This paper presents Learning4Detecting (L4D), an efficient learning framework designed for identifying unusual traffic incidents in Vehicular Ad hoc Networks (VANETs). L4D addresses the challenges of traffic outlier detection by combining advanced feature extraction techniques (LBP, GLCM, HOG) with a two-tier hybrid binary classification system, including event pattern recognition followed by a second verification classifier. This is complemented by a multi-class classification layer for event categorization. Unlike existing methods, L4D optimizes both preprocessing and classification to enhance detection accuracy, effectively handle unseen events, and capture spatio-temporal patterns, all while reducing computational overhead. Experimental results demonstrate that L4D outperforms existing techniques in both accuracy and efficiency when applied to real-world VANET datasets.
{"title":"L4D: An outlier-based learning framework for detecting event patterns in vehicular networks","authors":"Kawthar Zaraket , Hassan Harb , Ismail Bennis , Ali Jaber , Abdelhafid Abouaissa","doi":"10.1016/j.comcom.2026.108436","DOIUrl":"10.1016/j.comcom.2026.108436","url":null,"abstract":"<div><div>This paper presents Learning4Detecting (L4D), an efficient learning framework designed for identifying unusual traffic incidents in Vehicular Ad hoc Networks (VANETs). L4D addresses the challenges of traffic outlier detection by combining advanced feature extraction techniques (LBP, GLCM, HOG) with a two-tier hybrid binary classification system, including event pattern recognition followed by a second verification classifier. This is complemented by a multi-class classification layer for event categorization. Unlike existing methods, L4D optimizes both preprocessing and classification to enhance detection accuracy, effectively handle unseen events, and capture spatio-temporal patterns, all while reducing computational overhead. Experimental results demonstrate that L4D outperforms existing techniques in both accuracy and efficiency when applied to real-world VANET datasets.</div></div>","PeriodicalId":55224,"journal":{"name":"Computer Communications","volume":"248 ","pages":"Article 108436"},"PeriodicalIF":4.3,"publicationDate":"2026-01-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146079029","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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 : 2026-01-28DOI: 10.1016/j.comcom.2026.108435
Abdullah Khanfor , Raby Hamadi , Noureddine Lasla , Hakim Ghazzai
UAVs have the potential to revolutionize urban management and provide valuable services to citizens. They can be deployed across diverse applications, including traffic monitoring, disaster response, environmental monitoring, and numerous other domains. However, this integration introduces novel security challenges that must be addressed to ensure safe and trustworthy urban operations. This paper provides a structured, evidence-based synthesis of UAV applications in smart cities and their associated security challenges as reported in the literature over the last decade, with particular emphasis on developments from 2019 to 2025. We categorize these challenges into two primary classes: (1) cyber-attacks targeting the communication infrastructure of UAVs and (2) unwanted or unauthorized physical intrusions by UAVs themselves. We examine the potential of Artificial Intelligence (AI) techniques in developing intrusion detection mechanisms to mitigate these security threats. We analyze how AI-based methods, such as machine/deep learning for anomaly detection and computer vision for object recognition, can play a pivotal role in enhancing UAV security through unified detection systems that address both cyber and physical threats. Furthermore, we consolidate publicly available UAV datasets across network traffic and vision modalities suitable for Intrusion Detection Systems (IDS) development and evaluation. The paper concludes by identifying ten key research directions, including scalability, robustness, explainability, data scarcity, automation, hybrid detection, large language models, multimodal approaches, federated learning, and privacy preservation. Finally, we discuss the practical challenges of implementing UAV IDS solutions in real-world smart city environments.
{"title":"AI-driven intrusion detection for UAV in Smart Urban ecosystems: A comprehensive survey","authors":"Abdullah Khanfor , Raby Hamadi , Noureddine Lasla , Hakim Ghazzai","doi":"10.1016/j.comcom.2026.108435","DOIUrl":"10.1016/j.comcom.2026.108435","url":null,"abstract":"<div><div>UAVs have the potential to revolutionize urban management and provide valuable services to citizens. They can be deployed across diverse applications, including traffic monitoring, disaster response, environmental monitoring, and numerous other domains. However, this integration introduces novel security challenges that must be addressed to ensure safe and trustworthy urban operations. This paper provides a structured, evidence-based synthesis of UAV applications in smart cities and their associated security challenges as reported in the literature over the last decade, with particular emphasis on developments from 2019 to 2025. We categorize these challenges into two primary classes: (1) cyber-attacks targeting the communication infrastructure of UAVs and (2) unwanted or unauthorized physical intrusions by UAVs themselves. We examine the potential of Artificial Intelligence (AI) techniques in developing intrusion detection mechanisms to mitigate these security threats. We analyze how AI-based methods, such as machine/deep learning for anomaly detection and computer vision for object recognition, can play a pivotal role in enhancing UAV security through unified detection systems that address both cyber and physical threats. Furthermore, we consolidate publicly available UAV datasets across network traffic and vision modalities suitable for Intrusion Detection Systems (IDS) development and evaluation. The paper concludes by identifying ten key research directions, including scalability, robustness, explainability, data scarcity, automation, hybrid detection, large language models, multimodal approaches, federated learning, and privacy preservation. Finally, we discuss the practical challenges of implementing UAV IDS solutions in real-world smart city environments.</div></div>","PeriodicalId":55224,"journal":{"name":"Computer Communications","volume":"249 ","pages":"Article 108435"},"PeriodicalIF":4.3,"publicationDate":"2026-01-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146081614","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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 : 2026-01-28DOI: 10.1016/j.comcom.2026.108432
Tuan Le
Data delivery in Delay Tolerant Networks (DTNs) is fundamentally constrained by two scarce resources: node buffer space and contact duration. Conventional approaches typically address message scheduling and buffer management as isolated optimization problems, resulting in suboptimal resource utilization where high-utility messages are often dropped to accommodate lower-value traffic. To bridge this gap, this paper proposes a unified, utility-based framework that jointly optimizes these decisions by formulating the contact opportunity as a Multidimensional 0/1 Knapsack Problem (MKP). We derive rigorous, closed-form marginal utility functions for three distinct objectives: maximizing delivery probability, minimizing average latency, and a composite metric balancing both. Unlike static heuristics, these metrics are derived from a deadline-constrained probabilistic model that explicitly quantifies the marginal benefit of replication relative to the message’s remaining Time-to-Live (TTL). To solve the resulting NP-hard joint allocation problem in real time, we introduce a computationally efficient greedy heuristic based on Utility Density. Trace-driven simulations using real-world vehicular mobility datasets (San Francisco and Rome) demonstrate that our unified policy outperforms state-of-the-art baselines, including ReAR and OBSBM. Beyond enabling flexible Quality of Service (QoS) enforcement via a tunable weighting coefficient, the overall analysis demonstrates that the proposed framework effectively resolves resource contention, achieving up to a 40% improvement in Delivery Ratio in sparse environments and a reduction in Average Latency when optimized for speed compared to existing techniques.
{"title":"A unified utility-based framework for joint scheduling and buffer management in Delay Tolerant Networks","authors":"Tuan Le","doi":"10.1016/j.comcom.2026.108432","DOIUrl":"10.1016/j.comcom.2026.108432","url":null,"abstract":"<div><div>Data delivery in Delay Tolerant Networks (DTNs) is fundamentally constrained by two scarce resources: node buffer space and contact duration. Conventional approaches typically address message scheduling and buffer management as isolated optimization problems, resulting in suboptimal resource utilization where high-utility messages are often dropped to accommodate lower-value traffic. To bridge this gap, this paper proposes a unified, utility-based framework that jointly optimizes these decisions by formulating the contact opportunity as a Multidimensional 0/1 Knapsack Problem (MKP). We derive rigorous, closed-form marginal utility functions for three distinct objectives: maximizing delivery probability, minimizing average latency, and a composite metric balancing both. Unlike static heuristics, these metrics are derived from a deadline-constrained probabilistic model that explicitly quantifies the marginal benefit of replication relative to the message’s remaining Time-to-Live (TTL). To solve the resulting NP-hard joint allocation problem in real time, we introduce a computationally efficient greedy heuristic based on Utility Density. Trace-driven simulations using real-world vehicular mobility datasets (San Francisco and Rome) demonstrate that our unified policy outperforms state-of-the-art baselines, including ReAR and OBSBM. Beyond enabling flexible Quality of Service (QoS) enforcement via a tunable weighting coefficient, the overall analysis demonstrates that the proposed framework effectively resolves resource contention, achieving up to a 40% improvement in Delivery Ratio in sparse environments and a <span><math><mrow><mn>3</mn><mo>×</mo></mrow></math></span> reduction in Average Latency when optimized for speed compared to existing techniques.</div></div>","PeriodicalId":55224,"journal":{"name":"Computer Communications","volume":"249 ","pages":"Article 108432"},"PeriodicalIF":4.3,"publicationDate":"2026-01-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146071086","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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 : 2026-01-27DOI: 10.1016/j.comcom.2026.108434
Wei Wang , Panfeng He , Boyu Wan, Yong Chen, Yu Zhang
This paper proposes a Multi-Agent Reinforcement Learning (MARL) framework for joint spectrum and power optimization in unmanned aerial vehicle (UAV)-unmanned ground vehicle (UGV) clusters during dynamic reconfiguration. The system consists of heterogeneous agents (communication nodes and radars) operating in scenarios with malicious jamming and dynamic inter-cluster node transfers. The joint channel selection and power control problem is formulated as a partially observable Markov decision process (POMDP), with a Multi-Agent Proximal Policy Optimization (MAPPO)-based algorithm developed to address two key challenges: For fixed-topology networks, a MAPPO-based algorithm is proposed to maximum number of operational nodes while avoiding intra-cluster interference; For dynamic reconfiguration scenarios, an estimated maximum node capacity (EMNC)-based algorithm is proposed enabling rapid adaptation to topology changes. Simulation results demonstrate that the proposed approach achieves 93%–97% operational node ratios in static configurations (outperforming baseline methods by 15%–40%) while maintaining 90%–94% operational efficiency during dynamic reconfiguration events — a significant improvement over baseline methods that typically suffer 20%–30% performance degradation during topology changes. The proposed solution uniquely combines real-time decision-making with robust adaptation capabilities, offering a practical approach for resilient resource management in dynamic UAV–UGV networks where conventional methods fail to address both dynamic reconfiguration and adversarial interference simultaneously.
{"title":"Learning to reallocate: MAPPO-based spectrum and power optimization for UAV–UGV clusters with dynamic reconfiguration","authors":"Wei Wang , Panfeng He , Boyu Wan, Yong Chen, Yu Zhang","doi":"10.1016/j.comcom.2026.108434","DOIUrl":"10.1016/j.comcom.2026.108434","url":null,"abstract":"<div><div>This paper proposes a Multi-Agent Reinforcement Learning (MARL) framework for joint spectrum and power optimization in unmanned aerial vehicle (UAV)-unmanned ground vehicle (UGV) clusters during dynamic reconfiguration. The system consists of heterogeneous agents (communication nodes and radars) operating in scenarios with malicious jamming and dynamic inter-cluster node transfers. The joint channel selection and power control problem is formulated as a partially observable Markov decision process (POMDP), with a Multi-Agent Proximal Policy Optimization (MAPPO)-based algorithm developed to address two key challenges: For fixed-topology networks, a MAPPO-based algorithm is proposed to maximum number of operational nodes while avoiding intra-cluster interference; For dynamic reconfiguration scenarios, an estimated maximum node capacity (EMNC)-based algorithm is proposed enabling rapid adaptation to topology changes. Simulation results demonstrate that the proposed approach achieves 93%–97% operational node ratios in static configurations (outperforming baseline methods by 15%–40%) while maintaining 90%–94% operational efficiency during dynamic reconfiguration events — a significant improvement over baseline methods that typically suffer 20%–30% performance degradation during topology changes. The proposed solution uniquely combines real-time decision-making with robust adaptation capabilities, offering a practical approach for resilient resource management in dynamic UAV–UGV networks where conventional methods fail to address both dynamic reconfiguration and adversarial interference simultaneously.</div></div>","PeriodicalId":55224,"journal":{"name":"Computer Communications","volume":"249 ","pages":"Article 108434"},"PeriodicalIF":4.3,"publicationDate":"2026-01-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146071112","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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 : 2026-01-21DOI: 10.1016/j.comcom.2026.108430
Tahere Rahmati, Behrouz Shahgholi Ghahfarokhi
The exponential growth in traffic load and increasing number of connected devices have driven cellular networks to offer high capacity and to support massive access. Full-Duplex Ultra-Dense Networks (FD-UDNs) represent a promising technology to meet this demand in cellular networks. However, these networks encounter serious challenges concerning energy consumption and high levels of interference, which, if not properly managed, can adversely affect overall network performance. This paper presents a deep reinforcement learning-based solution for the problem of joint small base station (SBS) on/off switching and resource allocation, with the objective of maximizing energy efficiency and meeting quality of service (QoS) requirements. To reduce complexity, we decompose the problem into two sub-problems: 1) BS sleep management and 2) power and radio resource allocation. For BS sleep management, two approaches are proposed: centralized and distributed. In the centralized approach, the network decides about the sleep state of the SBSs. In the distributed approach, each SBS independently decides on its sleep state. Subsequently, by assigning users to the active stations, each BS allocates transmit power and radio resources to its users. The simulation results highlight performance of the proposed methods compared to the previous method in terms of both energy efficiency and user satisfaction rate. Additionally, the results show that our distributed sleep management method outperforms the centralized one.
{"title":"Deep reinforcement learning based energy management in full-duplex ultra dense networks with cell switching and radio resource allocation","authors":"Tahere Rahmati, Behrouz Shahgholi Ghahfarokhi","doi":"10.1016/j.comcom.2026.108430","DOIUrl":"10.1016/j.comcom.2026.108430","url":null,"abstract":"<div><div>The exponential growth in traffic load and increasing number of connected devices have driven cellular networks to offer high capacity and to support massive access. Full-Duplex Ultra-Dense Networks (FD-UDNs) represent a promising technology to meet this demand in cellular networks. However, these networks encounter serious challenges concerning energy consumption and high levels of interference, which, if not properly managed, can adversely affect overall network performance. This paper presents a deep reinforcement learning-based solution for the problem of joint small base station (SBS) on/off switching and resource allocation, with the objective of maximizing energy efficiency and meeting quality of service (QoS) requirements. To reduce complexity, we decompose the problem into two sub-problems: 1) BS sleep management and 2) power and radio resource allocation. For BS sleep management, two approaches are proposed: centralized and distributed. In the centralized approach, the network decides about the sleep state of the SBSs. In the distributed approach, each SBS independently decides on its sleep state. Subsequently, by assigning users to the active stations, each BS allocates transmit power and radio resources to its users. The simulation results highlight performance of the proposed methods compared to the previous method in terms of both energy efficiency and user satisfaction rate. Additionally, the results show that our distributed sleep management method outperforms the centralized one.</div></div>","PeriodicalId":55224,"journal":{"name":"Computer Communications","volume":"248 ","pages":"Article 108430"},"PeriodicalIF":4.3,"publicationDate":"2026-01-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146079028","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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 : 2026-01-20DOI: 10.1016/j.comcom.2026.108427
Yongcong Mou , Yinghui Tang , Miaomiao Yu
To effectively conserve energy in wireless sensor networks (WSNs) and reduce packet delay, we propose a ()-policy sleep scheme for each sensor node, functioning in four distinct states. We model the sensor node, which incorporates the sleep mechanism, as a discrete-time vacation queueing system that accounts for startup times and an activation threshold. We first employ a probabilistic analysis technique to conduct a transient analysis of the system, aiming to derive recursive formulas for the steady-state distribution of the number of packets. We further obtain explicit expressions for several essential system performance metrics, including the expected number of packets, mean delay, and average energy cost of the node. The simulation experiments on models with various service time distributions confirm the analytical results, and extensive numerical experiments evaluate the sensitivity of system performance to several parameters. A weighted-sum cost function integrating mean delay and average energy consumption is formulated, and optimal sleep-wake strategies that minimise the weighted sum cost are evaluated across diverse sleep time distributions, service time distributions, weight coefficients, and delay constraints. The results demonstrate the advantages of the -policy in achieving an ideal equilibrium between energy efficiency and mean delay in WSNs.
{"title":"Performance analysis and optimisation of wireless sensor networks with startup times and (V,N)-policy sleep scheduling","authors":"Yongcong Mou , Yinghui Tang , Miaomiao Yu","doi":"10.1016/j.comcom.2026.108427","DOIUrl":"10.1016/j.comcom.2026.108427","url":null,"abstract":"<div><div>To effectively conserve energy in wireless sensor networks (WSNs) and reduce packet delay, we propose a (<span><math><mrow><mi>V</mi><mo>,</mo><mi>N</mi></mrow></math></span>)-policy sleep scheme for each sensor node, functioning in four distinct states. We model the sensor node, which incorporates the sleep mechanism, as a discrete-time <span><math><mrow><mi>G</mi><mi>e</mi><mi>o</mi><mo>/</mo><mi>G</mi><mo>/</mo><mn>1</mn></mrow></math></span> vacation queueing system that accounts for startup times and an activation threshold. We first employ a probabilistic analysis technique to conduct a transient analysis of the system, aiming to derive recursive formulas for the steady-state distribution of the number of packets. We further obtain explicit expressions for several essential system performance metrics, including the expected number of packets, mean delay, and average energy cost of the node. The simulation experiments on models with various service time distributions confirm the analytical results, and extensive numerical experiments evaluate the sensitivity of system performance to several parameters. A weighted-sum cost function integrating mean delay and average energy consumption is formulated, and optimal sleep-wake strategies that minimise the weighted sum cost are evaluated across diverse sleep time distributions, service time distributions, weight coefficients, and delay constraints. The results demonstrate the advantages of the <span><math><mrow><mo>(</mo><mi>V</mi><mo>,</mo><mi>N</mi><mo>)</mo></mrow></math></span>-policy in achieving an ideal equilibrium between energy efficiency and mean delay in WSNs.</div></div>","PeriodicalId":55224,"journal":{"name":"Computer Communications","volume":"248 ","pages":"Article 108427"},"PeriodicalIF":4.3,"publicationDate":"2026-01-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146024173","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
An Electroencephalogram (EEG) signal plays a vital role in a healthcare communication system for recording the electrical activities of the human brain from the scalp. In recent times, the conventional IoT-based healthcare system uses the cloud computing paradigm to manage time-critical healthcare data. Moreover, switching to the fog computing, the fog-assisted EEG systems are for single EEG applications. However, the use of a fog computing paradigm for a single EEG system is not an efficient solution in terms of resource management and time consumption. Therefore, we introduce a Fog-enabled EEG architecture where multiple fog devices collaboratively process the data in a single integrated IoT platform. As the proposed architecture is new, we focus on developing the mathematical model of the architecture and discuss the crucial aspects. Additionally, we devise a dynamic optimal fog head selection within the network using a weighted multi-criteria decision-making approach. From the simulation, we observe that the average propagation delay is reduced by approximately 95% using 6G-enabled fog computing as compared to the cloud. Further, our method reduces the total delay by 83.87% compared to the existing baseline KCHE technique, showing the effectiveness of this work.
{"title":"An efficient master head selection for multi-EEG to multi-fog IoT network using 6G-driven FaaS","authors":"Rupalin Nanda , Sakthivel P. , Rama Krushna Rath , Abhishek Hazra","doi":"10.1016/j.comcom.2026.108429","DOIUrl":"10.1016/j.comcom.2026.108429","url":null,"abstract":"<div><div>An Electroencephalogram (EEG) signal plays a vital role in a healthcare communication system for recording the electrical activities of the human brain from the scalp. In recent times, the conventional IoT-based healthcare system uses the cloud computing paradigm to manage time-critical healthcare data. Moreover, switching to the fog computing, the fog-assisted EEG systems are for single EEG applications. However, the use of a fog computing paradigm for a single EEG system is not an efficient solution in terms of resource management and time consumption. Therefore, we introduce a Fog-enabled EEG architecture where multiple fog devices collaboratively process the data in a single integrated IoT platform. As the proposed architecture is new, we focus on developing the mathematical model of the architecture and discuss the crucial aspects. Additionally, we devise a dynamic optimal fog head selection within the network using a weighted multi-criteria decision-making approach. From the simulation, we observe that the average propagation delay is reduced by approximately 95% using 6G-enabled fog computing as compared to the cloud. Further, our method reduces the total delay by 83.87% compared to the existing baseline KCHE technique, showing the effectiveness of this work.</div></div>","PeriodicalId":55224,"journal":{"name":"Computer Communications","volume":"248 ","pages":"Article 108429"},"PeriodicalIF":4.3,"publicationDate":"2026-01-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145980580","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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 : 2026-01-14DOI: 10.1016/j.comcom.2026.108428
Yuqi Dai , Hua Zhang , Jingyu Wang , Jianxin Liao
Network configuration synthesis is essential for automated configuration management in large and complex networks. However, existing synthesizers face challenges in practical applications, including limited scalability, slow synthesis speed, insufficient support for various routing protocols, and difficulty in handling mixed vendor configurations.
To address these issues, this paper proposes MoCS, a modular configuration synthesizer that integrates multiple Large Language Models (LLMs) with Graph Neural Network (GNN)-enhanced recommendations to enable protocol-agnostic and vendor-compliant configuration synthesis. MoCS decomposes the synthesis pipeline into three LLM-based modules, each following a unified prompt engineering framework with task-specific adaptations. Specifically, the Intent Translation Module (IT-Module) translates natural language intents into structured configuration tasks, while the Configuration Graph Generation Module (CG-Module) constructs a Configuration Knowledge Graph (CKG) by incorporating semantic information from network topologies, structured tasks, and vendor-specific configuration templates. These two modules collaborate to support various protocols and mixed vendor configurations via a unified graph representation. The Configuration Recommendation Module (CR-Module) utilizes a heterogeneous GNN-based model (HGAT-CR) to perform type-aware reasoning over the CKG and generate top- candidate parameters. These candidates provide prior knowledge that narrows the search space and improves recommendation accuracy. Finally, they are refined through an LLM-guided optimization mechanism that combines formal verification feedback to produce the final configuration, ensuring maximal intent satisfaction while minimizing side effects.
Our evaluation demonstrates that MoCS outperforms existing synthesizers, including NetComplete, INCS, and ConfigReco. In large networks with complex intents, MoCS achieves a high coverage rate (88.23 ± 1.12%), low redundancy rate (7.89 ± 1.59%), perfect intent satisfaction rate (1.00 ± 0.00), and reasonable runtime (143.83 ± 21.89s). Furthermore, MoCS can synthesize mixed vendor configurations, which current synthesizers cannot handle.
{"title":"MoCS: Modular configuration synthesis via large language models and graph neural network-augmented recommendation","authors":"Yuqi Dai , Hua Zhang , Jingyu Wang , Jianxin Liao","doi":"10.1016/j.comcom.2026.108428","DOIUrl":"10.1016/j.comcom.2026.108428","url":null,"abstract":"<div><div>Network configuration synthesis is essential for automated configuration management in large and complex networks. However, existing synthesizers face challenges in practical applications, including limited scalability, slow synthesis speed, insufficient support for various routing protocols, and difficulty in handling mixed vendor configurations.</div><div>To address these issues, this paper proposes MoCS, a modular configuration synthesizer that integrates multiple Large Language Models (LLMs) with Graph Neural Network (GNN)-enhanced recommendations to enable protocol-agnostic and vendor-compliant configuration synthesis. MoCS decomposes the synthesis pipeline into three LLM-based modules, each following a unified prompt engineering framework with task-specific adaptations. Specifically, the Intent Translation Module (IT-Module) translates natural language intents into structured configuration tasks, while the Configuration Graph Generation Module (CG-Module) constructs a Configuration Knowledge Graph (CKG) by incorporating semantic information from network topologies, structured tasks, and vendor-specific configuration templates. These two modules collaborate to support various protocols and mixed vendor configurations via a unified graph representation. The Configuration Recommendation Module (CR-Module) utilizes a heterogeneous GNN-based model (HGAT-CR) to perform type-aware reasoning over the CKG and generate top-<span><math><mi>k</mi></math></span> candidate parameters. These candidates provide prior knowledge that narrows the search space and improves recommendation accuracy. Finally, they are refined through an LLM-guided optimization mechanism that combines formal verification feedback to produce the final configuration, ensuring maximal intent satisfaction while minimizing side effects.</div><div>Our evaluation demonstrates that MoCS outperforms existing synthesizers, including NetComplete, INCS, and ConfigReco. In large networks with complex intents, MoCS achieves a high coverage rate (88.23 ± 1.12%), low redundancy rate (7.89 ± 1.59%), perfect intent satisfaction rate (1.00 ± 0.00), and reasonable runtime (143.83 ± 21.89s). Furthermore, MoCS can synthesize mixed vendor configurations, which current synthesizers cannot handle.</div></div>","PeriodicalId":55224,"journal":{"name":"Computer Communications","volume":"248 ","pages":"Article 108428"},"PeriodicalIF":4.3,"publicationDate":"2026-01-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145980586","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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 : 2026-01-12DOI: 10.1016/j.comcom.2026.108425
Mahamadou Diawara , Andre Faye
With the introduction of the fifth generation of mobile networks (5G) in 3GPP Release 15, driven by the exponential growth in the number of mobile users, the emergence of new multimedia services and the proliferation of private networks, spectrum management has become a key challenge in the field of telecommunications. In light of the high costs associated with the acquisition and use of licensed frequency bands, the NR-U (New Radio-Unlicensed) standard has emerged as a strategic solution. It enables the extension of 5G services to unlicensed spectrum, thereby addressing the increasing demand for capacity and flexibility. Unlicensed bands, particularly those below 7 GHz, exhibit promising characteristics for supporting real-time critical applications. They offer a cost-effective, flexible communication infrastructure capable of dynamically adapting to network capacity demands. This paper presents an experimental study of 5G NR-U operation over sub-7 GHz unlicensed bands using the open-source OpenAirInterface (OAI) platform and USRP B210 software-defined radio. We integrated these bands into a 5G system and provided a reference framework for future research on communications over unlicensed spectrum with OAI. The implementation accounts for hardware constraints and the stringent requirements of real-time processing to emulate a realistic deployment environment. Performance, and power consumption analysis results confirm the relevance of using sub-7 GHz unlicensed bands for critical applications in private network scenarios or connectivity extensions in remote areas. The proposed implementation is validated through a drone-based application scenario.
{"title":"Design, Implementation, Performance evaluation of a Sub-7 GHz 5G NR-U system","authors":"Mahamadou Diawara , Andre Faye","doi":"10.1016/j.comcom.2026.108425","DOIUrl":"10.1016/j.comcom.2026.108425","url":null,"abstract":"<div><div>With the introduction of the fifth generation of mobile networks (5G) in <em>3GPP Release 15</em>, driven by the exponential growth in the number of mobile users, the emergence of new multimedia services and the proliferation of private networks, spectrum management has become a key challenge in the field of telecommunications. In light of the high costs associated with the acquisition and use of licensed frequency bands, the NR-U (New Radio-Unlicensed) standard has emerged as a strategic solution. It enables the extension of 5G services to unlicensed spectrum, thereby addressing the increasing demand for capacity and flexibility. Unlicensed bands, particularly those below 7 GHz, exhibit promising characteristics for supporting real-time critical applications. They offer a cost-effective, flexible communication infrastructure capable of dynamically adapting to network capacity demands. This paper presents an experimental study of 5G NR-U operation over sub-7 GHz unlicensed bands using the open-source OpenAirInterface (OAI) platform and USRP B210 software-defined radio. We integrated these bands into a 5G system and provided a reference framework for future research on communications over unlicensed spectrum with OAI. The implementation accounts for hardware constraints and the stringent requirements of real-time processing to emulate a realistic deployment environment. Performance, and power consumption analysis results confirm the relevance of using sub-7 GHz unlicensed bands for critical applications in private network scenarios or connectivity extensions in remote areas. The proposed implementation is validated through a drone-based application scenario.</div></div>","PeriodicalId":55224,"journal":{"name":"Computer Communications","volume":"248 ","pages":"Article 108425"},"PeriodicalIF":4.3,"publicationDate":"2026-01-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145980585","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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 : 2026-01-10DOI: 10.1016/j.comcom.2026.108423
Antonio M. Alberti , Epper Bonomo , Rodrigo H. Santos , Victor A. de J. Alberti , Marcelo E. Pellenz , Rodrigo da Rosa Righi
This work integrates NovaGenesis (NG), a clean-slate IoT architecture, with LoRa technology within low-power wide-area networks (LPWAN), extending previous efforts on NG connectivity with Wi-Fi. The research aims to update the embedded version of NG and develop devices for seamless LoRa and Wi-Fi IoT operation. It evaluates NG’s performance on LoRa and Wi-Fi, focusing on throughput, delay, and packet loss. Despite LPWAN limitations, the results show that the NG deployment is feasible, validating its self-organizing IoT life cycle to maintain service continuity between an ESP-32 and a data client. Performance meets the needs of IoT applications in agribusiness, logistics, and smart monitoring. In addition, a 24-hour environmental monitoring experiment was conducted in Santa Rita do Sapucaí(SRS), Minas Gerais, Brazil, where a commercial weather station was modified to integrate NG, allowing accurate collection of temperature, humidity, atmospheric pressure, wind conditions, solar radiation and UV index. The results met expected diurnal patterns in SRS, proving the reliability and precision of the sensors and communication infrastructure. This solution overcomes common IETF IoT stack limitations in devices naming, information provenance, entities identification, programmability via digital twins, programmability, services and devices self-organization, and trust formation, offering a robust platform for varied IoT scenarios in LPWAN environments. These are the key benefits of applying NovaGenesis for LoRa and Wi-Fi-based environmental monitoring.
这项工作将NovaGenesis （NG）这一全新的物联网架构与低功耗广域网（LPWAN）中的LoRa技术集成在一起，扩展了之前在NG连接Wi-Fi方面的努力。该研究旨在更新NG的嵌入式版本，并开发无缝LoRa和Wi-Fi物联网操作的设备。它评估了NG在LoRa和Wi-Fi上的性能，重点关注吞吐量、延迟和数据包丢失。尽管有LPWAN的限制，但结果表明，NG部署是可行的，验证了其自组织物联网生命周期，以保持ESP-32和数据客户端之间的服务连续性。性能满足物联网在农业综合企业、物流和智能监控领域的应用需求。此外，在巴西米纳斯吉拉斯州Santa Rita do Sapucaí（SRS）进行了一项24小时环境监测实验，在那里对一个商业气象站进行了改造，以整合NG，从而能够准确收集温度、湿度、大气压、风况、太阳辐射和紫外线指数。结果符合SRS的预期日模式，证明了传感器和通信基础设施的可靠性和精度。该解决方案克服了常见的IETF物联网堆栈在设备命名、信息来源、实体识别、通过数字双胞胎可编程性、可编程性、服务和设备自组织以及信任形成方面的限制，为LPWAN环境中的各种物联网场景提供了一个强大的平台。这些是将NovaGenesis应用于LoRa和基于wi - fi的环境监测的主要好处。
{"title":"Applying NovaGenesis: A service-oriented, self-organizing, and programmable IoT architecture for LoRa and Wi-Fi-based environmental monitoring","authors":"Antonio M. Alberti , Epper Bonomo , Rodrigo H. Santos , Victor A. de J. Alberti , Marcelo E. Pellenz , Rodrigo da Rosa Righi","doi":"10.1016/j.comcom.2026.108423","DOIUrl":"10.1016/j.comcom.2026.108423","url":null,"abstract":"<div><div>This work integrates NovaGenesis (NG), a clean-slate IoT architecture, with LoRa technology within low-power wide-area networks (LPWAN), extending previous efforts on NG connectivity with Wi-Fi. The research aims to update the embedded version of NG and develop devices for seamless LoRa and Wi-Fi IoT operation. It evaluates NG’s performance on LoRa and Wi-Fi, focusing on throughput, delay, and packet loss. Despite LPWAN limitations, the results show that the NG deployment is feasible, validating its self-organizing IoT life cycle to maintain service continuity between an ESP-32 and a data client. Performance meets the needs of IoT applications in agribusiness, logistics, and smart monitoring. In addition, a 24-hour environmental monitoring experiment was conducted in Santa Rita do Sapucaí(SRS), Minas Gerais, Brazil, where a commercial weather station was modified to integrate NG, allowing accurate collection of temperature, humidity, atmospheric pressure, wind conditions, solar radiation and UV index. The results met expected diurnal patterns in SRS, proving the reliability and precision of the sensors and communication infrastructure. This solution overcomes common IETF IoT stack limitations in devices naming, information provenance, entities identification, programmability via digital twins, programmability, services and devices self-organization, and trust formation, offering a robust platform for varied IoT scenarios in LPWAN environments. These are the key benefits of applying NovaGenesis for LoRa and Wi-Fi-based environmental monitoring.</div></div>","PeriodicalId":55224,"journal":{"name":"Computer Communications","volume":"248 ","pages":"Article 108423"},"PeriodicalIF":4.3,"publicationDate":"2026-01-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145980584","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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