Pub Date : 2025-02-07DOI: 10.1016/j.jnca.2025.104128
Xuan-Ha Nguyen, Kim-Hung Le
The rapid evolution of cyberattack techniques necessitates advanced intrusion detection systems (IDS) capable of multiclass novelty detection (MND), accurately classifying known attacks while identifying novel ones. Despite numerous successful studies focused on multi-class attack classification or novel attack detection separately, a significant research gap remains in achieving the effective MND for IDS. In this paper, we introduce the neighbour null Foley–Sammon transformation (nNFST), a novel single-model algorithm designed to address the MND challenge in IDS. nNFST employs a novel technique based on the inverse nearest neighbour algorithm to compute within-class and between-class variation. This technique preserves both the local distribution structure within each class and the global distribution structure across classes, thereby mitigating the impact of singular points on the algorithm and enhancing accuracy on complex data. Furthermore, nNFST leverages the kernel trick to improve detection accuracy and sparse matrix multiplication to reduce training costs. Comprehensive evaluation results on four public datasets demonstrate nNFST’s superior performance compared to related works in different tasks, achieving 97.12% to 99.56% accuracy in multiclass classification tasks, 94.53% to 99.33% accuracy in novel attack detection tasks, and a 0.825 to 0.975 Matthews correlation coefficient in MND tasks. These results highlight nNFST’s potential to significantly enhance IDS capabilities by concurrently classifying known attacks and identifying unknown attacks.
{"title":"nNFST: A single-model approach for multiclass novelty detection in network intrusion detection systems","authors":"Xuan-Ha Nguyen, Kim-Hung Le","doi":"10.1016/j.jnca.2025.104128","DOIUrl":"10.1016/j.jnca.2025.104128","url":null,"abstract":"<div><div>The rapid evolution of cyberattack techniques necessitates advanced intrusion detection systems (IDS) capable of multiclass novelty detection (MND), accurately classifying known attacks while identifying novel ones. Despite numerous successful studies focused on multi-class attack classification or novel attack detection separately, a significant research gap remains in achieving the effective MND for IDS. In this paper, we introduce the neighbour null Foley–Sammon transformation (nNFST), a novel single-model algorithm designed to address the MND challenge in IDS. nNFST employs a novel technique based on the inverse nearest neighbour algorithm to compute within-class and between-class variation. This technique preserves both the local distribution structure within each class and the global distribution structure across classes, thereby mitigating the impact of singular points on the algorithm and enhancing accuracy on complex data. Furthermore, nNFST leverages the kernel trick to improve detection accuracy and sparse matrix multiplication to reduce training costs. Comprehensive evaluation results on four public datasets demonstrate nNFST’s superior performance compared to related works in different tasks, achieving 97.12% to 99.56% accuracy in multiclass classification tasks, 94.53% to 99.33% accuracy in novel attack detection tasks, and a 0.825 to 0.975 Matthews correlation coefficient in MND tasks. These results highlight nNFST’s potential to significantly enhance IDS capabilities by concurrently classifying known attacks and identifying unknown attacks.</div></div>","PeriodicalId":54784,"journal":{"name":"Journal of Network and Computer Applications","volume":"236 ","pages":"Article 104128"},"PeriodicalIF":7.7,"publicationDate":"2025-02-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143377184","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}
Pub Date : 2025-02-05DOI: 10.1016/j.jnca.2025.104134
Murat Bakirci
Aerial monitoring assumes a pivotal role within the domain of Intelligent Transportation Systems (ITS), imparting invaluable data and discernments that ameliorate the efficacy, security, and holistic operability of transportation networks. Image processing, encompassing the derivation of valuable insights through the manipulation of visual data captured by imaging apparatus, resides at the core and is poised to establish a firm footing in forthcoming ITS applications. In this context, numerous machine learning methodologies have been devised to enhance image processing, with novel approaches continually emerging. YOLOv8 emerged earlier this year and is still in the process of assimilating its potential application within the domain of ITS. In this study, a comprehensive assessment was conducted on all constituent variants of YOLOv8, specifically within the context of its application in the domain of aerial traffic monitoring. Using a custom-modified commercial drone, extensive datasets were acquired encompassing a diverse range of flight scenarios and traffic dynamics. To optimize model performance, meticulous consideration was given to ensuring dataset inclusivity, encompassing the full spectrum of vehicular typologies, while maintaining a homogeneous structure that accommodates an array of environmental nuances, including illumination and shading variations. The outcomes evince that both YOLOv8l and YOLOv8x exhibit notable superiority over other variants, manifesting exceptional detection efficacy even amid high-density traffic scenarios and the presence of obstructive elements. Contrastingly, in comparison to earlier iterations of YOLO, the current models demonstrate heightened precision in vehicle classification, yielding a reduction in misclassification instances. Although YOLOv8n exhibits a relatively subdued performance relative to other models, its potential is discernible in real-time applications, particularly within the purview of ITS, owing to its commendable proficiency in detection rates.
{"title":"Advanced aerial monitoring and vehicle classification for intelligent transportation systems with YOLOv8 variants","authors":"Murat Bakirci","doi":"10.1016/j.jnca.2025.104134","DOIUrl":"10.1016/j.jnca.2025.104134","url":null,"abstract":"<div><div>Aerial monitoring assumes a pivotal role within the domain of Intelligent Transportation Systems (ITS), imparting invaluable data and discernments that ameliorate the efficacy, security, and holistic operability of transportation networks. Image processing, encompassing the derivation of valuable insights through the manipulation of visual data captured by imaging apparatus, resides at the core and is poised to establish a firm footing in forthcoming ITS applications. In this context, numerous machine learning methodologies have been devised to enhance image processing, with novel approaches continually emerging. YOLOv8 emerged earlier this year and is still in the process of assimilating its potential application within the domain of ITS. In this study, a comprehensive assessment was conducted on all constituent variants of YOLOv8, specifically within the context of its application in the domain of aerial traffic monitoring. Using a custom-modified commercial drone, extensive datasets were acquired encompassing a diverse range of flight scenarios and traffic dynamics. To optimize model performance, meticulous consideration was given to ensuring dataset inclusivity, encompassing the full spectrum of vehicular typologies, while maintaining a homogeneous structure that accommodates an array of environmental nuances, including illumination and shading variations. The outcomes evince that both YOLOv8l and YOLOv8x exhibit notable superiority over other variants, manifesting exceptional detection efficacy even amid high-density traffic scenarios and the presence of obstructive elements. Contrastingly, in comparison to earlier iterations of YOLO, the current models demonstrate heightened precision in vehicle classification, yielding a reduction in misclassification instances. Although YOLOv8n exhibits a relatively subdued performance relative to other models, its potential is discernible in real-time applications, particularly within the purview of ITS, owing to its commendable proficiency in detection rates.</div></div>","PeriodicalId":54784,"journal":{"name":"Journal of Network and Computer Applications","volume":"237 ","pages":"Article 104134"},"PeriodicalIF":7.7,"publicationDate":"2025-02-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143395510","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}
Pub Date : 2025-02-03DOI: 10.1016/j.jnca.2025.104126
Joseph Tolley , Cameron Makin , Kenneth King , Carl B. Dietrich
The goal of Spectrum Access Systems (SASs) and similar systems that coordinate access to shared radio frequency bands is to fairly and efficiently allocate spectrum use amongst users in a locality such as a county. In the US 3.5-GHz Citizens Broadband Radio Service (CBRS) band, a SAS communicates with secondary users (SUs) of the band through protocols as defined by the US Federal Communications Commission (FCC) and Wireless Innovation Forum (WInnForum) to allocate spectrum and update each endpoint on the status of an SU’s granted channel within the band. These protocols include heartbeats that regularly inform an SU of their permitted status and alert them to Primary User (PU) activity requiring an SU to vacate their occupied channel. The heartbeat protocol maintains continuity of communication between the SAS and the SUs to exchange minimal types of information. However, the reliance on this limited message format, combined with a synchronous protocol that enforces response times on the scale of minutes, significantly restricts the system’s capability. This limitation excludes many use cases, such as those requiring near real-time adaptation, more extensive management of transmission parameters, or the delivery of information like backup SU channel assignments.
We propose an alternative heartbeat protocol named the Enhanced Heartbeat Protocol (EHP) that uses asynchronous messaging and a strategically extended message format to enable a much wider range of use cases, including support for fast-moving swarms of unmanned aerial vehicles (UAVs) or other high-mobility applications. Simulations demonstrate that this asynchronous protocol is scalable, providing rapid and reliable notifications to a large number of SUs. Additionally, the protocol’s performance degrades gracefully as the number of SUs increases, making it more robust under various conditions compared to the current standard. The goal of our EHP is to provide SASs with more up-to-date spectrum information, enable scalable centralized systems, and introduce additional parameters to increase the utility and flexibility of SASs in diverse wireless communication scenarios.
{"title":"Enhanced heartbeat protocol for near real-time coordinated spectrum sharing with expanded use cases","authors":"Joseph Tolley , Cameron Makin , Kenneth King , Carl B. Dietrich","doi":"10.1016/j.jnca.2025.104126","DOIUrl":"10.1016/j.jnca.2025.104126","url":null,"abstract":"<div><div>The goal of Spectrum Access Systems (SASs) and similar systems that coordinate access to shared radio frequency bands is to fairly and efficiently allocate spectrum use amongst users in a locality such as a county. In the US 3.5-GHz Citizens Broadband Radio Service (CBRS) band, a SAS communicates with secondary users (SUs) of the band through protocols as defined by the US Federal Communications Commission (FCC) and Wireless Innovation Forum (WInnForum) to allocate spectrum and update each endpoint on the status of an SU’s granted channel within the band. These protocols include heartbeats that regularly inform an SU of their permitted status and alert them to Primary User (PU) activity requiring an SU to vacate their occupied channel. The heartbeat protocol maintains continuity of communication between the SAS and the SUs to exchange minimal types of information. However, the reliance on this limited message format, combined with a synchronous protocol that enforces response times on the scale of minutes, significantly restricts the system’s capability. This limitation excludes many use cases, such as those requiring near real-time adaptation, more extensive management of transmission parameters, or the delivery of information like backup SU channel assignments.</div><div>We propose an alternative heartbeat protocol named the Enhanced Heartbeat Protocol (EHP) that uses asynchronous messaging and a strategically extended message format to enable a much wider range of use cases, including support for fast-moving swarms of unmanned aerial vehicles (UAVs) or other high-mobility applications. Simulations demonstrate that this asynchronous protocol is scalable, providing rapid and reliable notifications to a large number of SUs. Additionally, the protocol’s performance degrades gracefully as the number of SUs increases, making it more robust under various conditions compared to the current standard. The goal of our EHP is to provide SASs with more up-to-date spectrum information, enable scalable centralized systems, and introduce additional parameters to increase the utility and flexibility of SASs in diverse wireless communication scenarios.</div></div>","PeriodicalId":54784,"journal":{"name":"Journal of Network and Computer Applications","volume":"236 ","pages":"Article 104126"},"PeriodicalIF":7.7,"publicationDate":"2025-02-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143377186","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-02-01DOI: 10.1016/j.jnca.2024.104072
Rakesh Kumar, Mayank Swarnkar
With the increasing popularity of IoT, there has been a noticeable surge in security breaches associated with vulnerable IoT devices. To identify and counter such attacks. Intrusion Detection Systems (IDS) are deployed. However, these IoT devices use device-specific application layer protocols like MQTT and CoAP, which pose an additional burden to the traditional IDS. Several Machine Learning (ML) and Deep Learning (DL) based IDS are developed to detect malicious IoT network traffic. However, in recent times, a variety of IoT devices have been available on the market, resulting in the frequent installation and uninstallation of IoT devices based on users’ needs. Moreover, ML and DL-based IDS must train with sufficient device-specific attack training data for each IoT device, consuming a noticeable amount of training time. To solve these problems, we propose QuIDS, which utilizes a Quantum Support Vector Classifier to classify attacks in an IoT network. QuIDS requires very little training data compared to ML or DL to train and accurately identify attacks in the IoT network. QuIDS extracts eight flow-level features from IoT network traffic and utilizes them over four quantum bits for training. We experimented with QuIDS on two publicly available datasets and found the average recall rate, precision, and f1-score of the QuIDS as 91.1%, 84.3%, and 86.4%, respectively. Moreover, comparing QuIDS with the ML and DL methods, we found that QuIDS outperformed by 37.7%, 24.4.6%, and 36.9% more average recall and precision rates than the ML and DL methods, respectively.
{"title":"QuIDS: A Quantum Support Vector machine-based Intrusion Detection System for IoT networks","authors":"Rakesh Kumar, Mayank Swarnkar","doi":"10.1016/j.jnca.2024.104072","DOIUrl":"10.1016/j.jnca.2024.104072","url":null,"abstract":"<div><div>With the increasing popularity of IoT, there has been a noticeable surge in security breaches associated with vulnerable IoT devices. To identify and counter such attacks. Intrusion Detection Systems (IDS) are deployed. However, these IoT devices use device-specific application layer protocols like MQTT and CoAP, which pose an additional burden to the traditional IDS. Several Machine Learning (ML) and Deep Learning (DL) based IDS are developed to detect malicious IoT network traffic. However, in recent times, a variety of IoT devices have been available on the market, resulting in the frequent installation and uninstallation of IoT devices based on users’ needs. Moreover, ML and DL-based IDS must train with sufficient device-specific attack training data for each IoT device, consuming a noticeable amount of training time. To solve these problems, we propose QuIDS, which utilizes a Quantum Support Vector Classifier to classify attacks in an IoT network. QuIDS requires very little training data compared to ML or DL to train and accurately identify attacks in the IoT network. QuIDS extracts eight flow-level features from IoT network traffic and utilizes them over four quantum bits for training. We experimented with QuIDS on two publicly available datasets and found the average recall rate, precision, and f1-score of the QuIDS as 91.1%, 84.3%, and 86.4%, respectively. Moreover, comparing QuIDS with the ML and DL methods, we found that QuIDS outperformed by 37.7%, 24.4.6%, and 36.9% more average recall and precision rates than the ML and DL methods, respectively.</div></div>","PeriodicalId":54784,"journal":{"name":"Journal of Network and Computer Applications","volume":"234 ","pages":"Article 104072"},"PeriodicalIF":7.7,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142790082","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}
Pub Date : 2025-02-01DOI: 10.1016/j.jnca.2024.104088
Annamaria Ficara, Hocine Cherifi, Xiaoyang Liu, Luiz Fernando Bittencourt, Maria Fazio
{"title":"Complex networks for Smart environments management","authors":"Annamaria Ficara, Hocine Cherifi, Xiaoyang Liu, Luiz Fernando Bittencourt, Maria Fazio","doi":"10.1016/j.jnca.2024.104088","DOIUrl":"10.1016/j.jnca.2024.104088","url":null,"abstract":"","PeriodicalId":54784,"journal":{"name":"Journal of Network and Computer Applications","volume":"234 ","pages":"Article 104088"},"PeriodicalIF":7.7,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142825324","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}
Pub Date : 2025-02-01DOI: 10.1016/j.jnca.2024.104046
Zheng Zhang , Xiangyu Su , Ji Wu , Claudio J. Tessone , Hao Liao
In the fraud detection, fraudsters frequently engage with numerous benign users to disguise their activities. Consequently, the fraud graph exhibits not only homogeneous connections between the fraudsters and the same labeled nodes, but also heterogeneous connections, where fraudsters interact with the legitimate nodes. Heterogeneous graph representation learning aims at extracting the structural and semantic information and embed it into the low-dimensional node representation. Recently, maximizing the mutual information between the local node embedding and the summary representation has achieved the promising results on node classification tasks. However, existing deep graph infomax methods still have the following limitations. Firstly, attribute information of nodes in the graph is not fully utilized for capturing the semantic relationships between nodes. Secondly, the local and global supervision signal are not simultaneously exploited for the node embedding learning. Thirdly, the multiplex heterogeneous relations among nodes are ignored. To address these issues, a heterogeneous graph representation learning model by mutual information estimation (MIE-HetGRL) is proposed in this paper to identify the fraudsters in the fraud review graph. Concretely, a high-order mutual information estimation is proposed to integrate the local and global mutual information as the supervision signal. Then we devise a semantic attention fusion module to aggregate the relation-aware node embeddings into a compact node representation. Finally, a joint contrastive learning is designed for facilitating the training and optimization of model. The experimental results show that our proposed model significantly outperforms state-of-the-art baselines for fraud detection.
{"title":"Heterogeneous graph representation learning via mutual information estimation for fraud detection","authors":"Zheng Zhang , Xiangyu Su , Ji Wu , Claudio J. Tessone , Hao Liao","doi":"10.1016/j.jnca.2024.104046","DOIUrl":"10.1016/j.jnca.2024.104046","url":null,"abstract":"<div><div>In the fraud detection, fraudsters frequently engage with numerous benign users to disguise their activities. Consequently, the fraud graph exhibits not only homogeneous connections between the fraudsters and the same labeled nodes, but also heterogeneous connections, where fraudsters interact with the legitimate nodes. Heterogeneous graph representation learning aims at extracting the structural and semantic information and embed it into the low-dimensional node representation. Recently, maximizing the mutual information between the local node embedding and the summary representation has achieved the promising results on node classification tasks. However, existing deep graph infomax methods still have the following limitations. Firstly, attribute information of nodes in the graph is not fully utilized for capturing the semantic relationships between nodes. Secondly, the local and global supervision signal are not simultaneously exploited for the node embedding learning. Thirdly, the multiplex heterogeneous relations among nodes are ignored. To address these issues, a heterogeneous graph representation learning model by mutual information estimation (MIE-HetGRL) is proposed in this paper to identify the fraudsters in the fraud review graph. Concretely, a high-order mutual information estimation is proposed to integrate the local and global mutual information as the supervision signal. Then we devise a semantic attention fusion module to aggregate the relation-aware node embeddings into a compact node representation. Finally, a joint contrastive learning is designed for facilitating the training and optimization of model. The experimental results show that our proposed model significantly outperforms state-of-the-art baselines for fraud detection.</div></div>","PeriodicalId":54784,"journal":{"name":"Journal of Network and Computer Applications","volume":"234 ","pages":"Article 104046"},"PeriodicalIF":7.7,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143143864","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}
Pub Date : 2025-01-31DOI: 10.1016/j.jnca.2025.104131
Mobasshir Mahbub , Mir Md. Saym , Sarwar Jahan , Anup Kumar Paul , Alireza Vahid , Seyyedali Hosseinalipour , Bobby Barua , Hen-Geul Yeh , Raed M. Shubair , Tarik Taleb
Due to the rapid progress in communication technologies, unmanned aerial vehicles (UAVs) have become increasingly capable of providing reliable and cost-effective wireless communication from aerial vantage points. Unlike conventional stationary infrastructure, UAVs exhibit attractive features such as high scalability and improved line-of-sight (LoS) connectivity. Consequently, UAV-assisted wireless communications have become a promising paradigm to enhance coverage and connectivity in terrestrial and non-terrestrial networks. Nevertheless, the efficient deployment of UAVs in continuously evolving wireless network scenarios has remained to be a challenging task. These challenges have attracted a large body of research literature and subsequently several survey papers on UAV-assisted wireless communications. One of the distinctive features of research in UAV-assisted wireless networks is its broad array of experimental and analytical tools and techniques. A thorough review of these methodologies can swiftly familiarize researchers with the most recent efforts within this expansive field. However, most of the existing review/survey papers in this domain lack a comprehensive discussion about the advanced technologies used in UAV-assisted wireless networks, such as rate-splitting multiple access (RSMA), simultaneous wireless information and power transfer (SWIPT), digital twin (DT), cognitive radio (CR), space-air-ground integrated network (SAGIN), cell-free massive multiple-input multiple-output (CF mMIMO), integrated sensing and communication (ISAC), quantum technology, holographic MIMO (HMIMO). Motivated by this limitation and considering the novel UAV-assisted communication scenarios that can benefit from the adoption of such technologies; this work provides a thorough analysis of state-of-the-art intertwined technologies relative to UAV-assisted communications along with a discussion of their effectiveness and limitations. Furthermore, this study provides a brief overview of the comprehensive challenges of UAV-assisted networks, along with their security challenges, and opens future direction in this domain. This work finally explores the unique challenges in each of the existing technologies developed for UAV-assisted wireless networks that have been limitedly explored in prior literature aimed at providing a set of directions for future works.
{"title":"A holistic survey of UAV-assisted wireless communications in the transition from 5G to 6G: State-of-the-art intertwined innovations, challenges, and opportunities","authors":"Mobasshir Mahbub , Mir Md. Saym , Sarwar Jahan , Anup Kumar Paul , Alireza Vahid , Seyyedali Hosseinalipour , Bobby Barua , Hen-Geul Yeh , Raed M. Shubair , Tarik Taleb","doi":"10.1016/j.jnca.2025.104131","DOIUrl":"10.1016/j.jnca.2025.104131","url":null,"abstract":"<div><div>Due to the rapid progress in communication technologies, unmanned aerial vehicles (UAVs) have become increasingly capable of providing reliable and cost-effective wireless communication from aerial vantage points. Unlike conventional stationary infrastructure, UAVs exhibit attractive features such as high scalability and improved line-of-sight (LoS) connectivity. Consequently, UAV-assisted wireless communications have become a promising paradigm to enhance coverage and connectivity in terrestrial and non-terrestrial networks. Nevertheless, the efficient deployment of UAVs in continuously evolving wireless network scenarios has remained to be a challenging task. These challenges have attracted a large body of research literature and subsequently several survey papers on UAV-assisted wireless communications. One of the distinctive features of research in UAV-assisted wireless networks is its broad array of experimental and analytical tools and techniques. A thorough review of these methodologies can swiftly familiarize researchers with the most recent efforts within this expansive field. However, most of the existing review/survey papers in this domain lack a comprehensive discussion about the advanced technologies used in UAV-assisted wireless networks, such as rate-splitting multiple access (RSMA), simultaneous wireless information and power transfer (SWIPT), digital twin (DT), cognitive radio (CR), space-air-ground integrated network (SAGIN), cell-free massive multiple-input multiple-output (CF mMIMO), integrated sensing and communication (ISAC), quantum technology, holographic MIMO (HMIMO). Motivated by this limitation and considering the novel UAV-assisted communication scenarios that can benefit from the adoption of such technologies; this work provides a thorough analysis of state-of-the-art intertwined technologies relative to UAV-assisted communications along with a discussion of their effectiveness and limitations. Furthermore, this study provides a brief overview of the comprehensive challenges of UAV-assisted networks, along with their security challenges, and opens future direction in this domain. This work finally explores the unique challenges in each of the existing technologies developed for UAV-assisted wireless networks that have been limitedly explored in prior literature aimed at providing a set of directions for future works.</div></div>","PeriodicalId":54784,"journal":{"name":"Journal of Network and Computer Applications","volume":"237 ","pages":"Article 104131"},"PeriodicalIF":7.7,"publicationDate":"2025-01-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143402546","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}
Pub Date : 2025-01-30DOI: 10.1016/j.jnca.2025.104125
Yali Yuan, Weiyi Zou, Guang Cheng
Website Fingerprinting (WF) attacks compromise the anonymity of Tor by analyzing traffic patterns. Multi-tab WF attacks, which aim to identify multiple categories of websites from obfuscated traffic, have achieved significant progress. However, existing methods often fail to fully exploit the relationships between traffic features. On the one hand, splitting-based methods have complex processes that result in the loss of local traffic features. On the other hand, end-to-end methods process complete traffic but perform poorly when relying on a single feature. To address these challenges, this paper proposes an effective Multi-tab Website Fingerprinting attack with Transformer-based Feature Fusion named MW3F. Specifically, MW3F first extracts high-level traffic features, including direction and inter-packet time. Subsequently, These new representations are fused using the multi-head self-attention, which captures both local dependencies and global interactions. Finally, to identify website categories adaptively, MW3F incorporates learnable label embeddings to probe and pool class-related features. Each website category prediction is associated with a corresponding label embedding. We evalute MW3F against state-of-the-art multi-tab WF attacks in both multi-tab and defense scenarios. In the closed-world scenario, MW3F achieves a mean average precision (mAP) of over 90% across all tab settings, outperforming the strongest baseline, ARES, by 7% in the 5-tab setting. In the defense scenario, MW3F achieves approximately 90% mAP against WTF-PAD and Front defenses, demonstrating superior performance and exceptional robustness.
{"title":"MW3F: Improved multi-tab website fingerprinting attacks with Transformer-based feature fusion","authors":"Yali Yuan, Weiyi Zou, Guang Cheng","doi":"10.1016/j.jnca.2025.104125","DOIUrl":"10.1016/j.jnca.2025.104125","url":null,"abstract":"<div><div>Website Fingerprinting (WF) attacks compromise the anonymity of Tor by analyzing traffic patterns. Multi-tab WF attacks, which aim to identify multiple categories of websites from obfuscated traffic, have achieved significant progress. However, existing methods often fail to fully exploit the relationships between traffic features. On the one hand, splitting-based methods have complex processes that result in the loss of local traffic features. On the other hand, end-to-end methods process complete traffic but perform poorly when relying on a single feature. To address these challenges, this paper proposes an effective Multi-tab Website Fingerprinting attack with Transformer-based Feature Fusion named MW3F. Specifically, MW3F first extracts high-level traffic features, including direction and inter-packet time. Subsequently, These new representations are fused using the multi-head self-attention, which captures both local dependencies and global interactions. Finally, to identify website categories adaptively, MW3F incorporates learnable label embeddings to probe and pool class-related features. Each website category prediction is associated with a corresponding label embedding. We evalute MW3F against state-of-the-art multi-tab WF attacks in both multi-tab and defense scenarios. In the closed-world scenario, MW3F achieves a mean average precision (mAP) of over 90% across all tab settings, outperforming the strongest baseline, ARES, by 7% in the 5-tab setting. In the defense scenario, MW3F achieves approximately 90% mAP against WTF-PAD and Front defenses, demonstrating superior performance and exceptional robustness.</div></div>","PeriodicalId":54784,"journal":{"name":"Journal of Network and Computer Applications","volume":"236 ","pages":"Article 104125"},"PeriodicalIF":7.7,"publicationDate":"2025-01-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143354819","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}
Pub Date : 2025-01-28DOI: 10.1016/j.jnca.2025.104116
Pau Ferrer-Cid, Jose M. Barcelo-Ordinas, Jorge Garcia-Vidal
The development of Internet of Things (IoT) technologies has led to the widespread adoption of monitoring networks for a wide variety of applications, such as smart cities, environmental monitoring, and precision agriculture. A major research focus in recent years has been the development of graph-based techniques to improve the quality of data from sensor networks, a key aspect of the use of sensed data in decision-making processes, digital twins, and other applications. Emphasis has been placed on the development of machine learning (ML) and signal processing techniques over graphs, taking advantage of the benefits provided by the use of structured data through a graph topology. Many technologies such as graph signal processing (GSP) or the successful graph neural networks (GNNs) have been used for data quality enhancement tasks. This survey focuses on graph-based models for data quality control in monitoring sensor networks. In addition, it introduces the technical details that are commonly used to provide powerful graph-based solutions for data quality tasks in sensor networks, such as missing value imputation, outlier detection, or virtual sensing. To conclude, different challenges and emerging trends have been identified, e.g., graph-based models for digital twins or model transferability and generalization.
{"title":"A review of graph-powered data quality applications for IoT monitoring sensor networks","authors":"Pau Ferrer-Cid, Jose M. Barcelo-Ordinas, Jorge Garcia-Vidal","doi":"10.1016/j.jnca.2025.104116","DOIUrl":"10.1016/j.jnca.2025.104116","url":null,"abstract":"<div><div>The development of Internet of Things (IoT) technologies has led to the widespread adoption of monitoring networks for a wide variety of applications, such as smart cities, environmental monitoring, and precision agriculture. A major research focus in recent years has been the development of graph-based techniques to improve the quality of data from sensor networks, a key aspect of the use of sensed data in decision-making processes, digital twins, and other applications. Emphasis has been placed on the development of machine learning (ML) and signal processing techniques over graphs, taking advantage of the benefits provided by the use of structured data through a graph topology. Many technologies such as graph signal processing (GSP) or the successful graph neural networks (GNNs) have been used for data quality enhancement tasks. This survey focuses on graph-based models for data quality control in monitoring sensor networks. In addition, it introduces the technical details that are commonly used to provide powerful graph-based solutions for data quality tasks in sensor networks, such as missing value imputation, outlier detection, or virtual sensing. To conclude, different challenges and emerging trends have been identified, e.g., graph-based models for digital twins or model transferability and generalization.</div></div>","PeriodicalId":54784,"journal":{"name":"Journal of Network and Computer Applications","volume":"236 ","pages":"Article 104116"},"PeriodicalIF":7.7,"publicationDate":"2025-01-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143083153","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}
Pub Date : 2025-01-27DOI: 10.1016/j.jnca.2025.104106
Waris Ali , Chao Fang , Akmal Khan
A Content Delivery Network (CDN) consists of a distributed infrastructure of proxy servers designed to deliver digital content to end users effectively. CDNs have gained popularity due to increasing Internet users and their growing demand for low-latency content delivery. However, several unexplored aspects within CDN technology, including management, standardization, and architecture of CDNs, are crucial to staying aligned with industry trends and advancements. Previous survey papers focus on CDN aspects and have not yet categorized state-of-the-art CDN architectures. In this survey, we categorize and analyze seven state-of-the-art CDN architectures, providing a detailed analysis of their components, benefits, and limitations. We highlight advancements such as the convergence of CDN and Content-Centric Networking (CCN) paradigms for improved data retrieval, fog-based CDN collaboration with MEC for edge processing optimization, and blockchain technology for secure content delivery. Additionally, we also identify research challenges within CDN architectures and discuss the effectiveness of proposed solutions. Finally, we propose future research directions, including the collaboration of reinforcement learning for adaptive edge responses in CDN-P2P, machine learning for CDN selection in multi-CDN setups, and federated learning for improved caching in software-based and fog-based CDNs.
{"title":"A survey on the state-of-the-art CDN architectures and future directions","authors":"Waris Ali , Chao Fang , Akmal Khan","doi":"10.1016/j.jnca.2025.104106","DOIUrl":"10.1016/j.jnca.2025.104106","url":null,"abstract":"<div><div>A Content Delivery Network (CDN) consists of a distributed infrastructure of proxy servers designed to deliver digital content to end users effectively. CDNs have gained popularity due to increasing Internet users and their growing demand for low-latency content delivery. However, several unexplored aspects within CDN technology, including management, standardization, and architecture of CDNs, are crucial to staying aligned with industry trends and advancements. Previous survey papers focus on CDN aspects and have not yet categorized state-of-the-art CDN architectures. In this survey, we categorize and analyze seven state-of-the-art CDN architectures, providing a detailed analysis of their components, benefits, and limitations. We highlight advancements such as the convergence of CDN and Content-Centric Networking (CCN) paradigms for improved data retrieval, fog-based CDN collaboration with MEC for edge processing optimization, and blockchain technology for secure content delivery. Additionally, we also identify research challenges within CDN architectures and discuss the effectiveness of proposed solutions. Finally, we propose future research directions, including the collaboration of reinforcement learning for adaptive edge responses in CDN-P2P, machine learning for CDN selection in multi-CDN setups, and federated learning for improved caching in software-based and fog-based CDNs.</div></div>","PeriodicalId":54784,"journal":{"name":"Journal of Network and Computer Applications","volume":"236 ","pages":"Article 104106"},"PeriodicalIF":7.7,"publicationDate":"2025-01-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143083154","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}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号