cover
Contact Name
Alfian Maarif
Contact Email
alfianmaarif@ee.uad.ac.id
Phone
-
Journal Mail Official
biste@ee.uad.ac.id
Editorial Address
-
Location
Kota yogyakarta,
Daerah istimewa yogyakarta
INDONESIA
Buletin Ilmiah Sarjana Teknik Elektro
ISSN : 26857936     EISSN : 26859572     DOI : 10.12928
Core Subject : Engineering,
Buletin Ilmiah Sarjana Teknik Elektro (BISTE) adalah jurnal terbuka dan merupakan jurnal nasional yang dikelola oleh Program Studi Teknik Elektro, Fakultas Teknologi Industri, Universitas Ahmad Dahlan. BISTE merupakan Jurnal yang diperuntukkan untuk mahasiswa sarjana Teknik Elektro. Ruang lingkup yang diterima adalah bidang teknik elektro dengan konsentrasi Otomasi Industri meliputi Internet of Things (IoT), PLC, Scada, DCS, Sistem Kendali, Robotika, Kecerdasan Buatan, Pengolahan Sinyal, Pengolahan Citra, Mikrokontroller, Sistem Embedded, Sistem Tenaga Listrik, dan Power Elektronik. Jurnal ini bertujuan untuk menerbitkan penelitian mahasiswa dan berkontribusi dalam pengembangan ilmu pengetahuan dan teknologi.
Arjuna Subject : -
Articles 295 Documents
Geographic-Origin Music Classification from Numerical Audio Features: Integrating Unsupervised Clustering with Supervised Models Pranolo, Andri; Sularso, Sularso; Anwar, Nuril; Putra, Agung Bella Utama; Wibawa, Aji Prasetya; Saifullah, Shoffan; Dreżewski, Rafał; Nuryana, Zalik; Andi, Tri
Buletin Ilmiah Sarjana Teknik Elektro Vol. 7 No. 4 (2025): December
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.12928/biste.v7i4.13400

Abstract

Classifying the geographic origin of music is a relevant task in music information retrieval, yet most studies have focused on genre or style recognition rather than regional origin. This study evaluates Support Vector Machine (SVM) and Convolutional Neural Network (CNN) models on the UCI Geographical Origin of Music dataset (1,059 tracks from 33 non-Western regions) using numerical audio features. To incorporate latent structure, we first applied K-means clustering with the optimal number of clusters (k=2) determined by the Elbow and Silhouette methods. The cluster assignments were used as auxiliary signals for training, while evaluation relied on the true region labels. Classification performance was assessed with Accuracy, Precision, Recall, and F1-score. Results show that SVM achieved 99.53% accuracy (95% CI: 97.38–99.92%), while CNN reached 98.58% accuracy (95% CI: 95.92–99.52%); Precision, Recall, and F1 mirrored these values. The differences confirm SVM’s superior performance on this dataset, though the near-perfect scores also suggest strong separability in the feature space and potential risks of overfitting. Learning-curve analysis indicated stable training, and cluster supervision provided small but consistent benefits. Overall, SVM remains a reliable baseline for tabular music features, while CNNs may require spectro-temporal representations to leverage their full potential. Future work should validate these findings across multiple datasets, apply cross-validation with statistical significance testing, and explore hybrid deep models for broader generalization.
Effective Analysis of Machine Learning Algorithms for Breast Cancer Prediction M, Vanitha; Anitha, V; Jackson, Beulah; F, Anne Jenefer
Buletin Ilmiah Sarjana Teknik Elektro Vol. 7 No. 4 (2025): December
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.12928/biste.v7i4.13663

Abstract

Early prognosis of Breast Cancer (BC) is significantly important to cure the disease easily so it is essential to develop methods that is able to aid doctors to get precise prognosis. Hence, a BC prognosis methodology is proposed utilizing Machine Learning (ML) approaches. The target of this paper is to utilize classification techniques to classify tumor types, or benign and malignant cells, using 569 samples from Wisconsin Diagnostic Breast Cancer (WDBC) database. Initially, preprocessing is employed to enhance the data’s quality, which includes data cleaning and min-max normalization. It improves the input breast cancer data's quality, accuracy, and suitability for further analysis. Followed by preprocessing, the ML approaches such as K-Nearest Neighbour (KNN), Random Forest (RF) and Support Vector Machine (SVM) methods are analyzed for the classification of BC data. Each algorithm offers a distinct approach to classification by capturing local patterns in data and handles high-dimensional spaces along with nonlinear boundaries through kernel tricks. The developed work is implemented in python software and comparative analysis is done with traditional methods. The outcomes demonstrates that the proposed KNN classifier shows better performance interms of precision, recall, F1-score with an accuracy of 96.49%, ensuring the earliest diagnosis of breast cancer compared with SVM and RF. This comparative approach enhances the reliability of the proposed methodology and supports the selection of the best-performing algorithm offering valuable insights for real-world clinical decision support systems.
Adaptive FLC-based Shunt Active Power Filter with a PV-Fed DC Link for Improved Current Compensation and THD Mitigation Budi, SH Suresh Kumar; Kiranmayi, R.
Buletin Ilmiah Sarjana Teknik Elektro Vol. 7 No. 4 (2025): December
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.12928/biste.v7i4.13804

Abstract

Power quality improvement with traditional controllers (PI, PID, fixed-parameter FLC) is difficult when dealing with nonlinear, time-varying loads and dynamic grid conditions. For microgrids that incorporate renewable energy sources, it is challenging to acquire the precise mathematical models that are necessary for this work. To address power quality challenges, such as distortion of current and Total Harmonic Distortion (THD), produced by nonlinear loads in PV fed systems, such as solar energy conversion, this publication proposes an Adaptive Fuzzy Logic Controller (FLC) based shunt Active Power Filter (APF). An analysis of the power quality enhancement achieved in a distribution power system using a single-stage solar PV integrated shunt APF is presented in this paper. In order to improve load side parameters, such as the elimination of even and odd current harmonics utilizing shunt APF is employed. This filter makes use of a shared DC-link voltage source. In addition, it transfers energy from the PV system's solar panels to the DC link voltage, which is an extra effort. In this paper, It looks at a single-phase inverter that uses an Adaptive FLC to improve parameters on the source and load sides, as well as harmonics, in grid-connected Distributed Generation systems. Also included is a detailed description of the active power filter's chosen current reference generator. Results that have been validated are attained using MATLAB/SIMULINK(R2023b).
Improving Performance and Safety in Mechanical Ventilation: A Robust Control Approach for Airway Pressure and Patient Flow Salman, Nooralhuda; Kadhim, Saleem K.
Buletin Ilmiah Sarjana Teknik Elektro Vol. 7 No. 4 (2025): December
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.12928/biste.v7i4.14283

Abstract

The mechanical ventilation system demand precise and highly responsive control for airway pressure (Paw) and the patient flow (Qpat) as system nonlinearities and time varying disturbances for example the changes in lung mechanics or patient effort it compromise patient safety and treatment efficacy. This study addresses to critical challenge of the robust regulation by comparing two advanced nonlinear control strategies which ar: the Nonlinear PID (NPID) controller and the Active Disturbance Rejection Control with Nonlinear PD structure (ADRC-NPD) controller. The research utilizes a state space model of the respiratory system that developed and simulated in the MATLAB/Simulink for rigorously test controller performance under abrupt changes in the desired pressure setpoint (Pset). The model incorporate clinically relevant lung mechanics that including fixed values for the airway resistance (Rl) and lung compliance (Cl) to represent specific patient scenario. Performance is assessed using key metrics are rise time, overshoot/undershoot, settling time and tracking error. The ADRC-NPD controller consistently demonstrated superior performance that attributed to it Extended State Observer (ESO) for real time estimation and compensation of total system disturbances. Specifically the ADRC-NPD achieved a significantly faster rise time reach to (0.174s vs. 0.38s) and minimal undershoot (-0.3025% vs. -16.573%) compared to the NPID controller that indicating exceptional tracking fidelity and stability crucial for patient well being. The findings strongly that suggest that the ADRC-NPD provides a more robust and clinically viable control solution. Future work will focus on the real-time clinical simulation and hardware in the-loop implementation to validate these results under dynamic and patient specific conditions.
Electrooculography and Camera-Based Control of a Four-Joint Robotic Arm for Assistive Tasks Rusydi, Muhammad Ilhamdi; Gultom, Andre Paskah; Jordan, Adam; Nurhadi, Rahmad Novan; Windasari, Noverika; Sasaki, Minoru; Ramlee, Ridza Azri
Buletin Ilmiah Sarjana Teknik Elektro Vol. 7 No. 4 (2025): December
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.12928/biste.v7i4.14305

Abstract

Individuals with severe motor impairments face challenges in performing daily manipulation tasks independently. Existing assistive robotic systems show limited accuracy (typically 85–92%) and low intuitive control, requiring extensive training. This study presents a control system integrating electrooculography (EOG) signals with real-time computer vision feedback for natural, high-precision control of a 4-degrees-of-freedom (4-DOF) robotic manipulator in assistive applications. The system uses an optimized K-Nearest Neighbors (KNN) algorithm to classify six eye-movement categories with computational efficiency and real-time performance. Computer-vision modules map object coordinates and provide feedback integrated with inverse kinematics for positioning. Validation with 10 able-bodied participants (aged 18–22) employed standardized protocols under controlled laboratory conditions. The KNN classifier achieved 98.17% accuracy, 98.47% true-positive and 1.53% false-negative rates. Distance-measurement error averaged 1.5 mm (± 1.6 mm). Inverse-kinematics positioning attained sub-millimeter precision with 0.64 mm mean absolute error (MAE) for frontal retrieval and 1.58 mm for overhead retrieval. Operational success rates reached 99.48% for frontal and 97.96% for top-down retrieval tasks. The system successfully completed object detection, retrieval, transport, and placement across ten locations. These findings indicate a significant advancement in EOG-based assistive robotics, achieving higher accuracy than conventional systems while maintaining intuitive user control. The integration shows promising potential for rehabilitation centers and assistive environments, though further validation under diverse conditions, including latency and fatigue, is needed.
A Review of EEG Applications in Neuromarketing: Methods, Insights, and Future Directions Pamungkas, Yuri; Thwe, Yamin; Karim, Abdul; Hashim, Uda
Buletin Ilmiah Sarjana Teknik Elektro Vol. 7 No. 4 (2025): December
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.12928/biste.v7i4.14375

Abstract

EEG is increasingly applied in neuromarketing as it provides direct insights into consumer cognition and emotion beyond traditional self-report measures. However, challenges such as small samples, low ecological validity, and methodological limitations hinder its broader real-world application. The research contribution is a comprehensive synthesis of 40 empirical studies that examine EEG applications in neuromarketing, highlighting methodological approaches, analytical techniques, key insights, and persistent gaps that define the current state of the field. This review applied a structured comparative method by extracting and analyzing details from published EEG-based neuromarketing studies, including sample characteristics, device specifications, stimuli types, analytical techniques, and outcomes. The data were organized into a review table and further examined for patterns, strengths, limitations, and emerging opportunities. The results reveal that EEG can reliably classify consumer preferences when paired with deep learning models, while EEG indices such as neural synchrony and frontal alpha asymmetry predict advertising effectiveness and purchase intention. Emotional and attentional processes were consistently reflected in ERP components, and multimodal integration with physiological and behavioral data improved predictive validity. Nonetheless, most studies relied on small, homogeneous samples and static laboratory stimuli, limiting generalizability. In conclusion, EEG holds strong potential for advancing neuromarketing research and practice, yet future work must address scalability, cross-cultural validation, and ecological realism to fully harness its promise.
Systematic Review of Lightweight Cryptographic Algorithms for IoT Security: Advances and Trends A, Shilpa Shetty; B, Sudeepa K; M, Chaithra K; G, Ananth Prabhu
Buletin Ilmiah Sarjana Teknik Elektro Vol. 7 No. 4 (2025): December
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.12928/biste.v7i4.14440

Abstract

The proliferation of the Internet of Things (IoT) has fundamentally transformed modern infrastructure, but has also intensified security risks due to device resource constraints and interconnected environments. This systematic review synthesizes research on lightweight cryptographic algorithms for IoT security, focusing on studies published from 2019 to 2025. Relevant articles were identified through comprehensive searches of IEEE Xplore, ScienceDirect, Springer, and ACM Digital Library using Boolean strings that targeted terms including “lightweight cryptography,” “IoT security,” “side-channel resistance,” and “NIST LWC Standard.” Only peer-reviewed works in English addressing cryptographic primitives suitable for constrained IoT platforms were included; gray literature and studies without benchmarking on IoT-class hardware were excluded. Selection adhered to PRISMA guidelines to reduce selection bias. This review maps algorithmic taxonomies, highlights advances such as ASCON (NIST LWC 2025), side-channel and post-quantum resistance, and discusses real-world hardware-software trade-offs. Limitations arise from database scope, language constraints, and potential exclusion of emerging industry preprints. The analysis identifies persistent gaps—side-channel mitigations, context-aware security, and privacy—with guidance for future research. Overall, the findings clarify current capabilities and boundaries, supporting the development of scalable, energy-efficient, and robust cryptographic frameworks for secure IoT deployments within documented methodological limits.
Evaluation of the Effectiveness of Hand Gesture Recognition Using Transfer Learning on a Convolutional Neural Network Model for Integrated Service of Smart Robot Umam, Faikul; Dafid, Ach.; Sukri, Hanifudin; Asmara, Yuli Panca; Morshed, Md Monzur; Maolana, Firman; Yusuf, Ahcmad
Buletin Ilmiah Sarjana Teknik Elektro Vol. 7 No. 4 (2025): December
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.12928/biste.v7i4.14507

Abstract

This study aims to develop and evaluate the effectiveness of a transfer learning model on CNN with the proposed YOLOv12 architecture for recognizing hand gestures in real time on an integrated service robot. In addition, this study compares the performance of MobileNetV3, ResNet50, and EfficientNetB0, as well as a previously funded model (YOLOv8) and the proposed YOLOv12 development model. This research contributes to SDG 4 (Quality Education), SDG 9 (Industry, Innovation and Infrastructure), and SDG 11 (Sustainable Cities and Communities) by enhancing intelligent human–robot interaction for educational and service environments. The study applies an experimental method by comparing the performance of various transfer learning models in hand gesture recognition. The custom dataset consists of annotated hand gesture images, fine-tuned to improve model robustness under different lighting conditions, camera angles, and gesture variations. Evaluation metrics include mean Average Precision (mAP), inference latency, and computational efficiency, which determine the most suitable model for deployment in integrated service robots. The test results show that the YOLOv12 model achieved an mAP@0.5 of 99.5% with an average inference speed of 1–2 ms per image, while maintaining stable detection performance under varying conditions. Compared with other CNN-based architectures (MobileNetV3, ResNet50, and EfficientNetB0), which achieved accuracies between 97% and 99%, YOLOv12 demonstrated superior performance. Furthermore, it outperformed previous research using YOLOv8 (91.6% accuracy.
Enhancing Facial Emotion Recognition on FER2013 Using Attention-based CNN and Sparsemax-Driven Class-Balanced Architectures Suwartono, Christiany; Bata, Julius Victor Manuel; Airlangga, Gregorius
Buletin Ilmiah Sarjana Teknik Elektro Vol. 7 No. 4 (2025): December
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.12928/biste.v7i4.14510

Abstract

Facial emotion recognition plays a critical role in various human–computer interaction applications, yet remains challenging due to class imbalance, label noise, and subtle inter-class visual similarities. The FER2013 dataset, containing seven emotion classes, is particularly difficult because of its low resolution and heavily skewed label distribution. This study presents a comparative investigation of advanced deep learning architectures against traditional machine-learning baselines on FER2013 to address these challenges and improve recognition performance. Two novel architectures are proposed. The first is an attention-based convolutional neural network (CNN) that integrates Mish activations and squeeze-and-excitation (SE) channel recalibration to enhance the discriminative capacity of intermediate features. The second, FastCNN-SE, is a refined extension designed for computational efficiency and minority-class robustness, incorporating Sparsemax activation, Poly-Focal loss, class-balanced reweighting, and MixUp augmentation. The research contribution is demonstrating how combining attention, sparse activations, and imbalance-aware learning improves FER performance under challenging real-world conditions. Both models were extensively evaluated: the attention-CNN under 10-fold cross-validation, achieving 0.6170 accuracy and 0.555 macro-F1, and FastCNN-SE on the held-out test set, achieving 0.5960 accuracy and 0.5138 macro-F1. These deep models significantly outperform PCA-based Logistic Regression, Linear SVC, and Random Forest baselines (≤0.37 accuracy and ≤0.29 macro-F1). We additionally justify the differing evaluation protocols by emphasizing cross-validation for architectural stability and held-out testing for generalization and note that FastCNN-SE contains ~3M parameters, enabling efficient inference. These findings demonstrate that architecture-level fusion of SE attention, Sparsemax, and Poly-Focal loss improves balanced emotion recognition, offering a strong foundation for future studies on efficient and robust affective-computing systems.
Cybersecurity and Privacy Governance in IoT-Enabled Social Work: A Systematic Review and Risk Framework Chen, Yih-Chang; Lin, Chia-Ching
Buletin Ilmiah Sarjana Teknik Elektro Vol. 7 No. 4 (2025): December
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.12928/biste.v7i4.14589

Abstract

Social work practice is rapidly integrating Internet of Things (IoT) technologies to expand service delivery, yet this integration introduces significant cybersecurity and privacy vulnerabilities that disproportionately threaten vulnerable populations. Existing literature predominantly emphasizes technical security solutions while neglecting the ethical considerations, protective needs of vulnerable groups, and governance frameworks specific to social work contexts. Research Contribution: This study develops the first systematic multidimensional framework integrating engineering and social science perspectives to evaluate IoT cybersecurity, privacy risks, and governance requirements in social work applications. Using a Systematic Literature Review following PRISMA guidelines, we searched five major databases from January 2020 to September 2024. We employed qualitative thematic analysis combined with an innovative quantitative assessment algorithm to score technologies, threats, and governance components across 55 primary studies. Key Findings: Mental health services and vulnerable population support face “very high” privacy risks (PRS > 8.0), primarily from systemic infrastructure weaknesses in consumer-grade devices rather than sophisticated cyberattacks. Homomorphic encryption achieves the highest security score (9.8/10) but exhibits the highest implementation complexity (9.0/10). Federated learning provides an optimal balance (security 8.5, complexity 8.0, cost 6.0). Ethical guidelines demonstrate the highest implementation difficulty (8.2/10), reflecting challenges in translating abstract principles into technical specifications. Quantitative gap analysis identifies vulnerable population protection as the highest research priority (gap score 3.7/10). This study offers an evidence-driven agenda for practitioners and policymakers, proposing context-specific technology selection criteria and adaptive governance models that prioritize interdisciplinary collaboration, ensuring IoT advancements effectively promote social welfare while protecting at-risk individuals.