Bulletin of Electrical Engineering and Informatics
Not a member yet
2885 research outputs found
Sort by
The use of fiber bragg grating coated with polyimide for CO2 gas sensor
Get PDFThis study presents the application of fiber bragg grating (FBG) sensors coated with polyimide for detecting carbon dioxide (CO₂) gas, employing both theoretical and experimental approaches. The basic FBG components were coated with polyimide layers of varying thicknesses. Subsequently, the fabricated FBG sensors were characterized using an optical interrogator system with four channels. Furthermore, the sensor was tested for CO₂ detection at a working temperature of 47 °C. Experimental data showed that the FBG sensor coated with polyimide layers of 10 nm, 15 nm, and 20 nm demonstrated sensitivities of 1.9 ppm, 1.84 ppm, and 1.8 ppm, respectively. In contrast, the uncoated FBG sensor exhibited a higher sensitivity of 3 ppm. Increasing the coating thickness beyond 20 nm leads to a decrease in sensor sensitivity. The findings suggest that an optimal polyimide coating thickness for CO₂ detection using FBG sensors is around 20 nm. Achieving high sensitivity in CO₂ gas sensors is crucial for their effective use across a broad range of applications
Iris-based lung cancer pre-scanning for mobile platforms
Get PDFLung cancer remains one of the leading causes of cancer-related mortality globally, with early detection being critical for improving survival rates. Traditional diagnostic methods such as computed tomography (CT) scans and biopsies are effective but often costly, invasive, and inaccessible in resource-limited settings. In this study, we evaluate suitable deep learning models for mobile platforms and propose an application for early detection of lung cancer based on iris images. Through experimentation and comparison, the results show that the MobileNet model family achieves high performance while maintaining a light-weight architecture. The positive results of this study further strengthen the potential application of iris in the pre-diagnosis of lung cancer via mobile platforms
Feasibility study of solar-diesel generation hybrid power systems: a case study of rural electrification in Papua, Indonesia
Get PDFThis study presents a contextually tailored off-grid hybrid energy system consisting of solar photovoltaic (PV), diesel generator (DG), and battery storage, designed for the remote mountains village of Jifak in Papua, Indonesia. The objective is to evaluate the technical, economic, environmental, and social feasibility of electrification in underserved regions. A comprehensive feasibility analysis was conducted using HOMER software, incorporating realistic communal load profiles, National Aeronautics and Space Administration (NASA) climate data, and field-based social assessments. The optimized system achieved a net present cost (NPC) of 0.5095/kWh, and annual emissions of 7,965 kg COâ‚‚. Sensitivity analysis was performed on fuel cost, discount rate, and inflation rate to assess system robustness. Beyond the technical metrics, the study assessed the socio-economic impacts of electrification, revealing improvements in lighting quality, education, productivity, income generation, and environmental awareness. The findings provide a replicable model and decision-making framework for policymakers and practitioners aiming to deploy low-carbon, sustainable electrification strategies in similarly remote regions
MAS-TENER: a modified attention score transformer encoder for Indonesian skill entity recognition
Get PDFSkill entity recognition is a crucial task for aligning educational curricula with the evolving needs of the industry, particularly in multilingual job markets. This study introduces modified attention score transformer encoder (MAS-TENER), a novel transformer-based model designed to enhance the recognition of skill entities from Indonesian job descriptions. The proposed model modifies the attention mechanism by integrating relative positional embeddings and removing the scaling factor in self-attention. These improvements enhance the context of tokens, allowing for the accurate establishment of hard skills, soft skills, and technology skills. The MAS-TENER model was pre-trained and fine-tuned using a combinF.ation of job description datasets and additional corpora, achieving an F1-score of 90.46% at the entity level. The experimental results demonstrate the model's ability to handle unstructured, mixed-language job descriptions, with significant potential for curriculum reform and the development of new workforce capabilities. The study demonstrates the efficacy of the MAS-TENER model as an effective response for any natural language processing (NLP) task in low-resource languages. Moreover, the scope of long-term job market analytics in action research has been a key skill set in the education policy arena, demonstrating collaborative workforce capabilities
Cyber security threats and web vulnerability analysis of higher educational institutions in Bangladesh
Get PDFThis paper presents a comprehensive analysis of cyber security threats and web vulnerabilities in the context of higher educational institutions in Bangladesh, including twenty public and private universities. Educational institutions are highly vulnerable due to their negligence in maintaining a functional network, mainly owing to budgetary constraints. As a result, they have become a hacker playground for many ambitious adversaries to boast their technical skills, regardless of the harm they may inflict. Through the use of vulnerability assessment and penetration testing (VAPT), we conducted a methodical analysis of the institutions’ web infrastructures, identify and categorize the prevalent security threats and vulnerabilities that may compromise the integrity, confidentiality, and availability of information systems. Our findings reveal significant disparities in the security strength of both public and private universities, with the latter demonstrating a higher degree of vulnerability due to varying factors, such as budget constraints, policy enforcement, and awareness levels. This study underscores the urgent need for robust cyber security frameworks tailored to the higher educational sector’s unique requirements, advocating for proactive measures to mitigate potential cyber threats. The implications of this research extend beyond the academic domain, offering insights into national cyber security strategies and the safeguarding of critical information infrastructures
Enhanced speech recognition in natural language processing
Get PDFSpeech recognition is crucial for helping individuals with physical disabilities access digital content. However, current systems have significant flaws that hinder user experience and complicate daily tasks. Environmental disturbances can cause misinterpretation, and existing automatic speech recognition (ASR) systems struggle with comprehending acoustic and linguistic nuances and handling diverse speaking styles and accents. To address these issues, a new model integrates bidirectional encoder representations from transformers (BERT) and transformer features with natural language processing (NLP) capabilities. This model aims to consolidate semantic, linguistic, and acoustic information extracted from the Kaldi speech recognition toolkit and improve accuracy by rescoring the list of N-best hypotheses. The innovative approach leverages advancements in NLP to enhance speech recognition's accuracy and robustness across various scenarios. Evaluations on the LibriSpeech dataset show that integrating BERT, transformer encoder, and generative pretrained transformer 2 for rescoring N-best hypotheses significantly improves transcription accuracy. The proposed model achieves a word error rate (WER) of 17.98%, outperforming other models. This development paves the way for advancements in speech recognition technology, offering better user experiences in real-world applications
An innovative design of a frequency-tunable UHF RFID antenna for identification applications
Get PDFThis paper introduces the design of a new frequency-reconfigurable ultra-high frequency radio frequency identification (UHF RFID) antenna, demonstrating an innovative approach that enables dynamic adjustment of its resonance frequency. The proposed antenna design features a central dipole structure, enhanced by two hexagonal split-ring resonators (H-SRR) at each end. A T-match network is integrated into the center of the dipole, which is essential for achieving impedance matching between the antenna and the Alien Gen2 H4 RFID microchip. The antenna is designed using a Rogers 4350B substrate, a high-performance dielectric material ideal for RFID applications. With dimensions of 68×32.6×1.524 mm3, the compact antenna maintains full UHF band (860 MHz to 930 MHz) coverage compliant with International Telecommunications Union (ITU) RFID standards. This ensures that the antenna can be used in different regions around the world, offering broad compatibility with various RFID systems. The antenna's frequency reconfigurability is achieved through the integration of localized capacitors with variable values, which plays a key role in enabling precise adjustments to the antenna's center frequency across the entire UHF band. Extensive simulation results validate the effectiveness of this reconfigurable design, demonstrating that the antenna can dynamically adjust its frequency while maintaining excellent performance metrics, including impedance matching, radiation efficiency, and bandwidth. This makes the proposed antenna an ideal choice for modern RFID applications
Optimizing job scheduling on cloud resources using the first-come, first-served-SlotFree method
Get PDFCloud computing environments encounter significant challenges in job scheduling, particularly due to excessive waiting times and inefficient resource utilization associated with conventional algorithms such as first-come, first-served (FCFS) and backfilling. This study introduces FCFS-SlotFree, a novel scheduling algorithm that enhances resource allocation efficiency by dynamically sorting jobs based on their arrival times and workloads, and subsequently assigning them to a fixed set of virtual machines (VMs) without relying on rigid time-slot constraints. This flexible scheduling approach facilitates better adaptation to heterogeneous workloads. Extensive experiments conducted under realistic cloud scenarios demonstrate that FCFS-SlotFree significantly reduces average waiting time (AWT) by approximately 32.78% compared to FCFS and by 9.68% compared to backfilling, while concurrently improving resource utilization by 3.58% and 1.27%, respectively. The results substantiate the algorithm’s effectiveness in optimizing scheduling performance and resource efficiency within complex cloud environments
A powerful machine learning method for detecting phishing threats
Get PDFPhishing threats exploit social engineering and deceptive web infrastructure to steal sensitive personal information, often by mimicking legitimate websites. With the proliferation of online services and the increasing prevalence of cybercrime, detecting phishing websites has become a critical challenge. This study presents a comprehensive machine learning (ML)-based approach for detecting phishing websites. A total of 48 discriminative features were extracted from 10,000 web pages—comprising 5,000 phishing and 5,000 legitimate sites. Nine ML classifiers were initially evaluated, including random forest (RF), support vector machine (SVM), and XGBoost. Ensemble models based on soft voting and stacking were then constructed to improve detection performance. Among the models, the soft voting classifier (VC) achieved the best performance with an accuracy and F1-score of 98.82%. The results indicate that ensemble learning offers a robust solution for the automated detection of phishing websites
Optimal tuning of robust proportional integral derivative based on sliding mode controller for an AVR system
Get PDFThe primary objective of the automated voltage regulator (AVR) is to maintain the terminal voltage of the synchronous generator at the specified level with great precision in power production systems. Accurate voltage regulation improves the longevity of equipment intended for operation at the specified voltage within a power system network. This study presents a robust control of an AVR system utilizing proportional integral derivative (PID) control based on sliding mode techniques. The suggested control method is implemented by utilizing the particle swarm optimization (PSO) technique to tune the parameters of the proposed controller in the AVR system. A comparative performance analysis is conducted between the proposed controller, PID controller, and (PSO-fractional order proportional integral derivative (FOPID) controller. The comparison is derived using transient response characteristics and parameter uncertainty. The results reveal that the proposed PSO-PID-sliding mode control (SMC) controller has superior performance, characterized by rapid convergence, reduced overshoot, stability achievements in time domains, and robustness against parameter fluctuations. The proposed controller has markedly enhanced the performance of the AVR system and can be effectively implemented inside it