Articles

Access the latest knowledge in applied science, electrical engineering, computer science and information technology, education, and health.

Advanced Search
Filter Icon

Filters article

Years

FAQ Arrow
Single Year
Range Year
0
0

Source Title

FAQ Arrow

Authors

FAQ Arrow

30,938 Article Results

Design and development of WIKIN: an interactive nuclear community website for Indonesia using Laravel framework

10.12928/telkomnika.v24i3.27663
Halim; National Research and Innovation Agency Hamadi , Hammam Ahmad; National Research and Innovation Agency Hanif , Muhtadan; National Research and Innovation Agency Muhtadan , Anhar Riza; Polytechnic Institute of Nuclear Technology Antariksawan , Aleksey G.; Tomsk Polytechnic University Goryunov
Despite its significant contributions to health, agriculture, and energy, the public perception of nuclear technology in Indonesia remains cautious and fragmented. Existing communication channels are largely one-way and regulatory, offering limited opportunities for public interaction and collaborative learning. This study investigates how an interactive web-based platform can enhance public engagement and knowledge sharing in nuclear science and technology. To address this challenge, a Nuclear Community Interactive Website (WIKIN) for Indonesia was designed and developed using the Laravel framework, following a structured waterfall methodology. The system integrates role-based access control, modular architecture, and responsive design to support community participation through the sharing of news, discussions, and documentation of service activities. The evaluation was conducted through black-box functional testing of 27 features (all passed) and a system usability scale (SUS) survey involving 51 users, which produced an average score of 74.8 (“Good”), indicating satisfactory usability and acceptance. These results demonstrate that WIKIN provides an effective model for fostering two-way communication, improving transparency, and strengthening public literacy regarding nuclear issues. This study contributes to digital public engagement research by demonstrating how user-centered design principles can be effectively applied to enhance trust, transparency, and community participation in nuclear science communication.
Volume: 24
Issue: 3
Page: 852-865
Publish at: 2026-06-01

Real-time flood forecasting with attention-enhanced hybrid deep learning using internet of things data

10.12928/telkomnika.v24i3.27501
Rissal; Satya Wacana Christian University Efendi , Indrastanti; Satya Wacana Christian University R. Widiasari
Floods are a frequent disaster in Semarang city, Indonesia, requiring an accurate and real-time forecasting system to support effective risk management. This study introduces a hybrid long short-term memory-gated recurrent unit (LSTM-GRU) model with an attention mechanism (attention-enhanced LSTM-GRU) designed to improve the accuracy of flood predictions based on multiparameter internet of things (IoT) data. The novelty of this study lies in the integration of the attention mechanism within the hybrid LSTM-GRU architecture, which allows the model to provide adaptive focus on features and time periods that most influence flood occurrences. The dataset used consists of 1,736 time series samples covering rainfall and water level data collected every 15 minutes from IoT sensors in the upstream and downstream areas of Semarang, Indonesia. Experimental results show that the hybrid model with the attention mechanism provides the best performance with a mean absolute percentage error (MAPE) value of 1.4%, root mean squared error (RMSE) of 1.05, and coefficient of determination (R²) reaching 0.96. This model also achieves 100% recall for the “Danger” class, demonstrating its reliability in detecting critical conditions. The practical implication of this research is the availability of a flood prediction model that is accurate, adaptive, and can be directly applied to IoT-based early warning systems in flood-prone urban areas.
Volume: 24
Issue: 3
Page: 866-882
Publish at: 2026-06-01

Classification of premature cardiac contractions based on RFECV and ensemble learning

10.12928/telkomnika.v24i3.27584
Elsa Sari Hayunah; Universitas Jenderal Soedirman Nurdiniyah , A’isya Nur Aulia; Universitas Jenderal Soedirman Yusuf , Norma; Universitas Jenderal Soedirman Amalia , Widhiatmoko Herry; Universitas Jenderal Soedirman Purnomo , Azizah Najda; Universitas Jenderal Soedirman Hafizha
Premature cardiac contractions, including premature atrial contractions (PACs) and premature ventricular contractions (PVCs), are common arrhythmias that may increase the risk of cardiovascular complications when they occur frequently. Accurate classification of these events from electrocardiogram (ECG) signals remains challenging due to noise and signal variability. This study proposes a machine learning–based classification framework that combines recursive feature elimination with cross-validation for feature selection and an ensemble learning strategy to improve classification robustness. The approach was evaluated using the Massachusetts Institute of Technology – Beth Israel Hospital (MIT-BIH) Arrhythmia database and achieved high classification performance, with an accuracy of 95.34%, F1-score of 92.11%, and balanced precision and recall for PVC and PAC. In addition, SHapley Additive exPlanations (SHAP) were employed to identify the most influential features, enhancing model interpretability. The results demonstrate that the proposed framework provides a reliable and interpretable solution for distinguishing premature cardiac contractions, highlighting its potential application in clinical decision support systems.
Volume: 24
Issue: 3
Page: 891-903
Publish at: 2026-06-01

Machine learning and deep learning for ransomware detection via feature decontamination

10.12928/telkomnika.v24i3.27833
Sriyanto; Institute Informatics and Business Darmajaya Sriyanto , Chairani; Institute Informatics and Business Darmajaya Fauzi , Mohd; Universiti Teknikal Malaysia Melaka (UTeM) Faizal Abdollah , Zuriati; Politeknik Negeri Lampung Zuriati
The continuous escalation of ransomware attacks poses a severe risk to network infrastructure and data integrity, highlighting the urgent requirement for dependable detection systems. This paper presents a comparative analysis of deep learning (DL) and machine learning (ML) techniques for identifying ransomware traffic using the UNSW-NB15 dataset. A significant obstacle in many intrusion detection investigations is feature contamination, where specific attributes inadvertently leak label data or reflect post-incident statistics, resulting in inflated and overly optimistic performance evaluations. To mitigate this concern, a feature decontamination protocol is implemented to isolate 29 reliable attributes, followed by the application of the synthetic minority over-sampling technique (SMOTE) to address the issue of class imbalance. Empirical results demonstrate that the random forest (RF) model achieves superior performance, reaching an accuracy of 0.9027 and a recall of 0.9507. Among the DL candidates, the multi-layer perceptron (MLP) delivers the most competitive outcomes with an accuracy of 0.8859 and an F1-score of 0.8996. These results suggest that ensemble-based ML frameworks offer more effective and computationally efficient ransomware detection when applied to decontaminated tabular datasets.
Volume: 24
Issue: 3
Page: 933-944
Publish at: 2026-06-01

Enhancing energy efficiency in wireless mesh networks through time-synchronized sleep scheduling and low-power hardware

10.12928/telkomnika.v24i3.27616
Rifki; Universitas Bhayangkara Jakarta Raya Muhendra , Dede; Universitas Bhayangkara Jakarta Raya Rukmayadi , Solihin; Universitas Bhayangkara Jakarta Raya Solihin
Energy efficiency remains a critical challenge in wireless mesh networks (WMNs), particularly for internet of things (IoT) deployments with battery powered nodes and multihop communication. This paper proposes a time synchronized sleep scheduling framework that integrates a lightweight regression-based time synchronization model with low-power hardware to reduce energy consumption in long range (LoRa)-based WMNs. The proposed mechanism aligns local node clocks with a global reference using slope and offset correction, enabling synchronized active and sleep states across nodes. This coordination significantly reduces idle listening and unnecessary radio-on time. The proposed approach is validated through real world experiments on a multihop LoRa mesh testbed with up to three hops. Results show a substantial improvement in energy efficiency, reducing cumulative energy consumption from 125.31 mWh to 28.18 mWh over 10 hours (77.5% reduction). The sleep-mode current is reduced to 0.01 mA, demonstrating effective duty cycling. Furthermore, the approach maintains stable routing, bounded latency, and high packet delivery ratio (PDR). These findings confirm that accurate time synchronization is a key enabler for energy-efficient and reliable multihop communication, providing a practical solution to extend the operational lifetime of IoT-based WMNs.
Volume: 24
Issue: 3
Page: 966-978
Publish at: 2026-06-01

Leveraging artificial intelligence for detection of denial-of service attacks in 5G network environments

10.12928/telkomnika.v24i3.27402
Baseel; Hakim Sabzevari University Al-Ali , Mina; Hakim Sabzevari University Malekzadeh
This research introduces an evaluation methodology that addresses the data leakage problem for detecting denial-of-service attacks in fifth-generation (5G) network slicing environments, and applies it to perform a benchmark comparison among twelve machine learning (ML), deep learning (DL), and probabilistic models using a publicly available 5G network slicing dataset for DoS/DDoS attacks. This methodology strictly enforces the execution of all preprocessing steps exclusively on the training data, where feature selection is performed using the mutual information (MI) metric, values are standardised via the z-score method, and synthetic samples are produced through the synthetic minority oversampling technique (SMOTE) technique on the training set only, with MI recalculated independently within each cross-validation (CV) cycle. Nine features out of eighty-four were retained at the elbow point where MI reached 0.51 or above. On the held-out test set containing approximately eighty percent benign data and twenty percent attack data, the convolutional neural network (CNN) model achieved the highest F1 value of 0.983 with a false discovery rate of 0.027, while the random forest model reached an F1 value of 0.968 at a considerably lower computational cost. All results remain tied to this particular dataset, and their generalisability to real-world 5G network traffic has not yet been validated.
Volume: 24
Issue: 3
Page: 926-932
Publish at: 2026-06-01

High-efficiency Doherty amplifier with metamaterial harmonic control for sub-6 GHz 5G base station

10.12928/telkomnika.v24i3.27618
Faycal; Moulay Ismail University El Hardouzi , Mohammed; Moulay Ismail University Lahsaini
This paper presents a broadband Doherty power amplifier (DPA) for Sub-6 GHz 5G wireless applications, integrating a metamaterial-based harmonic control circuit to suppress the 2nd and 3rd harmonics. The design employs a CGH40010F transistor on a Rogers RO3004C substrate. All results are obtained from simulations, while the resonator has been experimentally validated. The proposed approach enhances drain efficiency under output back-off (OBO) conditions without compromising linearity. Simulation results show a gain of 12.2–16.18 dB, with drain efficiency from 50.42 % to 79.83 %. Efficiency remains high at 3 dB and 6 dB back-off, ranging from 50.8–75.6 % and 36–62.8 %, respectively. These findings demonstrate the DPA’s potential for next-generation base stations requiring compact, efficient, and wideband amplifiers, offering a practical solution for modern wireless communication systems. The novelty lies in combining broadband DPA design with harmonic suppression via metamaterials, providing improved efficiency and spectral performance.
Volume: 24
Issue: 3
Page: 801-815
Publish at: 2026-06-01

Low-cost environmental chamber for battery calendar aging under tropical conditions: design and validation

10.12928/telkomnika.v24i3.27666
Uvi; University of Indonesia Desi Fatmawati , Iwa; University of Indonesia Garniwa , Faiz; Universitas Indonesia Husnayain , Danang; Universitas Gadjah Mada Lelono , Kuwat; Universitas Gadjah Mada Triyana , Amelia; Republic Indonesia Defense University Chandra Pratiwi
This study presents a low-cost environmental chamber designed to replicate tropical temperature and humidity conditions for calendar-aging studies of LiFePO₄ lithium iron phosphate (LFP) cells. The system integrates passive insulation with an Arduino-based active control system for real-time monitoring. The design’s novelty lies in its cost-effective ability to maintain tropical-specific profiles, validated against Meteorology, Climatology, and Geophysics Agency (BMKG) meteorological data to ensure correlation with diurnal cycles. Experimental results demonstrate stable operation for 13 consecutive days, with an average temperature of 29.47 °C and a stability metric of ±0.61 °C, keeping 100% of data within the 25–35 °C target. Although relative humidity (RH) showed an average of 76.52%, its stability was quantified by a 96.98% success rate in maintaining the 70–80% target range. The chamber’s suitability for long-term degradation studies was confirmed via a one-month calendar-aging test on a 15 Ah LFP cell, where tha battery capacity decreased from 100% to 99.58%. These results demonstrate that the proposed chamber reliably maintains tropical environmental conditions and is suitable for long-term, low-cost battery aging studies.
Volume: 24
Issue: 3
Page: 1014-1026
Publish at: 2026-06-01

Asegmentation based optical character recognition system for Bangla printed text

10.12928/telkomnika.v24i3.26961
Mahir; Bangabandhu Sheikh Mujibur Rahman Digital University Mahbub , Ahmedul; University of Dhaka Kabir
Bangla ranks as the fifth most spoken language globally, catalyzing significant interest in the development of Bangla optical character recognition (OCR) sys tems. The intricate structure of the Bangla script, including compound char acters, modifiers, and headlines, complicates the formation of words. This research introduces a complete OCR system pipeline for printed Bangla text. It employs a thinning-based segmentation approach combined with a convolu tional neural network (CNN) to recognize Bangla fonts. Additionally, a part of speech (POS)-aware spell checker is proposed that automatically corrects mis spelled words while considering their context within the sentence. We intro duce semi-generalized filters that adapt to new fonts, addressing conjunct for mation challenges in Bangla OCR. This flexible design allows for adaptation to new fonts. The ResNet50 model is utilized to accurately recognize segmented characters and modifiers. We achieve a character segmentation error of 3.354% and an overall segmentation error of 2.332%. The ResNet50 recognition model achieves an accuracy of 98.345%.
Volume: 24
Issue: 3
Page: 945-956
Publish at: 2026-06-01

Layered virtual-leader DMPC with SQP for scalable V formation tracking of omni-robots in cluttered maps

10.12928/telkomnika.v24i3.27589
Tuan Phu; Academy of Military Science and Technology Duong , Vinh Quang; Academy of Military Science and Technology Nguyen , Minh Tuan; Thai Nguyen University of Technology, Vietnam Nguyen
This paper proposes a hierarchical virtual leader based distributed model predictive control (DMPC) framework for V-formation control of omnidirectional mobile robots in static obstacle environments. The physical leader follows a predesigned reference trajectory, while the followers maintain the desired formation through distributed optimization. A layered communication topology is established, where only a subset of robots receives the leader’s predicted states and acts as virtual leaders for downstream followers. Each robot independently optimizes its control sequence using local neighbor information, enabling fully distributed coordination without centralized synchronization. The unified cost function considers formation maintenance, leader or virtual leader tracking, obstacle avoidance, and control effort. Static obstacles are represented on a grid map to ensure collision-free motion. Simulation results demonstrate that the proposed framework achieves accurate formation keeping, smooth trajectory tracking, and effective obstacle avoidance. The hierarchical virtual-leader architecture enhances scalability, coordination efficiency, and robustness for multi-robot formation systems.
Volume: 24
Issue: 3
Page: 1037-1047
Publish at: 2026-06-01

A study of constrained Bézier fitting curve with tangent continuity for quadruped walking robot gaits

10.12928/telkomnika.v24i3.27620
Hung T.; Thai Nguyen University of Technology Nguyen , Minh T.; Thai Nguyen University of Technology Nguyen , Mui D.; Thai Nguyen University of Technology Nguyen , Long Q.; Thai Nguyen University of Information and Communication Technology Dinh , Dung T.; Thai Nguyen University of Information and Communication Technology Nguyen
Smooth and stable gait generation is critical for quadruped robots operating in unstructured environments. This paper introduces a constrained Bézier fitting framework. It enforces tangent continuity (𝐶1 continuity) at the gait cycle junction. This continuity addresses the tangential discontinuities that commonly arise in unconstrained Bézier trajectories. The method formulates foot-trajectory design as a constrained least-squares problem solved via Lagrange multipliers, enabling the control points to simultaneously satisfy interpolation targets and matched-tangent conditions. The resulting curves retain the geometric flexibility of classical Bézier parametrizations while producing well-behaved velocity profiles suitable for legged locomotion. These trajectories are integrated into an impedance-controlled leg model and evaluated in the MuJoCo simulator. Simulation results indicate noticeable reductions in torque spikes and improvements in tracking accuracy when compared to unconstrained Bézier and spline baselines, with representative trials showing reductions on the order of 40% and tracking improvements of approximately 25%. The proposed approach combines mathematical rigor with practical applicability, providing an efficient and reliable solution for high-performance quadruped gait planning.
Volume: 24
Issue: 3
Page: 1048-1057
Publish at: 2026-06-01

Lightweight SDN/NFV-based framework for dynamic data-flow and network slice adaptation

10.12928/telkomnika.v24i3.27810
Sumbal; Wigan and Leigh College and University Centre Zahoor , Ali; Calrom Ltd. Mamoon
The increasing demand for responsive and reliable network services in next-generation communication systems has intensified the need for dynamic resource management and quality of service (QoS) assurance. Software-defined networking (SDN) and network function virtualization (NFV) provide programmability and flexibility for modern networks. However, practical platforms that demonstrate real-time adaptive behavior remain limited. This study differs from prior simulation-focused work by demonstrating real-time adaptive slice control in a reproducible container-based SDN/NFV emulation environment. A bottleneck-aware slice controller is developed to classify degradations as network-limited, server-limited, or service failure using joint indicators, and to select rerouting or service migration using stability constraints and a lightweight action-cost model. Experimental results show that throughput is restored to above 90% of nominal capacity. Recovery typically occurs within two to three control iterations. Service continuity is maintained with low control-plane overhead. The work provides a reproducible experimental baseline and a decision mechanism that reduces incorrect reroutes/migrations under ambiguous key performance indicator (KPI) drops.
Volume: 24
Issue: 3
Page: 779-785
Publish at: 2026-06-01

Smart classroom 4.0 using embedded systems for attendance, energy monitoring, and environmental control

10.12928/telkomnika.v24i3.27508
Septriandi; State Polytechnic of Malang Wirayoga , Nizar; State Polytechnic of Malang Fairuzaman , Moh Muzib; State Polytechnic of Malang Pratama , Wildan Ahmad; State Polytechnic of Malang Fauzi , Mohammad Alwi; State Polytechnic of Malang Ferdiansyah Alfarizi
The increasing demand for digitalization and energy efficiency in vocational education has encouraged the development of intelligent classroom systems. This study proposes an integrated smart classroom 4.0 system based on the Internet of things (IoT) and embedded systems to improve attendance management, electrical safety, energy efficiency, and environmental monitoring. The proposed system integrates smart attendance, power monitoring, automatic lighting control, and environmental sensing into a unified architecture using ESP32 microcontrollers and a Raspberry Pi embedded server. Attendance is automated using radio frequency identification (RFID), fingerprint recognition, and non-contact body temperature measurement, achieving an average accuracy of 92.5% with a system latency of 1.1–1.4 s. Electrical monitoring using the PZEM-004T sensor shows zero error for voltage and current measurements and a maximum power measurement error of 0.31%, while overload and abnormal voltage conditions are successfully handled through automatic protection. Automatic lighting control based on YOLOv5s image processing achieves approximately 90% detection accuracy under high occupancy conditions. All subsystems communicate via the message queuing telemetry transport (MQTT) protocol and are visualized through a real-time web dashboard. The results demonstrate that the proposed system provides a low-cost, scalable, and reliable solution for energy-efficient and intelligent classroom management in vocational education environments.
Volume: 24
Issue: 3
Page: 979-990
Publish at: 2026-06-01

Real-time classification of Pasaman oranges using Mamdani fuzzy inference system and ESP32 microcontroller

10.12928/telkomnika.v24i3.27407
Fahmi; Andalas University Fitrio Fauzi , Ifmalinda; Andalas University Ifmalinda , Azrifirwan; Andalas University Azrifirwan
Manual classification of Pasaman oranges based on visual assessment of size and color often produces inconsistent results due to human subjectivity. This study develops an automatic classification system using a Mamdani fuzzy inference system (FIS) implemented on an ESP32 microcontroller. Fruit diameter is measured using a high-frequency sound wave ranging module (HC-SR04) ultrasonic sensor, while surface color is detected using a TAOS color sensor 3200 (TCS3200) color sensor. The obtained data are processed through fuzzification, inference, and defuzzification to classify oranges into three quality grades (A, B, and C). System performance evaluation shows strong agreement between the developed system and matrix laboratory (MATLAB) simulation, with a coefficient of determination (R²) value of 0.9855, indicating reliable and consistent classification performance for automated agricultural grading applications.
Volume: 24
Issue: 3
Page: 1058-1067
Publish at: 2026-06-01

Interdigitized capacitive humidity sensor using FR4 printed circuit board material without sensitive layer

10.12928/telkomnika.v24i3.27525
Setyawan; Brawijaya University Purnomo Sakti , Jessica; Brawijaya University Alfa Aini , Shania; Brawijaya University Rachmi , Elsa; Brawijaya University Hedya Kusumaningtyas , Irma; Brawijaya University Azizah , Triswantoro; Brawijaya University Putro , Dewi; Brawijaya University Anggraeni
Humidity sensors are widely used across numerous applications, including agriculture, health, industry, environmental monitoring, and high-risk environments such as chip fabrication rooms. Interdigital capacitors (IDCs) are among several humidity-sensing designs that have been developed, and they are attractive due to simple fabrication and easy integration with electronic systems. We developed an IDC humidity sensor on a glass-epoxy substrate (FR4) using a standard commercial printed circuit board (PCB) production process. Without additional sensitive layer, the sensor production is simple. The sensor design was explored by varying electrode width. Calibration was performed in a self-constructed chamber over a relative humidity range from 15% to 90%. The IDC sensor was benchmarked against a standard humidity measurement instrument. The IDC capacitance showed a linear correlation with relative humidity, with a correlation coefficient of at least 0.98. With the finest track and spacing size of 0.254 mm achievable via commercial PCB manufacturing and a sensor footprint of 10×10 mm, the capacitance falls in the pF range, compatible with most available electronic devices. The sensitivity of the developed sensors is 2.361×10-4, 2.361×10-4, 2.361×10-4 picofarads for a track width of 0.254 mm, 0.3048 mm and 0.3556 mm respectively. These results indicate that mass production of IDC humidity sensors could be achieved at a low cost. The combination of small feature sizes and a compact sensor footprint support practical integration into existing electronic systems.
Volume: 24
Issue: 3
Page: 1027-1036
Publish at: 2026-06-01
1 2 3
...
2062 2063
Show 43 of 2063

Discover Our Library

Embark on a journey through our expansive collection of articles and let curiosity lead your path to innovation.

Explore Now
Library 3D Ilustration