Bulletin of Electrical Engineering and Informatics
Not a member yet
2885 research outputs found
Sort by
Efficient brain cancer identification using ResNet50 and ResNet50 V2: a comparative study with a primary MRI dataset
Get PDFPrimary malignant brain tumors along with central nervous system cause a significant amount of deaths every year, making brain cancer a major worldwide health problem. In South Asian countries, the number of patients suffering from brain cancer is steadily rising. Treatment effectiveness and improved patient outcomes depend on early detection. Using a dataset consisting of 6056 original raw MRI scans, this study evaluates how well convolutional neural networks (CNNs) diagnose brain cancer. We present ResNet50 and ResNet50V2 models assessed for their effectiveness in identifying brain cancers. Transfer learning and fine-tuning were employed to enhance model performance. The models demonstrated strong performance, with 87-99% accuracy rate. ResNet50V2 achieved the highest 99% accuracy. To detect tumor early, this work emphasizes how well the CNN-based machine learning methods help as timely intercession and patient care is necessary. Early prediction with 100% confidence and reliable precision is a critical issue in the modern world. Our goal is to use advanced algorithms to forecast images affected by cancer. Lastly, we will deploy an automated system that will enable us to confidently identify images affected by cancer. Our suggested methodology and its application could significantly impact the field of medical science by combining computer vision and health informatics
Performance of an internet of things based plant monitoring and irrigation system using solar energy
Get PDFThis paper aims to present an internet of things (IoT) based plant monitoring and irrigation system powered by solar energy. The system enhances plant care by continuously monitoring environmental conditions such as soil moisture, temperature and humidity with real data displaye via the Blynk platform. User can remotely monitor and control irrigation through interactive widgets, ensuring efficient plant management. By integrating solar energy, the system operates sustainably, and reduce reliance on conventional electricity. Performance evaluation demonstrates a temperature sensor accuracy of 98%, a humidity sensor accuracy of 95% and soil moisture sensor error margin of 2-3%. Experiment results indicate improved plant growth of 7-8% compared to traditional farming practices, showcasing the system’s potential for increased productivity and conversion. This research highlight the benefits of combining IoT and renewable energy to offer an innovative, and eco-friendly solution for agricultural management
5G cellular network planning in Parepare City
Get PDFThe telecommunications industry is rapidly advancing, particularly in cellular network communications that use air as the transmission medium, with 5G new radio (NR) emerging as a key global technology including in Indonesia. Defined by enhanced mobile broadband (eMBB) offering speeds up to 10 Gbps, ultra-reliable low-latency communications (URLLC) with latency below 1 millisecond, and massive machine-type communications (mMTC) supporting large-scale internet of things (IoT) connectivity, 5G plays a crucial role in modern digital infrastructure. This study focuses on the city of Parepare in South Sulawesi, an area driven by trade, port operations, fisheries, shipbuilding, and natural tourism highlighting the need for high-speed and reliable data services. The research aims to develop a comprehensive 5G NR network plan for Parepare through coverage and capacity analyses evaluating synchronization signal-reference signal received power (SS-RSRP), signal-to-interference-plus-noise ratio (SS-SINR), and throughput performance. Using Atoll software to design and map next-generation Node B (gNodeB) placements, the study offers a scientific approach to optimizing 5G deployment and supporting the city’s economic growth and tourism potential
A hybrid random forest and particle swarm optimization model for early preeclampsia detection
Get PDFPreeclampsia has become a serious medical problem in the world. Currently, there is no routine or comprehensive screening program in place for preeclampsia, which means that preventive measures are not as effective as they could be, potentially resulting in higher rates of illness and death among mothers and infant. The main purpose of this study is to predict early of preeclampsia using random forest algorithms. This study used a quantitative approach with samples 504. The data were analyzed using random forest with particle swarm optimization (PSO). Random forest have been an accuracy rate of 96.08%, for the area under the curve (AUC), precision, sensitivity, and specificity each (0.971; 97.06%; 97.06%; and 94.12%). Model significantly increased 1.39% after optimize from 94.69% to 96.08%. The design process model algorithm has been validated that have a high level of accuracy based on literature reviews. The quality of services offered will certainly influence people to utilize technology-based services more than conventional ones. Recommendation for field technology and health is building an application model for early prediction of preeclampsia based on machine learning (ML) which is an effort for health workers to provide optimal antenatal care and step in changing technology-based pregnancy checks as initial prevention for pregnant women so that preeclampsia can be avoided
Design and implementation of a solar-powered IoT-based real-time air quality monitoring system
Get PDFAir pollution has become a global issue due to rapid urbanization and industrialization. Air quality monitoring is essential for mitigating the adverse effects of air pollution on public health and the environment. This study presents a solar-powered internet of thing (IoT)-based air quality monitoring system designed for autonomous operation in outdoor settings. The prototype integrates an ESP32 microcontroller with low-cost sensors for PM2.5, PM10, temperature, humidity, and heat index. Powered by a solar panel and battery, the system ensures off-grid functionality, while Wi-Fi transmission to the Blynk platform, enables real-time visualization, historical record storage, and instant user access through mobile dashboards. The system was calibrated against reference instruments and deployed for 14 consecutive days. Results confirmed stable data transmission and reliable performance that suitability for outdoor use without reliance on grid power under real-world conditions. Furthermore, correlation analysis showed a strong relationship between PM2.5 and PM10, and moderate associations with humidity. Regression analysis further identified humidity and heat index as the most significant predictors, while temperature exhibited only minor influence. These findings demonstrate the feasibility of a low-cost, portable, and energy-autonomous IoT monitoring system, providing accurate real-time insights to support evidence-based air quality management
Online PID-neural network for tracking lower limb rehabilitation exoskeleton angular position
Get PDFGait trajectory tracking control is an essential component of a lower limb rehabilitation exoskeleton (LLRE). Meanwhile, the proportional-integral-derivative (PID) controller remains popular for a variety of applications, including LLRE. Nonetheless, employing PID presents a significant issue, namely determining how to choose or tune the parameters. This paper addresses the LLRE’s hipknee angular position tracking system based on an online PID-NN controller, i.e., a PID controller, whose parameters are online modified by a trained neural network (NN). A proposed framework for designing the PID-NN controller is elaborated. Numerical verifications are carried out by comparing the performance of the PID-based control system, whose parameters have been tuned using Ziegler-Nichols (ZN), without and using NN. Performance comparisons involving the presence of external disturbance are also carried out. The simulation results show that the proposed PID-NN-based control system provides better performance with lower mean squared error (MSE), root mean squared error (RMSE), and mean absolute error (MAE) values
Agrivoltaic systems: a literature review
Get PDFAgrivoltaic systems integrate solar energy generation with agricultural production to achieve efficient land use and mitigate climate change. This study presents a bibliometric analysis of the scientific literature on this topic, published from 2013 to 2023, to identify key trends, research areas, and emerging topics. Using the preferred reporting items for systematic reviews and meta-analyses (PRISMA) methodology and data from the Scopus database, the analysis was conducted with the R package bibliometrix and VOSviewer software. The results show remarkable growth in scientific output since 2020, with the United States, China, and Germany as the leading countries. The findings reveal the benefits of agrivoltaic systems, such as increased crop productivity, water-use efficiency, and income diversification for farmers. Emerging topics include the optimization of panel configurations and socioeconomic implications. Despite these benefits, challenges like high initial costs, social acceptance, and the need for adaptable designs persist. The conclusions underscore the importance of specific policies and incentives to support the adoption of these technologies. This analysis provides a comprehensive overview of the state of agrivoltaic systems, offering valuable insights for researchers, policymakers, and other stakeholders
Design of compact dual-resonance multiple-input-multiple-output antenna array for internet of things application
Get PDFThis work presents a compact dual-resonance multiple-input-multiple-output (MIMO) antenna array for internet of things (IoT)-enabled smart devices requiring both 5G and Wi-Fi connectivity. The antenna operates at 3.6 GHz (5G smartphones) and 5.4 GHz (high-speed Wi-Fi), using a dual-resonance MIMO configuration for reliable multi-device communication. An integrated electromagnetic band-gap (EBG) structure suppresses surface waves, reducing mutual coupling and achieving 35 dB isolation with an envelope correlation coefficient (ECC) of 0.05. A fabricated prototype validated the design, with measurements aligning closely with simulations. The proposed antenna offers compactness, dual-band operation, low coupling, and strong MIMO performance, making it well-suited for next-generation IoT systems
Generating data for predicting court decisions in Kazakhstan using machine learning
Get PDFThis study presents the development of a synthetic dataset and machine learning models for predicting court decisions in Kazakhstan. The dataset contains 100,000 cases generated from the Code of the Republic of Kazakhstan, covering both administrative and criminal offenses. Each record includes attributes such as the age of the accused, offense type and severity, and mitigating or aggravating factors. Regression models were applied to estimate offense severity, level of guilt, and likelihood of penalties, while classification models predicted the offense category, relevant law articles, and sentencing type. Predictions addressed both general outcomes—classifying cases as criminal or administrative—and specific judicial decisions, including fines, imprisonment terms, and other penalties. Classification models achieved 92% accuracy in determining offense category and sentencing type, and regression models reached a root mean squared error (RMSE) of 0.12 for offense severity. Using synthetic data preserves confidentiality while enabling pattern discovery for decision support. The results demonstrate the potential of artificial intelligence (AI) to improve sentencing prediction, prioritize case processing, and enhance transparency in Kazakhstan’s judicial system. Beyond transparency in decision support, the proposed approach also shows potential in crime prevention, workload optimization, and fostering digital transformation within judicial operations
From Interaction Flow Modeling Language to Symfony: an automated model transformation methodology
Get PDFWeb application development has become increasingly complex with the rise of modern frameworks and user-centric architectures. Ensuring efficient and reliable development processes requires adopting structured methodologies that bridge abstract models with platform specific implementations. This paper presents a methodology for automating the transformation of Interaction Flow Modeling Language (IFML) models into Symfony models using the Atlas Transformation Language (ATL). The proposed approach supports model-driven architecture (MDA) principles and bridges the gap between abstract user interaction models in platform independent model (PIM) level and platform specific model (PSM) level. By leveraging IFML’s user-centered modeling capabilities and Symfony’s model-view-controller (MVC) framework, our solution enables the automatic generation of reliable, structured front-end architectures. Two case studies demonstrate the feasibility of the approach. This work contributes to the automation of model driven web development by offering a scalable and reusable transformation process from IFML to Symfony