IEEE Access最新文献

{"title":"Corrections to “Data-Driven Driving State Control for Unmanned Agricultural Logistics Vehicle”","authors":"Xuesheng Zhou;sJun Zhou","doi":"10.1109/ACCESS.2025.3570781","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3570781","url":null,"abstract":"Presents corrections to the paper, (Corrections to “Data-Driven Driving State Control for Unmanned Agricultural Logistics Vehicle”).","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"94224-94224"},"PeriodicalIF":3.4,"publicationDate":"2025-06-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11022718","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144205872","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}

{"title":"Monitoring 5G Backhaul: An In-Band Telemetry Approach for Quality of Service","authors":"Eurico Dias;Duarte Raposo;Miguel Luís;Pedro Rito;Susana Sargento","doi":"10.1109/ACCESS.2025.3565274","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3565274","url":null,"abstract":"The use of Ultra-Dense Networks within 5G communications can be leveraged by the adoption of Millimeter-Wave (mmWave) technology for the backhaul, resulting in cost reductions and faster deployment of the infrastructure. However, this shift also introduces new concerns and restrictions. The wireless link susceptibility to signal strength and link loss degradation, due to loss of line-of-sight, makes the link quality inconsistent. The heterogeneity of network nodes poses an additional challenge for tracking link quality changes reliably. An effective network monitoring system using 5G Quality of Service (QoS) Indicators is necessary to correctly characterize and track channel and flow quality conditions in a 5G wireless backhaul. To tackle these challenges, we introduce an In-Band Telemetry (INT) approach, consisting of a P4-compatible dataplane model and an aggregation agent capable of gathering and processing per-packet measurements, exposing them as link and QoS flow quality metrics, suitable for integration with Software Defined Network (SDN) environments and 5G networks. Our study compares the accuracy achieved by the proposed in-band solution to a commercial network management system, in an outdoor test-bed with an obstructed mmWave backhaul link. The results demonstrate that this approach exhibits minimal measurement errors when assessing the throughput, latency, and Packet Error Rate (PER) of mmWave links. The solution attains an average forwarding overhead of approximately 17%, while maintaining a per-node aggregation processing total time upper-bound of 45 ms at 2.5 Gbps line rate.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"77990-78006"},"PeriodicalIF":3.4,"publicationDate":"2025-04-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10981470","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143925037","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}

{"title":"Z Packed U-Cell Modular Multilevel Converter for STATCOM Applications","authors":"Sandy Atanalian;Fadia Sebaaly;Rawad Zgheib;Kamal Al-Haddad","doi":"10.1109/ACCESS.2025.3566015","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3566015","url":null,"abstract":"The Modular Multilevel Converter (MMC) is a promising topology for STATCOM applications due to its key features, such as modularity, scalability, and reduced harmonic content. Increasing the number of voltage levels in MMC reduces harmonics but simultaneously increases the number of submodules (SMs) per arm, leading to larger sizes and higher costs, which presents a challenge. To address this, this article introduces a novel 17-level MMC-STATCOM based on the Z Packed U-Cell (ZPUC) converter as its SM, which enables the generation of more voltage levels with fewer components and reduced harmonic content, offering significant advantages in terms of size and cost. Given the complex structure of the proposed converter and the associated challenges in building a physical prototype, real-time (RT) simulation using FPGA technology is employed for validation. The key contributions include integrating the ZPUC-SM into a three-phase STATCOM for the first time and adapting the converter model and its control system to RT tools, including RT-LAB with an electric hardware solver for FPGA execution. In addition, capacitor voltage balancing and energy sorting algorithms are integrated within Phase-Shift Pulse Width Modulation, eliminating the need for an additional controller while maintaining the floating capacitors of ZPUC-SMs balanced and regulated. The specifications of the proposed converter are defined, the mathematical model and control system are derived, and a real-time implementation based on CPU and FPGA execution is built to verify the scheme. The obtained RT simulation results provide practical evidence confirming the effective operation of the proposed scheme in VAR compensation mode.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"78795-78807"},"PeriodicalIF":3.4,"publicationDate":"2025-04-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10981426","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143925255","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}

{"title":"Automatic Endoscopic Navigation for Monocular Depth and Ego-Motion Estimation in Wireless Capsule Endoscopy Through Transformer Network","authors":"Junjun Huang;Tianran He;Juan Xu;Weiting Wu;Wei Wu","doi":"10.1109/ACCESS.2025.3565749","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3565749","url":null,"abstract":"Gastrointestinal (GI) cancers are among the most prevalent globally. Wireless capsule endoscopy (WCE), a minimally invasive technology, offers a promising alternative for diagnosing and treating GI diseases. Accurate depth estimation from WCE but remains challenging due to the complexity of the GI environment and limited datasets. In this paper, we propose an automatic endoscopic navigation system for monocular depth and ego-motion estimation in wireless capsule endoscopy (WCE) through a Transformer-based encoder-decoder network. Minimally invasive surgeries, including gastrointestinal (GI) procedures, face unique challenges such as restricted field of view, illumination variation, and texture sparsity, which complicate depth estimation and pose estimation tasks. Traditional Structure from Motion (SfM) and SLAM methods are often inadequate for GI scenes due to these inherent complexities. To address these issues, we introduce a novel self-supervised neural network framework that integrates a dual-attention mechanism within a modified ResNet. This model simultaneously predicts depth maps and ego-motion from monocular GI images, without requiring ground truth depth data. Our approach enhances feature extraction through spatial and channel-wise attention, allowing the network to capture both local and global contextual information. Furthermore, a multi-scale structural similarity index combined with L1 loss function is employed to improve the accuracy of depth estimation in challenging endoscopic environments. The model leverages a multi-interval frame sampling strategy to simulate diverse ego-motion scenarios, making it robust to low frame rate inputs typically seen in WCE. For ego-motion estimation on the ColonSim dataset, our model achieves an Absolute Trajectory Error (ATE) of 0.09 m at 30 FPS, outperforming the next-best model, SC-SfMLearner, by 44.4%. Additionally, for depth estimation, our model records an Absolute Relative Error (Abs Rel) of 0.33, a Squared Relative Error (Sq Rel) of 0.27, and a Root Mean Square Error (RMSE) of 0.94 on the EndoSLAM dataset.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"77931-77951"},"PeriodicalIF":3.4,"publicationDate":"2025-04-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10980270","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143925327","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}

{"title":"Trending Machine Learning Methods for Vehicle, Pedestrian, and Traffic for Detection and Tracking Task in the Post-Covid Era: A Literature Review","authors":"João Vitor de Andrade Porto;Péter Tamás Szemes;Hemerson Pistori;József Menyhárt","doi":"10.1109/ACCESS.2025.3565901","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3565901","url":null,"abstract":"This study, aimed at professionals in research and development in the fields of computer vision, artificial intelligence, and intelligent transportation, presents a systematic literature review on recent machine learning methodologies applied to the detection and tracking of vehicles, pedestrians, and traffic flow. The analysis of articles published between 2022 and 2025 (early access) in the post-COVID era explored the integration of machine learning and deep learning to address traffic challenges, allowing for the comparison of different approaches and the formulation of hypotheses based on the 46 articles that comprised the review corpus. Furthermore, the evaluation of the reported metrics revealed inconsistencies in the methodologies employed, attributed to the lack of standardization across the studies. In light of this, this work proposes alternatives for future experiments, emphasizing the emerging potential of the field through the adoption of new standardization systems and the exploration of experimental combinations.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"77790-77803"},"PeriodicalIF":3.4,"publicationDate":"2025-04-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10980333","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143925189","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}

{"title":"TEDVIL: Leveraging Transformer-Based Embeddings for Vulnerability Detection in Lifted Code","authors":"Gary A. McCully;John D. Hastings;Shengjie Xu","doi":"10.1109/ACCESS.2025.3565980","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3565980","url":null,"abstract":"Ransomware and other malware inflict devastating financial and operational damage on organizations worldwide by exploiting deeply embedded, hard-to-detect vulnerabilities in their systems. Detecting these vulnerabilities in compiled code before malicious actors exploit them remains a critical challenge in cybersecurity. This research introduces TEDVIL (Transformer-based Embeddings for Discovering Vulnerabilities in Lifted Code), a novel framework which uses transformer-based embeddings to train neural networks to detect vulnerabilities in lifted code. The framework was implemented using bidirectional (BERT and RoBERTa) and unidirectional (GPT-1 and GPT-2) transformer-based models to generate embeddings for training Long Short-Term Memory (LSTM) neural networks to detect stack-based buffer overflows in Low-Level Virtual Machine (LLVM) intermediate representation code. For comparison, simpler word2vec models (Skip-Gram and Continuous Bag of Words) were also trained, and their embeddings were used to train LSTMs. The results show that the LSTMs using GPT-2 embeddings outperformed those using GPT-1, BERT, RoBERTa, and word2vec embeddings, achieving a top accuracy of 92.5% and an F1-score of 89.7%. Notably, these results are achieved when the embedding model is trained with a dataset of just 48,000 functions, demonstrating effectiveness in resource-constrained settings. The findings underscore the effectiveness of TEDVIL in identifying hard-to-detect vulnerabilities in compiled code, and lay the groundwork for future research in leveraging transformer-based models for vulnerability detection.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"76894-76913"},"PeriodicalIF":3.4,"publicationDate":"2025-04-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10980253","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143913430","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}

{"title":"Adaptive Node-Oriented Data Placement for Heterogeneous Hadoop Clusters","authors":"Vishnu Prasad Verma;Santosh Kumar;Nenavath Srinivas Naik","doi":"10.1109/ACCESS.2025.3565759","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3565759","url":null,"abstract":"Modern society is experiencing a data explosion thanks to rapid IT development and the increasing intelligence of devices. The vast and complex data can be utilized to extract actionable insight using a big data processing framework. Hadoop is a popular big data processing framework on heterogeneous commodity hardware. While Hadoop offers a robust framework for large-scale, data-intensive tasks via its MapReduce paradigm, hardware heterogeneity across nodes often leads to straggler effects that degrade Hadoop cluster performance. This paper introduces the Adaptive Node-Oriented Data placement for Efficient Hadoop Execution (ANODE) method, which leverages historical job execution data to dynamically assess each node’s processing capability. By employing an agent-based mechanism, ANODE optimizes block allocation within the data node, alleviating imbalances caused by Hadoop’s default uniform placement strategy. Experimental results on a heterogeneous eleven-node Hadoop cluster demonstrate that ANODE reduces job completion times by up to 25%, significantly enhancing data locality and resource utilization compared to the default approach.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"79548-79559"},"PeriodicalIF":3.4,"publicationDate":"2025-04-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10980281","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143932277","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}

{"title":"YOLO-Air: An Efficient Deep Learning Network for Small Object Detection in Drone-Based Imagery","authors":"Jigang Qiu;Fangkai Cai;Ning Fu;Yuanfei Yao","doi":"10.1109/ACCESS.2025.3565560","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3565560","url":null,"abstract":"UAV imagery is widely used in areas like traffic safety, disaster rescue, and airspace management, due to its small size and low cost. However, it poses unique challenges for object detection due to small objects, complex backgrounds, and noise interference. To tackle these challenges, we propose YOLO-Air, a novel small object detection network designed specifically for UAV imagery. We propose SECAConv (Squeeze-Excitation Convolution with Attention), which enhances the feature representation of small objects through dynamic weight allocation and channel attention mechanisms. Additionally, we design the novel AeroFPN (Aerial Feature Pyramid Network) to optimize feature transmission by alleviating shallow feature loss through the inclusion of the xsmall detection head. Furthermore, we develop ASFM (Adaptive Scale Fusion Module), which suppresses background noise interference through effective multi-scale feature fusion and adaptive channel attention mechanisms, thereby improving the network’s ability to detect small objects. Experimental results demonstrate that YOLO-Air achieves significant accuracy improvements on both the VisDrone-DET2019 and AI-TOD datasets. Compared to the baseline YOLOv8n, YOLO-Air improved <inline-formula> <tex-math>$mAP_{50}$ </tex-math></inline-formula> from 41.2% to 44.5% on the VisDrone-DET2019 dataset, and from 44.9% to 47.5% on the AI-TOD dataset, while maintaining computational efficiency. These results validate YOLO-Air as an effective solution for small object detection in UAV aerial imagery.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"79718-79735"},"PeriodicalIF":3.4,"publicationDate":"2025-04-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10980347","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143929700","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}

{"title":"Adaptive Ruminant Optimization With LoRa-Based Communication for Formation Control of Multiple UAVs","authors":"Muhammad Aamir Khan;Zain Anwar Ali;Muhammad Haris Muneer;Raza Hasan","doi":"10.1109/ACCESS.2025.3565815","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3565815","url":null,"abstract":"In a dynamic environment with mountains and hazardous peaks, avoiding collisions and maintaining the desired formation is a crucial problem. This paper addresses this problem by presenting a novel formation control strategy of a cluster of UAVs in three different scenarios. The first scenario is designed to test the designed algorithm and hence contains no obstacles. The second scenario introduces some obstacles in the form of mountains to see whether the proposed algorithm can avoid the obstacles while maintaining the formation. In the last scenario, all the UAVs join together in one big cluster and have to avoid the obstacles while maintaining the formation. To design the environment for the scenarios, this study uses graph theory. To address the aforementioned scenarios, this paper offers a novel strategy by integrating a bio-inspired algorithm called the Adaptive Ruminant Optimization Algorithm (AROA) with the Long Range (LoRa) communication to achieve the formation control of multiple UAVs. Initially, AROA offers the best agents of each of the swarm. Then, the proposed method helps choose the best agent to be the leader for each of the swarm. The leader of each swarm finds the best trajectory for each swarm. LoRa-based networking technique is used for the connectivity between the UAVs. In addition, this study uses basis splines (B-splines) to smooth the planned trajectories of UAVs. Lastly, simulations demonstrate the better convergence and efficiency of the designed strategy by comparing it with classic algorithms. The simulations also show that the proposed method successfully maintains formation control in all three scenarios.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"80076-80087"},"PeriodicalIF":3.4,"publicationDate":"2025-04-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10980250","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143929701","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}

{"title":"Detection of Malicious Clients in Federated Learning Using Graph Neural Network","authors":"Anee Sharma;Ningrinla Marchang","doi":"10.1109/ACCESS.2025.3565712","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3565712","url":null,"abstract":"Federated Learning (FL) facilitates decentralized model training without the exchange of raw data, thereby guaranteeing privacy. However, due to its distributed nature, this paradigm is susceptible to adversarial threats such as sign-flipping attacks, in which malicious clients reverse model parameter signs in order to poison the global aggregation process. This study introduces a detection framework that is graph-based and leverages Graph Attention Networks (GATs) to overcome these challenges. The framework detects malicious clients with high accuracy by representing FL local models as directed graphs and capturing layer-wise statistical features. The efficacy of the approach is demonstrated by extensive experiments on the FEMNIST dataset, which simulate varying attacker percentages (15%, 35%) and attack probabilities (0.5, 0.7, 1.0). The GAT model obtains a 100% detection rate with zero false positives within an optimal threshold range of 0.5–0.9, as demonstrated by the results. Furthermore, isolating detected attackers during targeted rounds (20-60) substantially maintains FL global model performance, thereby mitigating the cascading effects of poisoned updates and ensuring system stability. This work offers a practicable, scalable, and robust solution to improve the security of FL systems against adversarial behaviors.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"77952-77972"},"PeriodicalIF":3.4,"publicationDate":"2025-04-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10980311","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143925138","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}