Indonesian Journal of Electrical Engineering and Computer Science
Not a member yet
9109 research outputs found
Sort by
Integrated energy-efficient and location-aware routing in wireless sensor networks
Get PDFSensor nodes in wireless sensor networks are commonly distributed randomly across a given landscape, and their placement may be randomized for specific applications, even extending to national deployments. The energy consumption associated with data transmission and reception by the cluster’s leader is notably higher compared to other nodes. To address this issue, it is recommended that wireless sensor networks adopt a more energy-efficient routing technique. This proposed technique assumes a spatial separation between different node types. Elevating the threshold enhances the likelihood that nodes with ample remaining power will endure as cluster leaders. Ultimately, a hybrid data transfer strategy is formulated, wherein data is directly exchanged between the base station and cluster heads among the super nodes containing advanced nodes. Most nodes employ a combination of single-hop and multi-hop approaches for data transport, aiming to minimize the power required for transmission between the cluster’s control node and the base station. According to simulation results, this proposed method surpasses the stable election protocol (SEP), demonstrating superiority over the improved threshold-sensitive stable election protocol in terms of the operational duration of a wireless sensor network
Automated Alzheimer’s disease detection and classification based on optimized deep learning models using MRI
Get PDFAlzheimer’s disease (AD) is a devastating neurologic condition characterized by brain atrophy and neuronal loss, posing a significant global health challenge. Early detection is paramount to impede its progression. This study aims to construct an optimized deep learning (DL) framework for early AD detection and classification using magnetic resonance images (MRI) scans. The classification task involves distinguishing between four AD stages: mild demented (MD), very mild demented (VmD), moderate demented (MoD), and non-demented (ND). To achieve effective classification, three DL models (VGG16, InceptionV3, and ResNet50) are implemented and fine-tuned. A systematic evaluation is conducted to optimize hyper-parameters, with extensive experimentation. The results demonstrate superior classification performance of the customized DL models compared to state-of-the-art methods. Specifically, visual geometry group 16 (VGG16) achieves the highest accuracy of 95.85%, followed by ResNet50 with 89.38%, while InceptionV3 yields the lowest accuracy of 87.23%. This study highlights the critical role of selecting appropriate DL models and customizing them for accurate AD detection and classification across various stages, offering significant insights for advancing clinical diagnosis and treatment strategies
Water quality monitoring with an early warning system for enhancing the shrimp aquaculture production
Get PDFAquaculture provided 43% of the aquatic animal food consumed by humans in 2007, and it is anticipated to expand even more to meet the increasing future demand. Marine Science Techno Park (MSTP) is one of the Techno Parks in Indonesia and is located in Teluk Awur, Jepara. MSTP has an intensive system of Vannamei shrimp farming activities. The challenge that has been faced annually is the organic matter of the remaining shrimp feed that accumulates in the waters can cause a decrease in pond water quality. Ammonia-N anions derived from the decomposition of feed residues can cause toxins in shrimp culture ponds which can further interfere with shrimp survival. The objective of this study is to design a self-controlled water quality monitoring system that is equipped with an early warning system based on internet of things (IoT) and to implement the design for analyzing the water quality including dissolved oxygen (DO), pH, and temperature in the Vannamei farming pond of MSTP UNDIP through integrating IoT which is equipped with an early warning system. If the water quality reaches a threshold value, the monitoring system will send a sound signal to the buzzer followed by sending an alert notification in real-time conditions automatically to a smartphone
Unveiling visionary frontiers: a survey of cutting-edge techniques in deep learning for retinal disease diagnosis
Get PDFRetinal disorders impact millions of people globally. These disorders can be detected and diagnosed early enough to not only cure but also avoid permanent blindness. Manual identification of these diseases has always been tedious, time-consuming, and inconsistent. For ophthalmologists, retinal fundus images are a valuable source of information in diagnosing retinal diseases. Automatic identification of eye disorders using artificial intelligence (AI) based learning models has seen substantial development in the computer vision sector recently. Various models, particularly deep learning (DL) models are incredible in identifying and classifying diseases. In the presented review, we have performed an in-depth analysis of various existing DL models, involving preprocessing, classification, segmentation, and techniques to deal with data imbalance. We have also endeavored to gauge the effectiveness of these models by evaluating their performance using the metrics employed in their assessment. In addition, we explored various challenges along with the potential future scope in this domain.
Review of battery models and experimental parameter identification for lithium-ion battery equivalent circuit models
Get PDFThe growing use of electric vehicles has led to an ever-increasing demand for efficient and reliable management systems to control the behavior of lithium-ion batteries, especially with respect to heat generation and state-of-charge. Understanding these patterns constitutes a major new challenge for these batteries, as remaining ignorant of their behavior can result in decreased performance, shorter service life and even safety dangers. This review provides an overview of the different modeling techniques applied to simulate battery behavior. Different methods using equivalent electrical circuit models are discussed, covering both simple battery models and more complex equivalent electrical circuit models, with a focus on the 2RC-Thévenin circuit model. In this context, parameter approach methods for these systems are reviewed. In addition, laboratory tests are run to identify the various model parameters for a lithium-ion battery. This comprehensive study is designed to guide scientists and engineers in the selection and use of suitable tools for state-of-charge and battery health studies
Pneumonia stage analyzes through image processing
Get PDFA physical examination and diagnostic imaging techniques including lung biopsies, ultrasounds, and chest X-rays are typically used to make the diagnosis of pneumonia infection, an infectious disease that has the potential to be life-threatening. The objective of this research is to categorize the stages of pneumonia through image processing methods. Before that, an ensemble model for diagnosing pneumonia infections is created utilizing the transfer learning algorithms ResNet50V2 and DenseNet201. The 5,857 images were taken from the PAUL MOONEY dataset for this research. The proposed ensemble averaging model recognizes lung infection appropriately and accurately. By applying a contour detection approach, the left and right chests are separated and the affected pixels from there to analyze the stage of pneumonia. It is very crucial to identify the stage for treatment purposes
Artificial intelligence-based Karawo motif formation using genetic algorithm
Get PDFThis research explores the application of artificial intelligence in generating Karawo motifs, a traditional Indonesian pattern. The research involves collecting a dataset of existing Karawo motifs and utilizing genetic algorithms to evolve and create novel pattern variations. The generated motifs are evaluated based on their adherence to traditional design principles and aesthetic appeal. The formation of Karawo motifs begins with randomly selecting image data from a database. Then, the selection of transformation treatments is performed by optimizing the fitness function within the genetic algorithm. The applied types of transformations include geometric transformations, Boolean transformations, and arithmetic transformations. The outlined genetic algorithm steps include determining the fitness function, performing its evaluation, selecting fitness values, applying crossover, implementing mutation, managing survivor selection, and terminating iterations. The results indicate that the developed system is capable of creating diverse and appealing Karawo motif patterns, showcasing the potential of combining traditional artistry with artificial intelligence. This study has the potential to expand the possibilities of Karawo motif design and contribute to the preservation and promotion of Indonesian cultural heritage
Mutual information-MOORA based feature weighting on naive bayes classifier for stunting data
Get PDFOne effort to reduce stunting rates is to predict stunting status early in toddlers. This study applies Naive Bayes (NB) to build a stunting prediction model because it is simple and easy to use. This study proposes a filter-based feature weighting technique to overcome the NB assumption, which states that each feature has the same contribution to the target. The frequency of an event in a dataset influences the feature weighting using mutual information criteria. This is the gap in the filter-based ranking highlighted in this study. Therefore, this study proposes a feature-weighting method that combines mutual information with the MOORA (MI-MOORA) decision-making method. This technique makes it possible to include external factors as criteria for ranking important features. For stunting cases, the external consideration for ranking purposes is the assessment of nutrition experts based on their experience in dealing with stunted toddlers. The MI-MOORA technique makes the availability of clean water the most influential feature that contributes to the stunting status. In the ten best features, the MI-MOORA ranking results are dominated by family factors. Based on the performance evaluation results of NB and other classifiers, MI-MOORA can improve the performance of stunt prediction models
Real-time forest fire detection, monitoring, and alert system using Arduino
Get PDFEarly fire detection is critical to protecting forests from wildfires and enabling rapid responses to minimize fire spread. Existing forest fire detection methods cannot quickly detect forest fires and evaluate the fire risk of these sensitive areas. Hence, this research aims to develop a real-time forest fire detection, monitoring, and alert system. The development of the system started with assembling temperature and humidity sensors, a smoke sensor, an Arduino microcontroller, and a wireless fidelity module. Then, a fire monitoring and alert system was developed using Blynk. From the sensitivity flame sensor analysis with the fire, the flame sensor detected the presence of fire up to 60 cm. The sensor also indicated high temperature (45 °C) and low humidity (53.4%) at noon. Low temperature (29 ℃) and high humidity (88.4%) were identified in the morning. Moreover, the highest carbon dioxide (CO2) concentration of 1,800 ppm was recorded when the smoke from the fire was detected. The global positioning system module shows the accurate real-time location of the system displayed in the Blynk application. In conclusion, this system can detect and monitor early forest fires in real-time and can alert the authorities to protect forests from wildfires
Effects of Pr3+ -activated BaZrGe3O9@TiO2 phosphor compound on light emitting diodes validated by computer simulation
Get PDFThe Pr3+ -doped BaZrGe39 gallogermanate phosphors are reported to have a well-defined successive deep defect structure that effectively mitigates thermal carrier fading. This phosphor also presents a red emission with a peak at 615 nm, originating from the Pr3+ transtition from 1D2 to 3H4. We investigated the impact of Pr3+ -activated BaZrGe3O9 (referred to as BZG:Pr) on the lighting characteristics of light emitting diodes (LED) packages in this paper. By combining BZG:Pr with TiO2 particles and silicone, we produced a phosphor layer (designated as BZG:Pr@TiO2). The optical performance of the resulting LED was systematically examined by varying the TiO2 doping percentage. Our findings reveal that the incorporation of the BZG:Pr phosphor enhances the red spectral component, thereby contributing to improved homogeneity in color distribution. However, a progressive increase in TiO2 content within the phosphor layer corresponds to diminishing luminous output and decreased chromatic rendering efficiency of the LED. Employing a lower concentration of TiO2 proves advantageous, as it capitalizes on the scattering-enhancing attributes while leveraging the red emission of the BZG:Pr phosphor. This synergistic approach yields a favorable balance between luminosity and color quality, enhancing the LED’s overall performance