Computer Networks最新文献

{"title":"A novel optimization and deep learning enabled resource allocation algorithm using blockchain-based Edge computing in IoT","authors":"Lixia Xu ,&nbsp;Jie Yin","doi":"10.1016/j.comnet.2025.111618","DOIUrl":"10.1016/j.comnet.2025.111618","url":null,"abstract":"<div><div>By ensuring secure data transmission and communications in the Internet of Things (IoT), blockchain technology can enhance privacy and security aspects of edge computing, particularly in resource allocation, storage, and management. This work builds a novel optimization and deep learning-enabled resource allocation system using blockchain-based Edge computing in IoT. The effort focuses on using smart contracts to distribute computer resources among users. Edge computing nodes (ECNs) receive user requests from data service subscribers (DSSs) and assign them in a way that maximizes income for the ECNs while satisfying DSS requirements. An optimization algorithm, named Trend factor smoothing-based One-to-One-Based Optimizer (TFsOOBO), which was created by incorporating Trend factor smoothing into One-to-One-Based Optimizer (OOBO), is presented to allocate resources. Furthermore, deep Q network (DQN) is used to predict network delay. Therefore, resource allocation is carried out by the proposed TFsOOBO method using deep learning-based prediction. The experimental results indicate a network delay time of 33.3 %, computing delay time of 22.3 %, and reward of 54.8 %, demonstrate that the TFsOOBO-based resource allocation outperforms conventional models.</div></div>","PeriodicalId":50637,"journal":{"name":"Computer Networks","volume":"271 ","pages":"Article 111618"},"PeriodicalIF":4.6,"publicationDate":"2025-08-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144863397","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":"Power allocation optimization for hybrid IRS-assisted 6G V2V communication","authors":"Xiaoxuan Hu ,&nbsp;Xianyu Wei ,&nbsp;Liang Shan ,&nbsp;Jin Qi ,&nbsp;Zhenjiang Dong ,&nbsp;Yanfei Sun","doi":"10.1016/j.comnet.2025.111610","DOIUrl":"10.1016/j.comnet.2025.111610","url":null,"abstract":"<div><div>In response to the severe path loss caused by high-frequency communication in a 6G millimeter-wave/sub-terahertz vehicle-to-vehicle (V2V) communication system, a vehicle-based hybrid intelligent reflecting surface (HIRS) power allocation model was proposed by considering the high mobility of vehicles and the limited availability of power resources. An HIRS-assisted V2V communication model was constructed by incorporating the location information of communicating vehicles. According to the generalized Snell reflection law, a cylindrical IRS phase configuration that can reflect signals in any direction was preconfigured. The optimal power allocation of the HIRS was determined on the basis of the Dinkelbach algorithm. The simulation results indicated that the spectrum efficiency of the HIRS is much superior to that of a vehicle-based passive IRS, approaching that of a vehicle-based active IRS, and the energy efficiency of the HIRS is significantly higher than that of the active IRS.</div></div>","PeriodicalId":50637,"journal":{"name":"Computer Networks","volume":"271 ","pages":"Article 111610"},"PeriodicalIF":4.6,"publicationDate":"2025-08-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144831037","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":"Efficient missing tag identification for large-scale RFID systems via collision exploitation","authors":"Chu Chu ,&nbsp;Sun Mao ,&nbsp;Jinsong Wu ,&nbsp;Gang Li ,&nbsp;Zhenbing Li ,&nbsp;Guangjun Wen","doi":"10.1016/j.comnet.2025.111605","DOIUrl":"10.1016/j.comnet.2025.111605","url":null,"abstract":"<div><div>Radio Frequency Identification (RFID) has wide-ranging applications, including tracking inventory in warehouses and retail stores. Among those applications, efficiently identifying the missing items is a critical yet challenging issue. However, traditional methods are time-consuming due to lengthy ID transmissions and underutilization of collision data. To overcome these defects, we exploit information in collision slots and propose the SSMTI protocol. SSMTI employs lightweight vectors to convert all time slots into collision slots, enabling tags in each slot to simultaneously transmit a designed string. By using Manchester encoding to decode the aggregated string, the reader can verify multiple tags concurrently. To accelerate the identification process when most tags are missing, we further propose the ISMTI protocol. ISMTI adopts a probabilistic approach to design the string for each collided tag, facilitating the reader to identify more tags within shorter strings than SSMTI. Moreover, ISMTI can dynamically adjust the verification mechanism when the proportion of missing tags varies to maintain efficiency. We also provide theoretical analysis of the proposed protocols to optimize parameter settings and evaluate their performance through extensive simulations. Compared with state-of-the-art solutions, SSMTI and ISMTI can significantly improve time efficiency by 39.74% and 68.87%.</div></div>","PeriodicalId":50637,"journal":{"name":"Computer Networks","volume":"271 ","pages":"Article 111605"},"PeriodicalIF":4.6,"publicationDate":"2025-08-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144852599","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":"Improving mobility management in LEO Satellite networks utilizing Soft Actor-Critic algorithms","authors":"Bo Li,&nbsp;Hao Wang","doi":"10.1016/j.comnet.2025.111608","DOIUrl":"10.1016/j.comnet.2025.111608","url":null,"abstract":"<div><div>Low Earth Orbit (LEO) satellites offer promising opportunities for non-terrestrial networks due to their low latency, global coverage, and cost-effectiveness. However, their rapid movement and large number of user equipment (UE) access requests pose significant challenges for mobility management. This paper addresses these challenges by modeling UE access requests as an M/M/K queuing system, covering initial access, handover, and re-access scenarios. We propose a novel mobility management solution using a multi-attribute decision-making algorithm based on Soft Actor-Critic (SAC), which dynamically adjusts decision parameters in response to real-time environmental conditions. Our experimental analysis demonstrates that SAC improves throughput, reduces access failures, and minimizes unnecessary handovers, outperforming traditional and advanced algorithms in dynamic satellite communication environments. The results highlight the potential of SAC to significantly enhance network performance in LEO satellite networks.</div></div>","PeriodicalId":50637,"journal":{"name":"Computer Networks","volume":"271 ","pages":"Article 111608"},"PeriodicalIF":4.6,"publicationDate":"2025-08-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144831039","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":"On resource consumption of machine learning in communications network security","authors":"Md Muzammal Hoque ,&nbsp;Ijaz Ahmad ,&nbsp;Jani Suomalainen ,&nbsp;Paolo Dini ,&nbsp;Mohammad Tahir","doi":"10.1016/j.comnet.2025.111600","DOIUrl":"10.1016/j.comnet.2025.111600","url":null,"abstract":"<div><div>As the complexity of communication networks continues to increase, driven by a diverse array of devices, services and applications, the adoption of Machine Learning (ML) has seen a significant rise to address various challenges ranging from management to security. Regarding network security, the application of ML ranges from preventive measures to detection and remediation due to its ability to dynamically learn and adapt to evolving threat landscapes. However, ML requires a significant amount of resources, mainly due to the fact that ML operates on data, and the volumes of data are consistently rising. This review article explores the resource consumption aspect of ML techniques used for network security and provides a comprehensive review of the current state of research. Moreover, we propose a taxonomy that can be used to classify the methods through which the resource consumption can be reduced for different ML-based network security implementations. The focus of the study encompasses several key aspects related to resource consumption, including energy, computing, memory, latency, bandwidth, and human resources. These resources are critical in improving the efficiency and optimizing the reliability and sustainability of network security solutions. Furthermore, based on an extensive literature review, we summarize key points regarding optimizing resource consumption in ML-based network security solutions. Finally, the challenges and future research directions for resource-efficient, ML-based network security solutions are outlined to aid in the advancement of research in this area.</div></div>","PeriodicalId":50637,"journal":{"name":"Computer Networks","volume":"271 ","pages":"Article 111600"},"PeriodicalIF":4.6,"publicationDate":"2025-08-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144810460","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":"On the trustworthiness of federated learning models for 5G network intrusion detection under heterogeneous data","authors":"Vangelis Lamprou,&nbsp;George Doukas,&nbsp;Christos Ntanos,&nbsp;Dimitris Askounis","doi":"10.1016/j.comnet.2025.111598","DOIUrl":"10.1016/j.comnet.2025.111598","url":null,"abstract":"<div><div>The rapid advancement of 5G networks is reshaping wireless communications through ultra-fast speeds, low latency, and seamless connectivity. This shift is accompanied by emerging technologies such as edge computing, network traffic management, resource allocation, and network slicing, which distribute data processing across the network. These trends amplify the need for privacy-preserving and decentralized learning methods, particularly in security-critical applications where transmitting raw data to a central server may be infeasible or undesirable. Federated Learning has emerged as a promising paradigm to meet these demands by enabling model training across distributed data sources. In this study, we explore the trustworthiness of Federated Learning models compared to a centralized counterpart, under various heterogeneous client data distributions. The distributions are generated via label-based symmetric Latent Dirichlet Allocations, where the Dirichlet concentration parameter <span><math><mi>α</mi></math></span> controls the degree of class imbalance across clients. We use traffic flow data from a 5G Network Intrusion Detection task to design centralized and federated Artificial Neural Network architectures and extract feature importance scores using the Integrated Gradients algorithm. Our findings, based on the top-10 features, show that federated models trained on slightly-skewed (<span><math><mrow><mi>α</mi><mo>=</mo><mn>10</mn></mrow></math></span>) and mildly-skewed (<span><math><mrow><mi>α</mi><mo>=</mo><mn>2</mn></mrow></math></span>) data achieve trust scores closely aligned with the central model. The <span><math><mrow><mi>α</mi><mo>=</mo><mn>10</mn></mrow></math></span> model achieves an average importance score of 2.2 (7.2% lower than the central model at 2.37), with 91.3% feature overlap. The <span><math><mrow><mi>α</mi><mo>=</mo><mn>2</mn></mrow></math></span> model scores 2.44 (3% higher), with identical overlap. In contrast, highly-skewed models (<span><math><mrow><mi>α</mi><mo>=</mo><mn>0</mn><mo>.</mo><mn>1</mn><mo>,</mo><mn>0</mn><mo>.</mo><mn>5</mn></mrow></math></span>) show diminished trustworthiness, scoring 1.62 and 1.74 (31.6% and 26.6% lower), with overlaps of 80% and 82.5%, respectively. These results highlight the impact of client data heterogeneity on model trustworthiness and underscore the sensitivity of federated models to high levels of data heterogeneity.</div></div>","PeriodicalId":50637,"journal":{"name":"Computer Networks","volume":"271 ","pages":"Article 111598"},"PeriodicalIF":4.6,"publicationDate":"2025-08-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144810320","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":"Spatial–temporal Causal Fusion Graph Neural Networks for urban traffic prediction","authors":"Nianwen Ning ,&nbsp;Wenbo Zhang ,&nbsp;Wei Li ,&nbsp;Hengji Li ,&nbsp;Yanyu Zhang ,&nbsp;Yi Zhou ,&nbsp;Fuqiang Liu","doi":"10.1016/j.comnet.2025.111582","DOIUrl":"10.1016/j.comnet.2025.111582","url":null,"abstract":"<div><div>Urban traffic forecasting is essential for effective traffic management and road planning. Accurate traffic flow prediction can significantly alleviate congestion and enhance traffic safety. Despite recent advances in spatial–temporal modeling, many approaches overlook dynamic interactions between traffic lanes. In the spatial domain, road network connectivity is influenced by complex factors such as infrastructure maintenance and traffic accident patterns. In the temporal domain, potential causal dependencies are embedded within the dynamic propagation of traffic time series.To address these limitations, we propose the Spatial–temporal Causal Fusion Graph Neural Network (STCFGNN). Our framework includes a fusion graph learning module that extracts intricate spatial dependencies using Granger causality tests. This module identifies causal structures from historical traffic flow data, which are then integrated with adaptive graph frameworks for enhanced spatial fusion. Additionally, we implement a spatial–temporal attention module to assess the impact of various features across different nodes. The experimental results indicate that, compared to the best baseline model, STCFGNN reduces the MAE by 1.75% (METR-LA) and 7.94% (PEMS-BAY) in the 60 min prediction.</div></div>","PeriodicalId":50637,"journal":{"name":"Computer Networks","volume":"271 ","pages":"Article 111582"},"PeriodicalIF":4.6,"publicationDate":"2025-08-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144831038","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":"SPDTM: A shapley-value based preference discovery and task matching to maximize satisfaction in mobile crowd sensing","authors":"Zhongxin Huang ,&nbsp;Zanbo Sun ,&nbsp;Wei Tan ,&nbsp;Houbing Herbert Song ,&nbsp;Tian Wang ,&nbsp;Shaobo Zhang ,&nbsp;Anfeng Liu","doi":"10.1016/j.comnet.2025.111602","DOIUrl":"10.1016/j.comnet.2025.111602","url":null,"abstract":"<div><div>Mobile Crowd Sensing (MCS) is a widely adopted paradigm for task-centered data collection. Researchers have noticed that assigning workers to their preferred tasks can improve the task completion rate and satisfaction of MCS. However, there are still some key issues in current research that have not been addressed. First, MCS heavily rely on the impractical assumption that worker-task preferences are prior known for optimal matching. In actual MCS, such preferences are difficult to obtain, and dishonest workers may report false preferences to gain illegal rewards. Second, there are numerous eligible workers for each task in MCS, which makes task difficult to match with highest preference worker. To tackle these challenge issues, a Shapley-value based Preference Discovery and Task Matching (SPDTM) scheme is proposed to maximize task-worker satisfaction. The SPDTM scheme primarily includes three components. First, a Shapley-value based task to workers preference discovery is designed to obtain the true preferences of task to workers based on the contribution of workers. Then, we propose an optimized strategy for selecting workers by trade-off exploration and exploitation. Finally, based on worker preferences discovery and satisfaction improvement, we propose a matching algorithm for task-worker assignment. The outcomes of our experiments indicate that the SPDTM scheme improves task-worker matching satisfaction by 26.01 %, and the detection rate of workers’ true preferences has reached 94.79 %.</div></div>","PeriodicalId":50637,"journal":{"name":"Computer Networks","volume":"271 ","pages":"Article 111602"},"PeriodicalIF":4.6,"publicationDate":"2025-08-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144826340","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":"Resource optimization for minimizing latency and cost in UAV-assisted mobile edge computing (MEC) networks","authors":"Shamim Taimoor ,&nbsp;Muhammad Naeem ,&nbsp;Sherali Zeadally ,&nbsp;Waleed Ejaz","doi":"10.1016/j.comnet.2025.111587","DOIUrl":"10.1016/j.comnet.2025.111587","url":null,"abstract":"<div><div>Unmanned aerial vehicles (UAVs) enable a mobile edge computing (MEC) paradigm with reduced latency by bringing computational resources closer to the network edge. However, UAV-MEC servers have less computation and caching resources than ground base stations (BSs). The management of communication and control resources is crucial to coordinate communication, computing, and caching due to the involvement of aerial networks. Thus, managing joint caching, communication, computing, and control (4C) resources is vital in UAV-assisted MEC networks. To address these challenges, we developed a computational model for efficient resource management to reduce the linear combination of network cost and latency under constrained caching, computing, and offloading. We used binary decision variables for the allocation of computational and offloading resources. The formulated problem is a binary linear programming problem incorporating binary decision variables and linear constraints. We propose an interior point method-based heuristic to obtain a sub-optimal solution with low complexity. Simulation results demonstrate the effectiveness of our proposed approach compared to the branch and bound algorithm.</div></div>","PeriodicalId":50637,"journal":{"name":"Computer Networks","volume":"271 ","pages":"Article 111587"},"PeriodicalIF":4.6,"publicationDate":"2025-08-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144809549","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":"Threat impact analysis of man-in-the-middle attacks on delay-based geolocation on the internet","authors":"Bar Pincu ,&nbsp;Aviram Zilberman ,&nbsp;Ilia Leibovich ,&nbsp;Rami Puzis ,&nbsp;Andikan Otung ,&nbsp;Motoyoshi Sekiya ,&nbsp;Yuval Elovici","doi":"10.1016/j.comnet.2025.111551","DOIUrl":"10.1016/j.comnet.2025.111551","url":null,"abstract":"<div><div>Man-in-the-middle (MITM) attacks on the Internet pose significant security threats. These attacks enable hackers to intercept and manipulate traffic between routers, exploit vulnerabilities in the Internet’s trust model, divert traffic, eavesdrop on communications, or insert malicious data into the network. This paper investigates the abilities of MITM attackers to manipulate delay-based geolocation algorithms. The focus is on how attackers, controlling a series of routers and a host computer, can distort geolocation results by altering delay measurements obtained from ICMP ping messages sent to the host computer. This manipulation targets delay-based geolocation techniques used in location-based services (LBSs). Through a systematic investigation, this study evaluates the effectiveness and capabilities of MITM attackers in subverting geolocation algorithms to falsify a victim’s location. Our findings highlight the necessity of implementing robust security measures on the Internet to mitigate the risk of MITM attacks and protect the integrity of geolocation-based services.</div></div>","PeriodicalId":50637,"journal":{"name":"Computer Networks","volume":"271 ","pages":"Article 111551"},"PeriodicalIF":4.6,"publicationDate":"2025-08-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144831035","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}