IEEE Transactions on Network Science and Engineering最新文献_第10页

A Cooperative Kernel-Based Method for Task Offloading in Vehicular Edge Computing 基于协同核的车辆边缘计算任务卸载方法

IF 7.9 2区计算机科学

IEEE Transactions on Network Science and Engineering Pub Date : 2025-03-05 DOI: 10.1109/TNSE.2025.3566800

Kangli Zhao;Penglin Dai;Huanlai Xing;Xiao Wu

{"title":"A Cooperative Kernel-Based Method for Task Offloading in Vehicular Edge Computing","authors":"Kangli Zhao;Penglin Dai;Huanlai Xing;Xiao Wu","doi":"10.1109/TNSE.2025.3566800","DOIUrl":"https://doi.org/10.1109/TNSE.2025.3566800","url":null,"abstract":"Vehicular Edge Computing (VEC) has emerged as a promising paradigm for supporting real-time applications by deploying communication and computation resources at the network edge. However, existing task offloading strategies often suffer from performance degradation within VEC systems, where offloading outcomes are difficult to predict accurately due to fluctuating environmental parameters that are challenging to obtain in real-time. Moreover, current offloading methods rely on learning-based approaches, necessitating extensive training efforts to adapt to diverse vehicular applications. Accordingly, we investigate the Cooperative Task Offloading (CTO) problem by considering dynamic nature of vehicular environment, which aims to minimize overall task completion time. We reformulate CTO as a cooperative contextual multi-armed bandit problem and propose a Cooperative Kernel-based Server Selection (CK-SS) algorithm, which facilitates offloading decisions by enabling online reward estimation through information sharing among vehicles. Specifically, we develop a composite kernel function that captures both task characteristics and temporal correlations among historical context-action pairs and an efficient rule for online parameter update. The reward under a given context is estimated based on Gaussian Process and the action is determined using Upper Confidence Bound (UCB) policy. Finally, we implement a simulation model and comprehensive simulation results demonstrate the effectiveness of the CK-SS across various scenarios.","PeriodicalId":54229,"journal":{"name":"IEEE Transactions on Network Science and Engineering","volume":"12 5","pages":"3919-3932"},"PeriodicalIF":7.9,"publicationDate":"2025-03-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144891133","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Subgraph Structure Feature Learning for Triangle Clique Prediction in Complex Networks 复杂网络中三角团预测的子图结构特征学习

IF 7.9 2区计算机科学

IEEE Transactions on Network Science and Engineering Pub Date : 2025-03-05 DOI: 10.1109/TNSE.2025.3566227

Yabing Yao;Zhiheng Mao;Yangyang He;Zhipeng Xu;Ziyu Ti;Pingxia Guo;Fuzhong Nian;Ning Ma

{"title":"Subgraph Structure Feature Learning for Triangle Clique Prediction in Complex Networks","authors":"Yabing Yao;Zhiheng Mao;Yangyang He;Zhipeng Xu;Ziyu Ti;Pingxia Guo;Fuzhong Nian;Ning Ma","doi":"10.1109/TNSE.2025.3566227","DOIUrl":"https://doi.org/10.1109/TNSE.2025.3566227","url":null,"abstract":"Link prediction is a critical task in network analysis, widely used to infer potential relationships between nodes. While traditional methods focus on pairwise interactions, real-world networks often exhibit higher-order interactions involving multiple nodes, such as 3-clique (triangle), which play a crucial role in understanding tightly-knit groups and complex network dynamics. In this paper, we propose a triangle clique prediction method based on <underline>Subgraph <underline>Structure <underline>Feature <underline>Learning (SSFL), which focuses on triangle structures in a network for prediction. In detail, it extracts the one-hop neighborhood around a target 3-clique, encodes it as a enclosing subgraph, and represents its structural features as a vector. These feature vectors are then processed using a fully connected neural network to predict 3-clique formations effectively. Experimental results show that the proposed method outperforms similarity-based link prediction methods and demonstrates comparable performance to embedding-based and machine learning-based approaches across various datasets. Our work can not only directly predict 3-clique structures in a network, but also provides insights into better understanding the evolution mechanism of networks.","PeriodicalId":54229,"journal":{"name":"IEEE Transactions on Network Science and Engineering","volume":"12 5","pages":"3888-3900"},"PeriodicalIF":7.9,"publicationDate":"2025-03-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144891254","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

PCA: Semi-Supervised Segmentation With Patch Confidence Adversarial Training PCA：基于Patch置信度对抗训练的半监督分割

IF 6.7 2区计算机科学

IEEE Transactions on Network Science and Engineering Pub Date : 2025-03-05 DOI: 10.1109/TNSE.2025.3548416

Zhenghua Xu;Runhe Yang;Zihang Xu;Shuo Zhang;Yuchen Yang;Weipeng Liu;Weichao Xu;Junyang Chen;Thomas Lukasiewicz;Victor C. M. Leung

{"title":"PCA: Semi-Supervised Segmentation With Patch Confidence Adversarial Training","authors":"Zhenghua Xu;Runhe Yang;Zihang Xu;Shuo Zhang;Yuchen Yang;Weipeng Liu;Weichao Xu;Junyang Chen;Thomas Lukasiewicz;Victor C. M. Leung","doi":"10.1109/TNSE.2025.3548416","DOIUrl":"https://doi.org/10.1109/TNSE.2025.3548416","url":null,"abstract":"Deep-learning-based semi-supervised learning (SSL) methods have achieved a strong performance in medical image segmentation, which can alleviate doctors' expensive annotation by utilizing a large amount of unlabeled data. Unlike most existing semi-supervised learning methods, adversarial training methods distinguish samples from different sources by learning the data distribution of the segmentation map, leading the segmenter to generate more accurate predictions. We argue that the current performance restrictions for such approaches are the problems of feature extraction and learning preferences. In this article, we propose a new semi-supervised adversarial method called Patch Confidence Adversarial Training (PCA) for medical image segmentation. The PCA method's discriminator penalizes patch-level structures, guiding the generator to optimize different patch areas, by leveraging pixel context, the generator is driven to focus on high-frequency features, making it harder to deceive the discriminator and easy to converge to an ideal state, which more effectively guides the segmenter to generate high-quality pseudo-labels. Furthermore, at the discriminator's input, we supplement image information constraints, making it simpler to fit the expected data distribution. Extensive experiments on the Automated Cardiac Diagnosis Challenge (ACDC) 2017 dataset and the Brain Tumor Segmentation (BraTS) 2019 challenge dataset show that our method outperforms the state-of-the-art semi-supervised methods, which demonstrates its effectiveness for medical image segmentation.","PeriodicalId":54229,"journal":{"name":"IEEE Transactions on Network Science and Engineering","volume":"12 4","pages":"2473-2486"},"PeriodicalIF":6.7,"publicationDate":"2025-03-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144492255","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

ProxaDyn: A Proximity-Aware Dynamic Caching Approach for Named Data Networks ProxaDyn：命名数据网络的邻近感知动态缓存方法

IF 6.7 2区计算机科学

IEEE Transactions on Network Science and Engineering Pub Date : 2025-03-04 DOI: 10.1109/TNSE.2025.3547424

Matta Krishna Kumari;Nikhil Tripathi;Piyush Joshi

{"title":"ProxaDyn: A Proximity-Aware Dynamic Caching Approach for Named Data Networks","authors":"Matta Krishna Kumari;Nikhil Tripathi;Piyush Joshi","doi":"10.1109/TNSE.2025.3547424","DOIUrl":"https://doi.org/10.1109/TNSE.2025.3547424","url":null,"abstract":"Named Data Network (NDN), a future Internet architecture is introduced to address the shortcomings of the current Internet architecture. NDN supports in-network caching to facilitate scalable content distribution and enhance overall network performance. However, the known NDN caching strategies suffer from a few common drawbacks, such as inefficient cache utilization, high content redundancy, and overhead due to lookup repetition. To address these issues, in this paper, we propose a novel caching strategy called ProxaDyn for efficient content lookup, placement, and replacement. During the content lookup phase, ProxaDyn interacts exclusively with the router responsible for caching a particular content. This eliminates interaction with other intermediate routers, thereby significantly reducing content access latency. For content placement, ProxaDyn strategically selects an on-path router based on content popularity. Popular content is placed in the cache of a router closer to the consumer, while less popular content is cached in a router away from the consumer. This approach significantly improves the cache hits and reduces the access latency. We test ProxaDyn over a diverse range of real-world network topologies. Using extensive experiments, we show that ProxaDyn could achieve significantly better results compared to the state-of-the-art NDN caching strategies.","PeriodicalId":54229,"journal":{"name":"IEEE Transactions on Network Science and Engineering","volume":"12 3","pages":"2360-2372"},"PeriodicalIF":6.7,"publicationDate":"2025-03-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143870988","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Blockchain Assisted Industrial Data Registration and Reconstruction Management Scheme 区块链辅助工业数据登记和重建管理方案

IF 6.7 2区计算机科学

IEEE Transactions on Network Science and Engineering Pub Date : 2025-03-04 DOI: 10.1109/TNSE.2025.3547409

Zewei Liu;Chunqiang Hu;Ruifeng Zhao;Pengfei Hu;Arwa Alrawais;Tao Xiang

{"title":"Blockchain Assisted Industrial Data Registration and Reconstruction Management Scheme","authors":"Zewei Liu;Chunqiang Hu;Ruifeng Zhao;Pengfei Hu;Arwa Alrawais;Tao Xiang","doi":"10.1109/TNSE.2025.3547409","DOIUrl":"https://doi.org/10.1109/TNSE.2025.3547409","url":null,"abstract":"As a typical Industrial Internet of Things (IIOT) application, three-dimensional point cloud reconstruction brings us benefits and convenience. The reconstructed mathematical models can be employed to facilitate precise quality control, which is important for the usage of the reconstructed products. Conversely, traditional reconstruction methods are characterized by inefficiency, and the errors inherent in each phase of the reconstruction chain often remain opaque and vulnerable to tampering. Hence, we propose a blockchain assisted industrial data registration and reconstruction management scheme (BIRMS). First, the tamper-proof and distributed storage characteristics of blockchain are fully utilized to ensure the authenticity and transparency of output errors throughout the reconstruction process. It is worth noting that smart contracts are designed to facilitate the management and query of on-chain data. Then, a novel swarm intelligence algorithm called EGWODA is designed to handle the issue which is low efficiency in the registration step of reconstruction. Finally, simulation results indicate the feasibility and efficiency of the BIRMS.","PeriodicalId":54229,"journal":{"name":"IEEE Transactions on Network Science and Engineering","volume":"12 3","pages":"2345-2359"},"PeriodicalIF":6.7,"publicationDate":"2025-03-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143870841","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A Hybrid EKF/WUFIR Filter for Indoor Localization Integrating INS and UWB Data 结合INS和UWB数据的室内定位混合EKF/WUFIR滤波器

IF 6.7 2区计算机科学

IEEE Transactions on Network Science and Engineering Pub Date : 2025-03-03 DOI: 10.1109/TNSE.2025.3546918

Long Cheng;Jiahe Song;Wenhao Zhao

{"title":"A Hybrid EKF/WUFIR Filter for Indoor Localization Integrating INS and UWB Data","authors":"Long Cheng;Jiahe Song;Wenhao Zhao","doi":"10.1109/TNSE.2025.3546918","DOIUrl":"https://doi.org/10.1109/TNSE.2025.3546918","url":null,"abstract":"Due to the complex and variable indoor environment, ultra-wideband (UWB) signal transmission is often obstructed by walls and obstacles, resulting in non-line-of-sight (NLOS), which reduces localization accuracy. Inertial navigation system (INS) is an autonomous navigation system that does not rely on external information and is not affected by NLOS. Therefore, a hybrid EKF/WUFIR filter indoor localization algorithm that integrates INS and UWB data is proposed. The proposed algorithm is composed of three parts: INS localization, UWB localization and data fusion. In the INS localization part, the motion model is used to determine the state of the target in real time using measurement data obtained from the inertial measurement unit (IMU). In the UWB localization part, a resettable residual weighted particle filter algorithm is proposed to mitigate the effect of NLOS on the localization results. In the data fusion part, a hybrid filtering algorithm combining extended Kalman filter (EKF) and weighted unbiased finite impulse response (WUFIR) filtering is proposed to fuse the INS and UWB localization data. Simulation and experimental results show that the proposed algorithm outperforms other comparative algorithms in terms of robustness and localization accuracy.","PeriodicalId":54229,"journal":{"name":"IEEE Transactions on Network Science and Engineering","volume":"12 3","pages":"2266-2276"},"PeriodicalIF":6.7,"publicationDate":"2025-03-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143871070","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Achievable Rate Optimization of RIS-Assisted Multi-Antenna FD DF Relay Cooperation System With SWIPT Technology 基于SWIPT技术的ris辅助多天线FD - DF中继协同系统可实现速率优化

IF 6.7 2区计算机科学

IEEE Transactions on Network Science and Engineering Pub Date : 2025-03-03 DOI: 10.1109/TNSE.2025.3546759

Shunwai Zhang;Qingzhu Ma;Hao Cheng;Rongfang Song

{"title":"Achievable Rate Optimization of RIS-Assisted Multi-Antenna FD DF Relay Cooperation System With SWIPT Technology","authors":"Shunwai Zhang;Qingzhu Ma;Hao Cheng;Rongfang Song","doi":"10.1109/TNSE.2025.3546759","DOIUrl":"https://doi.org/10.1109/TNSE.2025.3546759","url":null,"abstract":"To pursue higher achievable rate and wider coverage transmission in wireless communications, this paper proposes a novel reconfigurable intelligent surfaces (RIS)-assisted multi-antenna full-duplex (FD) decode-and-forward (DF) relay cooperation system with simultaneous wireless information and power transfer (SWIPT) technology, which can fully enjoy the advantages of both RIS and SWIPT-based FD DF relay with multiple antennas. In order to maximize the achievable rate of the proposed system, the phase shifts of RIS, the precoding vector and the power splitting factor are jointly optimized. At first, optimal phase shifts of RIS are achieved via aligning the phases of received signals at the destination. Subsequently, the alternating optimization (AO)-based algorithm is adopted to decompose the original optimization problem into two sub-problems, i.e., the precoding vector optimization and the power splitting factor optimization. The sub-problems are still complicated and nonconvex, and the successive convex approximation (SCA) method is applied to reformulate them into convex problems which can be further solved by iterative method. Simulation results illustrate the advantages of the proposed system and reveal the effects of various factors on its performance. Simulation results also demonstrate the superiorities of the joint optimization algorithm compared with its counterparts.","PeriodicalId":54229,"journal":{"name":"IEEE Transactions on Network Science and Engineering","volume":"12 3","pages":"2243-2253"},"PeriodicalIF":6.7,"publicationDate":"2025-03-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143871036","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Enhancing Graph Structure Learning via Motif-Driven Hypergraph Construction 基于基元驱动的超图构建增强图结构学习

IF 6.7 2区计算机科学

IEEE Transactions on Network Science and Engineering Pub Date : 2025-03-03 DOI: 10.1109/TNSE.2025.3547349

Jia-Le Zhao;Xian-Jie Zhang;Xiao Ding;Xingyi Zhang;Hai-Feng Zhang

{"title":"Enhancing Graph Structure Learning via Motif-Driven Hypergraph Construction","authors":"Jia-Le Zhao;Xian-Jie Zhang;Xiao Ding;Xingyi Zhang;Hai-Feng Zhang","doi":"10.1109/TNSE.2025.3547349","DOIUrl":"https://doi.org/10.1109/TNSE.2025.3547349","url":null,"abstract":"Graph neural networks (GNNs), as a cutting-edge technology in deep learning, perform particularly well in various tasks that process graph structure data. However, their foundation on pairwise graphs often limits their capacity to capture latent higher-order topological semantic information. Thus, it is crucial to find a way to extract the latent higher-order information of graphs without missing the lower-order information of the original graph. To address this issue, we here develop a method to construct hypergraph based on motifs, and then a novel neural network framework, named MD-HGNN, is proposed for enhanced graph learning. Specifically, we first utilize motifs of the original graph to construct the hypergraph and eliminate nested structures within the hypergraph to prevent information redundancy. Subsequently, GNNs and hypergraph neural networks (HGNNs) are employed separately to extract the lower-order and higher-order topological semantic information of the graph. Finally, the lower-order and higher-order information are integrated to obtain an embedded representation of graph. Extensive experimental results demonstrate that MD-HGNN preserves the original lower-order graph structure information while effectively extracting higher-order features. Moreover, its performance and robustness are validated across different downstream tasks.","PeriodicalId":54229,"journal":{"name":"IEEE Transactions on Network Science and Engineering","volume":"12 3","pages":"2333-2344"},"PeriodicalIF":6.7,"publicationDate":"2025-03-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143870990","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Protocol-Based Model Predictive Control for Networked Switching Systems With Piecewise-Homogeneous Sojourn Probabilities 分段齐次逗留概率网络交换系统基于协议的模型预测控制

IF 6.7 2区计算机科学

IEEE Transactions on Network Science and Engineering Pub Date : 2025-03-03 DOI: 10.1109/TNSE.2025.3547324

Jun Cheng;Hongjie Pang;Huaicheng Yan;Ju H. Park;Wenhai Qi

{"title":"Protocol-Based Model Predictive Control for Networked Switching Systems With Piecewise-Homogeneous Sojourn Probabilities","authors":"Jun Cheng;Hongjie Pang;Huaicheng Yan;Ju H. Park;Wenhai Qi","doi":"10.1109/TNSE.2025.3547324","DOIUrl":"https://doi.org/10.1109/TNSE.2025.3547324","url":null,"abstract":"Networked switching systems, which integrate multiple subsystems controlled by switching signals, play a crucial role in modern cyber-physical applications such as industrial automation and smart grids. However, their performance is often limited by constrained communication bandwidth and complex dynamic interactions. To address these challenges, this paper proposes a protocol-based model predictive control (MPC) framework for networked switching systems with piecewise-homogeneous sojourn probabilities. A dynamically matching mechanism is designed to quantify mode mismatches caused by network-induced uncertainties. Additionally, an adaptive dynamic-memory event-triggered protocol (ADMETP) is developed, which leverages historical data to optimize triggering decisions and dynamically adjusts thresholds to reduce communication overhead while maintaining system stability. Sufficient conditions for mean-square exponential stability are derived using Lyapunov theory, providing rigorous theoretical guarantees. The effectiveness of the approach is validated through simulations of a numerical experiment and an RLC circuit, demonstrating superior resource utilization and control performance compared to existing methods. This work bridges the gap between adaptive resource management and robust control in networked switching systems, offering practical insights for applications with constrained communication resources.","PeriodicalId":54229,"journal":{"name":"IEEE Transactions on Network Science and Engineering","volume":"12 3","pages":"2322-2332"},"PeriodicalIF":6.7,"publicationDate":"2025-03-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143870880","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Exploring the Privacy-Accuracy Trade-Off Using Adaptive Gradient Clipping in Federated Learning 利用自适应梯度裁剪探索联邦学习中隐私与准确性的权衡

IF 6.7 2区计算机科学

IEEE Transactions on Network Science and Engineering Pub Date : 2025-02-28 DOI: 10.1109/TNSE.2025.3546777

Benteng Zhang;Yingchi Mao;Xiaoming He;Ping Ping;Huawei Huang;Jie Wu

{"title":"Exploring the Privacy-Accuracy Trade-Off Using Adaptive Gradient Clipping in Federated Learning","authors":"Benteng Zhang;Yingchi Mao;Xiaoming He;Ping Ping;Huawei Huang;Jie Wu","doi":"10.1109/TNSE.2025.3546777","DOIUrl":"https://doi.org/10.1109/TNSE.2025.3546777","url":null,"abstract":"In Differentially Private Federated Learning (DP-FL), gradient clipping can prevent excessive noise from being added to the gradient and ensure that the impact of noise is within a controllable range. However, state-of-the-art methods adopt fixed or imprecise clipping thresholds for gradient clipping, which is not adaptive to the changes in the gradients. This issue can lead to a significant degradation in accuracy while training the global model. To this end, we propose Differential Privacy Federated Adaptive gradient Clipping based on gradient Norm (DP-FedACN). DP-FedACN can calculate the decay rate of the clipping threshold by considering the overall changing trend of the gradient norm. Furthermore, DP-FedACN can accurately adjust the clipping threshold for each training round according to the actual changes in gradient norm, clipping loss, and decay rate. Experimental results demonstrate that DP-FedACN can maintain privacy protection performance similar to that of DP-FedAvg under member inference attacks and model inversion attacks. DP-FedACN significantly outperforms DP-FedAGNC and DP-FedDDC in privacy protection metrics. Additionally, the test accuracy of DP-FedACN is approximately 2.61%, 1.01%, and 1.03% higher than the other three baseline methods, respectively. DP-FedACN can improve the global model training accuracy while ensuring the privacy protection of the model. All experimental results demonstrate that the proposed DP-FedACN can help find a fine-grained privacy-accuracy trade-off in DP-FL.","PeriodicalId":54229,"journal":{"name":"IEEE Transactions on Network Science and Engineering","volume":"12 3","pages":"2254-2265"},"PeriodicalIF":6.7,"publicationDate":"2025-02-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143870877","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0