Computer Networks最新文献

{"title":"Energy consumption minimization for robotic systems in intelligent factories with the assistance of STAR-RIS: A reinforcement learning approach","authors":"Nguyen Thi Thanh Van ,&nbsp;Hoang Le Hung ,&nbsp;Nguyen Cong Luong ,&nbsp;Huy T. Nguyen ,&nbsp;Nguyen Tien Hoa ,&nbsp;Ngo Manh Duy ,&nbsp;Ngo Manh Tien","doi":"10.1016/j.comnet.2025.111441","DOIUrl":"10.1016/j.comnet.2025.111441","url":null,"abstract":"<div><div>The integration of wireless communication to factory robots to form connected robots has become an encouraging technology for intelligent factories due to their cost-effectiveness and high flexibility. However, due to obstacles, guaranteeing stable communication links between an access point (AP) and a robot is challenging. To address this, we propose to deploy a Simultaneously Transmitting and Reflecting Reconfigurable Intelligent Surface (STAR-RIS) which allows incident signals to be reflected and transmitted on both sides of the RIS surface, enabling comprehensive <span><math><mrow><mn>36</mn><msup><mrow><mn>0</mn></mrow><mrow><mn>0</mn></mrow></msup></mrow></math></span> coverage to improve data rate between the AP and the robot. Then, we define an optimization problem that seeks to minimize the system’s overall energy consumption including the AP communication energy and the robot energy consumption while satisfying the requirements of the robot’s safety distance, maximum movement duration, and data rate threshold. The problem involves optimizing the robot’s trajectory, the transmitted power of the AP, the phase shifts, and the transmitting/reflecting coefficient of the STAR-RIS. The optimization problem is nonconvex due to the nonconvex objective function, the nonconvex obstacle-robot distance constraint, the phase shifts and transmitting/reflecting coefficient of STAR-RIS, and the data rate requirement constraint. In addition, there are many dynamic factors in the working environment, such as the robot’s location, the channel between the AP and robot. Therefore, we first approximate the original optimization problem by a Markov Decision Process (MDP) model, then propose to use a DRL algorithm based on Proximal Policy Optimization (PPO) which uses an actor and critic network policy reinforcement to solve the optimization problem. We conducted extensive simulations under various scenarios, and the results show that the case with the use of the STAR-RIS significantly reduces the travel distance of the robot and the system energy consumption compared with the cases with A2C based algorithm, the conventional RIS or without RIS.</div></div>","PeriodicalId":50637,"journal":{"name":"Computer Networks","volume":"270 ","pages":"Article 111441"},"PeriodicalIF":4.4,"publicationDate":"2025-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144606050","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}

{"title":"Edge computing for IoT: Novel insights from a comparative analysis of access control models","authors":"Tao Xue ,&nbsp;Ying Zhang ,&nbsp;Yanbin Wang ,&nbsp;Wenbo Wang ,&nbsp;Shuailou Li ,&nbsp;Haibin Zhang","doi":"10.1016/j.comnet.2025.111468","DOIUrl":"10.1016/j.comnet.2025.111468","url":null,"abstract":"<div><div>IoT edge computing positions computing resources closer to the data sources to reduce the latency, relieve the bandwidth pressure on the cloud, and enhance data security. Nevertheless, data security in IoT edge computing still faces critical threats (e.g., data breaches). Access control is fundamental for mitigating these threats. However, IoT edge computing introduces notable challenges for achieving resource-conserving, low-latency, flexible, and scalable access control. To review recent access control measures, we organize them in a novel way according to different data lifecycles – data collection, storage, and usage – and, meanwhile, review blockchain technology in this novel organization. In this way, we provide novel insights and envisage several potential research directions. This survey can help readers find gaps systematically and prompt the development of access control techniques in IoT edge computing under the intricacy of innovations in access control.</div></div>","PeriodicalId":50637,"journal":{"name":"Computer Networks","volume":"270 ","pages":"Article 111468"},"PeriodicalIF":4.4,"publicationDate":"2025-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144606053","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}

{"title":"beam-align: Distributed user association for mmWave networks with multi-connectivity","authors":"L. Weedage,&nbsp;C. Stegehuis,&nbsp;S. Bayhan","doi":"10.1016/j.comnet.2025.111446","DOIUrl":"10.1016/j.comnet.2025.111446","url":null,"abstract":"<div><div>Since the spectrum below 6 GHz bands is insufficient to meet the high bandwidth requirements of 5G use cases, 5G networks expand their operation to mmWave bands. However, operation at these bands has to cope with a high penetration loss and susceptibility to blocking objects. Beamforming and multi-connectivity (MC) can together mitigate these challenges. But, to design such an optimal user association scheme leveraging these two features is non-trivial and computationally expensive. Previous studies either considered a fixed MC degree for all users or overlooked beamforming. Driven by the question <em>what is the optimal degree of MC for each user in a mmWave network,</em> we formulate a user association scheme that maximizes throughput considering beam formation and MC. Our numerical analysis shows that there is no one-size-fits-all degree of optimal MC; it depends on the number of users, their rate requirements, locations, and the maximum number of active beams at a BS. Based on the optimal association, we design <span>beam-align</span>: an efficient heuristic with polynomial-time complexity <span><math><mrow><mi>O</mi><mrow><mo>(</mo><mrow><mo>|</mo><mi>U</mi><mo>|</mo></mrow><mo>log</mo><mrow><mo>|</mo><mi>U</mi><mo>|</mo></mrow><mo>)</mo></mrow></mrow></math></span>, where <span><math><mrow><mo>|</mo><mi>U</mi><mo>|</mo></mrow></math></span> is the number of users. Moreover, <span>beam-align</span> only uses local BS information - i.e. the received signal quality at the user. Differing from prior works, <span>beam-align</span> considers beamforming, multiconnectivity and line-of-sight probability. Via simulations, we show that <span>beam-align</span> performs close to optimal in terms of per-user capacity and satisfaction while it outperforms frequently-used signal-to-interference-and-noise-ratio based association schemes. We then show that <span>beam-align</span> has a robust performance under various challenging scenarios: the presence of blockers, rain, and clustered users.</div></div>","PeriodicalId":50637,"journal":{"name":"Computer Networks","volume":"270 ","pages":"Article 111446"},"PeriodicalIF":4.4,"publicationDate":"2025-07-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144606086","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}

{"title":"Ultra lightweight post-quantum resistant 5G-AKA protocol","authors":"An Braeken","doi":"10.1016/j.comnet.2025.111444","DOIUrl":"10.1016/j.comnet.2025.111444","url":null,"abstract":"<div><div>As constrained devices like Internet of Things (IoT) devices become increasingly integrated with 5G networks, efficient and secure authentication and key management mechanisms are essential to ensure seamless and protected communication with the core 5G infrastructure. However, the current 5G-AKA (Authentication and Key Agreement) protocol lacks resistance against perfect forward secrecy (PFS) and post-quantum security, making it vulnerable to future adversarial threats, particularly quantum-enabled attacks. Most existing research provides partial solutions addressing either PFS or post-quantum security, no approach fully resolves both issues simultaneously in an efficient manner. This paper presents a novel authentication mechanism that relies solely on symmetric key cryptography, ensuring both high performance and robust security. Our innovation lies in replacing the conventional use of identical key material with a hybrid setup, where the user and 5G core each hold both common and distinct key components . By eliminating the computational overhead associated with asymmetric cryptography, our proposed solution offers an extremely efficient and scalable security solution, having almost 5 times lower energy consumption as the current 5G-AKA standard and requiring almost three times less security material to be sent during the protocol. As a consequence, this protocol offers a sustainable solution both with respect to energy and security aspects.</div></div>","PeriodicalId":50637,"journal":{"name":"Computer Networks","volume":"270 ","pages":"Article 111444"},"PeriodicalIF":4.4,"publicationDate":"2025-07-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144606087","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}

{"title":"Accelerating personalized federated learning via dynamic gradient substitution and client selection","authors":"Ziwei Zhan ,&nbsp;Weijie Liu ,&nbsp;Xiaoxi Zhang ,&nbsp;Chee Wei Tan ,&nbsp;Lei Xue ,&nbsp;Haisheng Tan ,&nbsp;Xu Chen","doi":"10.1016/j.comnet.2025.111428","DOIUrl":"10.1016/j.comnet.2025.111428","url":null,"abstract":"<div><div>Personalized federated learning (PFL) has gained widespread attention for its ability to preserve privacy and adapt to user-specific characteristics. Among the leading PFL methods, meta-learning based algorithms like Per-FedAvg offer a unified framework of gradient updates for all clients, eliminating the necessity of personalized model architectures that are common in other PFL approaches. However, their computation inefficiency and challenges in accommodating system heterogeneity are under-explored. This work proposes <em>pFedSara</em>, a novel PFL framework that accelerates the training of a target PFL method, Per-FedAvg, by exploiting the lightweight, vanilla FL algorithm, FedAvg. Instead of fervently creating marginally altered approaches, <em>pFedSara</em> is the first that strategically <em>reuses and blends</em> existing techniques for PFL training, navigating the runtime-accuracy trade-off, and it offers a comprehensive theoretical analysis. Specifically, it leverages dynamic gradient substitution and client selection by assessing runtime, loss, and gradient similarity between FedAvg and Per-FedAvg, the two candidate local update methods for each client. Additionally, it incorporates gradient scaling to accommodate incomplete Per-FedAvg computations that cannot be replaced by FedAvg, eliminating additional biases. A novel convergence analysis is provided, quantifying the biases introduced by both heterogeneous data and our employed hybrid update methods for computation speed-up. Extensive experiments demonstrate that <em>pFedSara</em> achieves superior training efficiency compared with state-of-the-art PFL methods.</div></div>","PeriodicalId":50637,"journal":{"name":"Computer Networks","volume":"270 ","pages":"Article 111428"},"PeriodicalIF":4.4,"publicationDate":"2025-07-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144634458","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}

{"title":"Task combination optimization via dual-view heterogeneous graph contrastive learning for mobile crowdsensing","authors":"Jian Wang ,&nbsp;Qi Zhang ,&nbsp;Guanzhi He ,&nbsp;Guosheng Zhao","doi":"10.1016/j.comnet.2025.111546","DOIUrl":"10.1016/j.comnet.2025.111546","url":null,"abstract":"<div><div>Mobile crowd sensing is an intelligent sensing technology that relies on mobile devices and user participation. Its core is to use widely distributed user groups to complete various sensing tasks. However, due to factors such as differences in user abilities, geographical locations and user participation motivations, task allocation often becomes imbalanced, leading to the reduction of the task completion ratio. To address the above problem, a task combination optimization method based on dual-view heterogeneous graph contrastive learning is proposed. First, by constructing a heterogeneous graph model, the task features and external dependencies (such as user behavior and scene conditions) are mapped to different nodes and edges in the graph structure. Second, from the perspectives of multi-path and edge features, the heterogeneous graph contrastive learning method is used to learn the representation of task nodes. The node classes are predicted based on these learned representations, and task nodes of the same class are combined to form a task cluster. Finally, based on the similarity of sensing users, multiple collaboration groups are formed. Task clusters are allocated according to the satisfaction of group members. Experiments using the Yelp, Freebase, Epinions and DBLP datasets show that our proposed method demonstrates relatively strong generalization ability and significantly improves the relevance of tasks in the combination. In addition, in the task allocation experiments, our proposed method successfully enhances user satisfaction and task completion ratio.</div></div>","PeriodicalId":50637,"journal":{"name":"Computer Networks","volume":"270 ","pages":"Article 111546"},"PeriodicalIF":4.4,"publicationDate":"2025-07-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144614795","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}

{"title":"Distributed learning based message dissemination approach for underwater surveillance in OUSN","authors":"Linfeng Liu,&nbsp;Xiangyu Yan,&nbsp;Jia Xu","doi":"10.1016/j.comnet.2025.111507","DOIUrl":"10.1016/j.comnet.2025.111507","url":null,"abstract":"<div><div>Opportunistic Underwater Sensor Network (OUSN) is deployed for various underwater surveillance. The nodes in OUSN always keep moving, and the movement laws are extremely complex, due to many factors such as the movement intentions and living habits of their carriers. Besides, the message dissemination in OUSN is implemented in an opportunistic manner, because the communication links between nodes are intermittent, making the stable communication paths hard to be formed. The link prediction technique can be applied to predict the future links that may appear in the network topologies, and these links indicate the probabilities of future encounters between nodes, which helps to improve the performance of message dissemination. Especially, because the communication links between nodes are varied over time, a centralized link prediction is not feasible, and the nodes should predict the future links and make dissemination decisions locally. In this paper, we propose a Distributed Learning based Message Dissemination Approach (DLMDA) for each node to disseminate the stored data messages distributedly. Specifically, the link prediction results are expressed by some adjacency matrices, based on which each node disseminates the stored data messages to several selected neighbours. By adopting a distributed learning framework, both the communication overhead and processing delay of DLMDA are significantly reduced by avoiding the uploads of historical links to a server. Simulation results demonstrate the superior performance of DLMDA, i.e., through predicting the future links with the distributed learning framework, DLMDA enhances the delivery ratio of data messages, and reduces the delivery delay of data messages and the communication overhead effectively.</div></div>","PeriodicalId":50637,"journal":{"name":"Computer Networks","volume":"270 ","pages":"Article 111507"},"PeriodicalIF":4.4,"publicationDate":"2025-07-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144595847","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}

{"title":"DQN-enhanced interference mitigation for NGSO satellites in NTN based on spectrum sensing","authors":"Lanyining Li,&nbsp;Jiaxin Zhang,&nbsp;Hao Yuan,&nbsp;Jianghua Long,&nbsp;Xing Zhang","doi":"10.1016/j.comnet.2025.111509","DOIUrl":"10.1016/j.comnet.2025.111509","url":null,"abstract":"<div><div>With the rapid development of non-terrestrial networks (NTN), the challenge of co-frequency interference within non-geostationary satellite orbit (NGSO) constellation networks has become increasingly significant. The dynamic nature of NGSO, coupled with frequent transitions in connections with earth stations (ES), presents a major obstacle for traditional interference mitigation techniques designed for geostationary (GSO) and NGSO constellations. This study proposes a novel approach based on deep q-network (DQN) reinforcement learning, which integrates spectrum sensing (SS) and a dynamic network training mechanism. By leveraging historical spectrum databases and real-time spectrum sensing, we optimize channel capacity while minimizing computational resource consumption under interference constraints. Simulation results show that the proposed method allows all NGSO links to rapidly select and adapt their actions, including access targets, spectrum selections, and transmission power. Moreover, this study also evaluates the impact of primary user (PU) mobility at different altitudes on interference. The dynamic network training mechanism proposed here can reduce the relative probability of interference by 23.1% to 25% at various altitudes.</div></div>","PeriodicalId":50637,"journal":{"name":"Computer Networks","volume":"270 ","pages":"Article 111509"},"PeriodicalIF":4.4,"publicationDate":"2025-07-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144634457","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}

{"title":"The design of an RIS-assisted FDMA wireless sensor network for sum throughput maximization","authors":"Omid Abachian Ghasemi ,&nbsp;Masoumeh Azghani ,&nbsp;Mehdi Chehel Amirani","doi":"10.1016/j.comnet.2025.111512","DOIUrl":"10.1016/j.comnet.2025.111512","url":null,"abstract":"<div><div>In this article, we consider a reconfigurable intelligent surface (RIS)-assisted wireless sensor network (WSN), where sensors transmit data to a fusion center (FC) via the frequency division multiple access (FDMA) protocol. This sensor network leverages the capabilities of a primary network equipped with an RIS during periods of primary network inactivity. Power and bandwidth allocation, along with RIS phase shifts, are optimized to maximize sum-throughput (the performance metric), subject to constraints on total power and minimum sensor throughput. This optimization enables higher sum-throughput without increasing total energy consumption or bandwidth requirements. The non-convex optimization problem is tackled using a block coordinate descent (BCD) technique. This technique decomposes the problem into the three subproblems: power allocation subproblem, which is a convex subproblem solved using the Lagrange Dual Method (LDM) and the Karush-Kuhn–Tucker (KKT) conditions; the bandwidth allocation subproblem, which, due to its convexity, is solved similarly; and finally, the RIS phase shift adjustment subproblem, which is solved using gradient ascent. The BCD algorithm iteratively optimizes these subproblems until convergence is achieved. Numerical results demonstrate the superiority of the proposed scheme over its counterparts across various simulation scenarios. For instance, the proposed method with a 100-element RIS can improve the sum-throughput by approximately 50% compared to a conventional FDMA system.</div></div>","PeriodicalId":50637,"journal":{"name":"Computer Networks","volume":"270 ","pages":"Article 111512"},"PeriodicalIF":4.4,"publicationDate":"2025-07-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144596300","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}

{"title":"EncryptoVision: A dual-modal fusion-based multi-classification model for encrypted traffic recognition","authors":"Zhiyuan Li ,&nbsp;Yujie Jin","doi":"10.1016/j.comnet.2025.111499","DOIUrl":"10.1016/j.comnet.2025.111499","url":null,"abstract":"<div><div>With the development of security, confidentiality, and data privacy technologies, the classification of fine-grained encrypted traffic has become increasingly important. Nowadays, existing deep learning methods, including CNN, LSTM, and transformer, have shown impressive classification performance. However, many of these methods merely utilize the raw packet bytes to generate traffic representations, resulting in the potential loss of crucial information, such as dynamic traffic patterns and changes in protocols. In this paper, we propose a dual-modal fusion-based multi-classification model for encrypted traffic recognition, called EncryptoVision. Firstly, we transform the encrypted traffic data into three-channel images and incorporate a triplet attention mechanism to enhance the interaction among the three channels. Then, we use the multi-head self-attention mechanism to expand the model’s global receptive field, allowing it to capture more detailed spatial feature information. Additionally, we also leverage the learning abilities of the transformer encoder to extract temporal feature information from the traffic for long-term time series prediction. Next, we use the spatial–temporal fusion features to obtain the fine-grained features for multi-classification. Experimental results show that our model outperforms state-of-the-art models in classification performance across four real-world encrypted traffic datasets.</div></div>","PeriodicalId":50637,"journal":{"name":"Computer Networks","volume":"270 ","pages":"Article 111499"},"PeriodicalIF":4.4,"publicationDate":"2025-07-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144571253","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}