ETRI Journal最新文献_第3页

Proactive handover optimization via multi-agent deep reinforcement learning in integrated LEO satellite–terrestrial networks for high-mobility terminals 基于多智能体深度强化学习的高机动性LEO星地融合网络主动切换优化

IF 2 4区计算机科学

ETRI Journal Pub Date : 2026-08-18 Epub Date: 2026-04-02 DOI: 10.4218/etrij.2025-0240

Huiyeon Jang, Junyoung Kim, In-Sop Cho, Minsu Shin, Joon Gyu Ryu, Soyi Jung

{"title":"Proactive handover optimization via multi-agent deep reinforcement learning in integrated LEO satellite–terrestrial networks for high-mobility terminals","authors":"Huiyeon Jang, Junyoung Kim, In-Sop Cho, Minsu Shin, Joon Gyu Ryu, Soyi Jung","doi":"10.4218/etrij.2025-0240","DOIUrl":"https://doi.org/10.4218/etrij.2025-0240","url":null,"abstract":"This paper proposes a predictive handover decision framework for optimizing handover planning in integrated low Earth orbit (LEO) satellite–terrestrial networks. To forecast the fluctuations in received signal reference power (RSRP) caused by the high mobility of mobile terminals (MTs) and satellites accurately, we employ a convolutional neural network–long short-term memory encoder–decoder architecture. This forecasting model enables each MT to predict its future RSRP levels independently over a given time horizon. By leveraging the predicted RSRP trajectories, each MT autonomously determines its handover strategy, including its optimal target and timing. To enhance the performance of these decentralized handover decisions, we integrate a multi-agent deep reinforcement learning approach based on the QMIX algorithm. This approach enables coordinated policy learning among MT agents while preserving decentralized execution, thereby improving overall system stability and handover efficiency in highly dynamic environments.","PeriodicalId":11901,"journal":{"name":"ETRI Journal","volume":"48 4","pages":"577-589"},"PeriodicalIF":2.0,"publicationDate":"2026-08-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.4218/etrij.2025-0240","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148783751","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Size-agnostic file fragment classification via fixed byte slicing 通过固定字节切片进行大小无关的文件片段分类

IF 2 4区计算机科学

ETRI Journal Pub Date : 2026-08-18 Epub Date: 2026-04-30 DOI: 10.4218/etrij.2025-0314

Hyeongsik Kim, Sisung Liu, Je Hyeong Hong

引用次数: 0

Operation-level scheduling framework for efficient deep learning inference on embedded systems using directed acyclic graphs 基于有向无环图的嵌入式系统高效深度学习推理的操作级调度框架

IF 2 4区计算机科学

ETRI Journal Pub Date : 2026-08-18 Epub Date: 2026-01-29 DOI: 10.4218/etrij.2025-0201

Mooseop Kim, SuGil Choi, Sungjun Wang, Chi Yoon Jeong

{"title":"Operation-level scheduling framework for efficient deep learning inference on embedded systems using directed acyclic graphs","authors":"Mooseop Kim, SuGil Choi, Sungjun Wang, Chi Yoon Jeong","doi":"10.4218/etrij.2025-0201","DOIUrl":"https://doi.org/10.4218/etrij.2025-0201","url":null,"abstract":"This study presents an operation-level scheduling framework for efficient deep learning inference on heterogeneous embedded systems. Motivated by the observation that deep neural networks comprise diverse operations in which the execution latency is highly dependent on the target hardware and input dimensions. The framework hypothesizes that accurate latency prediction and fine-grained scheduling of individual operations reduce end-to-end inference time. It follows a three-stage approach: (i) offline profiling of operation latencies across varying input sizes and devices; (ii) training latency prediction models using input-aware features; and (iii) directed acyclic graph-based runtime scheduling to assign each operation to a central processing unit, graphics processing unit, or both. The framework is evaluated on two embedded platforms (Jetson Nano and ODROID-XU4) and demonstrates an inference latency reduction of up to 74% across multiple deep learning models. These results indicate that the framework is adaptable, lightweight, and effective for resource-constrained artificial intelligence deployments.","PeriodicalId":11901,"journal":{"name":"ETRI Journal","volume":"48 4","pages":"635-649"},"PeriodicalIF":2.0,"publicationDate":"2026-08-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.4218/etrij.2025-0201","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148785242","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Enhancing secrecy in RIS–fountain code-aided multi-cluster satellite–terrestrial NOMA networks under eavesdropping threats 监听威胁下RIS-fountain码辅助多集群星地NOMA网络的保密性增强

IF 2 4区计算机科学

ETRI Journal Pub Date : 2026-08-18 Epub Date: 2026-05-26 DOI: 10.4218/etrij.2025-0234

Nguyen Van Toan, Pham Ngoc Son, Tran Trung Duy, Lam-Thanh Tu

{"title":"Enhancing secrecy in RIS–fountain code-aided multi-cluster satellite–terrestrial NOMA networks under eavesdropping threats","authors":"Nguyen Van Toan, Pham Ngoc Son, Tran Trung Duy, Lam-Thanh Tu","doi":"10.4218/etrij.2025-0234","DOIUrl":"https://doi.org/10.4218/etrij.2025-0234","url":null,"abstract":"This paper proposes a hybrid satellite–terrestrial relay network that integrates advanced technologies, such as physical layer security, fountain coding, non-orthogonal multiple access, reconfigurable intelligent surfaces (RISs), and partial relay selection (PRS). The satellite transmits fountain-coded packets to terrestrial relays, where the best relay selected using the PRS algorithm employs the decode-and-forward technique to deliver data to two user clusters with RIS support. This setup enhances both reliability and security. Meanwhile, eavesdroppers attempt to intercept the RIS-reflected signals. Exact closed-form expressions are derived for the outage probability (OP) and system OP of legitimate users, as well as the intercept probability (IP) and system IP of eavesdroppers. Monte Carlo simulations confirmed the analytical results and examined the effects of the key parameters. Additionally, two benchmark models were developed to facilitate performance and security comparisons with the proposed scheme.","PeriodicalId":11901,"journal":{"name":"ETRI Journal","volume":"48 4","pages":"590-603"},"PeriodicalIF":2.0,"publicationDate":"2026-08-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.4218/etrij.2025-0234","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148785102","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

SmartFed: Federated learning with collaborative optimization of the model algorithm and data enhancement SmartFed：采用模型算法协同优化和数据增强的联邦学习

IF 2 4区计算机科学

ETRI Journal Pub Date : 2026-08-18 Epub Date: 2025-11-26 DOI: 10.4218/etrij.2025-0028

Yan Zeng, Wanyu Hong, Chengchuang Huang, Changhui Jiang, Meiting Xue, Jilin Zhang, Li Zhou, Jian Wan

{"title":"SmartFed: Federated learning with collaborative optimization of the model algorithm and data enhancement","authors":"Yan Zeng, Wanyu Hong, Chengchuang Huang, Changhui Jiang, Meiting Xue, Jilin Zhang, Li Zhou, Jian Wan","doi":"10.4218/etrij.2025-0028","DOIUrl":"https://doi.org/10.4218/etrij.2025-0028","url":null,"abstract":"Federated learning enables collaborative model training across devices while preserving user privacy. However, the imbalanced class distribution in local datasets presents a challenge to global model accuracy. To address this problem, this paper proposes SmartFed, a novel optimization method integrating algorithm optimization and data enhancement. It introduces a confidence scoring mechanism to evaluate the divergence between global and local models, guiding local training to retain high-confidence knowledge and acquire low-confidence knowledge from local data. Additionally, SmartFed uses a variational autoencoder to generate virtual data features for fine-tuning the classifier layers of both models, complemented by feature distillation. Experiments show that SmartFed improves accuracy by 16.3%, reduces convergence time overhead by 22.2%, and decreases communication overhead by 33.1%, compared with the accuracy of state-of-the-art methods. The results show that SmartFed provides an effective solution for addressing class imbalance in federated learning, leading to enhanced model performance and more efficient training processes.","PeriodicalId":11901,"journal":{"name":"ETRI Journal","volume":"48 4","pages":"650-662"},"PeriodicalIF":2.0,"publicationDate":"2026-08-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.4218/etrij.2025-0028","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148785106","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

125-GHz-band RF receiver front end using 40 nm complementary metal–oxide–semiconductor technology 125 ghz频段射频接收器前端采用40 nm互补金属氧化物半导体技术

IF 2 4区计算机科学

ETRI Journal Pub Date : 2026-08-18 Epub Date: 2025-12-05 DOI: 10.4218/etrij.2025-0176

Seon-Ho Han, Dong-Woo Kang, Bon-Tae Koo

{"title":"125-GHz-band RF receiver front end using 40 nm complementary metal–oxide–semiconductor technology","authors":"Seon-Ho Han, Dong-Woo Kang, Bon-Tae Koo","doi":"10.4218/etrij.2025-0176","DOIUrl":"https://doi.org/10.4218/etrij.2025-0176","url":null,"abstract":"This paper presents a 125-GHz-band radio-frequency (RF) receiver front end designed for radar applications using 40 nm complementary metal–oxide–semiconductor (CMOS) technology. The receiver front-end chip integrates a five-stage single-to-differential low-noise amplifier, transconductance I/Q driver, quadrature I/Q passive mixer, quadrature I/Q transimpedance amplifier, single-to-differential local oscillator (LO) driver, and quadrature I/Q LO generator. Designing an I/Q receiver front end for the 125-GHz band is particularly challenging due to the difficulty of generating and maintaining well-matched I/Q signals at such high frequencies. This work demonstrates highly matched I/Q signal outputs using a simplified circuit architecture that achieves both low power consumption and a compact core chip size. Measurement results of the RF receiver front-end chip reveal a conversion gain of 30 dB, 3 dB bandwidth ranging from 121 GHz to 129 GHz, an input P1dB of −30 dBm, and a double-sideband noise figure of 11.4 dB. The measured gain and phase mismatches between the I and Q channels are only 0.1 dB and 3.6°, respectively. The chip operates with low power consumption, drawing only 80 mA from a 1 V supply. It also features a compact core area of 0.22 mm2.","PeriodicalId":11901,"journal":{"name":"ETRI Journal","volume":"48 4","pages":"723-730"},"PeriodicalIF":2.0,"publicationDate":"2026-08-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.4218/etrij.2025-0176","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148783503","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

3D Gaussian rasterization for direct rendering of view-multiplexed images on multiview displays 用于在多视图显示器上直接渲染视图复用图像的三维高斯光栅化

IF 2 4区计算机科学

ETRI Journal Pub Date : 2026-08-18 Epub Date: 2026-02-21 DOI: 10.4218/etrij.2025-0246

Joonsoo Kim, Kugjin Yun, Hyon-Gon Choo, Hyeongju Mun, Kyeongbo Kong

{"title":"3D Gaussian rasterization for direct rendering of view-multiplexed images on multiview displays","authors":"Joonsoo Kim, Kugjin Yun, Hyon-Gon Choo, Hyeongju Mun, Kyeongbo Kong","doi":"10.4218/etrij.2025-0246","DOIUrl":"https://doi.org/10.4218/etrij.2025-0246","url":null,"abstract":"Real-time rendering of three-dimensional scenes is essential for interactive, realistic display systems. Recently, representations based on 3D Gaussian primitives have emerged as a promising approach for novel view synthesis, offering both photorealistic image quality and real-time rendering. However, extending such renderers to autostereoscopic multiview displays is not straightforward because each subpixel must be rendered from a different viewpoint, whereas conventional renderers exploit shared viewport information to achieve efficiency. In this paper, we propose a rasterization method for 3D Gaussian-based representations that enables efficient rendering for multiview displays. By leveraging prior knowledge of the target viewport layout, the proposed method achieves an approximately 11 times speedup compared with the baseline without compromising image quality. We analyze the rendering performance of the proposed method under various combinations of renderer configurations and 3D scene datasets, demonstrating its potential for real-time multiview rendering.","PeriodicalId":11901,"journal":{"name":"ETRI Journal","volume":"48 4","pages":"711-722"},"PeriodicalIF":2.0,"publicationDate":"2026-08-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.4218/etrij.2025-0246","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148785076","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

RETRACTION: Comparing Automated and Non-Automated Machine Learning for Autism Spectrum Disorders Classification Using Facial Images 摘要：比较使用面部图像进行自闭症谱系障碍分类的自动化和非自动化机器学习

IF 2 4区计算机科学

ETRI Journal Pub Date : 2026-06-16 Epub Date: 2026-03-23 DOI: 10.4218/etr2.70112

{"title":"RETRACTION: Comparing Automated and Non-Automated Machine Learning for Autism Spectrum Disorders Classification Using Facial Images","authors":"","doi":"10.4218/etr2.70112","DOIUrl":"https://doi.org/10.4218/etr2.70112","url":null,"abstract":"RETRACTION: B. R. G. Elshoky, E. M. G. Younis, A. A. Ali, and O. A. S. Ibrahim, “Comparing Automated and Non-Automated Machine Learning for Autism Spectrum Disorders Classification Using Facial Images,” ETRI Journal 44, no. 4 (2022): 613–623, https://doi.org/10.4218/etrij.2021-0097.The above article, published online on 23 March 2022 in Wiley Online Library (wileyonlinelibrary.com), has been retracted by agreement between the journal Editor-in-Chief, Sungwon Yi; and John Wiley & Sons, Ltd.The retraction was agreed upon due to concerns regarding both the reliability of the study and significant privacy issues. The dataset used in the study contains images of children purported to have, or not have autism spectrum disorder (ASD), sourced from the internet without any documented clinical history or independent verification of an ASD diagnosis, raising concerns about the scientific validity of the dataset. In addition, the images of minors have been used and presented in the article without consent from the individuals or their parents/legal guardians. For these reasons, the article has been retracted and the corresponding images of children have been redacted. The corresponding author, Basma Ramdan Gamal Elshoky, agreed with the retraction decision; no confirmation was obtained by the remaining co-authors.","PeriodicalId":11901,"journal":{"name":"ETRI Journal","volume":"48 3","pages":""},"PeriodicalIF":2.0,"publicationDate":"2026-06-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.4218/etr2.70112","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148268989","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Multi-component framework for improved photovoltaic power forecasting performance 改进光伏发电预测性能的多组件框架

IF 2 4区计算机科学

ETRI Journal Pub Date : 2026-06-16 Epub Date: 2025-05-20 DOI: 10.4218/etrij.2024-0615

Jie Yang

引用次数: 0

Graph neural network-enhanced auxiliary classifier generative adversarial network framework for robust intrusion detection 图神经网络增强辅助分类器生成对抗网络框架的鲁棒入侵检测

IF 2 4区计算机科学

ETRI Journal Pub Date : 2026-06-16 Epub Date: 2025-10-29 DOI: 10.4218/etrij.2025-0152

Tianjing Wang, Qi Liu, Hang Shen, Xiaokang Luo, Guangwei Bai

{"title":"Graph neural network-enhanced auxiliary classifier generative adversarial network framework for robust intrusion detection","authors":"Tianjing Wang, Qi Liu, Hang Shen, Xiaokang Luo, Guangwei Bai","doi":"10.4218/etrij.2025-0152","DOIUrl":"https://doi.org/10.4218/etrij.2025-0152","url":null,"abstract":"To address the challenges of traffic diversity and data imbalance in network intrusion detection, we propose GraphACGAN, a novel detection framework that integrates auxiliary classifier generative adversarial networks (ACGANs) with graph neural networks (GNNs). In this architecture, the GNN is embedded in the ACGAN discriminator to exploit the latent graph structures inherent to network traffic, thereby improving the model's capacity to distinguish between benign and malicious behaviors. Simultaneously, the ACGAN generator was leveraged to synthesize minority-class attack traffic, effectively mitigating class imbalances and enhancing generalization. Comprehensive experiments conducted on three benchmark datasets, namely, NF-BoT-IoT-v2, NF-ToN-IoT-v2, and NF-UNSW-NB15-v2, demonstrate that GraphACGAN consistently outperforms all baselines, including E-GraphSAGE, GCN, ACGAN, LSTM, and KNN in terms of accuracy, precision, recall, and F1-score. An evaluation of resource-limited computing platforms further demonstrates GraphACGAN's inference efficiency for real-world deployment.","PeriodicalId":11901,"journal":{"name":"ETRI Journal","volume":"48 3","pages":"554-569"},"PeriodicalIF":2.0,"publicationDate":"2026-06-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.4218/etrij.2025-0152","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148269358","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0