops with loop antennas by varying the geometric parameters to achieve each approach. Two single-turn square loop antennas were designed, with a circumference of 30 cm and 75 cm, and compared with four multi-turn square inductive loops with an exterior side of 31.2 mm and varying geometric parameters. The figures of merit of all sensors, such as inductance, quality factor, and resonant frequency, are extracted through measurement and simulation to identify the most influential parameter that maximizes the sensing capability toward PD.

Using experimental results related to the biosorption of Fe(III) by activated carbon derived from olive pit waste, we developed and evaluated four artificial neural network (ANN) models in this study, namely MLP-ANN, RBF-ANN, GR-ANN, and PSO-ANN, to predict the removal efficiency of Fe(III) during the adsorption process. The purpose of these models was to forecast the effect of the following five important operational variables: the initial concentration, time, stirring speed, temperature, and biosorbent dose. We conducted a thorough assessment of the performance of these models and compared their ability to predict the removal capacity. Several statistical metrics have been used to quantify the quality of the different models. The results calculated by the machine learning models were analyzed and compared with the experimental results. The obtained values of the coefficient of determination were 0.9997 for the PSO-ANN model, 0.991 for the GR-ANN model, 0.983 for the RBF-ANN model, and 0.837 for the MLP-ANN model concerning the removal efficiency. All the studied models are able to accurately predict the adsorbed quantity. ANOVA analysis was used to evaluate the effect of the parameters inherent to the PSO-ANN model. The PSO-ANN model proves to be a powerful tool for estimating the efficiency of Fe(III) ion removal.

Impedance biosensing offers a highly sensitive and non-invasive method for detecting biomolecules and monitoring cellular activities, which is crucial for timely diagnosis and management of viral infections. Traditional methods, although effective, often involve costly and cumbersome equipment and require frequent hospital visits, making them less practical for continuous monitoring. This study introduces a novel biosensor based on SiNW-FET coupled with an advanced preamplifier designed to detect viruses through impedance changes caused by the interaction of antibodies and antigens. Utilizing COMSOL/MATLAB simulations, this research accurately models the sensor’s response to electrode functionalization with antibodies, and evaluates how nanoscale adjustments in electrode size and spacing can enhance biosensing capabilities. The proposed system promises continuous patient monitoring, alerting healthcare providers to critical changes that might indicate viral infections or significant shifts in cellular behavior. The performance of the sensor, validated through detailed simulations, demonstrates its potential as an effective tool for healthcare, ensuring timely interventions and improved patient outcomes. The proposed sensor design achieved an accuracy of approximately 92%, a sensitivity of about 85%, and a specificity of roughly 99%, demonstrating its high effectiveness in detecting viral infections and ensuring accurate monitoring. The sensor highlighted its potential as an effective tool for real-time, non-invasive healthcare monitoring.

In 2020, the world suffered from the COVID-19 pandemic. This situation highlighted the considerable need for systems to detect viruses quickly. Providing an accurate e-health system that detects, monitors, and controls virus spread became necessary. Countries attempt to reduce the virus's spread using basic processes such as quarantine and lockdown. Early virus detection is requested to reduce the threat and decrease the propagation of the pandemic. This paper proposes an attempt to avoid speed-up spreading based on airborne virus detection. The suggested detection based on the SiNW-FET biosensor is not approved as a standardized diagnostic test. To support the airborne detection methodology, a metasurface technology based on Reflector cylindrical antenna with a sensing unit is proposed to detect viruses. The proposal considered the propagation delays and concentration of a virus. Conducted results prove that the receiver's volume is considered a primary key to an accurate detection rather than the virus's concentration.

Precision agriculture, also referred to as smart farming, is one of the main pillars of modern society to achieve the Sustainable Development Goals (SDGs). Precision agriculture aims to improve the quality and quantity of production while conserving scarce natural resources. Smart farming has grown in recent years thanks to the adoption of modern technologies, including artificial intelligence (AI) and the Internet of Things (IoT). In this work, we consider an irrigation system for olive orchards based on unmanned aerial vehicles (UAVs). Specifically, UAVs ensure remote sensing (RS), which offers the advantage of collecting vital information on a large temporal and spatial scale (which cannot be achieved with traditional technologies). However, UAV-based irrigation systems face tremendous challenges due to the various requirements of a powerful computing ability, battery capacity, energy efficiency, and spectral efficiency for different connected devices. This paper addresses the energy efficiency and spectral efficiency trade-off problem of UAV-based irrigation systems. We propose to adopt massive multiple input, multiple output (M-MIMO) technology to ensure wireless communication. In fact, this technology plays a significant role in future sixth-generation (6G) wireless mobile networks and has the potential to enhance the energy efficiency as well as the spectral efficiency. We design a network model with a three-layered architecture and analytically compute the achievable spectral efficiency and the energy efficiency of the studied system. Then, we numerically determine the optimal number of ground base station antennas as well as the optimal number of IoT devices that should be used to ensure the maximum energy efficiency while guaranteeing a high spectral efficiency. The numerical results prove that the proposed UAV-based irrigation system outperforms conventional systems and demonstrate that the best spectral and energy efficiency trade-off is obtained by using the M-MMSE combiner.

Smart Agriculture, also known as Agricultural 5.0, is expected to be an integral part of our human lives to reduce the cost of agricultural inputs, increasing productivity and improving the quality of the final product. Indeed, the safety and ongoing maintenance of Smart Agriculture from cyber-attacks are vitally important. To provide more comprehensive protection against potential cyber-attacks, this paper proposes a new deep learning-based intrusion detection system for securing Smart Agriculture. The proposed Intrusion Detection System IDS, namely GMLP-IDS, combines the feedforward neural network Multilayer Perceptron (MLP) and the Gaussian Mixture Model (GMM) that can better protect the Smart Agriculture system. GMLP-IDS is evaluated with the CIC-DDoS2019 dataset, which contains various Distributed Denial-of-Service (DDoS) attacks. The paper first uses the Pearson’s correlation coefficient approach to determine the correlation between the CIC-DDoS2019 dataset characteristics and their corresponding class labels. Then, the CIC-DDoS2019 dataset is divided randomly into two parts, i.e., training and testing. 75% of the data is used for training, and 25% is employed for testing. The performance of the newly proposed IDS has been compared to the traditional MLP model in terms of accuracy rating, loss rating, recall, and F1 score. Comparisons are handled on both binary and multi-class classification problems. The results revealed that the proposed GMLP-IDS system achieved more than 99.99% detection accuracy and a loss of 0.02% compared to traditional MLP. Furthermore, evaluation performance demonstrates that the proposed approach covers a more comprehensive range of security properties for Smart Agriculture and can be a promising solution for detecting unknown DDoS attacks.

In this paper, we present a comprehensive analysis of two parallel conductors, fabricated on a printed circuit board (PCB), forming an elementary planar transformer, with an emphasis on extracting performance parameters such as self-inductance, resistance, coupling coefficient, mutual inductance, and inter-winding capacitance. Results issued from theoretical equations, FEM-based simulations, and characterizations were compared together to evaluate the mutual inductance. A procedure using the Open-Short de-embedding technique has been successfully applied to extract the correct performance values.

Abstract Introducing near Zero Energy Buildings (nZEBs) in the European Union involves integrating new renewable energy technologies into buildings. Geothermal energy is one of them that can be exploited through the application of the Earth-Air Heat Exchanger (EAHX). It is basically constituted of a series of pipes buried underground at a particular depth. It utilizes the soil as a heat source or sinks to supply cooling or heating to the building. This type of exchanger can offer many advantages in terms of energy savings for the air-conditioning requirements and for assuring the indoor thermal comfort. This paper presents a performance analysis of an EAHX applied to the ventilation of a near Zero Energy Building situated in Spain. The proposed considered the thermo-physical proprieties of the soil in the region under investigation. The purpose is to study the effect of changing the depth in the ground, the pipe length, the pipe diameter, and the air velocity on the outlet air temperature. The mean efficiency of the EAHX is detailed then. The numerical model is implemented using Matlab environment. Results showed that, for specific parameters, the daily average cooling capacity can reach 1.7 KWh, and the EAHX can provide an energy saving of approximately 152 MWh for one year.

Health has recently faced many challenges, including improving a healthy environment and reducing human life's dangers and economic crises. The last pandemic COVID-19 had badly affected survivor sectors with infection and lockdown exigence. Scientists proposed several solutions to reduce the negative impact of a such pandemic by proposing systems for earlier detection of viruses. The use of metamaterials as an emerging technology in the biosensors field allows a high accuracy. This paper presents a method for detecting and capturing airborne viruses using metasurface technology. The goal is to develop a system that can identify and capture these viruses using FET sensors. The accuracy of the detection is tied to the concentration of aerosols. The model proposes a guided flow of aerosols that positively impacts the detection of viruses through the FET biosensor. The simulation results based on Concentration and airflow velocity delays prove the proposed model's performance.

Health care systems have become essential parts of today's societies. However, the COVID-19 virus pandemic has highlighted the challenges that health care systems face. This need has encouraged scientists to improve these systems by supporting the Internet of Things (IoT) architectures and providing more contactless sensors. Still, traditional equipment based on using a contact sensor to measure vital signs, such as the heart rate, faces problems with accuracy and can have psychological effects on patients. The proposed solution aims to monitor the heart rate through non-contact measurement using a smart camera. This objective was achieved by applying an intelligent algorithm to a facial video sequence. This paper attempts to estimate the heart rate based on the color variation of the forehead skin. Specifically, the proposed method aimed to accurately estimate the heart rate in the standard condition and in the case of darker skin where the variation of the skin color was only slightly changed. The results proved that the estimation's accuracy could reach 98%.

The technological revolution affects the growth of systems in terms of functionality and complexity. Industries of embedded systems become increasingly an area of interest for researchers to develop Computer-Aided Design (CAD) environments to support at the same time the complexity in terms of different components and functionalities in terms of application programming interface and libraries. Mainly, CAD tools based on multi-level co-simulation are challenged by the time-to-market constraint. As known, the behavior description at a higher level provides a speedy simulation, but it suffers from bad accuracy. Therefore, describing a customized model for a system behavior with sufficient functional details at an earlier stage of modeling is a great challenge to researchers. As an attempt to overcome the last challenge, this paper presents a co-simulation model based on a synchronization methodology to ensure the verification between the conceptual level and the functional level. The proposed system-level co-simulation model is implemented to interfaces to provide the switch context in the case of the Arena and the Simulink/Matlab environments. The evaluation was performed by using two case studies with different domains to prove the effectiveness of the proposed system-level co-simulation interfaces.

Recent innovations in technology related to medical fields are widely wished to enhance prevention, diagnosis, and treatment. Seen that Alzheimer's Disease (AD) is hard to be identified at an earlier stage, many approaches and techniques are proposed. Detecting the AD-based in Magnetic Resonance Images (MRI) presents a great challenge. The recognition of AD helps to slow the effects of the disease when using an early treatment. An automated tool called Computer-Aided Diagnosis (CAD) is well invited to recognize and to identify AD. The motivation behind this work is to assess the features of how much explicit highlights the AD. The Gyrification index, the cortical thickness, and the Alzheimer's Disease Assessment Scale (ADAS) are studied in this paper. Many classifiers are implemented to highlight the best one. In this paper, we propose to use the classifier Data-driven Error Correcting Output Code (DECOC) prepared with Gyrification index, cortical thickness, and ADAS psychological grades. The proposed CAD framework identifies AD more accurately than others done by alternative classifiers. The outcomes prove that the cortical thickness and the ADAS psychological grades provide accurate identification of the AD instead of other features.

The revolution in technology affects many fields and among them the Healthcare system. The application-based computer was developed to help specialists to detect diseases, and to perform some basics operations. In this paper, focus is given on the proposed attempts to detect Epilepsy Disease (ED). Several Computer-Aided Diagnosis (CAD) methods were used to provide the brain’s disease status according to signals related to brain activities. These applications achieved acceptable results but still have their limitations. An intelligence CAD based on the Balanced Communication-Avoiding Support Vector Machine (BCA-SVM) is proposed to detect ED using Electroencephalogram (EEG) signals. This attempt is implemented on a Raspberry Pi 4 as a real board to ensure real-time processing. The CAD-based on BCA-SVM achieved an accuracy of 99.8% and the execution time was around 3.2s satisfying the real-time requirement.

As systems become increasingly complex, their simulation techniques have to be more accurate and enhanced. Despite the wide use of robotic arms in industries, they still encompass a wide number of complexities. The control of a flexible robot arm driven by a Brushless DC Motor (BDCM) for tracking problems is a great challenge, not only for its complex algorithms but also for its verification process. Robotic systems are designed heterogeneously by combining continuous and event discrete models. Therefore, computer-aided engineering tools have to be enhanced in order to support the verification of the control algorithms. Ensuring definite and rapid simulations is a challenging task for robotic application systems. Although control strategies can be tested using the Matlab/Simulink environment to assess their performance, their verification at a low-level still remains a very difficult task. This paper studies different simulation techniques based on Model In the Loop (MIL), Hardware In the Loop (HIL), and Hardware Software In the Loop (HSIL) on a flexible robot arm driven by BDCM in order to overcome each method's limits, focusing on performance and cost and simulation time reduction. The HSIL achieves the highest accuracy and speed than MIL and HIL and provides reusability and portability of the control unit compared to the other techniques.