Bulletin of Electrical Engineering and Informatics
Not a member yet
2885 research outputs found
Sort by
DEMNET NeuroDeep: Alzheimer detection using electroencephalogram and deep learning
Get PDFAlzheimer’s disease (AD) stands out as the most prevalent neurological brain disorder, and its diagnosis relies on various laboratory techniques. The electroencephalogram (EEG) emerges as a valuable tool for identifying AD, offering a quick, cost-effective, and readily accessible means of detecting early-stage dementia. Detecting AD in its early stages is crucial, as early intervention yields more successful outcomes and entails fewer risks than treating the disease at a later stage. The objective of this research is to create an advanced diagnosis system for AD using machine learning (ML) and EEG data. The proposed system utilizes a multilayer perceptron (MLP) and a deep neural network with bidirectional long short-term memory (BiLSTM) as the classifier. The feature extraction process involves incorporating Hjorth parameters, power spectral density (PSD), differential asymmetry (DASM), and differential entropy (DE). The BiLSTM classifier, particularly when combined with DE, exhibits outstanding performance with an accuracy of 97.27%. This amalgamation of DE and the deep neural network surpasses current state-of-the-art techniques, underscoring the substantial potential of this approach for precise and advanced diagnosis of AD
The use of generative adversarial network as a domain adaptation method for cross-corpus speech emotion recognition
Get PDFThe research of speech emotion recognition (SER) is growing rapidly. However, SER still faces a cross-corpus SER problem which is performance degradation when a single SER model is tested in different domains. This study shows the impact of implementing a generative adversarial network (GAN) model for adapting speech data from different domains and performs emotion classification from the speech features using a 1D convolutional neural network (CNN) model. The results of this study found that the domain adaptation approach using a GAN model could improve the accuracy of emotion classification in speech data from 2 different domain such as the ryerson audio-visual database of emotional speech and song (RAVDESS) speech corpus and the EMO-DB speech corpus ranging from 10.88% to 28.77%, with the highest average performance increase across three different class balancing method reaching 18.433%
Distributed denial-of-service attack detection short review: issues, challenges, and recommendations
Get PDFAn attacker can attack a network in several methods when there are a lot of device connections. Distributed denial-of-service (DDoS) attacks could result from this circumstance, which could damage resources and corrupt data. Therefore, irregularity in traffic data must be detected to identify malicious behavior in a network, which is critical for maintaining the integrity of current cyber-physical systems (CPS) as well as network security. This article attempts to study and compare various approaches to detecting DDoS attacks and expresses data paths for packet filtering for high-speed networks (HSN) performance, using machine or deep learning techniques used in intrusion detection systems (IDSs) and flow-based IDSs. The study presents a comprehensive DDoS attack taxonomy, categorizes detection strategies, and highlights the HSN accuracy assessment features. By exposing the problems and difficulties associated with DDoS attacks on HSN, several investigation paths are proposed to assist researchers in determining and developing the best solution
X-band and Ku-band PIN diode loaded reflectarray unit cells with adaptive frequency switching
Get PDFThe fast advancement of intelligent new applications has led to the creation of high-performance antennas. Reflectarrays (RAs), also known as planar reflectors, are seen as promising antennas for several such modern-day applications. This work presents a comprehensive investigation of frequency switchable RA antennas operating in the X-band and Ku-band frequency ranges. Various strategic configurations of combined slots have been suggested, using integrated P-layer, I-layer, and N-layer (PIN) diodes, with the purpose of creating unit cells in RAs that may switch frequencies and exhibit a gradual change in phase distribution. The frequency variation achieved in X-band for the ON state of PIN diodes is from 8.13 GHz to 11.69 GHz, whereas for the OFF state it is from 8.13 GHz to 11.68 GHz. Similarly, for Ku-band ON and OFF states of PIN diodes provided frequency variations of 13.6 GHz to 17.1 Ghz and 12.8 Ghz to 16.6 GHz respectively. Frequency tunability of 0.85 GHz and 0.72 GHz has been successfully achieved in X-band and Ku-band
Metamaterial inspired miniaturized ultra-wideband monopole hexagonal antenna with triple band-filter functions
Get PDFIn this letter, a new technique to the design of an ultra-wideband (UWB) monopole hexagonal antenna with triple band-rejected functions and to restrict the interferences with the exist bands is proposed, the design has the form of a hexagonal patch and a ground plane having rectangular shaped etched in the back side of the substrate to achieve the UWB behavior. The triple-band filter feature is generated by inserting a metamaterial (MTM) as a split ring resonator slots (SRRs) and a complementary split ring resonators (CSRRs) strip, thus no extra size is needed. The triple band-elimination is for 3.3-3.9 GHz centered at 3.5 GHz for 5G band, 4.99-5.4 GHz centered at 5.2 GHz for wireless local area network (WLAN) band, and 6.2-6.8 GHz centered at 6.5 GHz for IEEE INSAT/Supra-extended C-band. The antenna dimension has a compact size of 20×25×1.6 mm3. Current distribution on the antenna is used to analyze the effect of MTMs on the antenna operations. The simple structure and small size of the antenna makes it suitable for most of the wireless communication systems
Multimodal deep learning from sputum image segmentation to classify Mycobacterium tuberculosis using IUATLD assessment
Get PDFTuberculosis (TB) continues to be a major global health issue, especially in areas with limited resources where diagnostic tools are often insufficient. Traditional TB detection methods are slow and lack sensitivity, particularly for early-stage or low bacterial load cases. This study introduces a new multimodal deep learning model that integrates sputum image segmentation across RGB, hue, saturation, and value (HSV), and CIELAB color channels, using the YOLOv8 model for real-time detection and segmentation. The model uses the International Union Against Tuberculosis and Lung Disease (IUATLD) grading scale for accurate Mycobacterium tuberculosis (MTB) classification. Our approach shows high accuracy (92.24%) and precise forecasting (mean absolute percent error (MAPE) of 0.23%), greatly enhancing diagnostic speed and reliability. This research offers a novel method for classifying MTB using a multimodal deep learning model that integrates sputum image segmentation across RGB, HSV, and CIELAB color channels. By using the YOLOv8 model for real-time bounding box detection and segmentation, and the IUATLD grading scale for classification, our method achieves high accuracy and precision in identifying TB bacteria. Our findings indicate that this multimodal deep learning approach significantly improves diagnostic accuracy and speed, providing a reliable tool for early TB detection
Comparative simulation of phishing attacks on a critical information infrastructure organization: an empirical study
Get PDFNowadays, cybersecurity is crucial. Therefore, cybersecurity awareness should be a concern for businesses, particularly critical infrastructure organizations. The results of this study, using simulated phishing attacks, indicate that in the first attempt, workers of a Thai railway firm received a phony email purporting to inform recipients of a special deal from a reputable retailer of information technology (IT) equipment. The findings showed that 10.9% of the 735 workers fell for the scam. This demonstrates a good level of awareness regarding cyber dangers. The workers who were duped by the initial attack received awareness training. Next, a second attempt was carried out. This time, the strategy was for the workers to change their passwords through an email notification from the fake IT staff. According to the findings, 1.4% of the workers fell victim to both attacks (different email content), and a further 8.0% of the workers who did not fall victim to the first attack were deceived. Furthermore, after the statistical analysis, it was confirmed that there is a difference in the relationship between the workers and the two phishing attack simulations using different content. As a result, this study has demonstrated that different types of content can affect levels of awareness
Position control to expand the headlights’ angle of a car by DC motor drive system using PSO algorithm for speed loop
Get PDFToday, along with the robust development of society, cars are almost considered a primary means of transportation. This article focuses on designing headlight controls for older car models that are not equipped with adaptive headlight systems (AHS), which are different from modern cars such as Porsche, BMW, Audi, and Mercedes-Benz vehicles. The design is for a lighting system that operates during nighttime to improve illumination and enhance visibility in curves, with cost-effective and suitable solutions for average vehicles to ensure safety. This system uses a DC motor to control the headlight angle based on the steering wheel rotation. It is combined with the particle swarm optimization (PSO) algorithm to find the best response parameters for the proportional-integral-derivative (PID) controller. Research results on the MATLAB/Simulink and the experimental model show that the model established by this method has good accuracy, the controllers can significantly reduce the excessive deviation of the headlights’ operational precision, and traffic accidents can be minimized, increasing safety for users
Develop a quantum key distribution application based on the BB84 protocol combined with a classical channel
Get PDFAmid the escalating concerns over internet security, quantum cryptography stands out as a highly promising solution for significantly enhancing the security of networking systems, emerging among them is the quantum key distribution (QKD) with the function of creating secret session keys a breeze when leveraging the intriguing properties of quantum mechanics. This study is rooted in the BB84 QKD method, where in the distribution process in the quantum realm is simulated to derive a shared key via a public channel connecting two clients with the assistance of a server, utilizing the quantum inspire (QI) platform to generate qubits within the BB84 protocol. The results, the findings regarding the performance of BB84 reveal that when the server is set up, and the key size increases to 4000 bits, the process of sending module takes 16.215 sec, the transfer module takes approximately 5.2 hours, the receive module takes 1.257 sec to finish the process for the final session key share. This indicates a noteworthy enhancement in the execution speed of QKD employing the BB84 protocol, which now holds the potential for reinforcing network security using quantum computing systems
Recent developments in vehicle routing problem under time uncertainty: a comprehensive review
Get PDFThis review paper examines recent advancements in vehicle routing optimization under time uncertainty, focusing on the vehicle routing problem (VRP). It sys-tematically analyzes research papers to identify strategies for optimizing routes despite temporal uncertainties, covering key areas such as optimization algo-rithms, uncertainty modeling techniques, and simulation methods. The study investigates dynamic dispatching models, reliability considerations, and multi-objective optimization approaches. By synthesizing existing literature, this pa-per presents the current state of research in vehicle routing under time uncer-tainty and suggests potential future research directions. Our findings indicate that integrating robust optimization techniques with advanced simulation meth-ods could significantly enhance decision-making processes in uncertain envi-ronments. Additionally, the paper highlights the role of machine learning and artificial intelligence in developing adaptive algorithms that respond to dynamic changes in real-time. As the need for efficient logistics solutions grows, this comprehensive review underscores the importance of addressing uncertainties in vehicle routing to improve operational efficiency, reduce costs, and enhance customer satisfaction