IEEE Access最新文献_第3页

Machine Learning Approach to Aerodynamic Analysis of NACA0005 Airfoil: ANN and CFD Integration NACA0005翼型气动分析的机器学习方法：ANN和CFD集成

IF 3.6 3区计算机科学

IEEE Access Pub Date : 2025-07-24 DOI: 10.1109/ACCESS.2025.3592338

Taiba Kouser;Dilek Funda Kurtulus;Srikanth Goli;Abdulrahman Aliyu;Imil Hamda Imran;Luai M. Alhems;Azhar M. Memon

{"title":"Machine Learning Approach to Aerodynamic Analysis of NACA0005 Airfoil: ANN and CFD Integration","authors":"Taiba Kouser;Dilek Funda Kurtulus;Srikanth Goli;Abdulrahman Aliyu;Imil Hamda Imran;Luai M. Alhems;Azhar M. Memon","doi":"10.1109/ACCESS.2025.3592338","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3592338","url":null,"abstract":"This study presents a machine learning approach to predict the unsteady aerodynamic performance of a NACA0005 airfoil. Data generated by computational fluid dynamics (CFD) is used to train the model for Reynolds numbers <inline-formula> <tex-math>$Re in [{1000-5000}]$ </tex-math></inline-formula> and angles of attack ranging from 9° to 11°. A robust Scaled Conjugate Gradient (SCG) algorithm is employed for efficient training of data. The ANN has a two-layer architecture, with 9 fixed neurons in the first hidden layer and a varying number of neurons in the second layer to achieve optimal performance. The model yielded coefficients of determination (<inline-formula> <tex-math>$R^{2}$ </tex-math></inline-formula>) of 0.994 (Coefficient of lift (<inline-formula> <tex-math>$C_{l}$ </tex-math></inline-formula>)) and 0.9615 (Coefficient of drag (<inline-formula> <tex-math>$C_{d}$ </tex-math></inline-formula>)) for training, and 0.9563 (<inline-formula> <tex-math>$C_{l}$ </tex-math></inline-formula>) and 0.9085 (<inline-formula> <tex-math>$C_{d}$ </tex-math></inline-formula>) for testing. Overall mean errors are found to be less than 1%. It offers a powerful surrogate modeling approach for aerodynamic studies at ultra-low Reynolds numbers. Moreover, it provides rapid and reliable alternatives to traditional CFD simulations in aerodynamic analysis for unseen cases.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"131088-131101"},"PeriodicalIF":3.6,"publicationDate":"2025-07-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11095683","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144725280","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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

Performance Enhancement of Dual-Input Frequency-Periodic Load Modulated Power Amplifier at 1–5.7-GHz Bandwidth With Co-Designed Biasing Network 协同设计偏置网络对1 - 5.7 ghz双输入频率周期负载调制功率放大器性能的增强

IF 3.6 3区计算机科学

IEEE Access Pub Date : 2025-07-24 DOI: 10.1109/ACCESS.2025.3592238

Takuma Torii;Yuji Komatsuzaki;Shintaro Shinjo;Ryo Ishikawa

{"title":"Performance Enhancement of Dual-Input Frequency-Periodic Load Modulated Power Amplifier at 1–5.7-GHz Bandwidth With Co-Designed Biasing Network","authors":"Takuma Torii;Yuji Komatsuzaki;Shintaro Shinjo;Ryo Ishikawa","doi":"10.1109/ACCESS.2025.3592238","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3592238","url":null,"abstract":"This study proposes a novel dual input power amplifier (PA) with a frequency-periodic load modulated output matching network supported by a broadband biasing network. The output matching network consists of two transmission lines which enable the dual-input PA to operate in Doherty or outphasing modes depending on the frequency. The proposed broadband biasing network simply consists of a short stub circuit that cooperates with the output matching network. The biasing network not only provides a bias to PA, but also improves the bandwidth in the back-off operation over the broadband characteristic of 145%. The two independent input signals are utilized to optimize the operation of Doherty and outphasing mode. The dual-input PA is implemented using a <inline-formula> <tex-math>$0.15~mu $ </tex-math></inline-formula> m GaN HEMT process. The fabricated PA shows a saturated output power of 35.2 dBm to 38.1 dBm with a power added efficiency (PAE) of 36.6% to 62% for the broadband 1 GHz to 5.7 GHz. The fabricated PA demonstrated an averaged output power of 27.7 dBm to 31.8 dBm, a PAE of 35% to 56.2% and an adjacent channel power Ratio (ACPR) of -41 to -55 dBc.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"131856-131868"},"PeriodicalIF":3.6,"publicationDate":"2025-07-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11095680","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144725153","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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

Model-Driven Architecture for Accelerating the Development of Tourism Applications 加速旅游应用开发的模型驱动架构

IF 3.6 3区计算机科学

IEEE Access Pub Date : 2025-07-24 DOI: 10.1109/ACCESS.2025.3592552

Lahbib Naimi;Lamya Benaddi;Charaf Ouaddi;Adnane Souha;Hamza Abdelmalek;Abdeslam Jakimi

{"title":"Model-Driven Architecture for Accelerating the Development of Tourism Applications","authors":"Lahbib Naimi;Lamya Benaddi;Charaf Ouaddi;Adnane Souha;Hamza Abdelmalek;Abdeslam Jakimi","doi":"10.1109/ACCESS.2025.3592552","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3592552","url":null,"abstract":"The tourism sector has witnessed a significant shift towards digital transformation, with mobile and web applications becoming essential tools for enhancing tourist experiences. Despite their potential, the development of these applications is often hindered by challenges such as platform heterogeneity, rapid adaptability, and integration complexity. This paper proposes a Model-Driven Architecture (MDA) approach to address these challenges and accelerate the development of tourism applications. The approach introduces the use of the 6As framework as a domain model, alongside systematic model transformations, to streamline the development process. A detailed case study demonstrates the transformation of a class diagram into a functional backend, showcasing the advantages of automation, scalability, and maintainability. Validation results reveal significant improvements in development time compared to traditional manual methods. This contribution highlights the potential of MDA to simplify the development of complex systems while ensuring quality and adaptability. Future work aims to enhance the methodology by incorporating additional architectural patterns, expanding support for diverse platforms, and integrating advanced validation techniques.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"131696-131715"},"PeriodicalIF":3.6,"publicationDate":"2025-07-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11095692","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144725299","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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

Entity Pair Relation Classification Based on Contrastive Learning and Biaffine Model 基于对比学习和Biaffine模型的实体对关系分类

IF 3.6 3区计算机科学

IEEE Access Pub Date : 2025-07-24 DOI: 10.1109/ACCESS.2025.3592203

Songhua Hu;Ziming Zhang;Hengxin Wang;Lihui Jiang

{"title":"Entity Pair Relation Classification Based on Contrastive Learning and Biaffine Model","authors":"Songhua Hu;Ziming Zhang;Hengxin Wang;Lihui Jiang","doi":"10.1109/ACCESS.2025.3592203","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3592203","url":null,"abstract":"In natural language processing, the biaffine model can effectively captures sentence structure and word relationships for tasks like text classification and relation extraction. However, it struggles with entity pair relation classification, particularly in overlapping or complex scenarios. To address this, this paper proposes BERT-CL-Biaffine, an improved relation classification model integrating bidirectional entity contrastive learning and a global pointer network. The model enhances the biaffine architecture by training it to identify entity boundaries and leveraging contrastive learning to strengthen semantic associations between overlapping entity pairs. Experiments on the NYT and WebNLG datasets demonstrate that BERT-CL-Biaffine outperforms baseline models, achieving F1 score improvements of 1% and 1.2%, respectively. The model excels in classifying overlapping entity pairs and handles challenges like imbalanced relation types and ambiguous entity features, particularly in complex scenarios. The results validate that bidirectional entity contrastive learning and global pointer networks significantly enhance the biaffine model’s feature representation and classification performance. This approach offers a robust solution for relation extraction in intricate textual contexts.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"131289-131302"},"PeriodicalIF":3.6,"publicationDate":"2025-07-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11095676","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144725281","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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

Federated Mental Wellbeing Assessment Using Smartphone Sensors Under Unreliable Participation 在不可靠参与下使用智能手机传感器的联邦心理健康评估

IF 3.6 3区计算机科学

IEEE Access Pub Date : 2025-07-24 DOI: 10.1109/ACCESS.2025.3591310

Gavryel Martis;Ryan McConville

{"title":"Federated Mental Wellbeing Assessment Using Smartphone Sensors Under Unreliable Participation","authors":"Gavryel Martis;Ryan McConville","doi":"10.1109/ACCESS.2025.3591310","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3591310","url":null,"abstract":"Today’s smartphones are equipped with sensors that can track and collect data about users’ everyday activities, which can then be transformed into behavioural indicators of users’ health and wellbeing. Prior studies were focused on centralised machine learning techniques, which transfers all the data to a central server. With modern smartphones being powerful enough to process data locally on a user’s device, federated learning (FL) has emerged as a promising alternative that addresses privacy concerns inherent in centralised setups. This study explores the feasibility of FL models to predict mental wellbeing in a decentralised setting. We also closely evaluate how FL can be applied in such applications in the wild, i.e., where user participation may be inconsistent due to device limitations or privacy concerns inherent in mental health monitoring. To further alleviate demands on edge clients, we incorporate federated continual learning, allowing for adaptive, timely model updates that enhance robustness in real-world mental health applications. In our experiments, we trained tree-based, fully-connected and recurrent neural networks, comparing each time with the centralised approach and random baselines. We also assess the model’s ability to generalise across different users and adapt to temporal changes, ensuring reliability across diverse real-world contexts. The findings suggested that given the widespread use of such devices, FL holds great potential in mood and depression detection while protecting data privacy. Our continual FL achieves similar performance to standard FL, but with added benefit of faster model updates.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"131042-131052"},"PeriodicalIF":3.6,"publicationDate":"2025-07-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11096109","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144725285","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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

MT-EfficientNetV2: A Multi-Temporal Scale Fusion EEG Emotion Recognition Method Based on Recurrence Plots mt - effentnetv2：一种基于递归图的多时间尺度脑电情感识别方法

IF 3.6 3区计算机科学

IEEE Access Pub Date : 2025-07-24 DOI: 10.1109/ACCESS.2025.3592336

Zihan Zhang;Zhiyong Zhou;Jun Wang;Hao Hu;Jing Zhao

{"title":"MT-EfficientNetV2: A Multi-Temporal Scale Fusion EEG Emotion Recognition Method Based on Recurrence Plots","authors":"Zihan Zhang;Zhiyong Zhou;Jun Wang;Hao Hu;Jing Zhao","doi":"10.1109/ACCESS.2025.3592336","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3592336","url":null,"abstract":"Emotion recognition based on electroencephalography (EEG) signals has garnered significant research attention in recent years due to its potential applications in affective computing and brain-computer interfaces. Despite the proposal of various deep learning-based methods for extracting emotional features from EEG signals, most existing models struggle to effectively capture both long-term and short-term dependencies within the signals, failing to fully integrate features across different temporal scales. To address these challenges, we propose a deep learning model that combines multi-temporal-scale fusion, termed MT-EfficientNetV2. This model segments one-dimensional EEG signals using combinations of varying window sizes and fixed step lengths. The Recursive Plot (RP) algorithm is then employed to transform these segments into RGB images that intuitively represent the dynamic characteristics of the signals, facilitating the capture of complex emotional features. Additionally, a three-branch input feature fusion module has been designed to effectively integrate features across different scales within the same temporal domain. The model architecture incorporates DEconv and the SimAM attention mechanism with EfficientNetV2. This integration enhances the global fusion and expression of multi-scale features while strengthening the extraction of key emotional features at the local level, thereby suppressing redundant information. Experiments conducted on the public datasets SEED and SEED-IV yielded accuracies of 98.67% and 96.89%, respectively, surpassing current mainstream methods and validating the feasibility and effectiveness of the proposed approach.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"132079-132096"},"PeriodicalIF":3.6,"publicationDate":"2025-07-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11095664","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144750990","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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

An Optical and mm-Wave Converged, Dual-Band, Multi-Beam Rotman Lens Antenna Array System Enabling Simplified Designs of B5G/mmW Base Stations for Ultra Dense Wireless Networks 一种光学和毫米波融合、双频、多波束罗特曼透镜天线阵列系统，可简化超密集无线网络B5G/毫米波基站的设计

IF 3.6 3区计算机科学

IEEE Access Pub Date : 2025-07-24 DOI: 10.1109/ACCESS.2025.3592213

Lauryn P. Smith;Charles A. Lynch;C. Alex Kaylor;L. Alberto Campos;Lin Cheng;Stephen E. Ralph;Manos M. Tentzeris

{"title":"An Optical and mm-Wave Converged, Dual-Band, Multi-Beam Rotman Lens Antenna Array System Enabling Simplified Designs of B5G/mmW Base Stations for Ultra Dense Wireless Networks","authors":"Lauryn P. Smith;Charles A. Lynch;C. Alex Kaylor;L. Alberto Campos;Lin Cheng;Stephen E. Ralph;Manos M. Tentzeris","doi":"10.1109/ACCESS.2025.3592213","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3592213","url":null,"abstract":"With the exponential demand for additional capacity, ultra-dense wireless networks (UDNs) have become an interesting area of study. One of the major challenges in UDN deployment is the complexity and cost of the base stations, which affect scalability. This study proposes an integrated system combining optical and millimeter-wave (mm-Wave/mmW) technologies to address these challenges by centralizing processing tasks and simplifying the base station. The need for bulky, power-intensive components at each base station is significantly reduced, by transmitting signals over fiber from the central location to the base stations. This architecture opens the door to centralized artificial intelligence-based control of the base station. A dual-band Rotman lens antenna array is integrated into the system to provide flexible, passive beamforming capabilities, supporting multiple frequencies and multiple beams in a compact form, further reducing the overall number of devices required at the base station. The multi-layer Rotman lens antenna achieves a total -3-dB realized gain coverage of ±42° at both 28 GHz and 39 GHz. The maximum realized gain is 12.6 dBi and 12.9 dBi, at 28 GHz and 39 GHz respectively. To demonstrate the capabilities of the proposed optical and mm-Wave converged Rotman-lens enabled simplified base station architectures, a proof-of-concept experiment is performed integrating optical multi-carrier generation, optical modulation, fiber transmission, optical-to-electrical conversion and transmission through the presented dual-band, Rotman lens antenna array. The results demonstrate BER below the hard-decision FEC threshold, EVM meeting IEEE standard requirements, and open eye diagrams, confirming acceptable performance of the proposed architecture for simplified UDN base stations.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"132067-132078"},"PeriodicalIF":3.6,"publicationDate":"2025-07-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11095717","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144751009","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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

Compression-Aware Hybrid Framework for Deep Fake Detection in Low-Quality Video 基于压缩感知的低质量视频深度假检测混合框架

IF 3.6 3区计算机科学

IEEE Access Pub Date : 2025-07-24 DOI: 10.1109/ACCESS.2025.3592358

Lagsoun Abdel Motalib;Oujaoura Mustapha;Hedabou Mustapha

{"title":"Compression-Aware Hybrid Framework for Deep Fake Detection in Low-Quality Video","authors":"Lagsoun Abdel Motalib;Oujaoura Mustapha;Hedabou Mustapha","doi":"10.1109/ACCESS.2025.3592358","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3592358","url":null,"abstract":"Deep fakes pose a growing threat to digital media integrity by generating highly realistic fake videos that are difficult to detect, especially under the high compression levels commonly used on social media platforms. These compression artifacts often degrade the performance of deep fake detectors, making reliable detection even more challenging. In this paper, we propose a handcrafted deep fake detection framework that integrates wavelet transforms and Conv3D-based spatiotemporal descriptors for feature extraction, followed by a lightweight ResNet-inspired classifier. Unlike end-to-end deep neural networks, our method emphasizes interpretability and computational efficiency, while maintaining high detection accuracy under diverse real-world conditions. We evaluated four configurations based on input modality and attention mechanisms: RGB with attention, RGB without attention, grayscale with attention, and grayscale without attention. Experiments were conducted on the FaceForensics++ dataset (C23 and C40 compression levels) and Celeb-DF v2 (C0 and C40), across intra- and inter-compression settings, as well as cross-dataset scenarios. Results show that RGB inputs without attention achieve the highest accuracy on FaceForensics++, while grayscale inputs without attention perform best in cross-dataset evaluations on Celeb-DF v2, attaining strong AUC scores. Despite its handcrafted nature, our approach matches or surpasses the existing state-of-the-art (SOTA) methods. Grad-CAM visualizations further reveal both strengths and failures (e.g., occlusion and misalignment), offering valuable insights for refinement. These findings underscore the potential of our framework for efficient and effective deep fake detection in low-resource and real-time environments.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"131980-131997"},"PeriodicalIF":3.6,"publicationDate":"2025-07-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11095666","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144750984","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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

Metering Automation System 3.0 Base Version Based on Machine Learning 基于机器学习的计量自动化系统3.0基础版

IF 3.6 3区计算机科学

IEEE Access Pub Date : 2025-07-24 DOI: 10.1109/ACCESS.2025.3592384

Sheng Li;Leping Zhang;Hang Dai;Lukun Zeng;Yuan Ai;Shuang Qi;Yuanzhai Cui

{"title":"Metering Automation System 3.0 Base Version Based on Machine Learning","authors":"Sheng Li;Leping Zhang;Hang Dai;Lukun Zeng;Yuan Ai;Shuang Qi;Yuanzhai Cui","doi":"10.1109/ACCESS.2025.3592384","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3592384","url":null,"abstract":"The accurate identification of equipment base versions in Metering Automation System 3.0 (MAS 3.0) is critical for ensuring interoperability and maintenance efficiency in modern smart grids. However, traditional machine learning methods and standalone deep learning architectures struggle to balance spatiotemporal feature extraction, computational efficiency, and deployment constraints for high-frequency multivariate metering data. This study proposes a hybrid DSCNN-CBAM-BiLSTM framework that synergistically integrates depthwise separable convolutions, dual attention mechanisms, and bidirectional temporal modeling to address these challenges. The depthwise separable convolutional neural network (DSCNN) minimizes parameter overhead while capturing spatial correlations across distributed grid nodes, followed by convolutional block attention modules (CBAM) that dynamically recalibrate channel and spatial features to amplify discriminative patterns. Bidirectional LSTM (BiLSTM) layers then model long-range temporal dependencies in both forward and backward directions, enabling robust contextual analysis of energy consumption sequences. Validated on 14 TB of operational data from China Southern Power Grid, the framework achieves 96.7% classification accuracy with an inference latency of 8.9 ms—outperforming CNNs (89.2%), Transformers (90.5%), and GRUs (92.1%) while reducing GPU memory usage by 35.7–72.7%. Edge deployment tests on NVIDIA Jetson AGX Xavier demonstrate real-time compatibility with IEC 61850-7-420 protocols, maintaining <15 ms latency at 200-node resolution. These advancements establish a highly effective and resource-efficient framework. For resource-efficient, edge-deployable analytics in smart grid infrastructure, effectively bridging the gap between high-accuracy version identification and industrial computational constraints.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"132097-132108"},"PeriodicalIF":3.6,"publicationDate":"2025-07-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11095682","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144751007","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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

Gradient Flow Decoding 梯度流解码

IF 3.6 3区计算机科学

IEEE Access Pub Date : 2025-07-24 DOI: 10.1109/ACCESS.2025.3592296

Tadashi Wadayama;Lantian Wei

{"title":"Gradient Flow Decoding","authors":"Tadashi Wadayama;Lantian Wei","doi":"10.1109/ACCESS.2025.3592296","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3592296","url":null,"abstract":"This paper presents the Gradient Flow (GF) decoding for LDPC codes. GF decoding, a continuous-time methodology based on gradient flow, employs a potential energy function associated with bipolar codewords of LDPC codes. The decoding process of the GF decoding is concisely defined by an ordinary differential equation and thus it is well suited to an analog circuit implementation. We experimentally demonstrate that the decoding performance of the GF decoding for AWGN channels is comparable to that of the multi-bit mode gradient descent bit flipping algorithm. We further introduce the negative log-likelihood function of the channel for generalizing the GF decoding. The proposed method is shown to be tensor-computable, which means that the gradient of the objective function can be evaluated with the combination of basic tensor computations. This characteristic is well-suited to emerging AI accelerators, potentially applicable in wireless signal processing. The paper assesses the decoding performance of the generalized GF decoding in LDPC-coded MIMO channels. For LDPC-coded MIMO channels, our method achieves approximately 1.6 dB performance gain over MMSE + BP. Furthermore, an exploration of score-based channel learning for capturing statistical properties is also provided.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"131937-131956"},"PeriodicalIF":3.6,"publicationDate":"2025-07-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11095675","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144725275","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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