Computer NetworksPub Date : 2025-03-25DOI: 10.1016/j.comnet.2025.111239
Zheng Wan , Shenglu Zhao , Cuifang Wang , Yifeng Tan , Xiaogang Dong
{"title":"Joint DRL and GCN-based Cloud-Edge-End collaborative cache optimization for metaverse scenarios","authors":"Zheng Wan , Shenglu Zhao , Cuifang Wang , Yifeng Tan , Xiaogang Dong","doi":"10.1016/j.comnet.2025.111239","DOIUrl":"10.1016/j.comnet.2025.111239","url":null,"abstract":"<div><div>The rapid emergence of the Metaverse demands deeply immersive and highly responsive virtual experiences, necessitating ultra-high transmission speeds and extremely low latency. Centralized data processing methods are facing increasing constraints in managing large-scale user data due to limitations in computing power, storage capacity, and network bandwidth. To address these challenges, this paper presents a Cloud-Edge-End transmission framework for Metaverse scenarios aimed at optimizing resource allocation, reducing latency, and enhancing rendering efficiency. We propose a distributed trajectory prediction (DTP) algorithm and develop a distributed trajectory prediction cluster system that utilizes the FastDTW algorithm to predict trajectory segments and calculate subscene popularity. Additionally, a real-time collaborative cache optimization scheme (GCNAC), based on GCN and DRL, is introduced to dynamically adjust caching strategies according to subscene popularity, thereby improving cache hit rates and reducing cache replacement frequencies. Simulations demonstrate that the GCNAC scheme markedly outperforms existing methods in target subscene cache hit rates and transmission latency across varying capacities, achieving a 43.49% reduction in edge replacement frequency. This study provides a practical solution for terminal pre-rendering of Metaverse scenes, offering both theoretical support and practical guidance for the development of scene rendering technologies and the generative Metaverse.</div></div>","PeriodicalId":50637,"journal":{"name":"Computer Networks","volume":"263 ","pages":"Article 111239"},"PeriodicalIF":4.4,"publicationDate":"2025-03-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143706247","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}
Computer NetworksPub Date : 2025-03-24DOI: 10.1016/j.comnet.2025.111234
Hongyu Du , Shouhui Zhang , Xi Xv , Yimu Ji , Sisi Shao , Fei Wu , Shangdong Liu
{"title":"AF-MCDC: Active Feedback-Based Malicious Client Dynamic Detection","authors":"Hongyu Du , Shouhui Zhang , Xi Xv , Yimu Ji , Sisi Shao , Fei Wu , Shangdong Liu","doi":"10.1016/j.comnet.2025.111234","DOIUrl":"10.1016/j.comnet.2025.111234","url":null,"abstract":"<div><div>Federated Learning (FL) has long been known for separating the training and model construction processes, ensuring the privacy of participating clients. However, this separation also introduces a new attack surface. Due to the decentralization feature, Federal Learning is prone to Byzantine attacks. Attackers can deliberately corrupt or malfunction one or more participants in the federated network, disrupting the overall model training process. Researchers have proposed many defense mechanisms to mitigate Byzantine attacks. Their main ideas include eliminating malicious updates that deviate from the overall distribution through similarity detection and avoiding malicious parameters using statistical characteristics. Yet, these defense mechanisms are usually passive, detection only happens on the central server, neglecting the important role of clients. Thus we propose AF-MCDC: Active Feedback-Based Malicious Client Dynamic Detection, a byzantine-robust federated learning method taking advantage of valid clients. What sets AF-MCDC apart from existing robust federated learning methods is its three-pronged defense approach. First, a detection mechanism is deployed on each client to verify the integrity of the distributed global model. If the model fails the integrity check, it will not be used to initialize the local model. On the server side, a decision is made based on the detection results uploaded by the clients, followed by performance scoring using cosine similarity among federated clients. Finally, a dynamic weighting mechanism based on client score rankings is applied to weigh the local models uploaded by all clients, effectively filtering out malicious clients Evaluation of two datasets, MNIST and CIFAR-10, demonstrates that AF-MCDC is robust against a significant portion of malicious clients. Furthermore, even when over half of the clients are malicious, AF-MCDC can still train a global model with performance comparable to the global model learned by FedAvg under non-adversarial conditions.</div></div>","PeriodicalId":50637,"journal":{"name":"Computer Networks","volume":"263 ","pages":"Article 111234"},"PeriodicalIF":4.4,"publicationDate":"2025-03-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143706249","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}
Computer NetworksPub Date : 2025-03-24DOI: 10.1016/j.comnet.2025.111223
Yang Xu, Junhao Cheng, Hongli Xu, Changyu Guo, Yunming Liao, Zhiwei Yao
{"title":"Towards layer-wise quantization for heterogeneous federated clients","authors":"Yang Xu, Junhao Cheng, Hongli Xu, Changyu Guo, Yunming Liao, Zhiwei Yao","doi":"10.1016/j.comnet.2025.111223","DOIUrl":"10.1016/j.comnet.2025.111223","url":null,"abstract":"<div><div>Federated Learning (FL) has arisen to train deep learning models on massive private data, which are produced and possessed by geographically dispersed clients at the network edge. However, in edge computing scenarios, FL usually suffers from the constrained and heterogeneous communication resource. To achieve communication-efficient FL, we concentrate on the technique of model quantization. The existing researches in FL mainly perform model quantization at the grain of the entire model. However, according to our empirical analysis, when quantizing each layer of a model with the same quantization level, the amount of saved memory differs significantly across layers. Besides, the model exhibits different decreases in test accuracy when each layer is separately quantized to the same degree. To this end, we propose a more efficient and flexible Layer-wise Quantization scheme for FL, termed FedLQ. We further theoretically analyze the relationship between the convergence bound and the quantization level. Furthermore, considering that the quantization of each layer will yield different effects on the communication cost and model accuracy, we develop a joint metric (<em>i.e.</em>, layer significance) to evaluate the comprehensive influence of layer-wise quantization on model training, and design a significance-aware algorithm to determine adaptive layer-wise quantization levels for different clients. Extensive experiments in simulation environment illustrate that FedLQ is able to effectively reduce communication consumption while still achieving promising accuracy even with low-bit quantization. Compared to the baselines, FedLQ can achieve up to 5.77<span><math><mo>×</mo></math></span> speedup when reaching the target accuracy, or obtain at most 27% improvement in test accuracy under low-bits quantization scenarios.</div></div>","PeriodicalId":50637,"journal":{"name":"Computer Networks","volume":"264 ","pages":"Article 111223"},"PeriodicalIF":4.4,"publicationDate":"2025-03-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143738224","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}
Computer NetworksPub Date : 2025-03-23DOI: 10.1016/j.comnet.2025.111229
A. Alphiya , T. Latha
{"title":"An Efficient QC-LDPC channel encoder/decoder architecture with parallel vector-matrix computations for 5G wireless networks on FPGA","authors":"A. Alphiya , T. Latha","doi":"10.1016/j.comnet.2025.111229","DOIUrl":"10.1016/j.comnet.2025.111229","url":null,"abstract":"<div><div>The Consultative Committee for Space Data Systems (CCSDS) has selected quasi-cyclic low-density parity-check (QC-LDPC) codes to enhance error correction performance in diverse wireless communication systems. The growing use of QC-LDPC codes necessitates a low-latency and low-complexity encoder/decoder architecture that may be implemented in practical baseband chips. However, the storage of information bit widths with multiple expansion factors would deteriorate the encoder's throughput and adaptability in a QC-LDPC encoding process. In order to compress the expansion factors and improve the encoder's reconfiguration capability by minimizing the information bits width, a Field Programmable Gate Array (FPGA) enabled QC-LDPC encoder with an information bits reordering mechanism (IBRM) is developed in this study. Additionally, the complexity of calculating the parity check matrix (PCM) is reduced by utilizing the parallel vector-matrix computations (PVMC) method. The decoding process is then carried out by a new logarithmic likelihood-ratio (LLR) tagging method-enabled decoder architecture, which reduces the computational complexity of the suggested decoder architecture. Furthermore, the basic addition and subtraction operations are carried out by a reconfigurable unified adder and subtractor unit (RUAS) that reduces the overall complexity of the decoder architecture. Finally, using Xilinx Verilog coding and Matlab, the VLSI architectures of the proposed encoder/decoder are implemented in FPGA. The analysis of the results shows that the proposed encoder architecture achieves a throughput of 11.974 Gbps and a clock frequency (CF) of 455 .913MHz. Likewise, the decoder consumes 0.002193 <span><math><mrow><mo>(</mo><mrow><mi>μ</mi><mi>s</mi></mrow><mo>)</mo></mrow></math></span> latency to complete the decoding process.</div></div>","PeriodicalId":50637,"journal":{"name":"Computer Networks","volume":"264 ","pages":"Article 111229"},"PeriodicalIF":4.4,"publicationDate":"2025-03-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143768780","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}
Computer NetworksPub Date : 2025-03-22DOI: 10.1016/j.comnet.2025.111187
Elham Moeinaddini , Eslam Nazemi , Amin Shahraki
{"title":"A new approach on self-adaptive trust management for social Internet of Things","authors":"Elham Moeinaddini , Eslam Nazemi , Amin Shahraki","doi":"10.1016/j.comnet.2025.111187","DOIUrl":"10.1016/j.comnet.2025.111187","url":null,"abstract":"<div><div>The emergence of the Social Internet of Things (SIoT) represents a significant evolution of the traditional Internet of Things (IoT) paradigm. While earlier IoT systems prioritized the integration of intelligent devices, the SIoT paradigm shifts the focus toward enabling social interaction and collaboration among connected entities. Within the SIoT, various nodes can autonomously establish social connections to access the desired services. Given the dynamic and decentralized nature of this open environment, effective trust management becomes crucial. In recent years, machine learning (ML) techniques have made significant progress in enhancing trust computing within the SIoT ecosystem. However, challenges such as scalability, dynamic behavior, and device resource limitations continue to pose difficulties. This study introduces SATM-SIoT, a decentralized self-adaptive trust management model for SIoT that integrates the MAPE-K (Monitor, Analyze, Plan, Execute, Knowledge) control loop. By adjusting thresholds based on perceived hostile behavior, this model effectively identifies malicious devices, enhancing adaptivity in trust management. To address resource limitations and ensure scalability, our model assigns trust evaluation tasks to Fog nodes, leveraging their distributed computational power. We utilize ML techniques, specifically multi-layer perceptron (MLP), to analyze device behavior and assess trustworthiness. Federated learning (FL) enhances local ML models, enabling collaborative learning and incremental updates based on new data to adapt to the dynamics of SIoT. The primary goal of SATM-SIoT is to improve overall trustworthiness and accurately detect malicious devices while tackling challenges related to scalability, dynamic behavior, and resource constraints. We conducted experiments on a simulated SIoT network to validate our model. The results show that SATM-SIoT effectively identifies almost all malicious devices in a SIoT network. Utilizing the FL technique and MAPE-K loop led to an average increase in Success Rate of 7% and 9%, respectively.</div></div>","PeriodicalId":50637,"journal":{"name":"Computer Networks","volume":"263 ","pages":"Article 111187"},"PeriodicalIF":4.4,"publicationDate":"2025-03-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143706250","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}
Computer NetworksPub Date : 2025-03-22DOI: 10.1016/j.comnet.2025.111215
Wenjie Zhou, Linbo Zhai, Zekun Lu, Kai Xue, Tian Zhang
{"title":"Latency minimization in IRS-UAV assisted WPT-MEC systems: An ID-AOPDDQN-based trajectory and phase shift optimization approach","authors":"Wenjie Zhou, Linbo Zhai, Zekun Lu, Kai Xue, Tian Zhang","doi":"10.1016/j.comnet.2025.111215","DOIUrl":"10.1016/j.comnet.2025.111215","url":null,"abstract":"<div><div>Intelligent Reflectors (IRS) assisted Wireless Power Transmission and Mobile Edge Computing (WPT-MEC) are considered as solutions for implementing sustainable Internet of Things (IoT) networks that can effectively improve network performance and reduce data transmission latency. There are still challenges such as flexible deployment of IRSs and multivariate joint optimization remain. In this paper, we study the task offloading problem of unmanned aerial vehicles (UAVs) carrying IRSs (IRS-UAV) assisted WPT-MEC, which exploits the flexibility of the UAV to dynamically improve the energy harvesting and task offloading channel transmission between ground equipment (GD) and access point (AP). In this system, we consider the association relationship between the hovering points (HPs) of the IRS-UAV and the GDs, the phase shift of the IRS-UAV, the flight trajectory, the beamforming vector, and the offloading decision and transmit power of the GDs to minimize latency performance under the constraint of energy consumption. To solve this multivariable non-convex problem, we propose an ID-AOPDDQN (ISODATA clustering, successive convex approximation and parametric Dueling deep <span><math><mi>Q</mi></math></span>-network) algorithm. At first, we cluster the HPs and the association relationship between GDs and HPs through an efficient load balancing algorithm (ISODATA), so as to cover all GDs to the maximum extent. Secondly, on the basis of clustering, we divide the target problem into two sub-problems. For the first sub-problem, Successive Convex Approximation (SCA) is used to transform it into a convex problem, and the phase shift and beamforming vectors of radio energy transmission are optimized alternately. For the second subproblem, we design the PDDQN (a combination of DDPG and Dueling DQN) algorithm to process the mixed space based on the first problem of the solution, where DDPG processes continuous motion (such as phase shift) and Dueling DQN processes discrete action (such as offloading decisions). Simulation results show that the ID-AOPDDQN algorithm significantly improves the performance of the system in latency.</div></div>","PeriodicalId":50637,"journal":{"name":"Computer Networks","volume":"263 ","pages":"Article 111215"},"PeriodicalIF":4.4,"publicationDate":"2025-03-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143715571","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 traffic engineering optimization for segment routing: A recommendation perspective","authors":"Linghao Wang, Miao Wang, Chungang Lin, Yujun Zhang","doi":"10.1016/j.comnet.2025.111224","DOIUrl":"10.1016/j.comnet.2025.111224","url":null,"abstract":"<div><div>Traffic engineering (TE) is important for improving network performance. Recently, segment routing (SR) has gained increasing attention in the TE field. Many segment routing traffic engineering (SR-TE) methods compute optimal routing policies by solving linear programming (LP) problems, which suffer from high computation time. Therefore, various methods have been proposed for accelerating TE optimization. However, prior methods solve individual TE optimization problems from scratch, overlooking valuable information from existing historical solutions. We argue that these data can imply the distribution of optimal solutions for solving future TE problems. In this paper, we provide a new perspective on accelerating SR-TE optimization. First, we generated and analyzed historical solutions of a widely used LP model, and revealed two key findings from the data: Flows are predominantly routed through a small subset of intermediate nodes; similar decisions can be made for some flows. Then, inspired by the findings, we propose RS4SR, the first framework to our knowledge leveraging historical solutions for SR-TE acceleration. It can significantly reduce the size of LP model by performing candidate recommendation and flow clustering. Experiments on real-world topologies and various traffic matrices demonstrate that a simple implementation of RS4SR is sufficient to obtain near-optimal solutions within the time limit of two seconds on large-scale networks, utilizing a small number of historical solutions.</div></div>","PeriodicalId":50637,"journal":{"name":"Computer Networks","volume":"264 ","pages":"Article 111224"},"PeriodicalIF":4.4,"publicationDate":"2025-03-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143738225","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}
Computer NetworksPub Date : 2025-03-22DOI: 10.1016/j.comnet.2025.111198
Xiaolong Xu , Xuanhong Zhou , Muhammad Bilal , Sherali Zeadally , Jon Crowcroft , Lianyong Qi , Shengjun Xue
{"title":"Socially beneficial metaverse: Framework, technologies, applications, and challenges","authors":"Xiaolong Xu , Xuanhong Zhou , Muhammad Bilal , Sherali Zeadally , Jon Crowcroft , Lianyong Qi , Shengjun Xue","doi":"10.1016/j.comnet.2025.111198","DOIUrl":"10.1016/j.comnet.2025.111198","url":null,"abstract":"<div><div>In recent years, the maturation of emerging technologies such as Virtual Reality, Digital Twins and Blockchain has accelerated the realization of the metaverse. As a virtual world independent of the real world, the metaverse will provide users with a variety of virtual activities which bring great convenience to society. In addition, the metaverse can facilitate digital twins, which offers transformative possibilities for the industry. Thus, the metaverse has attracted the attention of the industry, and a huge amount of capital is about to be invested. However, the development of the metaverse is still in its infancy and little research has been undertaken so far. We describe the development of the metaverse. Next, we introduce the architecture of the socially beneficial metaverse (SB-Metaverse) and we focus on the technologies that support the operation of SB-Metaverse. In addition, we also present the applications of SB-Metaverse. Finally, we discuss several challenges faced by SB-Metaverse which must be addressed in the future.</div></div>","PeriodicalId":50637,"journal":{"name":"Computer Networks","volume":"263 ","pages":"Article 111198"},"PeriodicalIF":4.4,"publicationDate":"2025-03-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143706251","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}
Computer NetworksPub Date : 2025-03-22DOI: 10.1016/j.comnet.2025.111207
Jiqiang Zhai, Xinyu Wang, Zhonghui Zhai, Tao Xu, Zuming Qi, Hailu Yang
{"title":"Industrial IoT intrusion attack detection based on composite attention-driven multi-layer pyramid features","authors":"Jiqiang Zhai, Xinyu Wang, Zhonghui Zhai, Tao Xu, Zuming Qi, Hailu Yang","doi":"10.1016/j.comnet.2025.111207","DOIUrl":"10.1016/j.comnet.2025.111207","url":null,"abstract":"<div><div>The Industrial Internet of Things (IIoT) extends and optimizes IoT technology for industrial environments, playing a crucial role in industrial production, equipment monitoring, and supply chain management. However, the increasing diversity of devices at the IIoT application layer exacerbates network complexity, rendering IIoT systems more susceptible to malicious attacks and severe security risks. To address these challenges, we focus on unresolved security issues in the IIoT application layer, including poor generalization ability across different domains in detection, insufficient granularity in local feature recognition, and suboptimal performance in identifying diverse attack patterns. In response, we propose a Composite Attention-Driven Multi-Layer Pyramid Feature-Based Intrusion Detection Model (BCSP), which leverages a composite attention pyramid structure with a multi-scale attention mechanism to enhance semantic feature representation across different scales. This design enables the model to prioritize contextual semantic information while effectively capturing real-time traffic attributes and session-related features. To validate its effectiveness, we conduct extensive experiments using well-established public cybersecurity datasets and real-world network environments, where BCSP achieves a test accuracy of over 98%. Experimental results indicate that BCSP consistently outperforms conventional machine learning and deep learning models, demonstrating its effectiveness in IIoT intrusion detection.</div></div>","PeriodicalId":50637,"journal":{"name":"Computer Networks","volume":"263 ","pages":"Article 111207"},"PeriodicalIF":4.4,"publicationDate":"2025-03-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143687197","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}
Computer NetworksPub Date : 2025-03-22DOI: 10.1016/j.comnet.2025.111226
Na Fan , Yuxin Gao , Jialong Li , Zhiquan Liu , Wenjun Fan
{"title":"Multi-Attack Identification and Mitigation mechanism based on multi-agent collaboration in Vehicular Named Data Networking","authors":"Na Fan , Yuxin Gao , Jialong Li , Zhiquan Liu , Wenjun Fan","doi":"10.1016/j.comnet.2025.111226","DOIUrl":"10.1016/j.comnet.2025.111226","url":null,"abstract":"<div><div>This paper introduces a novel Multi-Attack Identification and Mitigation mechanism (MAIM) designed to enhance security within Vehicular Name Data Networking (VNDN), a derivative of Name Data Networking (NDN) optimized for the Internet of Vehicles (IoV). VNDN, while offering improved communication security for mobile networks, is vulnerable to interest flooding attacks. MAIM addresses this issue through a collaborative multi-agent system comprising detection algorithms, an identification model, and a mitigation model. The MAIM mechanism begins with vehicle nodes monitoring traffic and identifying potential threats, relaying this information to Road Side Units (RSUs), which utilize Random Forests to detect attacks. Detected threats are then communicated to the Base Station (BS), which employs Convolutional Neural Networks and Support Vector Machines to analyze and classify the attack type. The RSUs, informed by the BS, use Graph Convolution Networks to isolate malicious nodes, effectively mitigating the attack. Comparative simulation and real-world experiments demonstrate MAIM’s superior performance in attack recognition and mitigation, the average accuracy for attack detection is 97.5%, the average accuracy for attack identification reaches 85.2%, while the average interest satisfaction rate under attack suppression stands at 81%, highlighting its potential as a robust solution for securing VNDN environments.</div></div>","PeriodicalId":50637,"journal":{"name":"Computer Networks","volume":"263 ","pages":"Article 111226"},"PeriodicalIF":4.4,"publicationDate":"2025-03-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143697847","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}