IEEE Access最新文献

{"title":"Experimental Investigation of Ultra-Fast CDR Scheme for High-Speed Free-Space Optical Data Transmission With Multiple Access","authors":"Xinyu Wang;Meng Liu;Yuan Wan;Ampalavanapillai Nirmalathas;Christina Lim;Jianghao Li","doi":"10.1109/ACCESS.2025.3565527","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3565527","url":null,"abstract":"Time-division multiple access schemes such as time-slot coding (TSC) provide an effective and simple multi-user access framework for indoor free-space optical communications (FSO) with limited inter-user interference. For realistic multi-user scenarios employing TSC, clock data recovery is indispensable to achieve timing synchronization for different users, which lacks investigation in the previous study on TSC. In this paper, for the first time, we experimentally demonstrate a time-slot coded multi-user FSO system employing clock data recovery, in which a sum data rate of 40-Gb/s with bit-error-rate (BER) of each user around the 7% hard-decision forward error correction (HD-FEC) limit can be achieved. Moreover, to further improve the system performance, we propose a novel clock data recovery scheme for multi-user FSO systems employing TSC. Experimental results show that more than 3dB received optical power gain can be obtained to achieve the reference BER level of the 7% HD-FEC limit as compared to the system employing the conventional clock data recovery scheme. Higher-precision timing synchronization and ultra-fast convergence speed can also be achieved via our proposed clock data recovery scheme.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"78634-78640"},"PeriodicalIF":3.4,"publicationDate":"2025-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10979943","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143925040","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":"SV-SAE: Layer-Wise Pruning for Autoencoder Based on Link Contributions","authors":"Joohong Rheey;Hyunggon Park","doi":"10.1109/ACCESS.2025.3565296","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3565296","url":null,"abstract":"Autoencoders are a type of deep neural network and are widely used for unsupervised learning, particularly in tasks that require feature extraction and dimensionality reduction. While most research focuses on compressing input data, less attention has been given to reducing the size and complexity of the autoencoder model itself, which is crucial for deployment on resource-constrained edge devices. This paper introduces a layer-wise pruning algorithm specifically for multilayer perceptron-based autoencoders. The resulting pruned model is referred to as a Shapley Value-based Sparse AutoEncoder (SV-SAE). Using cooperative game theory, the proposed algorithm models the autoencoder as a coalition of interconnected units and links, where the Shapley value quantifies their individual contributions to overall performance. This enables the selective removal of less important components, achieving an optimal balance between sparsity and accuracy. Experimental results confirm that the SV-SAE reaches an accuracy of 99.25%, utilizing only 10% of the original links. Notably, the SV-SAE remains robust under high sparsity levels with minimal performance degradation, whereas other algorithms experience sharp declines as the pruning ratio increases. Designed for edge environments, the SV-SAE offers an interpretable framework for controlling layer-wise sparsity while preserving essential features in latent representations. The results highlight its potential for efficient deployment in resource-constrained scenarios, where model size and inference speed are critical factors.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"75666-75678"},"PeriodicalIF":3.4,"publicationDate":"2025-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10979845","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143925296","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":"Directed Fuzzy Edge Graphs Under q-ROF Environment: A Framework for Optimal Pathfinding","authors":"Nazia Nazir;Tanzeela Shaheen;Wajid Ali;Md Rafiul Hassan;Mohammad Mehedi Hassan","doi":"10.1109/ACCESS.2025.3565633","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3565633","url":null,"abstract":"This paper introduces a novel framework of Directed Edge q-Rung Orthopair Fuzzy Graphs (DEq-ROFGs), where graph vertices are crisp, and edges are characterized by q-rung orthopair fuzzy numbers (q-ROFNs). This structure captures the uncertainty in edge relationships while retaining deterministic node identities, making it ideal for applications in uncertain environments such as social networks, supply chains, healthcare systems, and recommendation systems. The paper defines foundational properties of DEq-ROFGs including subgraphs, completeness, and various degree-based metrics, and it establishes a proposition regarding the balance between in-degrees and out-degrees. The core contribution is a novel path-finding algorithm based on Hamacher operators and an improved score function, which identifies optimal paths between nodes under uncertainty. Unlike classical algorithms, it considers the suitability of a path, not just its length. Applied to an emergency road network scenario, the algorithm successfully determines the optimal route for service vehicles, and the choice between these routes can be made based on the score of the resulting path length. Comparative simulations show their effectiveness over traditional methods. Further analysis shows that increasing the q-value reduces both path score and length, and that Einstein operators yield higher destination scores than Hamacher and Dombi, confirming the model’s adaptability and robustness.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"81823-81834"},"PeriodicalIF":3.4,"publicationDate":"2025-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10979912","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144073043","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":"An Efficient DoA Estimation Approach for 2-D Planar Array Antennas via Axial Decomposition of Antenna Current Green’s Function","authors":"Jeong-Wan Lee;Sung-Jun Yang","doi":"10.1109/ACCESS.2025.3565641","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3565641","url":null,"abstract":"Array manifold, constructed from the received signals by incoming waves in antenna arrays, serves as a fundamental framework for characterizing electromagnetic behavior in direction finding systems. For densely arrayed 2-D planar antenna systems, strong inter-element mutual coupling distorts the array manifold, which directly degrading direction finding performance. While electromagnetic numerical techniques can be used to analyze mutual coupling effects in receiving antenna systems, large computational costs are required for 2-D array structures. This paper presents a fast reciprocal analysis method, based on a directional decomposition approach, for mutual coupling characterization of 2-D receiving antenna systems. The proposed method incorporates antenna current Green’s function theory to analyze the reciprocal property between transmit and receive modes of 2-D array antennas. Through directional decomposition, 2-D array problem is transformed into two separate 1-D array analyses, reducing computational complexity from <inline-formula> <tex-math>$mathcal {O}(M_{B}^{2}(N_{x}^{3} N_{y}^{3}))$ </tex-math></inline-formula> to <inline-formula> <tex-math>$mathcal {O}(M_{B}^{2}(N_{x}^{3} + N_{y}^{3}))$ </tex-math></inline-formula>. Building upon previous work that effectively characterized transmit-mode behavior, this study validates the directional decomposition approach in receiving modes based on reciprocity. The approach accurately predicts both receive-mode antenna current Green’s function and array manifolds while preserving mutual coupling and truncation effects. Validation is performed through direction finding scenarios. Validation through comparison with full-wave analysis demonstrates high correlation in array manifold components across all observation angles (<inline-formula> <tex-math>$theta : -63^{circ }$ </tex-math></inline-formula> to 63°, <inline-formula> <tex-math>$phi : -126^{circ }$ </tex-math></inline-formula> to 126°). In direction of arrival estimation applications with various multiple-source scenarios, the method achieves angular resolution comparable to conventional full-wave analysis while reducing computation time to 0.18% of the original requirement.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"76121-76134"},"PeriodicalIF":3.4,"publicationDate":"2025-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10980087","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143918655","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":"A Review of Advanced Deep Learning Methods of Multi-Target Segmentation for Breast Cancer WSIs","authors":"Qiaoyi Xu;Afzan Adam;Azizi Abdullah;Nurkhairul Bariyah","doi":"10.1109/ACCESS.2025.3565648","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3565648","url":null,"abstract":"Breast cancer is one of the most common cancers among women, with its heterogeneity posing significant challenges for diagnosis and treatment, profoundly impacting patient prognosis and quality of life. Whole Slide Imaging (WSI) in digital pathology provides high-resolution images that enable a comprehensive examination of the tumor microenvironment, offering advanced tools for breast cancer diagnosis and prognostic evaluation. However, manually reviewing whole slide images (WSIs) for tissue segmentation is time-consuming and prone to errors, highlighting the need for multi-target deep learning models to automate the segmentation of these complex structures. Multi-target segmentation offers distinct advantages by simultaneously processing multiple interrelated tissue regions within a single image, thereby enhancing accuracy and efficiency. Despite the potential of deep learning techniques in automating pathological analysis, their clinical adoption faces significant challenges. To address these, this paper proposes six criteria focused on clinical acceptability of deep learning methods: inherent limitations of WSIs, feature extraction, annotation requirements, efficiency, automated quantification, and interpretability. A rigorous review of publicly available datasets and deep learning methods identifies key challenges for clinical adoption. Following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, this review analyzes 29 core articles, highlighting the critical role of multi-target segmentation in breast cancer digital pathology while assessing the limitations of these techniques in clinical applications. Based on this analysis, this paper proposes six criteria to enhance the diagnostic performance of deep learning methods in multi-target segmentation for breast cancer digital pathology and to improve the clinical acceptability of deep learning methods.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"76016-76037"},"PeriodicalIF":3.4,"publicationDate":"2025-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10979932","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143918757","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":"Joint Encryption and Optimization for 6G MEC-Enabled IoT Networks","authors":"Manzoor Ahmed;Wali Ullah Khan;Fatma S. Alrayes;Yahia Said;Ali M. Al-Sharafi;Mi-Hye Kim;Khongorzul Dashdondov;Inam Ullah","doi":"10.1109/ACCESS.2025.3565415","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3565415","url":null,"abstract":"With the advent of advancements in future sixth-generation (6G) communication systems, Internet of Things (IoT) devices, characterized by their limited computational and communication capacities, have become integral in our lives. These devices are deployed extensively to gather vast amounts of data in real-time applications. However, their restricted battery life and computational resources present significant challenges in meeting the requirements of advanced communication systems. Mobile Edge Computing (MEC) has emerged as a promising solution to these challenges within the IoT realm in recent years. Despite its potential, securing MEC infrastructure in the context of IoT remains an open task. This study explores the operational dynamics of a secured IoT-enabled MEC infrastructure, focusing on providing real-time, on-demand, secure computational resources to low-powered IoT devices. It outlines a joint optimization problem to maximize computational throughput, minimize device energy consumption, reduce computational latency, and mitigate security overhead. An optimization algorithm is introduced to address these challenges by jointly allocating resources, thereby optimizing throughput, conserving energy, and meeting latency benchmarks through dynamic system adaptation. The effectiveness of the proposed model and algorithm is demonstrated through comparisons with relevant benchmark schemes, highlighting its efficiency in various scenarios. This work showcases the potential of advancements in encryption to deliver scalable security solutions with reduced resource consumption as the number of devices increases.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"79757-79770"},"PeriodicalIF":3.4,"publicationDate":"2025-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10979853","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143929815","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":"Short-Term Power Load Prediction Based on Level Processing Method and Improved GWO Algorithm","authors":"Yuntong Li","doi":"10.1109/ACCESS.2025.3565625","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3565625","url":null,"abstract":"In the field of short-term power load prediction, the current prediction methods have low prediction accuracy. To address this issue, this study introduces level processing method and improved grey wolf genetic algorithm to predict short-term power load and optimize the power load prediction accuracy. The genetic algorithm is applied to optimize the traditional grey wolf algorithm. Then, combined with the level set algorithm in the level processing algorithm, a genetic grey wolf hybrid model that integrates level processing is constructed. The variables in the load data are processed and analyzed through the level set algorithm. The final position of the population is determined based on the improved grey wolf genetic algorithm. Comparative experiments are conducted among the proposed model, the long short-term memory model, as well as the variational mode decomposition model. The average prediction accuracy remained within 0.652-0.859, significantly higher than the other two comparative models. The mean absolute error was 1.869, significantly lower than the other two models. The F1 score and accuracy were 0.891 and 90.32%, demonstrating that its predictive performance was significantly better than the other two models. Precision-recall curve, accuracy, mean absolute error, F1 score and other indicators are applied to evaluate the performance of the three models. The proposed model can accurately perform load prediction analysis in short-term power load prediction, and its prediction performance exceeds the other two prediction models. The prediction method can accurately predict short-term power load, providing useful references and inspirations for future researchers in power load prediction, and promoting the continuous development and progress of short-term power load prediction technology.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"78243-78256"},"PeriodicalIF":3.4,"publicationDate":"2025-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10979938","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143925103","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":"Safe Model Predictive Control of a Non-Holonomic Mobile Manipulator Under Multiple Constraints Using Control Barrier Functions","authors":"Zahra Kashi;Meysam Yadegar;Nader Meskin;Christos G. Cassandras","doi":"10.1109/ACCESS.2025.3565550","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3565550","url":null,"abstract":"This paper introduces a novel control strategy to ensure safety in the navigation of a mobile manipulator comprising a fixed-base manipulator mounted on a mobile platform. The approach initially addresses the trajectory tracking problem of a non-holonomic mobile manipulator (NH-MM) by employing decoupling dynamic control through model predictive control-optimizable control barrier function (MPC-OCBF). This allows for independent control of the end-effector and mobile platform trajectories, obstacle avoidance, and simultaneously adjusting the joint and control input limitations. The objective is to leverage the system’s redundancy to enable the mobile platform to effectively navigate feasible obstacle avoidance scenarios without affecting the primary task performance of the end-effector. Additionally, the method aims to perform obstacle avoidance when the redundancy of the manipulator is insufficient, addressing non-feasible obstacle avoidance scenarios. Through this approach, the mobile manipulator effectively avoids obstacles, allowing the end-effector to autonomously carry out its intended task. The effectiveness of the proposed method is validated through simulations and comparison with existing approaches. Additionally, a quantitative analysis is provided to evaluate and compare the performance of the controllers.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"76638-76653"},"PeriodicalIF":3.4,"publicationDate":"2025-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10979937","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143913347","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":"A Review on Research and Application of AI-Based Image Analysis in the Field of Computer Vision","authors":"Pan Wu;Xiaoqiang He;Wenhao Dai;Jingwei Zhou;Yutong Shang;Yourong Fan;Tao Hu","doi":"10.1109/ACCESS.2025.3565300","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3565300","url":null,"abstract":"The rapid development of artificial intelligence has significantly advanced the field of computer vision, particularly in image analysis and understanding. This paper provides a comprehensive review of the current state of the field, key technologies, and their applications across various real-world scenarios. It delves into the value of image analysis in critical areas such as personalized art, healthcare and medical image analysis, security monitoring and recognition technology, autonomous driving and traffic management, as well as industrial automation and quality control. This paper not only highlights the challenges and limitations, including dataset constraints, algorithm generalization, real-time computational costs, and privacy and ethical concerns, but also offers a forward-looking analysis of development trends such as interdisciplinary integration, weakly supervised and unsupervised learning, algorithm optimization and hardware advancements, and the protection of personal privacy and information. These insights provide a profound perspective on the future trajectory of computer vision and AI-driven image analysis.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"76684-76702"},"PeriodicalIF":3.4,"publicationDate":"2025-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10979847","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143913397","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":"An Interactive Framework for Personalized Navigation Based on Metacosmic Cultural Tourism and Large Model Fine-Tuning","authors":"Hang Guo;Zhiqiang Liu;Chuanqian Tang;Xiaodan Zhang","doi":"10.1109/ACCESS.2025.3565359","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3565359","url":null,"abstract":"With the wide application of large language models (LLMs) and the rapid growth of metaverse tourism demand, the digital tour and personalized interaction of historical sites have become the key to improving users’ digital travel experience. Creating an environment where users can access rich cultural information and enjoy personalized, immersive experiences is a crucial issue in the field of digital cultural travel. To this end, we propose a tourism information multimodal generation personalized question-answering interactive framework TIGMI (Tourism Information Generation and Multimodal Interaction) based on LLM fine-tuning, which aims to provide a richer and more in-depth experience for virtual tours of historical monuments. Taking Qutan Temple as an example, the framework combines LLM, retrieval augmented generation (RAG), and auto-prompting engineering techniques to retrieve accurate information related to the historical monument from external knowledge bases and seamlessly integrates it into the generated content. This integration mechanism ensures the accuracy and relevance of the generated answers. Through TIGMI’s LLM-driven command interaction mechanism in the 3D digital scene of Qutan Temple, users are able to interact with the building and scene environment in a personalized and real-time manner, successfully integrating historical and cultural information with modern digital technology. This integration significantly enhances the naturalness of interaction and personalizes the user experience, thereby improving user immersion and information acquisition efficiency. Evaluation results show that TIGMI excels in question-answering and multimodal interactions, significantly enhancing the depth and breadth of services provided by the personalized virtual tour. We conclude by addressing the limitations of TIGMI and briefly discuss how future research will focus on further improving the accuracy and user satisfaction of the generated content to adapt to the dynamically changing tourism environment.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"81450-81461"},"PeriodicalIF":3.4,"publicationDate":"2025-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10979851","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144099975","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}