Computer Networks最新文献_第7页

HGC-LSTM: A graph neural network-based model for HO forecasting in mobile networks HGC-LSTM：基于图神经网络的移动网络HO预测模型

IF 4.4 2区计算机科学

Computer Networks Pub Date : 2025-07-05 DOI: 10.1016/j.comnet.2025.111497

Gwladys Ornella Djuikom Foka , Razvan Stanica , Diala Naboulsi

{"title":"HGC-LSTM: A graph neural network-based model for HO forecasting in mobile networks","authors":"Gwladys Ornella Djuikom Foka , Razvan Stanica , Diala Naboulsi","doi":"10.1016/j.comnet.2025.111497","DOIUrl":"10.1016/j.comnet.2025.111497","url":null,"abstract":"<div><div>In the realm of mobile networks, the escalating growth in data traffic, primarily fueled by the proliferation of connected devices and the rising demand for data-intensive applications, poses an ongoing challenge. Ensuring seamless connectivity, mobility, and optimal user experience in such a dynamic environment is more complex than ever. In this context, effective user mobility management through handovers (HOs) emerges as a critical task. Well-managed HOs contribute to enhanced user quality of service (QoS), minimizing disruptions in connectivity as users move within the network. Conversely, poor HO management can lead to issues such as increased latency, network congestion, and elevated operational costs. Against this backdrop, our work focuses on the problem of HO forecasting, aiming to predict HOs among base stations over time. To address this challenge, we introduce the HO Graph Convolutional Long Short-Term Memory (HGC-LSTM) neural network forecasting approach. This innovative methodology incorporates Graph Neural Networks (GNNs) and Long Short-Term Memory (LSTM) networks to capture spatio-temporal dependencies and correlations among neighboring pairs of base stations during the forecasting process. Our evaluation, conducted on a real-world dataset, demonstrates that the proposed HGC-LSTM approach surpasses state-of-the-art methods in the likes of MuLSTM, transformer and ARIMA, reaching the desired trade-off between user QoS and overprovisioned resources.</div></div>","PeriodicalId":50637,"journal":{"name":"Computer Networks","volume":"270 ","pages":"Article 111497"},"PeriodicalIF":4.4,"publicationDate":"2025-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144654139","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

Improved efficient public/private cloud auditing scheme with dynamic updates 改进了具有动态更新的高效公共/私有云审计方案

IF 4.4 2区计算机科学

Computer Networks Pub Date : 2025-07-05 DOI: 10.1016/j.comnet.2025.111498

Yuanhang Zhang , Xu An Wang , Xiaoxuan Xu , Weiwei Jiang , Wei Wei , Xiaoyuan Yang , Hao Liu

{"title":"Improved efficient public/private cloud auditing scheme with dynamic updates","authors":"Yuanhang Zhang , Xu An Wang , Xiaoxuan Xu , Weiwei Jiang , Wei Wei , Xiaoyuan Yang , Hao Liu","doi":"10.1016/j.comnet.2025.111498","DOIUrl":"10.1016/j.comnet.2025.111498","url":null,"abstract":"<div><div>In the age of Big Data, cloud storage, a fundamental feature of cloud computing, plays a vital role in storing and managing data. Yet, cloud server data security encounters substantial challenges, and safeguarding the integrity of externally stored data in the cloud remains an imperative issue demanding immediate attention. Lightweight and efficient cloud auditing schemes have emerged. This paper introduces an enhanced protocol for secure cloud auditing, ensuring privacy in both public and private audit scenarios, and conducts a comprehensive security assessment. The scheme is built on the foundation of Wang et al.’s lightweight public/private auditing scheme. We find that the scheme is not secure, as the cloud server can forge the verification tags of outsourced data blocks. Therefore, this paper first designs an attack against Wang et al.’s scheme, which requires a number of files related to the number of sections per data block. Subsequently, we design a new tag generation algorithm to fix the security flaws. The results show that our scheme is not only lightweight and supports public/private audits but also resists the forgery attack, providing better security. At the same time, our scheme can also dynamically update the cloud data.</div></div>","PeriodicalId":50637,"journal":{"name":"Computer Networks","volume":"270 ","pages":"Article 111498"},"PeriodicalIF":4.4,"publicationDate":"2025-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144572767","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

Joint client–server selection and resource allocation based on split federated learning in Edge-to-Cloud computing environments 边缘到云计算环境下基于分离联邦学习的联合客户端-服务器选择和资源分配

IF 4.4 2区计算机科学

Computer Networks Pub Date : 2025-07-05 DOI: 10.1016/j.comnet.2025.111502

Yishan Chen , Xiangwei Zeng , Xiansong Luo , Zhiquan Liu

引用次数: 0

A chit-chat between Llama 2 and ChatGPT for the automated creation of exploits Llama 2和ChatGPT之间的聊天，用于自动创建漏洞

IF 4.4 2区计算机科学

Computer Networks Pub Date : 2025-07-05 DOI: 10.1016/j.comnet.2025.111501

Francesco Caturano, Jordan Ciotola, Simon Pietro Romano, Mario Varlese

{"title":"A chit-chat between Llama 2 and ChatGPT for the automated creation of exploits","authors":"Francesco Caturano, Jordan Ciotola, Simon Pietro Romano, Mario Varlese","doi":"10.1016/j.comnet.2025.111501","DOIUrl":"10.1016/j.comnet.2025.111501","url":null,"abstract":"<div><div>Software exploitation is the process of taking advantage of vulnerabilities in software systems in order to perform unintended activities. Its understanding leads to improved defensive measures and informed decision making about which security mechanisms to prioritize. However, creating a software exploit is typically a time-consuming and manual task that demands a deep understanding of programming, network protocols, operating system internals, and computer architectures. Additionally, it requires the ability to integrate this knowledge through complex reasoning and problem-solving techniques. This paper proposes an approach to tackle the aforementioned problems by encouraging a conversation between Large Language Models (LLMs) with the purpose of generating software exploits. First, the chosen LLMs are provided with the necessary context knowledge, through modern techniques of fine-tuning and prompt engineering. Then, the exploitation methodology is divided into several steps: vulnerable program analysis, identification of the exploit, planning of the exploitation process, discovery of architecture internals, and production of the exploit software.</div><div>A first Large Language Model (LLM) is designed to ask questions to a second LLM regarding the execution of the above mentioned steps. The final output from the second LLM provides fully automated, functional exploit code.</div><div>This method demonstrates how two LLMs – one possessing capabilities in exploitation and coding, and the other with expertise in computer architecture – can collaborate to successfully exploit Buffer Overflow vulnerabilities.</div></div>","PeriodicalId":50637,"journal":{"name":"Computer Networks","volume":"270 ","pages":"Article 111501"},"PeriodicalIF":4.4,"publicationDate":"2025-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144571238","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 comparative analysis of indoor localization technologies 室内定位技术的对比分析

IF 4.4 2区计算机科学

Computer Networks Pub Date : 2025-07-04 DOI: 10.1016/j.comnet.2025.111527

Koorosh Roohi , Atena Roshan Fekr

{"title":"A comparative analysis of indoor localization technologies","authors":"Koorosh Roohi , Atena Roshan Fekr","doi":"10.1016/j.comnet.2025.111527","DOIUrl":"10.1016/j.comnet.2025.111527","url":null,"abstract":"<div><div>Indoor localization holds great potential in various applications such as healthcare facilities, smart buildings, retail and shopping malls, museums, airports, parking lots, etc. Indoor localization systems aim to track and navigate targets in indoor spaces. These systems use various sets of technologies that can be categorized into four groups: Radio Frequency (RF) based, inertial based, optical based, and ultrasound based. To have a fair comparison between different technologies, in this review paper, we divide these technologies into wearable, contactless, and a fusion of different technology groups. All of these methods are proposed and used with different approaches such as machine learning, deep learning, geometric, and signal processing techniques. In this paper, we compare these methods in terms of localization performance, time complexity, coverage, and generalizability. Also, we determine which of these methods are suitable for different applications. It was observed that methods based on contactless RF based technologies outperformed others by showing centimeter level localization accuracy and preserving users' privacy. Additionally, fusing different types of technology can enhance performance compared to when they are used solely. Technologies and techniques that need further research are also discussed in details.</div></div>","PeriodicalId":50637,"journal":{"name":"Computer Networks","volume":"270 ","pages":"Article 111527"},"PeriodicalIF":4.4,"publicationDate":"2025-07-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144606052","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 secure lightweight identity authentication and key agreement scheme for internet of drones 无人机互联网安全轻量级身份认证与密钥协议方案

IF 4.4 2区计算机科学

Computer Networks Pub Date : 2025-07-04 DOI: 10.1016/j.comnet.2025.111503

Wenlong Dong, Xiujun Wang, Juyan Li

引用次数: 0

BCN: Enhanced Backpressure Flow Control with Rapid Notification in Datacenter Networks BCN：数据中心网络中具有快速通知的增强背压流量控制

IF 4.4 2区计算机科学

Computer Networks Pub Date : 2025-07-04 DOI: 10.1016/j.comnet.2025.111426

Xingyu Hu, Dinghuang Hu, Dezun Dong, Cunlu Li, Zejia Zhou, Guoyuan Yuan

{"title":"BCN: Enhanced Backpressure Flow Control with Rapid Notification in Datacenter Networks","authors":"Xingyu Hu, Dinghuang Hu, Dezun Dong, Cunlu Li, Zejia Zhou, Guoyuan Yuan","doi":"10.1016/j.comnet.2025.111426","DOIUrl":"10.1016/j.comnet.2025.111426","url":null,"abstract":"<div><div>With the growing demands for bandwidth and low latency in applications such as cloud computing, big data processing, and artificial intelligence, datacenter networks must support high-speed transmission and rapid response times. Current end-to-end congestion controls face delays in transmitting congestion notification, while intra-switch flow controls are often too coarse-grained and result in unfair traffic scheduling. This paper proposes BCN, a network management that employs precise flow control backpressure per hop, per flow to support rapid congestion notification. BCN utilizes fewer queues to dynamically allocate new flows and accurately manages the pause/resume of upstream congestion queues by identifying packets. BCN enables rapid rate adjustments for senders by actively providing deceleration/speedup congestion notification signals based on the current queuing status. Furthermore, BCN designs a global rate control model based on flow size. BCN is easy to deploy with low hardware overhead. To the best of our knowledge, BCN is the first work to offer precise flow control with rapid notification per hop, per flow. And we conducted a parallel simulation of an ultra-Clos network with a realistic workload. The results demonstrated that BCN reduced flow completion time (FCT) by 58.9% and achieved a 136.2% improvement in throughput compared to state-of-the-art methods. Under Incast traffic, BCN reduced FCT by 79.8% and improved throughput by 21.4%.</div></div>","PeriodicalId":50637,"journal":{"name":"Computer Networks","volume":"270 ","pages":"Article 111426"},"PeriodicalIF":4.4,"publicationDate":"2025-07-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144595849","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

LOGO-CL: Accelerating semi-supervised federated learning in edge computing LOGO-CL：加速边缘计算中的半监督联邦学习

IF 4.4 2区计算机科学

Computer Networks Pub Date : 2025-07-04 DOI: 10.1016/j.comnet.2025.111496

Yang Xu, Qianshu Wang, Hongli Xu, Yunming Liao, Liusheng Huang, Xin Hang

{"title":"LOGO-CL: Accelerating semi-supervised federated learning in edge computing","authors":"Yang Xu, Qianshu Wang, Hongli Xu, Yunming Liao, Liusheng Huang, Xin Hang","doi":"10.1016/j.comnet.2025.111496","DOIUrl":"10.1016/j.comnet.2025.111496","url":null,"abstract":"<div><div>Federated learning (FL) has gained important attention for training models across clients in edge computing. The scarcity of labeled data poses critical challenges in professional fields (e.g., medical diagnosis) for FL, motivating the development of federated semi-supervised learning (FSSL) to exploit unlabeled data. Though previous works can make full use of unlabeled data, they inevitably face two inherent challenges: limited communication resource and data heterogeneity. Existing FSSL approaches typically assign fixed values for global updating frequency and/or local updating frequency during training, and ignore the impact of unlabeled data distribution, which severely hinder convergence stability and training efficiency. To address these issues, we present a novel FSSL framework, termed LOGO-CL. Specifically, we jointly optimize both local and global updating frequencies by analyzing the coupled impact of these two hyper-parameters on the training process. We develop a multi-armed bandit (MAB) based online algorithm to adaptively determine diverse local updating frequencies as well as appropriate global updating frequency, so as to improve training efficiency. Furthermore, LOGO-CL leverages contrastive learning technique to alleviate the impact of unlabeled data heterogeneity by incorporating regularization terms into both global and local training processes, ensuring robustness, even with imbalanced data distribution. Extensive experiments demonstrate that LOGO-CL achieves the model training speedup by 2.3<span><math><mo>×</mo></math></span> and reduces communication cost by 44% when achieving the same target accuracy, compared with baselines. Moreover, LOGO-CL improves model accuracy up to 4.8% on the datasets with moderate noise.</div></div>","PeriodicalId":50637,"journal":{"name":"Computer Networks","volume":"270 ","pages":"Article 111496"},"PeriodicalIF":4.4,"publicationDate":"2025-07-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144595848","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

An optimization framework for joint wireless data-power transmission in distributed energy harvesting networks 分布式能量收集网络中联合无线数据-电力传输的优化框架

IF 4.4 2区计算机科学

Computer Networks Pub Date : 2025-07-04 DOI: 10.1016/j.comnet.2025.111506

Georgios Kallitsis , Vasileios Karyotis , Symeon Papavassiliou

引用次数: 0

PBAS: A blockchain-assisted priority-based batch authentication scheme for secure and scalable vehicular networks PBAS：一种区块链辅助的基于优先级的批量认证方案，用于安全和可扩展的车辆网络

IF 4.4 2区计算机科学

Computer Networks Pub Date : 2025-07-04 DOI: 10.1016/j.comnet.2025.111500

Juxia Li , Lejun Zhang , Chenglin Chen , Zuwei Li , Wenjie Long , Yuqian Wang , Ran Guo

{"title":"PBAS: A blockchain-assisted priority-based batch authentication scheme for secure and scalable vehicular networks","authors":"Juxia Li , Lejun Zhang , Chenglin Chen , Zuwei Li , Wenjie Long , Yuqian Wang , Ran Guo","doi":"10.1016/j.comnet.2025.111500","DOIUrl":"10.1016/j.comnet.2025.111500","url":null,"abstract":"<div><div>With the rapid advancement of Internet of Vehicles (IoV) technologies, traditional authentication mechanisms have exhibited diminishing efficacy. They struggle to meet strict requirements for security and performance, especially in large-scale, heterogeneous IoV environments. These conventional approaches typically suffer from high authentication delays, limited scalability, and insufficient resistance to security threats. To address these limitations, this paper proposes PBAS — a Blockchain-Assisted Priority-Based Batch Authentication Scheme that aims to enhance both authentication responsiveness and security guarantees in dynamic vehicular networks.PBAS introduces a hierarchical authentication framework capable of distinguishing among multiple role levels, employing end-to-end identity verification to ensure compatibility and mutual trust among heterogeneous nodes. Furthermore, by integrating environment-aware data, PBAS enables an adaptive authentication strategy that dynamically responds to fluctuating network conditions. The proposed architecture combines off-chain pre-authentication with on-chain grouped consensus verification in a dual-layered structure. This design significantly reduces authentication delay and computational overhead while improving scalability and resilience against network-based threats. Formal security analysis and theorem-based proofs demonstrate that PBAS can withstand various common attacks, including replay, impersonation, and collusion. Experimental results indicate that PBAS achieves substantial improvements in communication efficiency, authentication throughput, and system robustness. These advantages render it particularly suitable for high-density, role-diverse, and fast-changing vehicular environments, highlighting its strong potential for real-world deployment.</div></div>","PeriodicalId":50637,"journal":{"name":"Computer Networks","volume":"270 ","pages":"Article 111500"},"PeriodicalIF":4.4,"publicationDate":"2025-07-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144571254","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