Computer Communications最新文献

{"title":"FaaSBid: An auction-based model for Function as a Service in edge-fog environments using unallocated resources","authors":"Abdulrahman K. Al-Qadhi ,&nbsp;Rukshan Athauda ,&nbsp;Rohaya Latip ,&nbsp;Masnida Hussin","doi":"10.1016/j.comcom.2026.108413","DOIUrl":"10.1016/j.comcom.2026.108413","url":null,"abstract":"<div><div>The exponential growth of IoT devices has resulted in a need to process IoT workloads. Processing such workloads near the edge instead of the cloud has a number of advantages including lower latency, improved security and ability to meet many other Quality of Service attributes. Function as a Service (FaaS) is becoming a popular method to process such IoT workloads. In this paper, we propose a novel model, termed FaaSBid, that incentivise users to utilise serverless functions near the edge using unallocated resources. The service provider offers a discount range based on resource utilisation, where users offer bids to execute their functions near the edge resulting in cost savings while the service providers have a new revenue stream and higher resource utilisation near the edge. In this paper, a number of algorithms are proposed and evaluated for FaaSBid model. To initialise function placement, Fitness-Based Swap (FBSW) algorithm is proposed which places functions based on pre-defined information such as function size, function maximum execution time, and storage cost. Next, the Dynamic Demand Replacement Algorithm (DDRA) algorithm is used to place in-demand functions near the edge nodes periodically, while the proposed task scheduling algorithm - Maximum Revenue Bid (MRB) is used to give priority to tasks to maximise revenue near the edge. We have evaluated the FaaSBid model and the proposed algorithms and pricing model by comparing with a number of existing models and algorithms using real-world datasets. The results show that FaaSBid model provides higher resource utilisation, a new revenue stream for service providers while reducing costs for users. On average, in FaasBid, the proposed pricing model saved 12.9% and 6.5% compared to AWS fixed pricing and AuctionWhisk pricing respectively per function execution. Also, the results show that the proposed function placement and scheduling algorithms outperform many well-known function placement and scheduling algorithms in terms of revenue generated, resource utilisation, throughput, and latency with significant improvements near the edge. The results also demonstrated that dynamically placing functions based on demand has a significant impact. Overall, this paper outlines a new paradigm that uses unutilised resources near the edge, improving many QoS attributes from both service providers' and users’ perspectives.</div></div>","PeriodicalId":55224,"journal":{"name":"Computer Communications","volume":"248 ","pages":"Article 108413"},"PeriodicalIF":4.3,"publicationDate":"2026-01-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145980581","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}

{"title":"Large and reliable data transfer service for LoRa mesh network applications","authors":"Joan Miquel Solé,&nbsp;Roger Pueyo Centelles,&nbsp;Felix Freitag,&nbsp;Roc Meseguer,&nbsp;Roger Baig Viñas","doi":"10.1016/j.comcom.2025.108404","DOIUrl":"10.1016/j.comcom.2025.108404","url":null,"abstract":"<div><div>Recently, LoRa mesh networks have appeared as a communication technology for Internet of Things (IoT) devices. Through node-to-node communication, novel distributed IoT applications that extend beyond the capabilities of the LoRaWAN architecture can be enabled. However, current technologies for LoRa networks do not provide mechanisms for large and reliable data transfers between IoT nodes. This paper presents a service for such data transfers in LoRa mesh network applications, along with the protocol and formats used for inter-node communication. We explain the design choices and detail the implementation decisions to ensure that this service is practically usable. To this end, the service was integrated into the LoRaMesher library and is available as an open-source operational implementation. In experiments with ten real nodes and two network topologies, we observe that the service effectively achieves a large and reliable message delivery in an environment of concurrent transmissions and packet losses. In contrast, the cost of reliability for large data transfers is an increased number of messages and a higher delivery time. With the integration of the service into the LoRaMesher technology, developers now have a library that provides a reliable and large payload service for LoRa mesh network applications, eliminating the need to develop such capacity as a specific application-level solution.</div></div>","PeriodicalId":55224,"journal":{"name":"Computer Communications","volume":"248 ","pages":"Article 108404"},"PeriodicalIF":4.3,"publicationDate":"2026-01-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145929282","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}

{"title":"Sparse QoS prediction for cloud services via inductive subgraph pattern aware graph neural network","authors":"Jianlong Xu ,&nbsp;Caiyi Chen ,&nbsp;Qingcao Dai ,&nbsp;Guanchen Du ,&nbsp;Ruiqi Wang ,&nbsp;Mingtong Li ,&nbsp;Quanqing Guo ,&nbsp;Yuxiang Zeng","doi":"10.1016/j.comcom.2026.108415","DOIUrl":"10.1016/j.comcom.2026.108415","url":null,"abstract":"<div><div>Accurately predicting the Quality of Service (QoS) is a crucial issue for selecting suitable cloud services for each user. Collaborative prediction models have been successful in selecting suitable cloud services for users. However, they often struggle with extreme sparsity, where only a limited number of interactions are available for collaborative filtering. Some models excel at handling extreme sparsity but struggle with generalization at the same time. To overcome these challenges, we propose a sparse QoS prediction framework for cloud services via an inductive subgraph pattern-aware graph neural network (ISPA-GNN). Our framework employs a novel graph-based collaborative filtering approach combined with a subgraph sampling strategy to extract semantic information about user-service interactions more effectively. To optimize memory usage and enhance the generalization of unseen users or services, we utilize decoupled ID-based embeddings that maximize contextual information. Extensive experiments on a large-scale, real-world QoS dataset demonstrate that ISPA-GNN outperforms most current collaborative QoS prediction techniques in terms of mean absolute error (MAE) and root mean squared error (RMSE), while also achieving significant gains in memory efficiency.</div></div>","PeriodicalId":55224,"journal":{"name":"Computer Communications","volume":"248 ","pages":"Article 108415"},"PeriodicalIF":4.3,"publicationDate":"2026-01-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145929269","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}

{"title":"AgriSmart: An IoT-enabled framework for agricultural resource optimization","authors":"Xu Tao ,&nbsp;Jackson Butcher ,&nbsp;Christian Cumini ,&nbsp;Mounica Talasila ,&nbsp;Salmeron Cortasa Montserrat ,&nbsp;Alessio Sacco ,&nbsp;Michael Popp ,&nbsp;Guido Marchetto ,&nbsp;Simone Silvestri","doi":"10.1016/j.comcom.2026.108416","DOIUrl":"10.1016/j.comcom.2026.108416","url":null,"abstract":"<div><div>Efficient use of farming resources (e.g., nitrogen, water, pesticides) is key to maximizing productivity and promoting sustainable agriculture. Traditional methods, such as fixed-rate applications or soil sampling, often fail to adapt to changing in-season conditions and specific nutrient demands, leading to inefficiencies and environmental harm. In this work, we propose <em>AgriSmart</em>, an IoT-enabled framework that optimizes resource application strategies to maximize crop yield while minimizing resource usage within a given budget. AgriSmart formulates an optimization problem solved periodically using an enhanced Differential Evolution (DE) algorithm that balances exploration and exploitation, following a Model Predictive Control (MPC) approach. Crop yield responses to varying application timings and rates are estimated using the process-based crop simulation model DSSAT (Decision Support System for Agrotechnology Transfer). To improve flexibility and reduce computational complexity, we introduce <em>adjustable receding horizon</em> that allows multiple actions to be applied before re-optimization, enabling adaptation to resources with different application frequencies (e.g., water vs. nitrogen). As the time horizon advances, AgriSmart dynamically adjusts the resource applications to better match crop needs at each growth stage, responding to evolving weather and field conditions. We evaluate AgriSmart in two use cases: irrigation scheduling for soybean and nitrogen management for maize. Results show that AgriSmart outperforms existing methods, achieving up to 21.4% water savings for soybean without yield loss, and increasing maize yield by 20% while reducing nitrogen use by up to 32%.</div></div>","PeriodicalId":55224,"journal":{"name":"Computer Communications","volume":"248 ","pages":"Article 108416"},"PeriodicalIF":4.3,"publicationDate":"2026-01-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145980582","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}

{"title":"Context-aware anomaly detection by community detection in the Internet of Things","authors":"Fatemeh Stodt ,&nbsp;Christoph Reich ,&nbsp;Fabrice Theoleyre","doi":"10.1016/j.comcom.2026.108414","DOIUrl":"10.1016/j.comcom.2026.108414","url":null,"abstract":"<div><div>This paper introduces a novel context-aware anomaly detection framework for the Internet of Things, leveraging community detection in multi-edge graphs with a heterogeneous Graph Neural Network (HeteroGNN) architecture to enhance network security. The proposed framework detects anomalies such as unexpected communication patterns among devices that rarely interact, unusual traffic spikes during off-hours, or deviations in the contextual and knowledge-based interactions of devices. For example, in an industrial IoT environment, unauthorized access or malicious activity can be inferred from unexpected communication within a device community after working hours. Our detection approach uses multi-edge graphs to model diverse interactions (network communication, context, knowledge) and applies community detection to capture stable graph structures. By incorporating these insights into a HeteroGNN, the framework effectively distinguishes anomalous edges while maintaining scalability and adaptability to dynamic network conditions. Experimental evaluation on the CIC-ToN-IoT and CIC-IDS2017 dataset demonstrates the framework’s superior accuracy, precision, and robustness, establishing it as a practical and effective solution for securing IoT networks against both known and emerging threats.</div></div>","PeriodicalId":55224,"journal":{"name":"Computer Communications","volume":"248 ","pages":"Article 108414"},"PeriodicalIF":4.3,"publicationDate":"2026-01-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145904069","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}

{"title":"Joint optimization of UAV dual-task co-track and charging station location in large-scale IoT scenarios","authors":"Yi Zhang,&nbsp;Yi Hong,&nbsp;Chuanwen Luo,&nbsp;Xin Fan","doi":"10.1016/j.comcom.2026.108412","DOIUrl":"10.1016/j.comcom.2026.108412","url":null,"abstract":"<div><div>Unmanned Aerial Vehicles (UAVs), owing to their high flexibility and mobility, have emerged as efficient tools for data collection in fields such as environmental monitoring and agricultural mapping. However, their limited battery capacity significantly constrains flight range and mission duration. This limitation becomes particularly critical in large-scale Internet of Things (IoT) scenarios involving multiple cooperative UAVs. Existing studies often adopt fixed charging stations or costly mobile charging devices and treat data collection and energy replenishment as separate optimization problems, which hinders task continuity and reduces system energy efficiency. In this paper, we propose a joint optimization framework that integrates charging station placement with collaborative UAV scheduling for dual-task co-track data collection and charging, aiming to maximize data throughput and enhance energy efficiency. A multi-UAV system model is developed that incorporates various constraints, including task allocation, time, and energy. The objective is to jointly optimize the placement of fixed charging stations, the task assignments among UAVs, and the design of flight trajectories that unify data collection and charging operations. To solve this complex joint optimization problem, a path planning collaborative optimization algorithm (PCA) is designed. Simulation results show that, compared with greedy algorithms and fixed charging-station strategies, our method improves energy efficiency by about 31.28% and 15.18%, and reduces task completion time by 31.41% and 14.33%, respectively, clearly demonstrating the effectiveness and advantages of the proposed joint optimization strategy. This study offers a systematic solution for sustainable and efficient UAV-based data collection in complex operational environments.</div></div>","PeriodicalId":55224,"journal":{"name":"Computer Communications","volume":"247 ","pages":"Article 108412"},"PeriodicalIF":4.3,"publicationDate":"2026-01-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145925931","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}

{"title":"Unmanned aerial vehicle-enabled mobile edge computing for semantic communications","authors":"Liyuan Xie ,&nbsp;Wancheng Xie ,&nbsp;Huabing Lu ,&nbsp;Helin Yang","doi":"10.1016/j.comcom.2026.108411","DOIUrl":"10.1016/j.comcom.2026.108411","url":null,"abstract":"<div><div>With the rapid evolution of wireless networks and the increasing demand for flexible communication and computing services, unmanned aerial vehicles (UAVs) have emerged as a promising solution to enhance the performance of these networks. This paper investigates a UAV-assisted mobile edge computing (MEC) system with semantic communication (SemCom) to improve the efficiency of wireless networks by transmitting only meaningful information, thereby reducing bandwidth and computational resource requirements. We propose a resource scheduling approach to minimize the weighted sum of overall latency for task processing and energy consumption under malicious jamming attacks. The approach jointly optimizes device scheduling, UAV trajectory, task offloading ratio, bandwidth allocation, and the number of transmitted SemCom symbols under different constraints. The optimization problem is complex and non-convex, involving ongoing decision-making due to constantly changing parameters. To address this challenge, we present a proximal policy optimization (PPO)-based deep reinforcement learning (DRL) algorithm for real-time resource management. The proposed PPO-based resource scheduling approach effectively schedules both communication and computing resources to minimize the cost of the UAV-enabled wireless network against jamming attacks. Simulation-based performance analysis indicates that the PPO-based SemCom scheme reduces task execution latency and energy consumption compared to baseline approaches across various network scenarios. The proposed framework provides valuable insights into the design and optimization of UAV-assisted MEC systems with SemCom for enhanced wireless network performance in the presence of adversarial jamming.</div></div>","PeriodicalId":55224,"journal":{"name":"Computer Communications","volume":"248 ","pages":"Article 108411"},"PeriodicalIF":4.3,"publicationDate":"2026-01-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145980583","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}

{"title":"Dynamic resource allocation for digital twin-enhanced hierarchical federated learning in sustainable internet of things","authors":"Ze Wei ,&nbsp;Rongxi He ,&nbsp;Xiaojing Chen ,&nbsp;Chengzhi Song","doi":"10.1016/j.comcom.2025.108410","DOIUrl":"10.1016/j.comcom.2025.108410","url":null,"abstract":"<div><div>This paper proposes a digital twin (DT)-enhanced hierarchical federated learning framework for sustainable Internet of Things (IoT) networks. In this framework, mobile edge computing servers coordinate collaborative training, while DTs maintain real-time physical-virtual synchronization. Our core contributions are threefold. First, to tackle device heterogeneity, we propose two mechanisms: (1) an elastic time window that dynamically adapts aggregation deadlines based on median training times while incorporating distance-aware resource compensation to mitigate channel degradation, and (2) a DT-enhanced weighting strategy that dynamically balances energy sustainability, channel quality, and model freshness while guaranteeing convergence through closed-loop cross-layer coordination. Second, we derive a convergence bound explicitly linked to the device participation ratio, establishing a direct theoretical connection between resource allocation and learning performance. Then, through theoretical analysis, it can be found that reducing training latency and energy consumption by jointly optimizing computing and communication resources, as well as EH duration, is key to maximizing this ratio without compromising the reliability of the gradients, thereby indirectly enhancing convergence. Third, guided by this insight, we formulate a mixed-integer nonlinear programming problem that aims to maximize the participation ratio while jointly minimizing energy consumption and training latency, by optimizing the energy harvesting time, collaboration ratio, and communication/computation resources. To solve this NP-hard problem, we propose a DT-driven decomposition framework that partitions it into two subproblems, which are then solved by three DT-driven algorithms with provable near-optimality guarantees. Experimental results validate the superiority of our approach, demonstrating significant improvements in convergence performance, latency, energy efficiency, and participant sample rate, while also advancing the sustainability of FL.</div></div>","PeriodicalId":55224,"journal":{"name":"Computer Communications","volume":"248 ","pages":"Article 108410"},"PeriodicalIF":4.3,"publicationDate":"2026-01-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145929270","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}

{"title":"CISF: Consensus-based Information Sharing Framework for robust consistency in UAVs swarm disaster response","authors":"Xuefeng Du ,&nbsp;Yanqi Cheng ,&nbsp;Li Yin ,&nbsp;Ning Tong ,&nbsp;Fengqiang Xu ,&nbsp;Fengqi Li","doi":"10.1016/j.comcom.2025.108408","DOIUrl":"10.1016/j.comcom.2025.108408","url":null,"abstract":"<div><div>In disaster response scenarios, distributed unmanned aerial vehicle (UAV) swarms face substantial challenges in maintaining real-time information consistency due to network instability, communication delays, and potential Byzantine faults. Traditional approaches often fail to balance fault tolerance, communication latency, and task execution efficiency under such dynamic and adversarial conditions. This paper proposes the Consensus-based Information Sharing Framework (CISF), a novel solution specifically designed to ensure information consistency in dynamic disaster environments. CISF integrates a Stratified Parallel Byzantine Fault Tolerance (SPBFT) mechanism — optimized via a dynamic capability-reputation evaluation model — with a Multi-Round Search and Patrol Model (MSPM) based on an improved Cuckoo Search algorithm. MSPM employs a multi-objective fitness function to jointly optimize temporal efficiency, spatial coverage, and task priority, enabling comprehensive area exploration and continuous information validation. Theoretical analysis derives the optimal hierarchical ratio and the maximum fault tolerance threshold for CISF. Simulation results show that CISF maintains 93.8% consistency under Byzantine interference and reduces consensus latency by up to 56.2%, while remaining effective in highly dynamic environments. Overall, this study establishes a robust and efficient framework for achieving real-time, fault-tolerant information consistency in interference-prone UAV networks, offering broad applicability for future swarm-based disaster response systems.</div></div>","PeriodicalId":55224,"journal":{"name":"Computer Communications","volume":"247 ","pages":"Article 108408"},"PeriodicalIF":4.3,"publicationDate":"2026-01-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145925932","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}

{"title":"Joint design of resource allocation and QoS enhancement via serial optimization in UAV-NOMA communications","authors":"Zhongyu Wang ,&nbsp;Yanan Lian ,&nbsp;Jie Zeng ,&nbsp;Zheng Chang ,&nbsp;Tiejun Lv","doi":"10.1016/j.comcom.2025.108409","DOIUrl":"10.1016/j.comcom.2025.108409","url":null,"abstract":"<div><div>We investigate the challenges of user pairing, power allocation, and bandwidth allocation problems in unmanned aerial vehicle (UAV) systems that employ nonorthogonal multiple access (NOMA) for communication with multiple ground users. The primary objective is to maximize the system’s achievable transmission rate while ensuring the users’ quality of service (QoS) requirements under a constrained total power budget. Considering the nonconvexity of the original problem and the interdependencies among multiple optimization variables, the problem is decomposed into three subproblems to optimize power and bandwidth allocation. To increase resource utilization and address user pairing challenges, a serial-optimized communication scheme is proposed, which leverages an optimized block coordinate descent (OP-BCD) method to sequentially solve the subproblems. Specifically, the power allocation strategy is optimized using an optimized deep Q-network (DQN) combined with a gradient ascent approach, whereas the intergroup bandwidth is optimized via a sequential least squares programming (SLSQP). Simulation results demonstrate that the proposed group matching method significantly enhances resource utilization and fairness compared to other user pairing strategies. Moreover, the proposed scheme effectively increases the system transmission rate and resource efficiency.</div></div>","PeriodicalId":55224,"journal":{"name":"Computer Communications","volume":"247 ","pages":"Article 108409"},"PeriodicalIF":4.3,"publicationDate":"2025-12-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145884353","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}