Computer Networks最新文献

{"title":"XAInomaly: Explainable and interpretable Deep Contractive Autoencoder for O-RAN traffic anomaly detection","authors":"Osman Tugay Basaran,&nbsp;Falko Dressler","doi":"10.1016/j.comnet.2025.111145","DOIUrl":"10.1016/j.comnet.2025.111145","url":null,"abstract":"<div><div>Generative Artificial Intelligence (AI) techniques have become integral part in advancing next generation wireless communication systems by enabling sophisticated data modeling and feature extraction for enhanced network performance. In the realm of open radio access networks (O-RAN), characterized by their disaggregated architecture and heterogeneous components from multiple vendors, the deployment of generative models offers significant advantages for network management such as traffic analysis, traffic forecasting and anomaly detection. However, the complex and dynamic nature of O-RAN introduces challenges that necessitate not only accurate detection mechanisms but also reduced complexity, scalability, and most importantly interpretability to facilitate effective network management. In this study, we introduce the XAInomaly framework, an explainable and interpretable Semi-supervised (SS) Deep Contractive Autoencoder (DeepCAE) design for anomaly detection in O-RAN. Our approach leverages the generative modeling capabilities of our SS-DeepCAE model to learn compressed, robust representations of normal network behavior, which captures essential features, enabling the identification of deviations indicative of anomalies. To address the black-box nature of deep learning models, we propose reactive Explainable AI (XAI) technique called fastshap-C, which is providing transparency into the model’s decision-making process and highlighting the features contributing to anomaly detection.</div></div>","PeriodicalId":50637,"journal":{"name":"Computer Networks","volume":"261 ","pages":"Article 111145"},"PeriodicalIF":4.4,"publicationDate":"2025-02-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143521043","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":"MTCR-AE: A Multiscale Temporal Convolutional Recurrent Autoencoder for unsupervised malicious network traffic detection","authors":"Mukhtar Ahmed ,&nbsp;Jinfu Chen ,&nbsp;Ernest Akpaku ,&nbsp;Rexford Nii Ayitey Sosu","doi":"10.1016/j.comnet.2025.111147","DOIUrl":"10.1016/j.comnet.2025.111147","url":null,"abstract":"<div><div>The increasing sophistication of network attacks, particularly zero-day threats, underscores the need for robust, unsupervised detection methods. These attacks can flood networks with malicious traffic, overwhelm resources, or render services unavailable to legitimate users. Existing machine learning methods for zero-day attack detection typically rely on statistical features of network traffic, such as packet sizes and inter-arrival times. However, traditional approaches that depend on manually labeled data and linear structures often struggle to capture the intricate spatiotemporal correlations crucial for detecting unknown attacks. This paper introduces the Multiscale Temporal Convolutional Recurrent Autoencoder (MTCR-AE), an innovative framework designed to detect malicious network traffic by leveraging Multiscale Temporal Convolutional Networks (TCN) and Gated Recurrent Units (GRU). The MTCR-AE model captures both short- and long-range spatiotemporal dependencies while incorporating a temporal attention mechanism to dynamically prioritize critical features. The MTCR-AE operates in an unsupervised manner, eliminating the need for manual data labeling and enhancing its scalability for real-world applications. Experimental evaluations conducted on four benchmark datasets — ISCX-IDS-2012, USTC-TFC-2016, CIRA-CIC-DoHBrw2020, and CICIoT2023 — demonstrate the model’s superior performance, achieving an accuracy of 99.69%, precision of 99.63%, recall of 99.69%, and an F1-score of 99.66%. The results highlight the model’s capability to deliver state-of-the-art detection performance while maintaining low false positive and false negative rates, offering a scalable and reliable solution for dynamic network environments.</div></div>","PeriodicalId":50637,"journal":{"name":"Computer Networks","volume":"261 ","pages":"Article 111147"},"PeriodicalIF":4.4,"publicationDate":"2025-02-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143512489","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":"Spectrum efficiency through data: A methodology for evaluating local licensing strategies","authors":"Mohamad Alkadamani,&nbsp;Colin Brown,&nbsp;Kareem Baddour,&nbsp;Mathieu Châteauvert,&nbsp;Janaki Parekh,&nbsp;Adrian Florea","doi":"10.1016/j.comnet.2025.111115","DOIUrl":"10.1016/j.comnet.2025.111115","url":null,"abstract":"<div><div>The emerging demand for localized private networks tailored to specific and diverse use cases has increased interest in developing local spectrum licensing approaches that differ from traditional broad-coverage schemes. There is a significant need for forward-looking quantitative studies to guide technical decisions and ensure that licensing conditions align with overarching goals, such as supporting high spectrum reuse in potentially dense network deployments. To address this gap in the research literature, a novel data-driven methodology is introduced to evaluate the potential effectiveness of local spectrum licensing schemes from a regulatory perspective. This methodology utilizes real-world data to simulate prospective local deployment scenarios, capturing critical geographic details such as high-demand market areas, realistic industry locations, and high-resolution clutter information. The practical application of this methodology is demonstrated through a case study focused on the 3.9 GHz band in Canada, highlighting the importance of incorporating contextually relevant geospatial datasets to better inform local licensing regulations.</div></div>","PeriodicalId":50637,"journal":{"name":"Computer Networks","volume":"261 ","pages":"Article 111115"},"PeriodicalIF":4.4,"publicationDate":"2025-02-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143508268","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":"Hybrid attacks collaborative defense model using an ensemble honey badger algorithm","authors":"Guosheng Zhao ,&nbsp;Zhiwen Li ,&nbsp;Jian Wang","doi":"10.1016/j.comnet.2025.111149","DOIUrl":"10.1016/j.comnet.2025.111149","url":null,"abstract":"<div><div>Crowdsensing networks can be attacked by multiple malicious attacks at the same time, which seriously affects the reliability and security of the network. However, the existing defense technology mainly defends against specific types of attacks and cannot meet the security requirements of hybrid malicious attacks. Based on this, a hybrid attack collaborative defense model based on the integrated honey badger algorithm is proposed. First, the Crowdsensing network is formally described, and the collaborative mechanism of users participating is generalized. Through defense skill matching, efficient collaboration between users is achieved to meet the defense requirements of hybrid attacks. Then, the integrated honey badger algorithm is designed to solve the collaborative defense multi-objective optimization problem. The search intensity between multiple objectives is balanced by individual fitness evaluation, and the advantages and disadvantages of individuals in the collaborative defense scheme on the defense objectives are evaluated, and the Pareto optimal defense scheme is selected. At the same time, the algorithm continuously updates the collaborative defense scheme by selecting the exploration method suitable for the current defense state, and iteratively obtains the global optimal collaborative defense scheme. Finally, hybrid attack simulation and collaborative defense method performance analysis are carried out in the shared bicycle scheduling scenario. The experimental results show the feasibility and effectiveness of the proposed collaborative defense method.</div></div>","PeriodicalId":50637,"journal":{"name":"Computer Networks","volume":"261 ","pages":"Article 111149"},"PeriodicalIF":4.4,"publicationDate":"2025-02-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143488830","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":"ADPF: Anti-inference differentially private protocol for federated learning","authors":"Zirun Zhao,&nbsp;Zhaowen Lin,&nbsp;Yi Sun","doi":"10.1016/j.comnet.2025.111130","DOIUrl":"10.1016/j.comnet.2025.111130","url":null,"abstract":"<div><div>With the popularity of commercial artificial intelligence (AI), the importance of individual data is constantly increasing for the construction of large models. To ensure the utility of the released model, the security of individual data must be guaranteed with high confidence. Federated learning (FL), as the common paradigm for distributed learning, are usually subjected to various external attacks such as inversion attack or membership inference attack. Some solutions based on differential privacy (DP) are proposed to resist data revelation. However, the intelligence and collusion of adversaries are often underestimated during the training process. In this paper, an anti-inference differentially private federated learning protocol ADPF is proposed for data protection in an untrusted environment. ADPF models the attacker-defender scenario as a two-phase complete information dynamic game and designs optimization problems to find optimal budget allocations in different phases of training. Comparative experiments demonstrate that the performance of ADPF outperforms state-of-the-art differentially private federated learning protocol in both attack resistance and model utility.</div></div>","PeriodicalId":50637,"journal":{"name":"Computer Networks","volume":"261 ","pages":"Article 111130"},"PeriodicalIF":4.4,"publicationDate":"2025-02-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143471222","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":"A fine-grained compression scheme for block transmission acceleration over IPFS network","authors":"Changpeng Zhu ,&nbsp;Bincheng Fan ,&nbsp;Nan Xiang ,&nbsp;Bo Han ,&nbsp;Tian Zhou","doi":"10.1016/j.comnet.2025.111131","DOIUrl":"10.1016/j.comnet.2025.111131","url":null,"abstract":"<div><div>Interplanetary File System (IPFS) network, a prominent peer-to-peer decentralized file-sharing network, is extensively utilized by decentralized applications such as Blockchain and Metaverse for data sharing. With a substantial upsurge in data, compression schemes have been incorporated into the native IPFS protocol to enhance the efficiency of block transmission over an IPFS network. However, the protocols extended by these schemes drastically impair data shareability with the native protocol, contravening the core principles of IPFS-based distributed storage.</div><div>To address the issue, this paper presents a fine-grained compression scheme for the native IPFS protocol. Its primary concept is that blocks rather than files are considered as the smallest compressible units for block transmission, thereby ensuring the preservation of the data shareability with the protocol. To achieve the scheme, a three-layer architecture is firstly proposed and incorporated into the protocol as an independent compression component to support a variety of compression algorithms. Consequently, <em>Exchange</em> layer and <em>Storage</em> layer in the protocol are extended by leveraging the component to achieve block-level compression and decompression in the workflows of <em>upload</em> and <em>download</em> for the acceleration of block transmission over an IPFS network. Furthermore, a block pre-request approach is proposed and incorporated into <em>Exchange</em> layer to improve the block request mechanism in the layer which frequently causes block request awaiting, reducing block provision speed for compression algorithms, thus downgrading the acceleration. A comprehensive evaluation indicates that this extended IPFS protocol by our scheme has the same level of data shareability as the native IPFS protocol, and contributes to block transmission performance enhancement in <em>download</em> by up to 69% and in <em>upload</em> by as much as 201% relative to the current leading coarse-grained compression scheme, referred to as IPFSz.</div></div>","PeriodicalId":50637,"journal":{"name":"Computer Networks","volume":"261 ","pages":"Article 111131"},"PeriodicalIF":4.4,"publicationDate":"2025-02-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143471457","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":"FedShelter: Efficient privacy-preserving federated learning with poisoning resistance for resource-constrained IoT network","authors":"Tingyu Fan ,&nbsp;Xiaojun Chen ,&nbsp;Xudong Chen ,&nbsp;Ye Dong ,&nbsp;Weizhan Jing ,&nbsp;Zhendong Zhao","doi":"10.1016/j.comnet.2025.111125","DOIUrl":"10.1016/j.comnet.2025.111125","url":null,"abstract":"<div><div>Federated Learning (FL) enables decentralized IoT devices to collaboratively train a global model, but its distributed nature makes it vulnerable to poisoning and inference attacks, threatening security and privacy. Previous approaches combine Robust Defense Mechanisms with cryptographic techniques for Robust Secure Aggregation (SecAgg) to ensure security and privacy. However, schemes with strong resistance (robustness) against poisoning attacks often incur high complexity, increasing ciphertext communication and computational overhead, reducing efficiency, and limiting scalability for resource-constrained IoT networks. On the other hand, low-complexity schemes offer higher efficiency but weaker robustness, failing to counter stealthy poisoning attacks like backdoors. To address this, we propose FedShelter, an efficient privacy-preserving FL framework with poisoning resistance for IoT scenarios. It achieves lightweight Robust SecAgg utilizing Secure Two-party Computation (2PC) and incorporates customized encoding techniques to reduce communication overhead and defend against various poisoning attacks. Compared to state-of-the-art solutions like FLAME (USENIX Security’22) and RoFL (S&amp;P’23), FedShelter offers effective robustness against poisoning attacks while reducing communication by up to 56<span><math><mo>×</mo></math></span> and run-time by up to 37<span><math><mo>×</mo></math></span>, providing a fast and trustworthy training environment for distributed devices under resource-constrained IoT network.</div></div>","PeriodicalId":50637,"journal":{"name":"Computer Networks","volume":"261 ","pages":"Article 111125"},"PeriodicalIF":4.4,"publicationDate":"2025-02-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143488831","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":"Towards wireless time-sensitive networking: Multi-link deterministic scheduling via deep reinforcement learning","authors":"Xiaolin Wang ,&nbsp;Jinglong Zhang ,&nbsp;Xuanzhao Lu ,&nbsp;Fangfei Li ,&nbsp;Cailian Chen ,&nbsp;Xinping Guan","doi":"10.1016/j.comnet.2025.111119","DOIUrl":"10.1016/j.comnet.2025.111119","url":null,"abstract":"<div><div>Wireless time-sensitive networking (WTSN) has gained significant popularity because of its potential flexibility, scalability and bounded latency. Nevertheless, fewer studies have investigated specific scheduling strategies under unreliable wireless media. Besides, the lack of interoperability with current industrial networks and the ability to handle complex wireless network scheduling pose great challenges in WTSN design. To handle these challenges, we first propose a WiFi-based WTSN architecture. The wireless time-aware shaper (WTAS) and deep reinforcement learning based flow scheduling strategy are studied as the core mechanism of the proposed architecture. Specifically, the functionality of WTSN is designed through wireless gate control, control-aware retransmission, deterministic time slot design and delay analysis, and dynamic queue priority mapping. Then, we formulate a multi-link time slot scheduling problem to achieve load balance while considering transmission and control constraints. The soft actor-critic-based flow scheduling optimization (SAC-FSO) algorithm with elaborately designed decision variables mapping and dynamic constraint detection is proposed to efficiently achieve the wireless deterministic transmission. The performance evaluation demonstrates the effectiveness of the proposed WTSN architecture and WTAS modules. Comparison results with other algorithms demonstrate that the SAC-FSO algorithm is more efficient and scalable.</div></div>","PeriodicalId":50637,"journal":{"name":"Computer Networks","volume":"261 ","pages":"Article 111119"},"PeriodicalIF":4.4,"publicationDate":"2025-02-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143548808","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":"Quantum internet building blocks state of research and development","authors":"Aref Meddeb","doi":"10.1016/j.comnet.2025.111151","DOIUrl":"10.1016/j.comnet.2025.111151","url":null,"abstract":"<div><div>Breakthrough developments in quantum information technology is paving the way to the Quantum Internet. This article aims to provide an overview of the current state of research and development in the field of Quantum Communications and Quantum Internet. We examine the main building blocks and the latest achievements. By providing a holistic overview of recent achievements, this article fosters a deeper understanding of these building blocks, the related opportunities, and challenges associated with the realization of the Quantum Internet. We also provide some simple pedagogical models aimed to give the reader a clear insight on the interactions between quantum networking elements. A simple Quantum Internet 5Level-3D communication model is also provided, which integrates with, rather than parallels, the classical OSI and TCP/IP models. We further describe deployed testbeds and trials and highlight potential opportunities and challenges for Communication Service Provider's. We also provide some KPIs of quantum technologies and their evolution over the last decade. This article is intended as a self-contained entry level tutorial for newcomers who are not familiar with quantum information theory, but with some background knowledge about classical networking. Academics, students, as well as professionals can use this article to become able to begin researching the Quantum Internet, without having to read a plethora of articles, web pages, and books on the subject.</div></div>","PeriodicalId":50637,"journal":{"name":"Computer Networks","volume":"261 ","pages":"Article 111151"},"PeriodicalIF":4.4,"publicationDate":"2025-02-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143488832","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}

{"title":"Stochastic geometry analysis for information integration and communication in cellular and D2D-based heterogeneous IoT","authors":"Yulei Wang ,&nbsp;Li Feng ,&nbsp;Yalin Liu ,&nbsp;Zhongjie Li","doi":"10.1016/j.comnet.2025.111126","DOIUrl":"10.1016/j.comnet.2025.111126","url":null,"abstract":"<div><div>The heterogeneous Internet of Things (HetIoT), consisting of densely coexisted cellular and device-to-device (D2D) devices, has become a foundational component of 5th generation and beyond networks. Considering the high density of devices, guard zones around base stations (BSs) are introduced to mitigate interference with each other. However, previous studies neglect the impact of cellular devices on D2D transmissions, which is also crucial in HetIoT. To address this issue, this paper presents the use of guard zones around both transmitting BSs and cellular devices. By leveraging a stochastic geometry approach, a novel spatial distribution model is constructed to effectively capture the location randomness and interdependencies between BSs, cellular, and D2D devices. Based on this spatial model, an analytical model is developed to derive expressions for the successful transmission probabilities of BSs, cellular, and D2D devices. This model accounts for the inherent randomness and interrelations between cellular and D2D transmissions, while also characterizing the complex mutual interference resulting from the two types of guard zones. Extensive Monte Carlo simulations confirm the high accuracy of the proposed theoretical model. This analytical model provides practical guidance for practitioners in selecting suitable parameter configurations to enhance performance in HetIoT.</div></div>","PeriodicalId":50637,"journal":{"name":"Computer Networks","volume":"261 ","pages":"Article 111126"},"PeriodicalIF":4.4,"publicationDate":"2025-02-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143454689","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}