Indonesian Journal of Electrical Engineering and Computer Science
Not a member yet
9109 research outputs found
Sort by
Mutual coupling reduction between antennas array for 5G mobile applications
Get PDFThis paper introduces the design of a multiple-input-multiple-output (MIMO) antenna optimized for low-profile applications supporting sub-6 GHz fifth-generation (5G) wireless applications. We have started the design from a single antenna with a square patch shape, each antenna array is composed from 4-element radiators fed by using power dividers and quarter microstrip lines. Mounted on a single Rogers RT5880 substrate, the MIMO antenna functions at 3.5 GHz. In order to miniature and to decrease the mutual coupling between the both antennas array we have optimised a magnetic wall based on periodic structures permitting to decrease the mutual coupling between the both antenna array. The unit element from the wall was optimised, studied and validated in order to absorb the surface current and to enhance the isolation between the different radiating elements. The dimensions of the proposed MIMO antenna are 154×220×0.578 mm³. The MIMO antenna final circuit achieves a peak gain of 9 dBi and an isolation around -30 dB. The introduction of the magnetic wall permits to enhance the isolation between the antenna array from -20 dB to -30 dB at 3.5 GHz band. This advancement contributes to the overall performance improvement of the MIMO antenna system
An interactive visualization tool for the exploration and analysis of multivariate ocean data
Get PDFOcean data exhibits great heterogeneity from variances in measuring methods, formats, and quality, making it extremely complicated and diverse due to a variety of data kinds, sources, and study elements. A few examples of data sources are satellites, buoys, ships, self-driving cars, and distant systems. The processing of data is made more challenging by the significant regional and temporal variations in oceanic characteristics including temperature, salinity, and currents. This work presents an interactive tool for multivariate ocean parameter visualisation, specifically overlays, based on Python. In ocean data visualisation, overlays are extra visual layers or data points that are layered to improve comprehension over a basic map. Based on the available data and the visualisation goals, these overlays are chosen and blended. Users can customise overlays with this tool, which also supports formatting, 2D and 3D visualisation, and data preparation. In order to reduce artefacts, it uses kriging interpolation for 3D visualisation and a modified version of the ray casting algorithm for representing octree data. By integrating overlays like as bathymetry, currents, temperature, and marine life, users can produce visually appealing and comprehensive depictions of ocean data. This method provides a thorough grasp of intricate marine processes by making it easier to see patterns, trends, and abnormalities in the data
Development of a payload for monitoring biological samples in microgravity and hypergravity conditions
Get PDFThis research aims to address the need for monitoring the behavior of organic and inorganic materials in hypergravity conditions. To fulfill this objective, a container with specific features was designed. The container has a box with a lid, measuring 10×10×10 cm, conforming to the 1U volume of the CubeSat standard. It includes four cylindrical spaces to accommodate the sample wells. The container was 3D printed using polylactic acid (PLA) wire. For the electronic components, four ESP32-CAM modules were utilized, with two programmed to capture and upload photos to the cloud, and the other two programmed to capture and store photos on a micro SD memory card. Additionally, four light emitting diodes (LEDs) were incorporated to illuminate the well spaces. The total weight of the container is 450 grams, and it has a maximum wireless upload distance of 10 meters to the cloud. The storage capacity of the SD memory card determines the number of images that can be saved
Non-contact power system fault diagnosis: a machine learning approach with electromagnetic current sensing
Get PDFModern power system protection schemes incorporate artificial intelligence (AI) techniques. However, in a conventional way, most of these schemes rely on the data of current and voltage collected from current transformer (CT) and potential transformer (PT) respectively. CTs suffer from the drawback of core saturation and impact the accuracy and effectiveness of intelligent methods. Also, it has the constraints of size, safety, and economy. The research here explores the effectiveness of magnetic sensors in advanced power system protection schemes as an alternative to traditional current sensing. In the presented work, a novel dataset is prepared by transforming transmission line currents into magnetic field components. Several supervised as well as unsupervised machine learning algorithms have been applied to this data instead of traditional currents and voltage for fault prediction. The paper discusses the comparative evaluation of these algorithms based on various performance metrices which reveals that Gaussian Naïve Bayes (GNB), K-nearest neighbor (KNN), random forest (RF), and extreme gradient boost (XGB) algorithms excel in fault detection, while multilayer perceptron (MLP) and KNN performs better fault classification. The findings promise the potential for developing compact, safe, and cost-effective protection schemes utilizing magnetic field sensors
An Intrusion Detection System against RPL-based Routing Attacks for IoT Networks
Get PDFThe significant improvement in the Internet, Internet of Things (IoT), communication, and cloud computing have created considerable challenges in providing security for data and devices. In IoT networks, “Routing Protocol for Low power and Lossy networks”- (RPL) is a communication protocol that enables devices to exchange information and communicate with limited resources like low processing capabilities, less memory and energy. Through the Internet, unauthorised users can access RPL-based IoT networks, making these networks susceptible to routing attacks. Therefore, it is crucial to design Intrusion Detection System-(IDS) to address attacks from IoT communication devices. In this paper, we have proposed GCNConv, a Graph Neural Network (GNN) method that allows capturing the edge and node features of a graph to identify routing attacks. The proposed system has experimented on the RADAR dataset and experimental findings proved that, our approach performs well compared to state-of-the-art method with reference to precision, F1-score, accuracy and recall
The synergistic effect of QR decomposition with t-SNE
Get PDFThe study utilized non-parametric tests, specifically, the Mann-Whitney U test, to evaluate the performance of a proposed model called QRPCA-t-SNE, along with two other models, MDS and UMAP. The study compared these three models with two datasets on performance metrics such as model accuracy, training accuracy, testing accuracy, mean square error, AUC scores, precision, recall, and F1 scores. Once the model's performance was conducted, the Anderson-Darling test was to check for data normality before applying the hypothesis for model proof. The analysis revealed that Model 1 (QRPCA-t-SNE) significantly outperformed Model 2 (UMAP) and Model 3 (MDS) in terms of accuracy, with p-values of 0.0027 and 0.0003, respectively. This finding suggests that Model 1 (QRPCA-t-SNE) is suitable for high-accuracy and reliability applications, providing valuable insights into predictive analytics with a 95% confidence interval (confidence level α= 0.05)
Enhanced diabetic retinopathy detection and classification using fundus images with ResNet50 and CLAHE-GAN
Get PDFDiabetic retinopathy (DR), a progressive eye disorder, can lead to irreversible vision impairment ranging from no DR to severe DR, necessitating precise identification for early treatment. This study introduces an innovative deep learning (DL) approach, surpassing traditional methods in detecting DR stages. It evaluated two scenarios for training DL models on balanced datasets. The first employed image enhancement via contrast limited adaptive histogram equalization (CLAHE) and a generative adversarial network (GAN), while the second did not involve any image enhancement. Tested on the Asia pacific tele-ophthalmology society 2019 blindness detection (APTOS-2019 BD) dataset, the enhanced model (scenario 1) reached 98% accuracy and a 99% Cohen kappa score (CKS), with the non-enhanced model (scenario 2) achieving 95.4% accuracy and a 90.5% CKS. The combination of CLAHE and GAN, termed CLANG, significantly boosted the model's performance and generalizability. This advancement is pivotal for early DR detection and intervention, offering a new pathway to prevent irreversible vision loss in diabetic patients
A randomized blockchain consensus algorithm for enhancing security in health insurance
Get PDFHealth insurance fraud is a significant problem affecting insurance providers and policyholders. To address the rising problem of fraudulent activities in the health insurance sector, this paper proposes a pioneering blockchain-based system aimed at increasing transparency and security. Utilizing a hybrid Blockchain architecture, the system incorporates a consensus algorithm influenced by practical byzantine fault tolerance (PBFT) and proof of activity (PoA) to ensure reliability and efficiency in distributing mining power. Developed using Python, extensive testing confirms the system's performance and security metrics. Results show that a block size containing one transaction is 1.63 KB, with 1.2 KB for data and 0.43 KB for identification and hashing. Operational tests demonstrate that a single participant can upload 850 transactions to the transaction pool, with validation completed in just 7.49 seconds. Block appending time for these transactions is a swift 10 seconds. Notably, the system exhibits resilience against data tampering, detecting unauthorized changes within 881.3 milliseconds across 10,000 blocks and identifying irregularities in the transaction pool within 8.78 seconds. Additionally, to enhance data privacy, patient information is accessible only through a unique QR code, providing an extra layer of security; this research represents a significant advancement in combatting fraud and safeguarding data privacy
System availability assessment and optimization of a series-parallel system using a genetic algorithm
Get PDFTo optimize the operational availability of the series-parallel system and provide useful insights for maintenance planning, the study attempts to investigate the availability of a ball mill unit. These four different components make up the ball mill production system: “drum,” “ring-gear,” “gearbox,” and “electric motor.” There is a chain mechanism connecting all four components. The “ring gear” and “electric motor” components are composed of two independent units, one of which serves the desired purpose and the other is maintained in cold standby. The “drum” and “gearbox” of the components each contain only one unit. Therefore, a novel mathematical model is designed and implemented in this work by assuming arbitrary repair rates and exponentially distributed failure rates using the Markov process and Chapman-Kolmogorov equations. This study explored the availability with a normalization method and used genetic algorithm techniques to optimize ball mill availability. Putting this article into practice is of great benefit when developing an appropriate maintenance program. Through this, the study achieves maximum production. To investigate the behavior of several performance characteristics of the ball mill production system, numerical results and corresponding graphs are also specifically created for specific values of subsystem parameters, such as failure rate, and repair rate to increase the system’s overall efficiency
Multimodal approach for early prediction of COVID-19 disease using convolutional neural network
Get PDFThe latest human coronavirus is COVID-19. Chest radiography imaging is essential for screening, early detection, and monitoring COVID-19 infections since the virus resides in the lungs. Classical real time reverse transcriptase polymerase chain reaction (RT-PCR) data and chest X-ray pictures will become more important for COVID-19 identification as the pandemic spreads due to their affordability, wide availability, and infection control benefits, which reduce cross-contamination. This work presents multi-modal hybrid automated approaches to classify COVID-19 illness into three clinical categories: normal, pathogenic, and COVID-19 utilising RT-PCR test data and online chest X-ray datasets. The RT-PCR and chest X-ray image datasets were processed using supervised machine learning and convolutional neural networks (CNN). Together, these measures help us separate COVID-19 patients, those with similar symptoms, and healthy persons. The author improved detection times and classification accuracy with extra tree classifier’s feature selection and openCV’s image sharpening. The proposed approaches were tested using a research dataset. The proposed methods allowed reliable COVID-19 disease categorization for clinical decision-making, with random forest (RF) classifier global precision values of 91.58% on the RT-PCR dataset and CNN model accuracy of 95.46% on improved sharpened images