Bulletin of Electrical Engineering and Informatics
Not a member yet
2885 research outputs found
Sort by
Design and implementation of an industrial security system using color cameras
Get PDFThis paper examines the design, development, and implementation of a modern industrial security system that integrates color cameras to enhance surveillance and improve safety. The system leverages cutting-edge technologies to detect intrusions and incidents with greater accuracy, which significantly strengthens the security of industrial sites. The study focuses on key stages of the project, including the design, installation, and operational processes, while addressing the challenges encountered during these phases. The integration of color cameras provides clearer and more precise monitoring, allowing for quicker detection and response to potential threats, thus reducing risks effectively. Our results demonstrate that the system greatly improves surveillance efficiency, providing a reliable and robust solution tailored to the security demands of industrial environments. This research offers in-depth insights into the system’s design and functionality, showcasing its critical role in safeguarding industrial facilities. Overall, the proposed solution is an essential tool for enhancing safety protocols and risk management strategies, contributing to more secure industrial sites
Heart disease detection using machine learning
Get PDFHeart disease continues to be a major worldwide health issue, requiring accurate prediction models to improve early identification and treatment. This research aims to address two main objectives in light of the increasing prevalence of heart-related disorders. Firstly, it aims to determine the most efficient classifier for identifying heart disease among twenty-nine different classifiers that represent six distinct learning strategies. Furthermore, the research seeks to identify the most effective method for selecting features in heart disease datasets. The results show how well different classifiers and feature selection methods work by using two datasets with different features and judging performance using four important criteria. The evaluation results demonstrate that the RandomCommittee classifier outperforms in diagnosing heart illness, displaying strong skills across various learning strategies. This classifier exhibits favorable results in terms of accuracy, precision, recall, and F1-score metrics, hence confirming its appropriateness for predictive modeling in heart-related datasets. Moreover, the paper examines feature selection methods, specifically aiming to determine the most effective method for enhancing the predicted accuracy of heart disease models. The prediction models' overall performance is enhanced by their capacity to accurately identify and prioritize pertinent variables, thereby facilitating the early detection and management of heart-related problems
Comparative analysis of PoS and PoA consensus in Ethereum environment for blockchain based academic transcript systems
Get PDFMany educational institutions worldwide now use blockchain to verify electronic document, often relying on Ethereum 1.0, which uses proof of work (PoW) or proof of authority (PoA). However, Ethereum 2.0, launched in 2022 by Ethereum Foundation operates on proof of stake (PoS). This study provides comparative analysis of PoS and PoA consensus in Ethereum environment specifically focusing on performance and scalability in the context of academic transcript databases. To demonstrate this, a student academic reputation information system was developed using two different blockchain technologies: Ethereum 1.0 with PoA and Ethereum 2.0 with PoS. This setup was used to obtain comparative analysis data for the two blockchain systems by measuring the throughput and latency. We observed how these platforms responded to an increasing number and frequency of transactions with Hyperledger Caliper. Results indicates that in performance testing, both consensus mechanisms exhibited. Scalability tests revealed that both consensus mechanisms experienced increased latency with higher loads. However, PoA system was superior in average throughput and latency than PoS system except in high transaction of data addition. The experiment result show that PoA system better than PoS system in context of academic transcript databases, making it more suitable to be implemented on that context
Enhancing manufacturing efficiency: leveraging CRM data with Lean-based DL approach for early failure detection
Get PDFIn the pursuit of enhancing manufacturing competitiveness in India, companies are exploring innovative strategies to streamline operations and ensure product quality. Embracing Lean principles has become a focal point for many, aiming to optimize profitability while minimizing waste. As part of this endeavour, researchers have introduced various methodologies grounded in Lean principles to track and mitigate operational inefficiencies. This paper introduces a novel approach leveraging deep learning (DL) techniques to detect early failures in manufacturing systems. Initially, realtime data is collected and subjected to a normalization process, employing the weighted adaptive min-max normalization (WAdapt-MMN) technique to enhance data relevance and facilitate the training process. Subsequently, the paper proposes the utilization of a triple streamed attentive recalling recurrent neural network (TSAtt-RRNN) model to effectively identify Leanbased manufacturing failures. Through empirical evaluation, the proposed approach achieves promising results, with an accuracy of 99.23%, precision of 98.79%, recall of 98.92%, and F-measure of 99.2% in detecting early failures. This research underscores the potential of integrating DL methodologies with customer relationship management (CRM) data to bolster early failure detection capabilities in manufacturing, thereby fostering operational efficiency and competitive advantage
Securing patient data and access control in electronic health records with Ethereum blockchain
Get PDFBlockchain technology has become an essential tool for enhancing reliability and security across several industries, including the healthcare sector. In this work, we propose and implement an Ethereum-based blockchain framework to decentralize electronic health records (EHRs) at Tumakuru Siddaganga Hospital. The system establishes an append-only chain of transaction blocks that guarantees the confidentiality, auditability, and integrity of patient health records. By design, only authorized healthcare professionals can access patient data, and even then, only with the patient’s explicit consent—ensuring a privacy-preserving access model. Our approach demonstrated a 40% reduction in data access delays and eliminated unauthorized access attempts through smart contract-based access control. The decentralized nature of the framework reduces reliance on centralized databases, significantly lowering the risk of data tampering and breaches. Additionally, the implemented consensus protocol ensures that only verified transactions are recorded, maintaining consistency across distributed nodes. Compared to traditional systems, our blockchain-based solution improved the traceability of health data access events by 100%, ensuring transparency and accountability. These findings validate that blockchain technology can substantially enhance data sharing, integrity, and patient control in modern healthcare systems
Analysis of unmanned aerial vehicle airframe materials on circularly polarized antenna radiation characteristics
Get PDFThis paper presents an experimental examination of how unmanned aerial vehicle (UAV) airframe materials affect the electromagnetic characteristics of the airborne circularly polarized (CP) payload antenna. This study specifically investigates the received signal from the circularly polarized synthetic aperture radar (CP-SAR) antenna installed within the fuselage of the lapan surveillance UAV (LSU). In the airborne CP-SAR experiment, broadband CP microstrip subarray antennas were used along with LSU series airframe material composites comprising E-glass EW-185 and Carbon C522 Twill. The composite specimens were prepared to have the same size and thickness to minimize variability in the comparative analysis. The experimental study measures the transmission loss using S-parameters. At 5.3 GHz, the E-glass EW-185 fiber composite exhibits a material attenuation of -1.5 dB and a circular depolarization of 0.32 dB. The E-glass EW-185 fiber composite exhibits a material attenuation of -1.5 dB and a circular depolarization of 0.32 dB. In contrast, the Carbon C522 Twill fiber composite demonstrates a significantly higher material attenuation of -31.24 dB and a circular depolarization of 10.70 dB. Additionally, this paper examines the radiation pattern measurements of the CP-SAR antenna at various frequencies, providing a comprehensive analysis of the materials' impact on antenna performance
Adaptive fuzzy sliding-mode control for robot manipulator with uncertain model and external disturbance
Get PDFIn practice, robots operate as nonlinear systems and often encounter factors like nonlinear friction, load variations, and external disturbances during tasks. To address these challenges, a smart control approach has been developed that combines the strengths of fuzzy logic and sliding mode control (SMC) for precise robot manipulator positioning. The key benefit of SMC lies in its robustness, maintaining stability despite noise or parameter changes in the system. However, designing an SMC system often faces difficulties due to practical limitations, making deployment not always feasible in real-world applications. Additionally, a large control law amplitude can lead to chattering around the sliding surface. To overcome these issues, the study introduces a fuzzy logic-based method to adaptively estimate the control law's magnitude, guided by Lyapunov stability principles. This control scheme is tested on a four-degree-of-freedom robot manipulator, with simulation results confirming its effectiveness in MATLAB
COMATS: a cuckoo-mimicking data anonymization scheme for preserving sensitive preferences in transaction data
Get PDFSharing customer transaction data is becoming more perceived in e-commerce and retail industries. Even though the act derives benefits for companies, it may end up in certain privacy threats, such as sensitive personal preferences disclosure. Therefore, the data owner should take measures to minimize the threats. Data anonymization is one of the solutions that has been suggested to address the issue. However, there are still underlying problems, specifically in diminishing the amount of information loss and item loss, as well as maintaining data properties of the anonymized dataset. This paper proposes a unique data anonymization scheme called COMATS. It adopts the brood parasitism behavior of cuckoo birds in laying their eggs into host nests. The scheme incorporates item insertion technique and item suppression technique. The robustness of the proposed scheme lies in its strategy for selecting suppressed items and determining the inserted items. To ensure its efficacy, the proposed method is evaluated in several experiments. The experimental results suggest that the COMATS can guarantee privacy protection by reducing the probability of a successful attack. Additionally, it can also reduce the number of item losses and preserve better data utility in comparison to existing data anonymization schemes
Unmasking effects of feature selection and SMOTE-Tomek in tree-based random forest for scorch occurrence detection
Get PDFScorch occurrence during the production of flexible polyurethane foam has been a menace that consistently, jeopardize a foam’s integrity and resilience. It leads to foam suppression and compactness integrity failure due to scorch. There is always the increased likelihood of scorching, and makes crucial the utilization of methods that seek to avert it. Studies predict that the formation of foam constituent processes via optimization using machine learning have adequately trained models to effectively identify scorch occurrence during the profiling in the polyurethane foam production. Our study utilizes the random forest (RF) ensemble with feature selection (FS) and data balancing technique to identify production predictors. Study yields accuracy of 0.9998 with F1-score of 0.9819. Model yields 2-distinct cases for (non)-occurrence of scorch respectively, and the ensemble demonstrates that it can effectively and efficiently predict the occurrence of scorch in the production of flexible polyurethane foam manufacturing process
The adoption of online food delivery in facing COVID-19 among the Indonesian food MSMEs
Get PDFThis study investigates the factors influencing the Indonesian food micro, small, and medium enterprises (MSMEs) in adopting online food delivery (OFD) during the corona virus disease-2019 (COVID-19) pandemic, by employing the technology-organization-environment (TOE) framework. Through a quantitative approach involving 378 respondents, this research explores the multi-dimensional factors affecting OFD service adoption, there are innovation compatibility, innovation complexity (IC), innovation cost, owner’s self-efficacy, owner’s commitment, customer pressure (CSP), competitive pressure (CMP), government support (GS), and health protocol guarantee. Employing covariance-based structural equation modeling (CBSEM), the study reveals interesting relationships among the proposed factors. The findings underscore the role of GS and health protocol guarantees in enhancing owner's self-efficacy and commitment towards OFD adoption. Moreover, it challenges the presumed barriers of IC, suggesting a nuanced understanding of adoption process amid a crisis. This study not only enriches the theoretical discourse on technology adoption in the context of a pandemic but also provides practical implications for stakeholders in navigating the post-pandemic business landscape. Future research directions are proposed to explore the continuous intention of food MSMEs towards OFD services postpandemic, highlighting the evolving nature of the global business environment and the enduring impact of the COVID-19 pandemic on food industries