Kinetik: Game Technology, Information System, Computer Network, Computing, Electronics, and Control

Enhancing Plant Recommendation through IoT-integrated LLM Systems

2025-03-03T05:33:10+00:00

Over the past decade, artificial intelligence has experienced phenomenally rapid and extensive expansion across a wide range of industries. Alongside these developments, the agricultural sector stands to benefit significantly from the integration of technology. A significant challenge encountered by farmers is selecting the appropriate crop to plant. The selection of crops is influenced by various factors. Despite advancements in agricultural technology, a considerable gap remains in the integration of IoT with large language models (LLM) for delivering context-specific and data-driven plant recommendation. This study evaluates the reliability of plant recommendations produced by Internet of Things (IoT) devices utilizing the Llama 3.2 model. The model leverages real-time environmental data, including soil pH, altitude, and temperature, to recommend appropriate plant. The recommendations from the base model and a fine-tuned model were compared using precision, recall and F1-score metrics, and were further assessed against established agricultural literature on plant compatibility and growth requirements through human evaluation. The results show substantial performance improvements. The proposed approach achieved an AUC value 59% higher than that of the base model. Precision increased by 40%, recall improved by 105%, and the F1 score rose by 80% compared to the base model.

Design of 2x1 Single-band Microstrip Antenna Array with Proximity Coupling for Enhanced CCTV Performance

2025-05-05T05:22:09+00:00

The increasing demand for reliable wireless communication in modern surveillance systems, particularly Closed-Circuit Television (CCTV), requires the development of antennas with high efficiency, wide bandwidth, and stable signal performance. To meet these requirements, this study presents the design and analysis of a 2×1 microstrip antenna array with rectangular patches that use proximity coupling, optimized for operation in the 2.4 GHz ISM band. The antenna was designed and simulated using CST Studio Suite to evaluate its electromagnetic characteristics, while measurements using a Vector Network Analyzer (VNA) were performed to validate the performance of the manufactured prototype. Simulation results show that the antenna achieves a reflection loss of −24.62 dB, a voltage standing wave ratio (VSWR) of 1.12, and a frequency bandwidth of 159 MHz, indicating good impedance matching and wide operational capability. Meanwhile, measurement results showed a reflection loss of −12.59 dB, a VSWR of 1.15, and a frequency bandwidth of 86 MHz. Both simulation and measurement results showed directional radiation patterns, ensuring efficient energy radiation and better signal focus for monitoring coverage. The designed antenna also shows a measured gain of 9.25 dBi, exceeding the simulated gain of 6.99 dBi, confirming improved performance. The difference between simulation and measurement is mainly due to variations in substrate thickness, material tolerance, and environmental factors during testing. Overall, the proximal coupling approach has proven effective in improving coupling efficiency without adding design complexity. This antenna is well-suited for reliable and efficient data transmission in CCTV applications. Furthermore, the findings contribute significantly to advancements in antenna technology, particularly in the domains of wireless communication, IoT, and smart city-based surveillance systems.

Design of a Real-Time User Feedback for Mitigating Spurious SpO₂ Readings in Pulse Oximetry for Outpatient Monitoring

2025-08-20T01:59:19+00:00

Spurious SpO₂ readings—arising from motion artifacts, environmental interference, or device variability—remain a major limitation in wearable pulse oximetry, potentially triggering false alarms or missing hypoxemia during outpatient monitoring. Conventional devices often lack real-time mechanisms to detect and mitigate such errors, with previous reports indicating measurement biases of 11.2 - 24.5% across different models, underscoring the need for improved accuracy and user guidance. To address this gap, we present the design of an IoT-enabled wearable pulse oximeter with real-time user feedback, delivered through a mobile application. The system integrates a pulse oximetry and heart rate sensor (MAX30100) with a carbon monoxide gas sensor (MQ-7) and provides targeted notifications to guide corrective actions such as repositioning the probe, removing nail polish, or moving to fresh air. Validation involved controlled scenario testing (undetected SpO₂, CO >40 ppm, nail polish, and loose contact) and user trials with 15 healthy volunteers from varied academic backgrounds. The prototype demonstrated high accuracy, with low relative errors—0.92% (HR), 0.93% (SpO₂), and 0.015% (CO)—and strong usability, achieving 93.3% compliance with corrective prompts, an average response time of 4.0±0.7 seconds, and a satisfaction score of 4.3/5. Compared with commercial oximeters, the proposed system improved reliability by reducing measurement errors by at least 87% through real-time corrective feedback. Future work will focus on energy-efficient power management and large-scale community-based trials to further validate performance across diverse patient populations.

Adaptive EKF-Based Ship Trajectory Estimation with Earth Curvature Modeling and Dynamic Noise Tuning

2025-08-01T05:37:02+00:00

Accurate position estimation is critical for the effectiveness of automated weapon and navigation systems. Standard Extended Kalman Filter (EKF) models typically adopt flat-Earth assumptions and static noise covariances, which limit their accuracy in operational environments. This study proposes an optimized EKF framework that integrates two complementary approaches. First, ship trajectories are represented in Earth-Centered Earth-Fixed (ECEF) coordinates with a WGS-84 reference to account for Earth’s curvature. Second, process (Q) and measurement (R) covariances are adaptively determined using Joint Likelihood Maximization (JLM) with logarithmic scale exploration, enabling the filter to automatically identify the most accurate configuration. Each Q/R setting is evaluated within the EKF framework using root mean square error (RMSE) derived from radar data logs. The method was tested under short-history scenarios (5 and 10 data points) within an operational range of ±15 km, reflecting conditions commonly encountered in Combat Management Systems (CMS). The results show that while coordinate transformation alone provides only marginal improvements at short ranges, the combination of curvature modelling and adaptive Q/R tuning significantly reduces RMSE, achieving average errors approaching zero with high repeatability as measured by standard deviation. This research demonstrates a novel integration of geometric and statistical optimization in EKF design and highlights its applicability to ship trajectory estimation and defence systems.

Hybridization of PSO-SSA for Photovoltaic System MPPT Under Dynamic Irradiance and Temperature

2025-08-01T07:06:09+00:00

Maximum Power Point Tracking (MPPT) has become an important area of research to optimize the power generated by photovoltaic (PV) systems, particularly under various configurations such as series and parallel. Conventional methods, including Perturb and Observe (P&O) and Incremental Conductance (InC), often fail under dynamic or partial shading conditions, while metaheuristic algorithms such as Particle Swarm Optimization (PSO) and Salp Swarm Algorithm (SSA) provide global optimization but still suffer from slow convergence and power oscillations. This study proposes a hybrid MPPT approach by combining PSO and SSA to overcome these limitations. The algorithm was implemented in MATLAB/Simulink and tested under 96 scenarios covering series and parallel configurations with irradiance and temperature variations that change both suddenly (< 1 s) and gradually (> 1 s). Simulation results demonstrate that the hybrid PSO–SSA consistently achieves faster convergence compared to standalone PSO or SSA, with an average convergence time of 0.286 s in the series configuration (25–36% faster) and 0.282–0.284 s in parallel configuration, while achieving comparable power output to PSO. Overall, the proposed hybrid PSO–SSA algorithm provides a faster, more adaptive, and robust MPPT strategy under realistic PV operating conditions, contributing to reducing energy losses in fluctuating environments.

YOLOv9-Assisted Vision System for Health Assessment in Poultry Using Deep Neural Networks

2025-08-04T03:49:19+00:00

Poultry farming represents one of the fastest-growing sectors in global food production, yet disease outbreaks, high mortality, and labor shortages continue to threaten its sustainability. Conventional health monitoring methods based on visual inspection are time-consuming, subjective, and inadequate for early anomaly detection. In response, computer vision and deep learning have emerged as transformative tools for livestock management. While prior implementations of the YOLO object detection family, such as YOLOv5 and YOLOv8, have achieved notable success, their performance often deteriorates in dense flocks, low-light conditions, and occlusion-prone environments. This study introduces a YOLOv9-assisted vision framework tailored for poultry health assessment in commercial farm settings. The system integrates smart cameras with edge computing to enable real-time detection of behavioral and physiological anomalies without dependence on high-bandwidth or cloud-based resources. A dataset of 903 annotated poultry images, categorized into healthy and sick classes, was employed for model development. The trained model achieved 88.7% precision, 97% recall, an F1-score of 0.82, and a mAP@0.5 of 0.88, demonstrating robustness under variable illumination, bird occlusion, and high-density environments. Comparative evaluation confirmed that YOLOv9 provides a superior balance of accuracy, generalization, and computational efficiency relative to YOLOv8–YOLOv11, supporting practical deployment on edge devices. Limitations include the binary scope of health classification and reliance on a single dataset. Future directions involve extending the framework to multi-class disease recognition, cross-dataset validation, behavior-based temporal modeling, and multimodal fusion, thereby advancing predictive analytics and welfare-oriented poultry farming.

From Digital Literacy to Public Trust: The Strategic Role of E-Government Service Quality

2025-07-30T05:59:51+00:00

The transformation of public services in the digital era necessitates a synergistic alignment between e-governance practices and the digital competencies of the community to ensure services that are both high in quality and satisfactory to users. This study investigates the effect of e-governance and digital literacy on public satisfaction, with digital service quality serving as a mediating variable. The research focuses on the utilization of the S-Kepuharjo village digital service platform. Employing a quantitative approach, data were collected through a survey of 385 respondents and analyzed using Structural Equation Modeling (SEM) with AMOS software. The findings reveal that e-governance has a significant impact on satisfaction, both directly and indirectly via service quality. On the other hand, digital literacy does not directly influence satisfaction but exerts a significant indirect effect when mediated by digital service quality. The study confirms that service quality acts as a critical intermediary linking governance to user satisfaction. These results highlight that the success of village-level digital transformation is largely determined by the responsiveness and effectiveness of digital services. Accordingly, enhancing the inclusiveness, accessibility, and user-oriented nature of these services is essential for fostering public satisfaction and engagement in the digital landscape.

Website Quality Evaluation of OKE Garden using WebQual, The Marketing Mix, and Importance-Performance Analysis

2025-08-01T03:53:50+00:00

The rapid advancement of digital technologies has significantly impacted various service sectors, including the garden landscaping industry. In response to this development, OKE Garden has implemented a website-based e-commerce platform aimed at improving service accessibility and operational efficiency. This study seeks to evaluate the usability and service quality of this digital platform from the user’s perspective by adopting the Technology Acceptance Model (TAM) as the analytical framework. Within this framework, Perceived Ease of Use (PEOU) is assessed using WebQual 4.0 indicators, while Perceived Usefulness (PU) is measured through the four elements of the marketing mix, namely Product, Price, Place, and Promotion. To analyze the alignment between user expectations and actual service performance, the Importance-Performance Analysis (IPA) method was utilized. Data were obtained from 57 respondents in the Greater Jakarta area (Jabodetabek), primarily first-time users who had previously interacted with the OKE Garden website. Prior to analysis, the data underwent validity and reliability testing to ensure robustness. The findings show that users rated the importance of website attributes higher than their actual performance, indicating a gap that highlights areas requiring improvement. Several key indicators were identified—including ease of navigation, clarity of information, data security, and pricing strategy—which were categorized in Quadrant I (high importance, low performance), indicating areas that require immediate attention. Overall, the results suggest that although digital technology adoption has taken place, user acceptance remains suboptimal. Therefore, a more comprehensive enhancement of usability and service quality is necessary to meet user expectations and improve overall satisfaction.

Performance Analysis of Position Estimation in a Quarter-Car Suspension System Using Kalman-Bucy as a State Observer

2025-08-19T13:06:17+00:00

This study explores the implementation of the Kalman-Bucy observer for state estimation in a quarter-car suspension system operating under various real-world conditions. The research focuses on evaluating the observer’s performance in the presence of road surface disturbances, such as speed bumps, humps, and potholes, combined with stochastic noise and parameter variations. To test its robustness, the system is subjected to Gaussian white noise with an intensity of 10% in both the process and measurement signals. A sensitivity analysis is also carried out by varying the vehicle mass between 400 kilograms under unloaded conditions and 600 kilograms when fully loaded, thereby simulating different passenger and cargo scenarios. Simulation results demonstrate that the Kalman-Bucy observer consistently provides accurate and stable estimations of vehicle position, even in noisy and dynamically changing environments. The observer achieves a Root Mean Square Error (RMSE) of 3.3885 × 10⁻⁵ m, indicating near-perfect estimation accuracy. When integrated into a PID control framework, the proposed observer significantly improves system performance by reducing rise time from 9.76 s to 0.16 s, decreasing undershoot from −0.22 m to −0.15 m, and maintaining a similar settling time of approximately 25 s. Overall, the Kalman-Bucy observer proves to be a reliable and efficient method for state estimation and control enhancement in active suspension systems, showing strong potential for real-world automotive applications.

Mathematical Modeling of an Adaptive Lighting System Based on Solar Panels and Digital Communication for Microalgae Synthesis

2025-08-19T13:10:06+00:00

Microalgae are promising photosynthetic microorganisms widely used in biofuel, pharmaceutical, and environmental applications. Their cultivation efficiency is highly influenced by light intensity, temperature, and pH. This study presents a mathematical model of an adaptive lighting system powered by solar energy and controlled through digital communication for sustainable microalgae synthesis. The system dynamically regulates LED illumination using real-time environmental feedback from temperature and pH sensors integrated into an IoT network. The model combines first-order ordinary differential equations (ODEs) to describe solar input, LED power consumption, environmental response, and communication delay. Numerical simulations performed in MATLAB show that the adaptive control algorithm maintains optimal illumination while minimizing unnecessary energy use. Compared to conventional static lighting, the proposed model achieves a 35% reduction in energy consumption and improved environmental stability despite communication latency. The study provides a foundational framework for developing intelligent, energy-efficient photobioreactor systems that align with the Sustainable Development Goals (SDG 7 and SDG 13). Future work may extend the model toward real-time, predictive, and machine-learning-based control for field-scale implementation.

Weighted ANOVA and Mutual Information for Enhanced Intrusion Detection System

2025-08-19T14:19:02+00:00

The rapid escalation in the sophistication of network attacks has exposed the limitations of traditional Intrusion Detection Systems (IDS). While machine learning has shown great promise in enhancing IDS performance, its success often hinges on the effectiveness of feature selection. Standard feature selection techniques, however, struggle in cybersecurity applications due to the highly imbalanced nature of network traffic datasets. In such settings, minority attack classes—though critical—are often overshadowed by majority classes, leading to reduced detection of rare intrusions. To address this challenge, we propose a hybrid feature selection framework that integrates Analysis of Variance (ANOVA) and Mutual Information (MI) with a novel class-frequency weighting mechanism. This weighting scheme adjusts the relevance score of each feature according to the distribution of classes, ensuring that features associated with rare attacks are more strongly emphasized during the selection process. We evaluate our method on the UNSW-NB15 dataset using a Support Vector Machine classifier. The results show that our approach achieves substantial gains in recall for underrepresented classes while simultaneously reducing feature dimensionality and maintaining efficiency. By improving the visibility of features tied to minority attacks, the proposed framework provides a more balanced and reliable solution for modern IDS. This contribution advances the detection of rare but impactful threats and highlights a scalable pathway for building more resilient cybersecurity defenses.

Integrating Meteorological and PV Data for Short-Term Solar Irradiance Forecasting Using BPNN

2025-08-22T00:25:30+00:00

Solar power plants are highly dependent on solar radiation intensity, which fluctuates due to changes in atmospheric conditions. To maintain system stability and efficiency, an accurate short-term solar radiation prediction model is essential. This study developed a model for forecasting global solar radiation one hour ahead using the Backpropagation Neural Network (BPNN) method. The dataset was obtained from a photovoltaic (PV) system at Building A8 of Surabaya State University, recorded over four days (June 14-17, 2025) at two-minute intervals. Five input variables were used: clearness index, solar radiation, air temperature, air humidity, and PV output power, resulting in a total of 3,020 data samples. The model was trained through a trial-and-error process by varying the number of neurons, hidden layers, and epochs to determine the optimal configuration. The forecast capability of the model was assessed through four statistical indicators: Mean Squared Error (MSE), Root Mean Squared Error (RMSE), Mean Absolute Percentage Error (MAPE), and the coefficient of determination (R²). The best performance was achieved with a network architecture of 15 input neurons representing input variables resulting from data transformation using the sliding window method, one hidden with 25 neurons, and a single unit in the output layer trained for 2000 epochs, resulting in R2 = 0.98, MAPE = 5.89%, and MSE = 0.00027. The novelty of this research lies in the integration of meteorological data with actual PV power output as model input, enabling the network to capture more realistic nonlinear temporal relationships. The proposed short-term forecasting model provides a practical approach to predicting solar radiation based on historical data and can support efficient energy management and photovoltaic system performance analysis.

ANFIS-Controlled High Step Up DC DC Converter for Fuel Cell Systems with Enhanced Efficiency Against Load Variation

2025-08-22T00:34:47+00:00

The primary challenge in utilizing Fuel Cell (FC) systems lies in their inherently low and fluctuating output voltage, which contrasts with the requirements of a Direct Current (DC) bus network that demands a stable and relatively high voltage level. Ensuring consistent voltage regulation in the DC bus network is essential for reliable system performance. An interface converter is required to elevate and stabilize the voltage output under dynamic operating conditions. This paper introduces a high step-up DC–DC converter integrated with an Adaptive Neuro-Fuzzy Inference System (ANFIS)-based control scheme for enhancing the performance of FC power systems. The proposed work encompasses the modeling, analytical design, and structural development of the converter and its intelligent control mechanism. The proposed high step-up converter exhibits a novel structural configuration that integrates a clamp unit, a Multiplier Cell (MC), and cascaded Quadratic Boost Converter (QBC) stages. The contribution of this converter topology lies in its ability to enhance the reliability of fuel cell–based renewable energy systems, achieve high voltage amplification, ensure optimal efficiency, and maintain dynamic stability. This topology is specifically developed to attain an ultra-high voltage conversion ratio, achieving a significant voltage gain of up to 9.65 times, thereby effectively increasing the input voltage from 45 V to 400 V. The ANFIS controller effectively maintains a stable output voltage of 400 V with a maximum deviation of only ±3.5%. The proposed converter achieves a peak efficiency of 87% under varying load conditions, demonstrating its suitability for fuel cell-based energy systems.

Speed Synchronization of Multi-Conveyor System Using Bidirectional Interaction Topologies

2025-08-26T08:30:09+00:00

Long production lines composed of multiple stand-alone mode controllers often face challenges when speed synchronization is required, as each setpoint must be manually adjusted one by one. The issue can be addressed by designating one conveyor as the leader, while the others operate as followers that continuously adjust their speeds to match the leader. The main objective of this study is to develop a multi-conveyor leader–follower system based on distributed cooperative control, allowing all follower conveyors to maintain synchronized speeds with the designated leader unit. In this setup, the leader is equipped with the ability to command all followers to align their speeds to its own, which is governed by a fuzzy logic controller (FLC). Each follower operates in one of two modes: a stand-alone FLC mode or a synchronization mode using cooperative control. The cooperative control mechanism relies on speed information shared among neighboring conveyors, as defined by the system topology. Two types of bidirectional interaction topologies are explored in this work: The Bidirectional Coordinated Conveyor Topology (BCCT) and the Bidirectional Leader Coordinated Conveyor Topology (BLCCT). The proposed control strategy was tested on a mini multi-conveyor setup with one leader and four followers. Synchronization tests on two topologies produced RMSE values of 30.88 RPM for BCCT and 43.87 RPM for BLCCT. A brief disturbance was also applied to one follower to assess the controller’s resilience and its effect on overall system coordination. The study confirms that combining fuzzy logic with cooperative control enhances synchronization and coordination across conveyors.

Leveraging Green IoT to Enhance Energy-Saving Efficiency in Fairness-Oriented Residential Photovoltaic Charging Stations

2025-09-12T21:52:40+00:00

As electric vehicles (EVs) continue to gain global popularity, residential photovoltaic (PV) charging stations are becoming more common, providing a sustainable way to charge EVs. However, the intermittent nature of solar energy creates challenges in ensuring consistent and fair charging, making fairness-based charging scheduling essential. To automate this process, residential PV charging stations require a customized Internet of Things (IoT) system. A significant concern is the substantial energy consumption due to the high volume of data transmission within the IoT system. This research aims to enhance energy efficiency by leveraging green IoT strategies suitable for such applications. The study proposes the use of edge computing, optimized data transmission scheduling, and delta compression techniques at the edge to minimize energy use. The results demonstrate that these strategies are effective in achieving energy savings. Energy-saving efficiency on the source side ranges from 1.96% to 7.84%, while on the load side, it ranges from 57.5% to 61.3%. These findings highlight the effectiveness of the proposed strategies in reducing energy consumption, providing an efficient solution for optimizing data transmission in residential PV charging stations. Overall, the strategies contribute to the sustainable operation of electric vehicle charging infrastructure by improving energy efficiency and ensuring fair distribution of charging resources.

Comparative Evaluation of BM25–FAISS and Small-LLM–GPT in Retrieval-Augmented Generation Concept Map Assessment

2025-12-22T04:38:07+00:00

Concept map-based assessment is a practical approach to measure students’ conceptual understanding, but manual assessment still faces challenges such as subjectivity, inconsistency, and limited scalability. This study proposes the application of Retrieval-Augmented Generation (RAG) as an artificial intelligence-based automated assessment solution in an educational context. The objectives of this study are to compare the effectiveness of two retrieval methods, BM25 and FAISS, and to analyse the trade-off between large-scale generative models (GPT) and Small-LLM in assessing concept map propositions. This study uses a quantitative experimental approach by combining a retriever and a generator in the RAG system. Performance evaluation is carried out using the Macro-F1 and QWK metrics to measure agreement with expert judgment, and the Explanation Relevance Score (ERS) to assess explanation quality. The experimental results show that the FAISS–GPT combination achieves the best performance, with a Macro-F1 of 0.338 and a QWK of 0.146, slightly superior to the BM25–GPT combination. In contrast, the use of Small-LLM, both with BM25 and FAISS, showed lower performance with Macro-F1 values in the range of 0.167–0.221 and QWK close to zero. This finding confirms that semantic-based retrieval plays a vital role in improving the accuracy of automated assessment, while large-scale generative models are more effective in representing conceptual relationships in depth. This study contributes through a comparative analysis of retrievers and generators, and by introducing ERS as an additional metric for RAG-based automated assessment in the field of education.