Ad Hoc Networks最新文献

{"title":"Edge assisted heterogeneity aware vehicular selection using Federated Learning","authors":"Akshay Singh ,&nbsp;Alok Ranjan ,&nbsp;Guru Prasad A.S.","doi":"10.1016/j.adhoc.2025.103896","DOIUrl":"10.1016/j.adhoc.2025.103896","url":null,"abstract":"<div><div>Recently, artificial intelligence (AI) enabled solutions are getting popular to solve several challenges in Vehicular networks. Among several, AI based solutions are leveraged for road safety applications including driver behavior monitoring (DBM). Although traditional AI solutions have been adopted to build such solutions, user privacy remained a significant challenge. In addition, majorly the AI solution offerings are limited to centralized server computing. To address this, Federated Learning (FL) has been recently adopted in vehicular networks to train the model locally on the edge devices without revealing the private information and need for centralized server computing. However, its large-scale implementations are restricted due to practical considerations such as data heterogeneity &amp; quality of data which further results in inferior model performance. This paper presents a Federated Heterogeneity Aware Vehicle Selection (FedHAVS) approach which incorporates edge computing and serverless computing to optimize scalability and efficiency. Specifically, we leverage edge computing to decentralize model training at the vehicle level and enables distributed learning to improve decision making. The novelty of our work lies in the proposed algorithm used to measure the degree of heterogeneity across the clients which helps in segregating optimal and sub-optimal clients. Then a global model is obtained using the FL framework among optimal and sub-optimal set of clients which is tested on two different datasets namely SFDDD and RevItsOne. To evaluate the robustness of our method, we incorporated various state-of-the-art models. The experimental results demonstrate that FedHAVS achieves 99.24% &amp; 98.9% accuracy with Resnet18 on Independent and Identically Distributed (IID) settings for both SFDDD and RevItsOne datasets respectively. On Non-Independent and Identically Distributed (Non-IID) settings, FedHAVS achieved 98.84% on SFDDD and 97.66% on RevItsOne dataset with Densenet. Additionaly, compared to baselines, our proposed method reports least drop in accuracy when degree of heterogeneity increases ensuring better performance while the convergence time is minimum across both datasets.</div></div>","PeriodicalId":55555,"journal":{"name":"Ad Hoc Networks","volume":"177 ","pages":"Article 103896"},"PeriodicalIF":4.4,"publicationDate":"2025-05-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144116622","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":"Teacher–student learning based low complexity relay selection in wireless powered communications","authors":"Aysun Gurur Onalan ,&nbsp;Berkay Kopru ,&nbsp;Sinem Coleri","doi":"10.1016/j.adhoc.2025.103894","DOIUrl":"10.1016/j.adhoc.2025.103894","url":null,"abstract":"<div><div>Radio Frequency Energy Harvesting (RF-EH) networks are pivotal in enabling massive Internet-of-Things by facilitating controlled, long-distance energy transfer to energy-constrained devices. Relays, which assist in either energy or information transfer, significantly enhance the performance of such networks. However, the relay selection problem in multiple-source–multiple-relay RF-EH networks poses substantial computational challenges. To address these, this paper proposes a novel deep-learning-based relay selection framework that integrates convolutional neural networks (CNNs) and teacher–student learning. Specifically, the joint relay selection, time allocation, and power control problem are studied under non-linear EH conditions. First, the optimal solution to the time and power allocation problem for a given relay selection is derived. Then, the relay selection problem is formulated as a classification task, and two CNN-based architectures are proposed. To further improve computational efficiency without compromising accuracy, the teacher–student learning paradigm is employed, wherein a smaller student network is trained with the distilled knowledge of a larger teacher network. A novel dichotomous search-based algorithm is introduced to determine the optimal architecture of the student network. Simulation results demonstrate that the proposed solutions achieve lower complexity compared to state-of-the-art iterative approaches while maintaining optimality.</div></div>","PeriodicalId":55555,"journal":{"name":"Ad Hoc Networks","volume":"176 ","pages":"Article 103894"},"PeriodicalIF":4.4,"publicationDate":"2025-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144068964","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":"A fire hawk optimizer-based energy-efficient clustering scheme in underwater acoustic sensor networks (UASNs)","authors":"Sang-Woong Lee ,&nbsp;Musaed Alhussein ,&nbsp;Khursheed Aurangzeb ,&nbsp;Mohammad Sadegh Yousefpoor ,&nbsp;Efat Yousefpoor ,&nbsp;Mehdi Hosseinzadeh","doi":"10.1016/j.adhoc.2025.103889","DOIUrl":"10.1016/j.adhoc.2025.103889","url":null,"abstract":"<div><div>An underwater acoustic sensor network (UASN) is suitable for gathering data from aquatic environments, including lakes, rivers, seas, and oceans. This network faces several issues due to the distinct features of underwater environments and the limitations of acoustic channels. These challenges include energy limitations, unreliable communication links, and dynamic network topologies. Additionally, the difficulty of recharging or replacing batteries in underwater conditions makes energy optimization essential for prolonging the network lifespan. Currently, many energy-efficient approaches in UASNs emphasize node clustering and multi-hop communication, but most of these methods rely on distributed algorithms. This paper introduces a novel energy-efficient clustering framework called FHOEEC (<u><strong>F</strong></u>ire <u><strong>H</strong></u>awk <u><strong>O</strong></u>ptimization-based <u><strong>E</strong></u>nergy-<u><strong>E</strong></u>fficient <u><strong>C</strong></u>lustering), which integrates both distributed and centralized strategies. The clustering process is divided into three stages: (1) cluster formation, (2) selection of cluster heads, and (3) cluster maintenance. During the periodic neighbor discovery phase, FHOEEC examines two key aspects: the format of the hello packet and its propagation process. FHOEEC aims to create an energy-efficient, cluster-based network structure. To achieve this, the sink node utilizes the fire hawk optimization (FHO) algorithm to decide on the optimal range and number of clusters. To establish these clusters, a fitness function considers a weighted combination of three sub-functions: intra-cluster and inter-cluster distances, the proportion of isolated clusters compared to others, and cluster density. In the final stage, intra-cluster and inter-cluster communication paths are established by focusing on energy balance. This ensures that nodes with energy levels below a specified threshold are excluded from serving as intermediate nodes. Simulation results and performance evaluations show that FHOEEC outperforms three existing clustering methods–CCCS, GTC, and EULC–in terms of energy efficiency and network performance. Therefore, FHOEEC significantly enhances network lifespan, balances energy usage among the nodes, and offers better scalability than other schemes.</div></div>","PeriodicalId":55555,"journal":{"name":"Ad Hoc Networks","volume":"176 ","pages":"Article 103889"},"PeriodicalIF":4.4,"publicationDate":"2025-05-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143946872","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":"UAV-Edge Cloud collaboration for online offloading and trajectory control in multi-layer Mobile Edge Computing","authors":"Siyang Xu ,&nbsp;Jingyi Ma ,&nbsp;Qiuyu Lu ,&nbsp;Zhigang Xie ,&nbsp;Xin Song","doi":"10.1016/j.adhoc.2025.103866","DOIUrl":"10.1016/j.adhoc.2025.103866","url":null,"abstract":"<div><div>Integrating Mobile Edge Computing (MEC) with Unmanned Aerial Vehicles (UAVs) offers enhanced coverage and computational support for mobile IoT devices (MIDs). Due to the inherent computational capacity and energy constraints of UAV, existing UAV-assisted MEC systems struggle to satisfy computation-intensive network services for numerous MIDs. To address this issue, this paper proposes a UAV and ground-based Edge Cloud (EC) collaboration MEC system. Specifically, the EC is equipped with an energy transmitter to provide wireless power transfer (WPT) to the UAV, thereby collaboratively managing backlog tasks in scenarios where task and energy arrive stochastically. To achieve optimal service delivery, we formulate a long-term stochastic optimization problem aiming to jointly optimize UAV energy consumption and system throughput while ensuring task queue stability. However, this NP-hard problem posed by the stochastic nature of task arrivals and energy constraints, we develop an online offloading and trajectory control (OOTC) algorithm. This algorithm uses Lyapunov optimization theory to transform the long-term stochastic optimization problem into a deterministic per-slot optimization problem. The OOTC algorithm decouples control decisions across consecutive time slots, reducing computational complexity and ensuring queue stability without relying on statistical knowledge. We further decompose it into three subproblems, and derive suboptimal solutions by the Successive Convex Approximation (SCA) and the Lagrangian duality. Simulations show OOTC algorithm outperforms benchmarks and maintains stability.</div></div>","PeriodicalId":55555,"journal":{"name":"Ad Hoc Networks","volume":"176 ","pages":"Article 103866"},"PeriodicalIF":4.4,"publicationDate":"2025-05-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143937351","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":"Energy-saving and security-enhanced task offloading strategies in D2D-integrated MEC networks","authors":"Xiaowei Shi,&nbsp;Linyu Huang","doi":"10.1016/j.adhoc.2025.103890","DOIUrl":"10.1016/j.adhoc.2025.103890","url":null,"abstract":"<div><div>With the rapid development of the Internet of Things, the demand for low latency and efficient computation has increased significantly. Mobile edge computing (MEC) has become a key technology for improving the performance of Internet of Things (IoT) systems. In cellular networks, computation offloading through D2D communication between edge devices can effectively reduce task latency and improve energy efficiency. However, most of the existing works focus on energy efficiency and delay optimization and often ignores security issues. Hence, we study the D2D offloading problem of edge devices under centralized scheduling of MEC servers, and propose a risk assessment criterion based on security level to jointly optimize energy consumption and security. To solve the problem, it was first modeled as a Mixed-Integer Nonlinear Programming (MINLP) problem. By optimizing the constraints, the problem was transformed into an Integer Linear Programming (ILP) problem and the theoretical optimal solution was obtained. Considering the requirements of practical engineering applications, a low-complexity heuristic algorithm is designed. The proposed strategies can be widely used in D2D-Integrated MEC Networks to improve system energy efficiency and security.</div></div>","PeriodicalId":55555,"journal":{"name":"Ad Hoc Networks","volume":"176 ","pages":"Article 103890"},"PeriodicalIF":4.4,"publicationDate":"2025-05-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143928324","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":"Schedulability analysis in time-sensitive networking: A systematic literature review","authors":"Zitong Wang,&nbsp;Feng Luo,&nbsp;Yunpeng Li,&nbsp;Haotian Gan,&nbsp;Lei Zhu","doi":"10.1016/j.adhoc.2025.103897","DOIUrl":"10.1016/j.adhoc.2025.103897","url":null,"abstract":"<div><div>Time-Sensitive Networking (TSN) is a set of standards that provide low-latency, high-reliability guarantees for the transmission of traffic in networks, and it is becoming an accepted solution for complex time-critical systems such as those in industrial automation and the automotive. In time-critical systems, it is essential to verify the timing predictability of the system, and the application of scheduling mechanisms in TSN can also bring about changes in system timing. Therefore, schedulability analysis techniques can be used to verify that the system is scheduled according to the scheduling mechanism and meets the timing requirements. In this paper, we provide a clear overview of the state-of-the-art works on the topic of schedulability analysis in TSN in an attempt to clarify the purpose of schedulability analysis, categorize the methods of schedulability analysis and compare their respective strengths and weaknesses, point out the scheduling mechanisms under analyzing and the corresponding traffic classes, clarify the network scenarios constructed during the evaluation and list the challenges and directions still needing to be worked on in schedulability analysis in TSN. To this end, we conducted a systematic literature review and finally identified 123 relevant research papers published in major conferences and journals in the past 15 years. Based on a comprehensive review of the relevant literature, we have identified several key findings and emphasized the future challenges in schedulability analysis for TSN.</div></div>","PeriodicalId":55555,"journal":{"name":"Ad Hoc Networks","volume":"176 ","pages":"Article 103897"},"PeriodicalIF":4.4,"publicationDate":"2025-05-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143942231","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":"Quantum GA-driven Digital Twin for task urgency-aware partitioning and offloading in multi UAV-Aided MEC systems","authors":"Santanu Ghosh,&nbsp;Pratyay Kuila","doi":"10.1016/j.adhoc.2025.103891","DOIUrl":"10.1016/j.adhoc.2025.103891","url":null,"abstract":"<div><div>Unmanned aerial vehicle (UAV)-assisted mobile edge computing (MEC) empowers smart mobile devices (SMDs) to efficiently handle computation- and resource-intensive applications, particularly in critical scenarios. The integration of Digital Twin (DT) technology enhances scalability and streamlines the management of multi-user, multi-UAV-assisted MEC systems. This research focuses on partial task offloading within DT-enabled UAV-assisted MEC, addressing the joint problem of task partitioning and offloading using a quantum-inspired genetic algorithm (QIGA). The quantum chromosome is encoded and decoded through linear hashing. Task partitioning is performed to optimize system efficiency in terms of energy, latency, and load distribution across the MEC, while also considering task urgency. The fitness function incorporates two penalty factors to eliminate solutions that violate task deadlines or exceed the energy constraints of SMDs and edge servers. The QIGA is demonstrated to operate in polynomial time across all phases. Extensive simulations under various scenarios reveal that the proposed QIGA outperforms other algorithms in terms of energy efficiency, delay reduction, and load balancing within UAV-assisted MEC. Statistical analyses further validate the reliability and effectiveness of the results.</div></div>","PeriodicalId":55555,"journal":{"name":"Ad Hoc Networks","volume":"176 ","pages":"Article 103891"},"PeriodicalIF":4.4,"publicationDate":"2025-05-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143916149","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":"PDRL-CM: An efficient cooperative caching management method for vehicular networks based on deep reinforcement learning","authors":"Pingjie Ou ,&nbsp;Ningjiang Chen ,&nbsp;Long Yang","doi":"10.1016/j.adhoc.2025.103888","DOIUrl":"10.1016/j.adhoc.2025.103888","url":null,"abstract":"<div><div>In vehicular networks, onboard devices face the challenge of limited storage, and computational resources constrain their processing and storage capabilities. This limitation is particularly significant for applications that require complex computations and real-time responses. Additionally, limited storage capacity reduces the range of cacheable data, which can impact the immediate availability of data and the continuity of services. Therefore, improving cache utilization and meeting vehicles’ real-time data demands pose significant challenges. Deep reinforcement learning can optimize the issues arising from agents’ continuously changing state and action spaces due to increasing request demands. However, training the network may encounter instability and convergence difficulties in dynamic and complex environments or situations with sparse rewards. In response to these issues, this paper proposes a Priority-based Deep Reinforcement Learning Collaborative Cache Management method (PDRL-CM). PDRL-CM first designs a lightweight cache admission strategy that leverages data’s inherent and combined attributes. It then makes cache admission decisions Using Monte Carlo sampling and a max-value search strategy combined with a feedforward neural network. Secondly, the method considers minimizing system latency and reducing vehicle energy consumption as joint optimization problems. An improved deep reinforcement learning algorithm solves this problem and makes cache-sharding decisions. A prioritized experience replay mechanism is incorporated to adjust the network prediction model quickly and accelerate the convergence process. Experimental results indicate that, compared to existing DRL-based caching methods, PDRL-CM offers faster data processing efficiency and higher cache hit rates under varying vehicle density, storage capacity, and content volume conditions.</div></div>","PeriodicalId":55555,"journal":{"name":"Ad Hoc Networks","volume":"176 ","pages":"Article 103888"},"PeriodicalIF":4.4,"publicationDate":"2025-05-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143928323","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":"ESFMTO: A reliable task offloading strategy based on edge server failure model in IIoT","authors":"Yubin Yang ,&nbsp;Yan Chen ,&nbsp;Ningjiang Chen ,&nbsp;Juan Chen","doi":"10.1016/j.adhoc.2025.103887","DOIUrl":"10.1016/j.adhoc.2025.103887","url":null,"abstract":"<div><div>The extensive use of automation equipment and sensors in the Industrial Internet of Things (IIoT) has led to a significant increase in data volume, which has placed higher demands on the real-time processing capability. Edge computing enables real-time response and rapid decision-making by offloading data processing to the edge of the network. However, the complexity of the industrial production environment leads to edge server failures, which seriously affects the system stability and security. To address this issue, this paper develops an edge server failure model for IIoT, analyzing the interaction between the hardware failure and the container failure during the failure occurrence and recovery. Further, based on the edge server failure model, a task offloading strategy named ESFMTO is proposed, which employs the SAC-BNN (Soft Actor-Critic with Bayesian Neural Network) algorithm. The probability distribution of the task completion time is updated through the Bayesian Neural Network (BNN), accurately evaluating the Conditional Value at Risk (CVaR). During the training process of Deep Reinforcement Learning (DRL), a perturbation neural network is introduced to perturb the input state, which enhances the system robustness under uncertain failure conditions. Unlike existing approaches that often assume each edge server hosts a single container, the paper explicitly considers multi-container deployment in IIoT, bridging the gap between theoretical assumptions and real-world industrial requirements. Experimental results demonstrate that SAC-BNN outperforms existing methods in dealing with the edge server failure, reducing CVaR by at least around 0.9 and improving task completion rates by at least 13.86% compared to the baseline algorithms.</div></div>","PeriodicalId":55555,"journal":{"name":"Ad Hoc Networks","volume":"176 ","pages":"Article 103887"},"PeriodicalIF":4.4,"publicationDate":"2025-05-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143911784","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":"Swarm intelligence-inspired localization and power control for terahertz (THz) UAV-vehicle networks","authors":"Enis Körpe ,&nbsp;Mustafa Alper Akkaş ,&nbsp;Yavuz Öztürk","doi":"10.1016/j.adhoc.2025.103892","DOIUrl":"10.1016/j.adhoc.2025.103892","url":null,"abstract":"<div><div>In recent years, terahertz (THz) communication has gained significant attention as a transformative technology for high-speed wireless networks, addressing the limitations of conventional frequency bands in meeting the escalating demand for data transmission. THz communication is particularly critical in vehicle-to-everything (V2X) and unmanned aerial vehicle (UAV)-based communication networks, where ultra-low latency, high bandwidth, and reliable connectivity are essential. Operating in the frequency spectrum between the microwave and infrared bands, THz communication offers the potential for multi-gigabit data transmission rates, rendering it a promising enabler for next-generation intelligent transportation systems, autonomous vehicles, and UAV-supported applications. Furthermore, artificial intelligence (AI) emerges as a pivotal tool to enhance the reliability and efficiency of THz-based V2X and UAV communication networks by enabling the prediction of network traffic patterns and mobility dynamics. This study introduces a swarm intelligence-based AI approach designed to optimize system performance by minimizing latency and transmission power requirements while ensuring the required signal-to-noise ratio (SNR) within a UAV-assisted vehicular network operating in the THz band. The proposed methodology employs a dual-objective optimization framework that balances latency and transmission power within a predefined communication time frame. Comparative analysis is conducted between a baseline network with randomly distributed UAVs and a network employing UAV deployment guided by the proposed AI scheme. Also, the performance of proposed method is compared with existing swarm intelligence algorithms. Performance metrics, including SNR and latency, are evaluated to assess the system’s efficacy. The channel modeling process leverages the Line-by-Line Radiative Transfer Model (LBLRTM) to characterize the propagation environment in the UAV-assisted vehicular network.</div></div>","PeriodicalId":55555,"journal":{"name":"Ad Hoc Networks","volume":"176 ","pages":"Article 103892"},"PeriodicalIF":4.4,"publicationDate":"2025-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143906974","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}