Computer Networks最新文献_第8页

A botnet detection method for encrypted DNS traffic based on multi-branch knowledge distillation 一种基于多分支知识蒸馏的加密DNS流量僵尸网络检测方法

IF 4.6 2区计算机科学

Computer Networks Pub Date : 2026-03-01 Epub Date: 2026-01-28 DOI: 10.1016/j.comnet.2026.112060

Zhipeng Qin , Hanbing Yan , Xiangyu Li , Peng Wang

{"title":"A botnet detection method for encrypted DNS traffic based on multi-branch knowledge distillation","authors":"Zhipeng Qin , Hanbing Yan , Xiangyu Li , Peng Wang","doi":"10.1016/j.comnet.2026.112060","DOIUrl":"10.1016/j.comnet.2026.112060","url":null,"abstract":"<div><div>With advancements in encrypted network communication technologies, botnets increasingly use encrypted DNS traffic to spread covertly and execute attacks. Botnet traffic exhibits diverse and complex behaviors, and detecting botnets within encrypted DNS traffic poses challenges, such as high concealment, low detection efficiency, and difficulties in feature matching. To address these issues, this paper proposes a botnet detection method for encrypted DNS traffic based on multi-branch knowledge distillation. This method utilizes an adaptive feature extraction algorithm to capture encrypted DNS traffic features, applies spatial clustering based on traffic characteristics for multi-classification of botnets, and adopts a multi-level knowledge distillation strategy to develop several specialized botnet detection models. These models operate in parallel, enhancing detection efficiency and accuracy. Experimental results demonstrate that this approach significantly reduces computational complexity while maintaining high precision, improving detection efficiency and real-time capabilities.</div></div>","PeriodicalId":50637,"journal":{"name":"Computer Networks","volume":"277 ","pages":"Article 112060"},"PeriodicalIF":4.6,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146090329","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 explainable transformer-based model for phishing email detection: A large language model approach 用于网络钓鱼电子邮件检测的可解释的基于转换器的模型：大型语言模型方法

IF 4.6 2区计算机科学

Computer Networks Pub Date : 2026-03-01 Epub Date: 2026-01-28 DOI: 10.1016/j.comnet.2026.112061

Mohammad Amaz Uddin , Md Mahiuddin , Iqbal H. Sarker

{"title":"An explainable transformer-based model for phishing email detection: A large language model approach","authors":"Mohammad Amaz Uddin , Md Mahiuddin , Iqbal H. Sarker","doi":"10.1016/j.comnet.2026.112061","DOIUrl":"10.1016/j.comnet.2026.112061","url":null,"abstract":"<div><div>Phishing email is a serious cyber threat that tries to deceive users by sending false emails with the intention of stealing confidential information or causing financial harm. Attackers, often posing as trustworthy entities, exploit technological advancements and sophistication to make the detection and prevention of phishing more challenging. Despite extensive academic research, phishing detection remains an ongoing and formidable challenge in the cybersecurity landscape. In this research paper, we present a fine-tuned transformer-based masked language model, RoBERTa (Robustly Optimized BERT Pretraining Approach), for phishing email detection. In the detection process, we employ a phishing email dataset and apply the preprocessing techniques to clean and address the class imbalance issues, thereby enhancing model performance. The results of the experiment demonstrate that our fine-tuned model outperforms traditional machine learning models with an accuracy of 98.45%. To ensure model transparency and user trust, we propose a hybrid explanation approach, LITA (LIME-Transformer Attribution), which integrates the potential of Local Interpretable Model-Agnostic Explanations (LIME) and Transformers Interpret methods. The proposed method provides more consistent and user-friendly insights, mitigating local attribution inconsistencies between the two explanation approaches. Moreover, the study highlights the model’s ability to generate its predictions by presenting positive and negative contribution scores using LIME, Transformers Interpret, and LITA.</div></div>","PeriodicalId":50637,"journal":{"name":"Computer Networks","volume":"277 ","pages":"Article 112061"},"PeriodicalIF":4.6,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146090331","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

Hierarchical federated learning algorithm with synchronous and asynchronous aggregation collaboration in internet of vehicles 基于同步和异步聚合协作的车联网分层联邦学习算法

IF 4.6 2区计算机科学

Computer Networks Pub Date : 2026-03-01 Epub Date: 2026-01-15 DOI: 10.1016/j.comnet.2026.111994

Ling Xing, Jingjing Cui, Kaikai Deng, Honghai Wu, Huahong Ma

{"title":"Hierarchical federated learning algorithm with synchronous and asynchronous aggregation collaboration in internet of vehicles","authors":"Ling Xing, Jingjing Cui, Kaikai Deng, Honghai Wu, Huahong Ma","doi":"10.1016/j.comnet.2026.111994","DOIUrl":"10.1016/j.comnet.2026.111994","url":null,"abstract":"<div><div>Federated Learning (FL) has emerged as a key distributed framework for processing real-time vehicular data in the Internet of Vehicles (IoV), aiming to enhance driving experiences and service quality. However, challenges such as network congestion, limited signal coverage leading to latency, and inefficiencies arising from data and device heterogeneity hinder FL performance in IoV environments. To address these issues, we propose <em>HierFed</em>, a hierarchical federated learning algorithm that combines synchronous and asynchronous aggregation strategies, which comprises two components: (i) an <em>asynchronous cache aggregation mechanism</em>, where client-side cache updates are transmitted to the edge server for asynchronous aggregation. By calibrating the relationship between cache updates and model deviations, clients can update models without synchronous waiting, thereby mitigating latency; (ii) a <em>synchronous aggregation mechanism</em>, where the cloud server optimizes global model updates by computing aggregation weights under gradient normalization and constraint conditions, followed by reducing inter-client model inconsistency and enhances overall model convergence. Experimental results show that <em>HierFed</em> outperforms all baselines, achieving an 8.83% improvement in model accuracy at the edge tier (asynchronous FL), a 13.84% improvement at the cloud tier (synchronous FL), and a 10.79% overall improvement in hierarchical FL.</div></div>","PeriodicalId":50637,"journal":{"name":"Computer Networks","volume":"277 ","pages":"Article 111994"},"PeriodicalIF":4.6,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146039820","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

PathWeaver: Enhancing LEO satellite networks resilience via topology design and randomized routing against link flooding attacks PathWeaver：通过拓扑设计和针对链路泛洪攻击的随机路由增强LEO卫星网络的弹性

IF 4.6 2区计算机科学

Computer Networks Pub Date : 2026-03-01 Epub Date: 2026-01-30 DOI: 10.1016/j.comnet.2026.112067

Tao Hu , Luxin Bai , Zinuo Yin

{"title":"PathWeaver: Enhancing LEO satellite networks resilience via topology design and randomized routing against link flooding attacks","authors":"Tao Hu , Luxin Bai , Zinuo Yin","doi":"10.1016/j.comnet.2026.112067","DOIUrl":"10.1016/j.comnet.2026.112067","url":null,"abstract":"<div><div>Low Earth Orbit (LEO) Satellite Networks (LSNs) are becoming potential targets for cyberattacks. Their publicly available constellation topology and nearly deterministic routing algorithms expose a larger attack surface to link flooding attack (LFA) adversaries. Existing active defense methods primarily focus on single-faceted optimizations, such as topology obfuscation or resilient routing, which are not only difficult to apply to large-scale LSNs but also come at the cost of network performance. To this end, this paper introduces an active defense scheme called PathWeaver, which integrates dynamic topology design and randomized routing strategies to enhance LFA attack costs while ensuring network performance (latency). At the topology level, PathWeaver limits each satellite to a uniform +Grid connection model, reducing the global topology design problem to a limited search within the local view of individual satellites. Under a given traffic matrix, the optimal design is selected to achieve load balancing. At the routing level, leveraging the grid-like topology structure of the +Grid connection model, PathWeaver constructs a minimal-hop region for given source-destination pairs and introduces a randomized routing algorithm to determine the next hop randomly, increasing the uncertainty of forwarding paths while maintaining bounded end-to-end latency. Simulation results show that compared to existing methods, PathWeaver reduces the maximum link utilization (MLU) by up to 25.54% across different constellation sizes. Taking the attack cost under shortest-path routing as a baseline, PathWeaver increases the attack cost by up to 523%, enhancing network resilience and offering a new paradigm for the security design of LSNs.</div></div>","PeriodicalId":50637,"journal":{"name":"Computer Networks","volume":"277 ","pages":"Article 112067"},"PeriodicalIF":4.6,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147398081","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

BVPM: A blockchain-based scheme for secure and efficient vehicle platoon management BVPM：基于区块链的安全高效车辆排管理方案

IF 4.6 2区计算机科学

Computer Networks Pub Date : 2026-03-01 Epub Date: 2026-01-10 DOI: 10.1016/j.comnet.2026.112009

Piaoxiong Huang , Lin Wan , Zhiquan Liu , Libo Wang , Heng Sun , Feixiang Ye , Jianfeng Ma

{"title":"BVPM: A blockchain-based scheme for secure and efficient vehicle platoon management","authors":"Piaoxiong Huang , Lin Wan , Zhiquan Liu , Libo Wang , Heng Sun , Feixiang Ye , Jianfeng Ma","doi":"10.1016/j.comnet.2026.112009","DOIUrl":"10.1016/j.comnet.2026.112009","url":null,"abstract":"<div><div>Blockchain technology has considerable potential to enhance vehicle platoon management by providing robust security guarantees. Nevertheless, existing blockchain-based platoon management schemes suffer from inefficiency and poor scalability, imposing excessive computation and communication overheads on the resource-constrained vehicle side. To overcome these limitations, in this paper, we propose a Blockchain-based scheme for secure and efficient Vehicle Platoon Management (BVPM). Specifically, we introduce an enhanced Practical Byzantine Fault Tolerance algorithm with Collective Signing (CS-PBFT), which leverages a communication tree structure and Schnorr multisignatures to improve blockchain performance. In addition, we propose a distributed voting algorithm for leader election based on the Subpath Signature and Raft (SS-Raft) mechanism, where a novel subpath signature quantifies path similarity among vehicles. Security analysis demonstrates that the BVPM scheme provides robust security for platoon operations against several common attacks. Extensive simulations further reveal that, compared with the existing schemes, the BVPM scheme significantly reduces computation and communication overheads.</div></div>","PeriodicalId":50637,"journal":{"name":"Computer Networks","volume":"277 ","pages":"Article 112009"},"PeriodicalIF":4.6,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145981893","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

ABO For anomaly detection and computer network optimization ABO用于异常检测和计算机网络优化

IF 4.6 2区计算机科学

Computer Networks Pub Date : 2026-03-01 Epub Date: 2026-01-02 DOI: 10.1016/j.comnet.2025.111911

Grace Mupoyi Ntuala , Qi Xia , Qiufang Li , Jiaqin Liu , Patrick Mukala , Ansu Badjie , Lan Ma , Hu Xia , Jianbin Gao

{"title":"ABO For anomaly detection and computer network optimization","authors":"Grace Mupoyi Ntuala , Qi Xia , Qiufang Li , Jiaqin Liu , Patrick Mukala , Ansu Badjie , Lan Ma , Hu Xia , Jianbin Gao","doi":"10.1016/j.comnet.2025.111911","DOIUrl":"10.1016/j.comnet.2025.111911","url":null,"abstract":"<div><div>Blockchain networks face critical challenges in anomaly detection, scalability, and resilience under adversarial attacks. Existing solutions often lack integrated approaches that combine time-domain and frequency-domain analyses, failing to detect periodic patterns and provide real-time correction mechanisms. This paper introduces ABO (Advanced Blockchain Optimization), a novel framework that integrates Eigenvalue Theory, Discrete Fourier Transform (DFT), and Gaussian Mixture Models (GMM) for comprehensive anomaly detection, correction, and network optimization. ABO employs DFT for frequency-domain analysis to uncover hidden periodic transaction patterns, GMM for probabilistic real-time anomaly scoring, and Eigenvalue Theory to model transaction dependencies and identify critical nodes. A key innovation is the adaptive anomaly correction mechanism that recalibrates transaction flows, reallocates resources, and isolates malicious nodes to restore normal operations. Additionally, Fatou’s Lemma provides rigorous long-term transaction volume estimation for optimal resource allocation. Extensive experiments on real-world Ethereum datasets (70 million transactions) demonstrate that ABO achieves superior performance: 95.5% average detection accuracy (6% improvement over CNN/GRU/LSTM baselines), 44.2ms detection time (45% faster), 8460 tx/s throughput (98% higher), and 3.58J energy consumption (49% lower). Under adversarial attack scenarios including evasion, poisoning, structural, perturbation, and model-specific attacks, ABO maintains robust detection accuracy above 94% across all attack types. These results confirm that ABO provides a scalable, resilient, and efficient solution for blockchain security and optimization in large-scale deployments.</div></div>","PeriodicalId":50637,"journal":{"name":"Computer Networks","volume":"277 ","pages":"Article 111911"},"PeriodicalIF":4.6,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145981826","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

STEAM: Securing the connections in wireless sensor networks with varying target priority STEAM：确保无线传感器网络中具有不同目标优先级的连接

IF 4.6 2区计算机科学

Computer Networks Pub Date : 2026-03-01 Epub Date: 2026-01-14 DOI: 10.1016/j.comnet.2026.112024

Truong Q. Vu , Minh T. Trinh , Hien Q. Ta , Hanh T. Nguyen , Binh T.T. Huynh

{"title":"STEAM: Securing the connections in wireless sensor networks with varying target priority","authors":"Truong Q. Vu , Minh T. Trinh , Hien Q. Ta , Hanh T. Nguyen , Binh T.T. Huynh","doi":"10.1016/j.comnet.2026.112024","DOIUrl":"10.1016/j.comnet.2026.112024","url":null,"abstract":"<div><div>Achieving multi-connections in wireless sensor networks (WSNs) has been drawing attention from the research community. This paper addresses the problem of providing secure and reliable multi-connectivity in wireless sensor networks (WSNs) when different targets require different connectivity priorities and security levels. Existing multi-connectivity approaches do not incorporate secured communication constraints, operate only in simplified 2D settings, or employ high-complexity algorithms. We formalize the <em>Q</em>-Secure Connectivity problem, which extends <em>Q</em>-Connectivity by enforcing target-dependent secure communication ranges to prevent eavesdropping. To solve this problem efficiently, we propose Sensor Tree Establishment and Adjacency Mapping (STEAM), a novel graph-based heuristic that minimizes the number of relay nodes while ensuring node-disjoint and secure paths. STEAM builds a minimum spanning tree over targets and applies a Hungarian-based mapping to construct feasible connections. Experiments on realistic 3D datasets show that STEAM significantly reduces relay-node count by 3 times and computation time by 15 times, and achieves 30% lower energy consumption, 20% lower outage probability, 40% lower latency, and 50% higher throughput compared to the state-of-the-art method. This work provides the first practical solution to <em>Q</em>-Secure Connectivity in realistic 3D environments.</div></div>","PeriodicalId":50637,"journal":{"name":"Computer Networks","volume":"277 ","pages":"Article 112024"},"PeriodicalIF":4.6,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146039290","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

Throughput maximization for UAV-assisted IoT with NOMA 无人机辅助物联网与NOMA的吞吐量最大化

IF 4.6 2区计算机科学

Computer Networks Pub Date : 2026-03-01 Epub Date: 2026-01-14 DOI: 10.1016/j.comnet.2026.111999

Tianyi Zheng , Mengping Zhong , Zhe Sun , Jihan Feng , Xin Liu

引用次数: 0

O-RAN Architecture-based distributed learning framework for multi-RIS-aided vehicular networks 基于O-RAN架构的多ris辅助车辆网络分布式学习框架

IF 4.6 2区计算机科学

Computer Networks Pub Date : 2026-02-01 Epub Date: 2025-12-20 DOI: 10.1016/j.comnet.2025.111940

Madyan Alsenwi, Mehran Abolhasan, Justin Lipman

{"title":"O-RAN Architecture-based distributed learning framework for multi-RIS-aided vehicular networks","authors":"Madyan Alsenwi, Mehran Abolhasan, Justin Lipman","doi":"10.1016/j.comnet.2025.111940","DOIUrl":"10.1016/j.comnet.2025.111940","url":null,"abstract":"<div><div>This paper explores the utilization of Reconfigurable Intelligent Surfaces (RISs) within multi-cell vehicular open radio access networks to redirect signals toward users with wireless link blockages. To address this, a stochastic optimization problem is formulated to determine the optimal transmit precoding for Road Side Units (RSUs) and adjust the phase shifts of corresponding RISs, considering random obstacles in wireless links. The objective is to enhance long-term data throughput while guaranteeing the Quality of Service (QoS) for each user. To solve the problem, a federated learning framework is introduced that employs multi-agent Deep Reinforcement Learning (DRL) and aligns with the O-RAN architecture. Specifically, deep learning agents are deployed at network edge servers, integrated within the Near-Real-Time Radio Access Network Intelligent Controller (Near-RT RIC), where they gather network information, train local models, and perform online executions. A global model is constructed in the Non-Real-Time RIC, which resides on a central server, by aggregating the local models received from edge servers. Simulation results confirm that the proposed method significantly improves average network data rates while ensuring users receive adequate link quality.</div></div>","PeriodicalId":50637,"journal":{"name":"Computer Networks","volume":"275 ","pages":"Article 111940"},"PeriodicalIF":4.6,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145839797","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

Thales: An orientation-aware AS embedding for anomaly detection in dynamic BGP network 动态BGP网络异常检测的方向感知AS嵌入

IF 4.6 2区计算机科学

Computer Networks Pub Date : 2026-02-01 Epub Date: 2025-12-20 DOI: 10.1016/j.comnet.2025.111947

Zheng Wu , Yaoyu Zhou , Shiyu Liu , Zulong Diao , Yanmeng Wang , Fu Xiao

{"title":"Thales: An orientation-aware AS embedding for anomaly detection in dynamic BGP network","authors":"Zheng Wu , Yaoyu Zhou , Shiyu Liu , Zulong Diao , Yanmeng Wang , Fu Xiao","doi":"10.1016/j.comnet.2025.111947","DOIUrl":"10.1016/j.comnet.2025.111947","url":null,"abstract":"<div><div>Although the Border Gateway Protocol (BGP) is a critical protocol for Internet routing, its lack of robust security mechanisms leaves it vulnerable to anomalies. Current detection methods struggle to balance accuracy and computational efficiency in BGP’s complex and dynamic environment. To address these challenges, we propose Thales, a path encoding-based approach that focuses on deriving accurate and semantically meaningful AS embeddings from BGP data. Our method consists of three key components. First, by leveraging the hierarchical characteristics of the BGP network, we design an orientation-aware node embedding model that captures both vertical (between provider and customer ASes) and horizontal (between peer ASes) structural information for each AS. Second, to adapt to the evolving network topology, we introduce a spectral energy-triggered update mechanism that responds to structural perturbations in BGP update streams. Third, we design a lightweight detection framework that utilizes these embeddings to identify anomalies with low latency while enabling root-cause analysis through AS path comparisons. We evaluate Thales on both of the closed and open datasets. Experimental results show that Thales detects all known anomaly events with less false alarms while achieving the fastest training and inference times among baseline methods. Moreover, it is able to provide interpretable analysis by computing the differences between AS paths to locate the root causes.</div></div>","PeriodicalId":50637,"journal":{"name":"Computer Networks","volume":"275 ","pages":"Article 111947"},"PeriodicalIF":4.6,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145839793","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