Indonesian Journal of Electrical Engineering and Informatics (IJEEI)
Not a member yet
776 research outputs found
Sort by
Clustering the Addiction Levels of Drug Users Using Fuzzy C-Mean
Full text linkRecently, the number of drugs abused, such as Narcotics, Psychotropics, and Addictive Substances have been linear increased with the drug users. The increasing number of these cases triggers the difficulties for rehabilitation associations in diagnosticating the abuse level for medical and health prevention. Herein, data mining with a Fuzzy C-Mean clustering approach is employed to delve 506 drug users’ addiction into three clusters by considering several indicators including age, urine test, duration of use, physical effects, and psychological effects. As a result, 215 data are recorded in clusters 1 as high optimum addiction, 105, and 186 data in clusters 2, and 3 as medium and regular addiction levels, respectively. The Silhouette Coefficient, Calinski-Harabasz Index, and Davies-Bouldin Index algorithms reveal high potential values to indicate the proper achievement of this clustering structural test. A clustering software has been successfully developed and tested to aid the calculation and analysis. Hence, the rehabilitation associations in Riau province as end-user of this case are aided in identifying the addiction level of drug users in order to ensure the proper therapeutic prevention and curative action
Dual Band Circular Polarized Design of Rectangular Microstrip Antenna For GPS L-band and Galileo E-band Applications
Full text linkDesign of rectangular microstrip antenna employing three rectangular slots of unequal lengths on one of the patch edges, is presented to achieve dual band and dual sense circular polarized characteristics. Circular polarized response in the two bands is attributed to the optimum inter-spacing in between the rectangular patch’s TM10, TM11 and TM12 resonant modes. For axial ratio less than 3 dB, an optimum design offers axial ratio bandwidth of 26 MHz (2.05%) and 73 MHz (4.59%) in the dual bands, bearing frequency ratio of 1.25. This circular polarized bandwidth lies inside the VSWR < 2 bandwidth of 665 MHz (49.83%). Antenna offers radiation pattern maximum in the broadside direction across axial ratio and VSWR bandwidth, with a gain of more than 6 dBic. For the obtained antenna characteristics, the three rectangular slot cut design is suitable in variety of applications like, GPS L1 & L2 bands, and Galileo E1 & E6 bands. The experimental verification has been carried out for the proposed configuration that shows close agreement against the simulated results
Conversational Assistant with Large Language Model Agent for Natural Interaction in Home Automation Environments
Full text linkDigital transformation has driven the development of smart home automation systems for healthcare and remote assistance, which can integrate large language models, foundational models, and Retrieval-Augmented Generation techniques to design conversational assistants accessible to people with reduced mobility. In these cases, it is necessary to overcome usability barriers to control home devices because traditional home automation interfaces are not adapted to their specific needs. Previous research has identified systems such as voice control, gestures, and conversational agents based on generative artificial intelligence to facilitate assisted interaction in the home through home automation systems. This article describes a conversational agent, aimed at people with reduced mobility, to facilitate natural interaction with home automation environments, using technologies such as OpenAI, Langchain, and LlamaIndex. The results demonstrate the successful deployment of a web-based telecare platform based on conversational agents capable of retrieving technical documents through embedding, generating SQL queries, and MQTT topics for integration into real-world monitoring and assistance environments for people with reduced mobility
Enhanced Dung Beetle Optimization Algorithm Combined Harris Hawks Optimizer and Nelder Mead Simplex Method for Solar-Connected Smart-grid Application
Full text linkThe Dung Beetle Optimization algorithm (DBO) is a swarm-based intelligence algorithm with competitive performance against other popular optimization algorithms. However, its process is often fallen in local optimum, insufficient accuracy, and slow convergence speed due to a lack of combination or collaboration between search agents. This research proposes an advanced DBO approach combining the Harris Hawks Optimizer (HHO) and Nelder Mead method to improve slow convergence speed, insufficient accuracy, and premature convergence. The Nelder Mead method is used in the subpopulation of ball-rolling to reduce the probability of falling in local optimum, along with “seven kills” strategy of HHO method that is combined in the former iterations of the DBO algorithm to enhance its global search capacity and convergence speed. The performance of the proposed enhanced dung beetle optimization (EDBO) algorithm is evaluated via 30 CEC-2017 benchmark functions and compared with several representative meta-heuristic algorithms, including the original DBO and HHO, as well as three recently proposed methods: RUN, SMA, and AO. The result shows that EDBO consistently achieves superior performance over most of the C-test functions in terms of solution quality and robustness. Additionally, when applied to the optimization of the operating cost of a solar-connected residential power system, the proposed EDBO attains the best or highly competitive global optimum compared with the competing algorithms
Closed-Form Solution for Energy Efficiency Maximization in Uplink IRS-Assisted Multi-User NOMA Network
Full text linkGiven the growing concerns about energy consumption and its negative impact on the ecosystem, energy efficiency (EE) has become one of the most important key performance indicators in current and future wireless communication tech nologies. In this paper, we address the EE maximization problem in an uplink intelligent reflective surface (IRS)-assisted multi-user non-orthogonal multiple access (NOMA) network. This problem is formulated as a trade-off between the spectral efficiency (SE) and total power consumption, and it appears to be non convex. To avoid the complexity associated with the traditional iteration-based Dinkelbach method, we opt for an alternative closed-form solution for the users’ transmit power based on partial derivative analysis and Lambert function. Simulation results with a realistic power consumption models confirm the accuracy of our theoretical findings
HybridPPI: A Hybrid Machine Learning Framework for Protein-Protein Interaction Prediction
Full text linkProtein-protein interactions (PPIs) are key to cellular functions and disease mechanisms and are crucial for drug discovery and systems biology. Though experimental approaches, including yeast two-hybrid systems, provide informative discoveries, they are time-consuming, costly, and frequently yield significant false-positive rates. Newer computational tools, including DeepPPI and PIPR, have demonstrated their potential, but their reliance on single-modal features or specific machine-learning models limits their generalization and robustness. These limitations highlight the need for an enhanced framework that assimilates different types of features while integrating a diverse array of machine learning models to exploit the strengths offered by each model class. In this paper, we present a hybrid machine learning framework, HybridPPI, to effectively incorporate the power of sequence-based, structure-based, and network-based features based on wellknown ensemble learning techniques to predict PPIs. Our proposed algorithm is a stacking ensemble of multiple models (Support Vector Machines (SVM), Random Forest (RF), Convolutional Neural Networks (CNN), and Long Short-Term Memory Networks (LSTM)), with Gradient Boosting as the metamodel. Results show that HybridPPI (94.5% accuracy, 95.2% precision, and Area Under Curve of 0.97) outperforms the most advanced methods, indicating its robustness for PPI prediction. This scalable and generalizable framework can accommodate various biological applications. HybridPPI overcomes significant shortcomings of current methodologies and contributes to biological discovery.
Enhancing GRU-Based DRL with Delta-LiDAR for Robust UAV Navigation in Partially Observable Dynamic Environments
Full text linkPartial observability and sensor limitations are challenging for the navigation of autonomous Unmanned Aerial Vehicles (UAVs). Deep Reinforcement Learning (DRL) algorithms have emerged as potential tools in advancing this field. However, their effectiveness degrades in challenging environments, particularly in the presence of dynamic obstacles. Recent research trends emphasize the need for new DRL variants that guarantee robustness, real-time adaptability, and improved generalization under uncertainty. This paper proposes a lightweight DRL architecture that combines Proximal Policy Optimization (PPO) with a Gated Recurrent Unit (GRU), extended with a temporal LiDAR differencing feature called Delta-LiDAR. The difference between consecutive LiDAR scans is computed to provide the velocity and directional cues without the computational burden of Long Short-Term Memory (LSTM) networks. We evaluate three models, PPO-LSTM, PPO-GRU, and Delta-LiDAR augmented PPO-GRU in a 3D simulated UAV navigation environment characterized by noise, clutter, and dynamic obstacles. We considered several metrics, including success rate, collision frequency, trajectory smoothness, and computational efficiency, to determine the effectiveness of each architecture. The experimental results demonstrate that Delta-LiDAR improves GRU-based temporal reasoning. The deployment complexity is reduced compared with the LSTM-based architecture, which makes it ideal for real-time UAV operation in partially observable environments
A Privacy-Enhanced Scheme Within The Public Key Infrastructure For The Internet Of Things, Employing Elliptic Curve Diffie-Hellman (ECDH)
Full text linkThe Public Key Infrastructure (PKI) serves as the foundation for online security, particularly within the realm of the Internet of Things (IoT). It operates based on certified public keys that remain permanent but can be revoked when necessary, such as in the case of a change in ownership, compromise of the private key, or malicious activities. Although this method ensures secure key utilization with traceability, it also introduces a potential privacy risk due to the traceability and utilization of identity-based certificates. This approach is considered an innovative strategy for ensuring user confidentiality, integrity, authentication, and privacy in the context of the Internet of Things. The proposed solution integrates elliptic curves (ECDH) and traditional PKI to safeguard user privacy. It introduces two types of elliptic curve keys: long-term identity-based certified keys and dynamically generated temporary anonymous aliases. These aliases are seamlessly recorded by the certification authority, which maintains distinct directories for long-term and temporary keys. This dual-key approach enhances security while addressing the specific requirements of the Internet of Things
Plant-Disease Relation Model through BERT-BiLSTM-CRF Approach
Full text linkPlant Disease Relations (PDR) is one of the Information Extraction (IE) subtasks that reveals the relationship between plant entities and diseases that appear together in a sentence. Previous studies have proposed methods for detecting the extraction of relationships between plant diseases (PDR). Previous research has proposed a Short Dependency Path-Convolutional Neural Network (SDP-CNN) method to predict relationships. However, the proposed method has limitations when faced with long and complex sentences. To overcome these limitations, this study proposes the BERT-BiLSTM-CRF method to improve the model performance in detecting PDR. First, the data is processed into the BERT Encoder layer after the tokenization process. After the BERT Encoder calculates the hidden information, the next step is to enter the linear layer to obtain word embedding. Calculation results in the bilinear layer are forwarded to the softmax layer to predict the relationship of each pair. Computation results in the softmax layer are sent to the BiLSTM layer. Finally, the CRF layer is entered to improve the prediction process. An 80:20 ratio for training and testing data was used to build the model using the same parameter values over ten attempts. GridSearch hyperparameter tuning is also involved in improving model performance. Experimental results show that the architecture proposed in this research can increase the F1 score by 0.790, which proved to be higher than SDP-CNN with a micro F1 score of 0.764. The problem of predicting PDR was overcome by the BERT-BILSTM-CRF method. The issue of forecasting PDR was resolved using the BERT-BILSTM-CRF approach
Synthesis of Bandpass Filter as a Four-Pole Based on a Non-Homogeneous Line
Full text linkThe article deals with the synthesis of band-pass filters (BPF) for the design of microwave filtering devices, by using non-homogeneous lines (NL) with the selection of the appropriate wave impedance W. For this purpose, equivalent NL substitution circuits were created in the region of resonant and antiresonant frequencies, and four-pole matrices of the transmission line were determined, whose matrix of impedances and admittances does not have partial poles, and the transmission admittance and transmission impedance do not have zeros. BPF prototypes were synthesized with two parallel plumes based on a closed homogeneous line and one plume based on three NLs. A band-pass filter with an extended blocking band was implemented, and its amplitude-frequency characteristics were obtained. The use of NLs as resonators allows the choice of wave impedance to increase the blocking band of the BPF compared to the BPF on resonators based on homogeneous lines