cover
Article Per Year (5 Year)
All
Home Page
OAI Link
Editorial Team
Contact
Reviewer
Google Scholar
Contact Name
Nurul Khairina
Contact Email
nurulkhairina27@gmail.com
Phone
+6282167350925
Journal Mail Official
nurul@itscience.org
Editorial Address
Jl. Setia Luhur Lk V No 18 A Medan Helvetia Tel / fax : +62 822-5158-3783 / +62 822-5158-3783
Location
Kota medan,
Sumatera utara
INDONESIA
Journal of Computer Networks, Architecture and High Performance Computing
Published by Yayasan Cita Cendekiawan Al Khwarizmi
ISSN : 26559102     EISSN : 26559102     DOI : 10.47709
Core Subject : Science, Education,
Computer Science & IT Education
Journal of Computer Networks, Architecture and Performance Computing is a scientific journal that contains all the results of research by lecturers, researchers, especially in the fields of computer networks, computer architecture, computing. this journal is published by Information Technology and Science (ITScience) Research Institute, which is a joint research and lecturer organization and issued 2 (two) times a year in January and July. E-ISSN LIPI : 2655-9102 Aims and Scopes: Indonesia Cyber Defense Framework Next-Generation Networking Wireless Sensor Network Odor Source Localization, Swarm Robot Traffic Signal Control System Autonomous Telecommunication Networks Smart Cardio Device Smart Ultrasonography for Telehealth Monitoring System Swarm Quadcopter based on Semantic Ontology for Forest Surveillance Smart Home System based on Context Awareness Grid/High-Performance Computing to Support drug design processes involving Indonesian medical plants Cloud Computing for Distance Learning Internet of Thing (IoT) Cluster, Grid, peer-to-peer, GPU, multi/many-core, and cloud computing Quantum computing technologies and applications Large-scale workflow and virtualization technologies Blockchain Cybersecurity and cryptography Machine learning, deep learning, and artificial intelligence Autonomic computing; data management/distributed data systems Energy-efficient computing infrastructure Big data infrastructure, storage and computation management Advanced next-generation networking technologies Parallel and distributed computing, language, and algorithms Programming environments and tools, scheduling and load balancing Operation system support, I/O, memory issues Problem-solving, performance modeling/evaluation
Arjuna Subject : Ilmu Komputer (semua kategori) - Ilmu Komputer (Lain-Lain)
Articles 820 Documents
 78   79   80   81   82 
Generalized Chatterjea Type Contractions on Integrated Matrix Graph Metric Spaces Vinsensia, Desi; Utami, Yulia; G.S, M. Kurniawan; Virna, Lira
Journal of Computer Networks, Architecture and High Performance Computing Vol. 8 No. 1 (2026): Articles Research Januari 2026
Publisher : Information Technology and Science (ITScience)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47709/cnahpc.v8i1.7882

Abstract

This paper proposes a computationally verifiable integrate fixed point framework on the integrated metric space , where combines a continuous component endowed with the matrix induced metric with invertible and a discrete component defined by the shortest-path metric of a finite weighted graph. The objective is to obtain verifiable conditions that guarantee existence, uniqueness, and predictable convergence of fixed points for coupled continuous–discrete dynamics, while embedding the graph geometry directly into the metric via the scaling parameter . Our method studies the coupled operator and derives explicit sufficient inequalities ensuring that satisfies a Chatterjea-type contraction on , yielding an effective contraction factor . In particular, the threshold implies that admits a unique fixed point and that the hybrid Picard iteration converges geometrically in . Numerical experiments support these findings and clarify the integrate mechanism, when maps every vertex to a fixed node, the discrete mode stabilizes after the first iterate, and the successive iterate error decays exponentially at a rate consistent with , with numerical and analytic fixed points agreeing up to floating-point tolerance. Practically, the bound provides an a priori, computable convergence for implementations of matrix graph iterations relevant to graph structured computing and networked models. Future work includes reducing conservatism in the sufficient bounds, exploring richer couplings, and extending the analysis to broader graph classes.
Evaluation of Data Exposure Risks on Unencrypted Application Layer Protocols in RT/RW Net "X" Community Network Using NIST SP 800-86 Framework Reza Febriana; Muhammad Sidik Asyaky
Journal of Computer Networks, Architecture and High Performance Computing Vol. 8 No. 2 (2026): Research Paper April 2026
Publisher : Information Technology and Science (ITScience)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47709/cnahpc.v8i2.8014

Abstract

Security vulnerabilities in community-based networks, such as RT/RW Net, remain a critical concern due to the widespread use of unencrypted protocols. This study presents a quantitative evaluation of data exposure risks in application-layer protocols, focusing on HTTP traffic in local community networks. Using a network forensics approach based on the NIST SP 800-86 framework, traffic was captured and analyzed to measure the frequency and magnitude of sensitive data leaks using automated tools for network traffic analysis. The study quantified exposure across four key indicators: user credentials, session tokens, cookies, and personal information. The results indicated a high level of exposure, with analyzed HTTP packets successfully revealing sensitive data in plaintext, including usernames and passwords. Furthermore, statistical analysis of communication patterns identified significant opportunities for eavesdropping and session hijacking due to the lack of encryption standards. This evaluation provides empirical evidence of critical security gaps in RT/RW Net infrastructure and emphasizes the urgent need to transition to encrypted protocols (HTTPS). The findings provide a quantifiable risk assessment that can serve as a basis for implementing mitigation strategies in community-scale network management.
Deep Learning–Based Forest Fire Classification Using MobileNetV3, ResNet50, and YOLOv8 Djarot Hindarto
Journal of Computer Networks, Architecture and High Performance Computing Vol. 8 No. 2 (2026): Research Paper April 2026
Publisher : Information Technology and Science (ITScience)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47709/cnahpc.v8i2.8112

Abstract

Forest and land fires pose significant environmental, economic, and public health challenges worldwide, particularly in regions with extensive forest coverage and prolonged dry seasons. Early and accurate detection is essential to mitigate damage and support rapid response efforts. This study proposes a deep learning–based approach for forest fire image classification using three prominent models: MobileNetV3, ResNet50, and YOLOv8. A curated dataset of forest fire images was employed, consisting of fire and non-fire scenes captured under diverse environmental conditions, including variations in illumination, smoke density, and background complexity. Prior to model training, all images underwent preprocessing steps such as resizing, normalization, and data augmentation to improve robustness and generalization. The performance of each model was evaluated using standard classification metrics, including accuracy, precision, recall, F1-score, Matthews Correlation Coefficient (MCC), and Cohen’s Kappa. Experimental results indicate that YOLOv8 achieved the best overall performance, with an accuracy of 0.952, precision of 0.9566, recall of 0.952, F1-score of 0.9519, MCC of 0.9412, and Kappa of 0.9400. ResNet50 demonstrated competitive performance with an accuracy of 0.940, slightly outperforming MobileNetV3, which achieved an accuracy of 0.938. The findings highlight that while lightweight architectures such as MobileNetV3 provide efficient performance suitable for resource-constrained environments, more advanced detection frameworks like YOLOv8 offer superior classification capability. Overall, this research demonstrates the effectiveness of modern deep learning models for automated forest fire image classification and supports their potential deployment in real-time early warning and environmental monitoring systems.
Multi-Scale Hierarchical Diffusion Networks for Efficient Layout Generation: Improving Efficiency via Hierarchical Framework and Multi-Decoder Architectures Chakravarthy, Kalyan
Journal of Computer Networks, Architecture and High Performance Computing Vol. 8 No. 2 (2026): Research Paper April 2026
Publisher : Information Technology and Science (ITScience)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47709/cnahpc.v8i2.7497

Abstract

Layout generation remains a challenging task in automated design systems, where existing diffusion models often require extensive computational resources and numerous sampling steps. This work presents a novel multi-scale hierarchical diffusion architecture that achieves state-of-the-art performance through explicit three-level processing with progressive dimensional reduction (128d ? 64d ? 32d). The proposed framework demonstrates 92.5% loss reduction (0.496 to 0.037) over 50 training epochs with only 21,862 parameters, representing a 2.1× reduction compared to existing diffusion-based methods while maintaining superior generation quality. Experimental validation demonstrates the efficiency benefits of hierarchical design across multiple metrics including FID scores (12.3 vs 18.7), precision (0.87 vs 0.79), and training time (0.049s vs 0.127s per epoch). Comprehensive ablation studies quantify the contribution of each hierarchical level and validate architectural design choices.
THE DESIGN OF A HYBRID LSB STEGANOGRAPHY FRAMEWORK WITH ADAPTIVE PIXEL SELECTION AND CHAOS ENCRYPTION FOR SOCIAL MEDIA IMAGES Putra, M Ardi Wiratama; Rahman, Abdul
Journal of Computer Networks, Architecture and High Performance Computing Vol. 8 No. 2 (2026): Research Paper April 2026
Publisher : Information Technology and Science (ITScience)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47709/cnahpc.v8i2.7930

Abstract

The exchange of sensitive data via social media platforms faces dual challenges: the risk of third-party interception and distortion due to image compression. Conventional steganography methods based on Least Significant Bit (LSB) often fail to balance embedding capacity with visual quality and are vulnerable to statistical steganalysis attacks. This research proposes a hybrid steganography framework that integrates multidomain adaptive pixel selection and layered cryptographic security. The pixel selection method combines Canny Edge Detection, Local Binary Pattern (LBP), and Local Entropy to determine optimal Regions of Interest (ROI). Data security is reinforced through content encryption using Advanced Encryption Standard (AES-256) and pixel position scrambling using Arnold Cat Map (ACM). Validation was conducted on 100 images from the ALASKA2 and Dresden datasets. Experimental results demonstrate the system's superior performance in balancing quality and capacity under standard load, the system achieves an average Peak Signal-to-Noise Ratio (PSNR) of 77.85 dB and a Structural Similarity Index (SSIM) of 1.0000. Stress tests confirmed the system's scalability, accommodating a maximum capacity of 3.00 bpp while maintaining safe visual quality (PSNR 51.26 dB). Although the system is fragile against JPEG compression on public timelines, this characteristic is validated to function effectively as a tamper sensitivity feature to detect illegal manipulation. Therefore, this framework is recommended as a solution for secure covert communication via document transmission channels (file sharing) on social media, ensuring high confidentiality and data authenticity.
A COMPARATIVE ANALYSIS OF THE NAIVE BAYES AND C4.5 ALGORITHMS IN DETERMINING ASSISTANT BRANCH MANAGERS AT PT. BANK XYZ Absony, Absony; Yuilistia, Yuilistia
Journal of Computer Networks, Architecture and High Performance Computing Vol. 8 No. 2 (2026): Research Paper April 2026
Publisher : Information Technology and Science (ITScience)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47709/cnahpc.v8i2.7938

Abstract

Determining the right leader is a crucial factor in organizational success, including in the banking sector. This study aims to compare the performance of two popular classification algorithms, namely Naïve Bayes and C4.5, in the selection process of Branch Sub-Leaders at PT. Bank XYZ. Using a data mining approach, the research analyzes historical employee data encompassing personal attributes, competencies, and strategic priorities. The evaluation was conducted using a confusion matrix and ROC curve to measure accuracy, precision, recall, and F1-score for each algorithm. The experimental results show that C4.5 delivers superior performance, achieving an accuracy of 0.985 and an AUC of 1.000 in the binary scenario, while Naïve Bayes only reached an accuracy of 0.296 and an AUC of 0.8365. This study confirms that C4.5 is recommended as the primary model to support decision-making by providing the most suitable classification method for objective and transparent leadership placement. Furthermore, it contributes to sustainable managerial strategies through high accuracy and strong interpretability
ANALYSIS OF SEGMENTATION AND CLIENT TARGET MARKET BUSINESS DECISIONS IN CONSTRUCTION SERVICE COMPANY USING K-MEANS AND DECISION TREE ALGORITHMS : CASE STUDY AT CV JOWON SOLUSINDO Setia, Wondho; Mardiani, Mardiani
Journal of Computer Networks, Architecture and High Performance Computing Vol. 8 No. 2 (2026): Research Paper April 2026
Publisher : Information Technology and Science (ITScience)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47709/cnahpc.v8i2.7942

Abstract

High competition in the construction service industry requires companies to adopt efficient marketing strategies to reduce Customer Acquisition Costs (CAC). CV Jowon Solusindo faces challenges regarding marketing inefficiency due to the implementation of a one size fits all strategy and a high number of unconverted leads (Lost Prospects). This study aims to classify customer characteristics and discover decision rules to formulate personalized marketing strategies. This research employs a quantitative approach with Data Mining methods based on the CRISP-DM framework. The dataset consists of 576 historical transaction records that have undergone data cleaning processes. The method used is a hybrid approach, combining the K-Means Clustering algorithm for customer segmentation and the Decision Tree (C4.5) for rule pattern extraction. The results indicate that the K-Means algorithm with k=3 successfully mapped customers into three distinctive segments, Young Emerging Clients (Average age 33 years with the highest project value), Established Senior Clients (Average age 54 years with stable frequency), and Lost Prospects (Average age 42 years with the lowest offer value). The Decision Tree analysis yielded an accuracy of 67% and identified Age as the primary determinant factor with a split point at 43.5 years. Based on these findings, it is recommended to differentiate marketing strategies into digital visual approaches for customers under 43.5 years and personal approaches for those above that age, as well as pricing strategy adjustments to minimize failure in the Lost Prospects segment.
ANALYSING SENSITIVE DATA SECURITY USING TOKENIZATION-BASED DATA MASKING IN THE EXTRACT-TRANSFORM-LOAD (ETL) PROCESS Dana, Wayan; Rahman, Abdul
Journal of Computer Networks, Architecture and High Performance Computing Vol. 8 No. 2 (2026): Research Paper April 2026
Publisher : Information Technology and Science (ITScience)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47709/cnahpc.v8i2.7949

Abstract

Growth of internet users in Indonesia by 8.7% in 2025, reaching 212 million or 74.6% of the total population, has created significant opportunities for the digital economy but also raised concerns regarding data security. Inadequate management of digital data poses risks to public privacy, prompting the Indonesian government to enact Law Number 27 of 2022 on Personal Data Protection (UU PDP) as a legal framework for data protection. Effective data security measures are necessary across various stages, including safeguarding data during the Extract-Transform-Load (ETL) process. This study aims to develop a data masking technique based on tokenization during the transformation stage of the ETL process to enhance the security of sensitive data. The ETL process, which involves extracting data from diverse sources, transforming it into the required format, and loading it into a database, is particularly vulnerable to sensitive data exposure during the transformation stage, especially in unprotected staging environments. Tokenization replaces original data with tokens that hold no intrinsic value, ensuring data confidentiality throughout the transformation and staging phases. The study's findings indicate that tokenization effectively protects sensitive data before it is loaded into the database, while also minimizing the need for duplicate tables or additional storage for masked data. This research contributes practically to supporting the implementation of the UU PDP, strengthening data security in ETL systems, and fostering a secure digital ecosystem in Indonesia.
Hyperparameter Sensitivity of Vanilla Knowledge Distillation for Compact CNNs on CIFAR-100 Fauzan, Mochamad Rizal; Rachman, Raden Muhammad Rafi; Saputra, Shifa Rangga; Nugraha, Daffa Irsyad
Journal of Computer Networks, Architecture and High Performance Computing Vol. 8 No. 2 (2026): Research Paper April 2026
Publisher : Information Technology and Science (ITScience)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47709/cnahpc.v8i2.8239

Abstract

Knowledge distillation has become an effective strategy for improving compact convolutional neural networks, yet the performance of vanilla knowledge distillation in lightweight image classification is still often reported using default hyperparameter settings without systematic justification. This study addresses the limited empirical understanding of how two core vanilla knowledge distillation hyperparameters, temperature scaling (T) and loss balancing (?), affect compact convolutional neural networks under a unified experimental setting. Using CIFAR-100 as the benchmark dataset, a ResNet-50 teacher was employed to distill knowledge into two lightweight student models, MobileNetV2 and ShuffleNetV2 ×1.0. Performance was evaluated using top-1 accuracy, top-5 accuracy, parameter count, and inference latency. The teacher achieved 81.24% top-1 accuracy and 96.05% top-5 accuracy. Under the default distillation setting, MobileNetV2 improved from 79.18% to 80.83% top-1 accuracy and from 95.77% to 96.40% top-5 accuracy, while reducing latency from 3.98 ms to 3.44 ms. ShuffleNetV2 ×1.0 improved from 77.00% to 78.36% top-1 accuracy and from 94.81% to 95.45% top-5 accuracy, with only a marginal latency increase from 4.23 ms to 4.29 ms. To examine hyperparameter sensitivity, an ablation study was conducted on MobileNetV2 with T = 2, 4, and 6, and ? = 0.3, 0.5, and 0.7. The best configuration was obtained at T = 4 and ? = 0.3, yielding 80.88% top-1 accuracy and 96.51% top-5 accuracy. These results show that vanilla knowledge distillation consistently improves compact convolutional neural networks, but its effectiveness depends strongly on careful hyperparameter selection rather than inherited default settings.
Design and Implementation of a Real-Time IoT-Enabled Embedded Monitoring Architecture for an Off-Grid Infant Incubator Candra, Joni; Aritonang, Mhd Adi Setiawan; Nazwan, Muhammad
Journal of Computer Networks, Architecture and High Performance Computing Vol. 8 No. 2 (2026): Research Paper April 2026
Publisher : Information Technology and Science (ITScience)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47709/cnahpc.v8i2.8323

Abstract

Reliable real-time monitoring of infant incubators is essential in off-grid and resource-limited environments, where unstable power supply and limited infrastructure often compromise continuous operation and data reliability. This study aims to design and implement a real-time IoT-enabled embedded monitoring architecture that addresses the lack of dependable data acquisition and remote monitoring for infant incubators operating under off-grid conditions. The proposed system is developed using a microcontroller-based embedded platform integrated with temperature and environmental sensors, wireless communication modules, and a cloud-based data service. An off-grid photovoltaic power system supports continuous operation, while the embedded architecture is designed with power-aware and real-time constraints. The system adopts an edge-to-cloud approach, enabling local data acquisition and processing at the embedded level and real-time data transmission to a remote monitoring interface. The research methodology includes system architecture design, embedded firmware development, IoT communication implementation, and experimental performance evaluation under continuous off-grid operation. System performance is quantitatively evaluated in terms of data acquisition reliability, communication latency, real-time responsiveness, and operational stability. Experimental results show that the system achieves stable real-time monitoring with an average end-to-end communication latency below 200 ms, a sampling rate of 1 Hz, and continuous operation reliability exceeding 99% uptime during extended off-grid testing. The results demonstrate that integrating real-time embedded systems with IoT-based architecture significantly enhances monitoring reliability for infant incubators in off-grid environments. This study contributes a scalable embedded–IoT monitoring framework that can be extended to other cyber-physical systems operating under constrained energy and infrastructure conditions

Page 82 of 82 | Total Record : 820

 78   79   80   81   82 

Filter by Year

2019 2026


Reset
Filter By Issues
All Issue Vol. 8 No. 2 (2026): Research Paper April 2026 Vol. 8 No. 1 (2026): Articles Research Januari 2026 Vol. 7 No. 2 (2025): Research Article, Volume 7 Issue 2 April, 2025 Vol. 7 No. 2 (2025): Forthcoming: Research Article, Volume 7 Issue 2 April, 2025 Vol. 7 No. 4 (2025): Articles Research October 2025 Vol. 7 No. 3 (2025): Articles Research July 2025 Vol. 7 No. 1 (2025): Article Research January 2025 Vol. 6 No. 3 (2024): Articles Research Volume 6 Issue 3, July 2024 Vol. 6 No. 2 (2024): Articles Research Volume 6 Issue 2, April 2024 Vol. 6 No. 1 (2024): Article Research Volume 6 Issue 1, January 2024 Vol. 6 No. 4 (2024): Articles Research October 2024 Vol. 5 No. 2 (2023): Article Research Volume 5 Issue 2, July 2023 Vol. 5 No. 1 (2023): Article Research Volume 5 Issue 1, January 2023 Vol. 4 No. 2 (2022): Article Research Volume 4 Number 2, July 2022 Vol. 4 No. 1 (2022): Article Research Volume 4 Number 1, Januay 2022 Vol. 3 No. 1 (2021): Computer Networks, Architecture and High Performance Computing, January 2021 Vol. 3 No. 2 (2021): Journal of Computer Networks, Architecture and High Performance Computing, July Vol. 2 No. 2 (2020): Computer Networks, Architecture and High Performance Computing Vol. 2 No. 1 (2020): Computer Networks, Architecture and High Performance Computing Vol. 1 No. 2 (2019): Computer Networks, Architecture and High Performance Computing Vol. 1 No. 1 (2019): Computer Networks, Architecture and High Performance Computing More Issue