Ad Hoc Networks最新文献_第8页

A blockchain-enhanced trust-driven batch authentication scheme for secure VANETs 一种用于安全vanet的区块链增强的信任驱动批认证方案

IF 4.8 3区计算机科学 Q1 COMPUTER SCIENCE, INFORMATION SYSTEMS

Ad Hoc Networks

Pub Date : 2025-10-10 DOI: 10.1016/j.adhoc.2025.104045

Longxia Liao , Junhui Zhao , Qingmiao Zhang , He Fang

Ensuring secure and efficient message authentication in vehicular ad hoc networks (VANETs) is a significant challenge, particularly in the face of insider threats and high mobility. This paper proposes a trust-driven batch authentication scheme, dynamically optimized through real-time trust updates. We introduce a lightweight certificateless elliptic curve cryptography (CL-ECC) protocol for efficient batch verification, combined with a blockchain-based dynamic trust evaluation model that incorporates a Gompertz-based trust update function and an enhanced proof-of-work (PoW) algorithm for decentralized trust management. To ensure tamper-resilience and decentralization, Practical Byzantine Fault Tolerance (PBFT) consensus is employed. Security analysis demonstrates that the proposed scheme effectively defends against forgery, message tampering, replay attacks, and internal threats such as on-off and collusion attacks. Experimental results show that our scheme reduces computational overhead by 7.9-56.9% and improves average message utilization by 114.2-198.3% compared to conventional schemes under various scenarios. These results demonstrate that the proposed scheme is secure, efficient, and well-suited for highly mobile and trust-sensitive VANET environments.

在车载自组织网络（vanet）中确保安全高效的消息认证是一项重大挑战，特别是在面对内部威胁和高移动性的情况下。提出了一种基于信任驱动的批量认证方案，该方案通过实时的信任更新进行动态优化。我们引入了一种轻量级的无证书椭圆曲线加密（CL-ECC）协议，用于高效的批量验证，并结合了基于区块链的动态信任评估模型，该模型结合了基于gompertz的信任更新功能和用于分散信任管理的增强工作量证明（PoW）算法。为了确保防篡改和去中心化，采用了实用拜占庭容错（PBFT）共识。安全性分析表明，该方案能够有效防御伪造、消息篡改、重放攻击以及开关攻击和合谋攻击等内部威胁。实验结果表明，在各种场景下，与传统方案相比，该方案的计算开销降低了7.9 ~ 56.9%，平均消息利用率提高了114.2 ~ 198.3%。这些结果表明，该方案安全、高效，非常适合高度移动和信任敏感的VANET环境。

{"title":"A blockchain-enhanced trust-driven batch authentication scheme for secure VANETs","authors":"Longxia Liao , Junhui Zhao , Qingmiao Zhang , He Fang","doi":"10.1016/j.adhoc.2025.104045","DOIUrl":"10.1016/j.adhoc.2025.104045","url":null,"abstract":"<div><div>Ensuring secure and efficient message authentication in vehicular ad hoc networks (VANETs) is a significant challenge, particularly in the face of insider threats and high mobility. This paper proposes a trust-driven batch authentication scheme, dynamically optimized through real-time trust updates. We introduce a lightweight certificateless elliptic curve cryptography (CL-ECC) protocol for efficient batch verification, combined with a blockchain-based dynamic trust evaluation model that incorporates a Gompertz-based trust update function and an enhanced proof-of-work (PoW) algorithm for decentralized trust management. To ensure tamper-resilience and decentralization, Practical Byzantine Fault Tolerance (PBFT) consensus is employed. Security analysis demonstrates that the proposed scheme effectively defends against forgery, message tampering, replay attacks, and internal threats such as on-off and collusion attacks. Experimental results show that our scheme reduces computational overhead by 7.9-56.9% and improves average message utilization by 114.2-198.3% compared to conventional schemes under various scenarios. These results demonstrate that the proposed scheme is secure, efficient, and well-suited for highly mobile and trust-sensitive VANET environments.</div></div>","PeriodicalId":55555,"journal":{"name":"Ad Hoc Networks","volume":"180 ","pages":"Article 104045"},"PeriodicalIF":4.8,"publicationDate":"2025-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145322000","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}

引用次数: 0

Semantic communication-driven offloading for MEC-integrated D2D networks: A security and energy perspective 语义通信驱动的mec集成D2D网络卸载：安全和能源视角

IF 4.8 3区计算机科学 Q1 COMPUTER SCIENCE, INFORMATION SYSTEMS

Ad Hoc Networks

Pub Date : 2025-10-10 DOI: 10.1016/j.adhoc.2025.104046

Xiaowei Shi, Linyu Huang

Mobile Edge Computing (MEC) improves system response speed and efficiency by offloading computing tasks to the edge of the network. However, with the rapid development of the Internet of Things (IoT), the number of IoT devices and traffic surge, and the pressure on the computing and communication resources of MEC increases dramatically. Traditional computation offloading methods have been difficult to meet the needs for low latency and high efficiency. As a new communication paradigm, semantic communication significantly reduces the amount of data in communication by transmitting the “semantics” of information instead of redundant raw data, thus compensating for the shortcomings of MEC in the field of computation offloading. However, semantic communication is still in its early development stage, and there are relatively few studies on semantic communication-based computation offloading. To this end, this paper introduces semantic communication based on the Device-to-Device (D2D) assisted MEC system, and incorporates user mobility awareness to better adapt to dynamic environments and improve the accuracy of offloading decisions. By jointly optimizing offloading decision-making and resource allocation, the proposed system aims to minimize energy consumption and security risk. The optimization problem is modeled as an Integer Linear Programming (ILP) problem, and a low-complexity heuristic algorithm is designed to meet practical engineering requirements. The simulation results show that the schemes designed in this paper can flexibly adjust energy consumption and security risks, and show good performance in different scale scenarios, which verifies the potential and practicability of semantic communication in computation offloading.

移动边缘计算（MEC）通过将计算任务转移到网络边缘，提高系统响应速度和效率。然而，随着物联网（IoT）的快速发展，物联网设备数量和流量激增，MEC的计算和通信资源压力急剧增加。传统的计算卸载方法已经难以满足低延迟和高效率的要求。语义通信作为一种新的通信范式，通过传递信息的“语义”而不是冗余的原始数据，大大减少了通信中的数据量，从而弥补了MEC在计算卸载领域的不足。然而，语义通信仍处于早期发展阶段，基于语义通信的计算卸载研究相对较少。为此，本文引入了基于设备到设备（Device-to-Device, D2D）辅助MEC系统的语义通信，并结合用户移动性感知，以更好地适应动态环境，提高卸载决策的准确性。通过共同优化卸载决策和资源配置，实现能源消耗和安全风险最小化。将优化问题建模为整数线性规划（ILP）问题，设计了一种满足实际工程要求的低复杂度启发式算法。仿真结果表明，本文设计的方案能够灵活调整能耗和安全风险，并在不同规模场景下表现出良好的性能，验证了语义通信在计算卸载中的潜力和实用性。

{"title":"Semantic communication-driven offloading for MEC-integrated D2D networks: A security and energy perspective","authors":"Xiaowei Shi, Linyu Huang","doi":"10.1016/j.adhoc.2025.104046","DOIUrl":"10.1016/j.adhoc.2025.104046","url":null,"abstract":"<div><div>Mobile Edge Computing (MEC) improves system response speed and efficiency by offloading computing tasks to the edge of the network. However, with the rapid development of the Internet of Things (IoT), the number of IoT devices and traffic surge, and the pressure on the computing and communication resources of MEC increases dramatically. Traditional computation offloading methods have been difficult to meet the needs for low latency and high efficiency. As a new communication paradigm, semantic communication significantly reduces the amount of data in communication by transmitting the “semantics” of information instead of redundant raw data, thus compensating for the shortcomings of MEC in the field of computation offloading. However, semantic communication is still in its early development stage, and there are relatively few studies on semantic communication-based computation offloading. To this end, this paper introduces semantic communication based on the Device-to-Device (D2D) assisted MEC system, and incorporates user mobility awareness to better adapt to dynamic environments and improve the accuracy of offloading decisions. By jointly optimizing offloading decision-making and resource allocation, the proposed system aims to minimize energy consumption and security risk. The optimization problem is modeled as an Integer Linear Programming (ILP) problem, and a low-complexity heuristic algorithm is designed to meet practical engineering requirements. The simulation results show that the schemes designed in this paper can flexibly adjust energy consumption and security risks, and show good performance in different scale scenarios, which verifies the potential and practicability of semantic communication in computation offloading.</div></div>","PeriodicalId":55555,"journal":{"name":"Ad Hoc Networks","volume":"180 ","pages":"Article 104046"},"PeriodicalIF":4.8,"publicationDate":"2025-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145322055","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}

引用次数: 0

Cooperative task offloading and resource allocation for sequential constraint tasks in satellite edge computing networks 卫星边缘计算网络中顺序约束任务的协同任务卸载与资源分配

IF 4.8 3区计算机科学 Q1 COMPUTER SCIENCE, INFORMATION SYSTEMS

Ad Hoc Networks

Pub Date : 2025-10-08 DOI: 10.1016/j.adhoc.2025.104044

Peng Deng, Xiangyang Gong, Ziyi Wang, Xirong Que

Satellite remote sensing technology has promoted the emergence of a variety of earth observation (EO) tasks with Ultra HD resolution and/or high real-time requirements. Limited by the bandwidth of space-ground links and the computing power of a single satellite, on-orbit collaborative edge computing improves task processing efficiency. The task is split into subtasks and offloaded to different computing nodes. Multi-layer AI structure leads to sequential dependency among some subtasks. With the challenges of sequential constraints subtasks in dynamic satellite collaborative edge computing scenarios, satellite visible model based on spatial geometry is first proposed in this paper to characterize the communication window between satellites. And the objective function is formulated to minimize the weighted cost of system delay and energy consumption by jointly optimizing cooperative task offloading and resource allocation. This non-convex problem is further decomposed into the subtask offloading and resource allocation subproblems, which are solved by Tabu search algorithm and successive convex approximation algorithm respectively. The simulation results show that the proposed cooperative edge computing scheme reduces the latency and energy consumption weighted cost by 29.3% and 69.4%, respectively, compared with the satellite local computing scheme and the method of calculation on the ground after full download.

卫星遥感技术促进了各种超高清分辨率和/或高实时性要求的地球观测（EO）任务的出现。受空间-地面链路带宽和单个卫星计算能力的限制，在轨协同边缘计算提高了任务处理效率。任务被分解成子任务，并卸载到不同的计算节点。多层人工智能结构导致一些子任务之间的顺序依赖。针对动态卫星协同边缘计算场景中序列约束子任务的挑战，本文首次提出了基于空间几何的卫星可见性模型来表征卫星间的通信窗口。通过共同优化协同任务卸载和资源分配，建立了以系统延迟和能耗加权代价最小为目标函数。该非凸问题进一步分解为任务卸载子问题和资源分配子问题，分别采用禁忌搜索算法和逐次凸逼近算法求解。仿真结果表明，与卫星本地计算方案和全下载后地面计算方法相比，所提出的协同边缘计算方案的时延和能耗加权代价分别降低了29.3%和69.4%。

{"title":"Cooperative task offloading and resource allocation for sequential constraint tasks in satellite edge computing networks","authors":"Peng Deng, Xiangyang Gong, Ziyi Wang, Xirong Que","doi":"10.1016/j.adhoc.2025.104044","DOIUrl":"10.1016/j.adhoc.2025.104044","url":null,"abstract":"<div><div>Satellite remote sensing technology has promoted the emergence of a variety of earth observation (EO) tasks with Ultra HD resolution and/or high real-time requirements. Limited by the bandwidth of space-ground links and the computing power of a single satellite, on-orbit collaborative edge computing improves task processing efficiency. The task is split into subtasks and offloaded to different computing nodes. Multi-layer AI structure leads to sequential dependency among some subtasks. With the challenges of sequential constraints subtasks in dynamic satellite collaborative edge computing scenarios, satellite visible model based on spatial geometry is first proposed in this paper to characterize the communication window between satellites. And the objective function is formulated to minimize the weighted cost of system delay and energy consumption by jointly optimizing cooperative task offloading and resource allocation. This non-convex problem is further decomposed into the subtask offloading and resource allocation subproblems, which are solved by Tabu search algorithm and successive convex approximation algorithm respectively. The simulation results show that the proposed cooperative edge computing scheme reduces the latency and energy consumption weighted cost by 29.3% and 69.4%, respectively, compared with the satellite local computing scheme and the method of calculation on the ground after full download.</div></div>","PeriodicalId":55555,"journal":{"name":"Ad Hoc Networks","volume":"180 ","pages":"Article 104044"},"PeriodicalIF":4.8,"publicationDate":"2025-10-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145269960","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}

引用次数: 0

Enhanced hybrid PSO-FA: Joint optimization of resource allocation for VFC in 6G networks 增强型混合PSO-FA: 6G网络中VFC资源分配的联合优化

IF 4.8 3区计算机科学 Q1 COMPUTER SCIENCE, INFORMATION SYSTEMS

Ad Hoc Networks

Pub Date : 2025-10-08 DOI: 10.1016/j.adhoc.2025.104048

Fuqi Zhang , Huilin Jiang , Fu Liu , Kevin I-Kai Wang , Tao Hou , Yujia Liu , Xingtong Mu

Mobile vehicular ad hoc networks with fog computing in 6 G present a novel distributed architecture to enhance network computing capabilities in dynamic environments. However, these self-organizing networks face significant quality-of-service challenges including large average latency, high total energy consumption, and poor load balancing among fog nodes during multi-task offloading in mobile scenarios. This paper proposes a cross-layer resource and information management approach based on enhanced hybrid particle swarm optimization and firefly algorithm (EH-PSO-FA) for joint optimization. A fitness function incorporating load balancing and local computation penalties is designed to address mobility-aware resource allocation. To enhance global search capability and robustness in dynamic ad hoc environments, both PSO and FA algorithms are enhanced with adaptive mechanisms to avoid premature convergence and local optima. Extensive performance analysis and simulation demonstrate that EH-PSO-FA significantly outperforms baseline protocols, achieving up to 26.18% reduction in average latency and 57.74% improvement in energy-efficient design. The scalable solution effectively supports power-aware computing in mobile wireless ad hoc networks for urban vehicular scenarios.

基于雾计算的6g移动车辆自组网为增强动态环境下的网络计算能力提供了一种新的分布式架构。然而，这些自组织网络面临着显著的服务质量挑战，包括大的平均延迟、高的总能耗以及移动场景中多任务卸载过程中雾节点之间的负载平衡差。提出了一种基于增强混合粒子群优化和萤火虫算法（EH-PSO-FA）的跨层资源信息管理方法，用于联合优化。设计了一个结合负载平衡和局部计算惩罚的适应度函数来解决移动性感知的资源分配问题。为了增强动态环境下的全局搜索能力和鲁棒性，粒子群算法和遗传算法都加入了自适应机制，以避免过早收敛和局部最优。广泛的性能分析和仿真表明，EH-PSO-FA显著优于基线协议，平均延迟降低26.18%，节能设计提高57.74%。可扩展的解决方案有效地支持城市车辆场景的移动无线自组织网络中的功率感知计算。

{"title":"Enhanced hybrid PSO-FA: Joint optimization of resource allocation for VFC in 6G networks","authors":"Fuqi Zhang , Huilin Jiang , Fu Liu , Kevin I-Kai Wang , Tao Hou , Yujia Liu , Xingtong Mu","doi":"10.1016/j.adhoc.2025.104048","DOIUrl":"10.1016/j.adhoc.2025.104048","url":null,"abstract":"<div><div>Mobile vehicular ad hoc networks with fog computing in 6 G present a novel distributed architecture to enhance network computing capabilities in dynamic environments. However, these self-organizing networks face significant quality-of-service challenges including large average latency, high total energy consumption, and poor load balancing among fog nodes during multi-task offloading in mobile scenarios. This paper proposes a cross-layer resource and information management approach based on enhanced hybrid particle swarm optimization and firefly algorithm (EH-PSO-FA) for joint optimization. A fitness function incorporating load balancing and local computation penalties is designed to address mobility-aware resource allocation. To enhance global search capability and robustness in dynamic ad hoc environments, both PSO and FA algorithms are enhanced with adaptive mechanisms to avoid premature convergence and local optima. Extensive performance analysis and simulation demonstrate that EH-PSO-FA significantly outperforms baseline protocols, achieving up to 26.18% reduction in average latency and 57.74% improvement in energy-efficient design. The scalable solution effectively supports power-aware computing in mobile wireless ad hoc networks for urban vehicular scenarios.</div></div>","PeriodicalId":55555,"journal":{"name":"Ad Hoc Networks","volume":"180 ","pages":"Article 104048"},"PeriodicalIF":4.8,"publicationDate":"2025-10-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145321997","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}

引用次数: 0

Secret-key generation from wireless channels using variational autoencoders and domain adversarial neural networks 利用变分自编码器和域对抗神经网络从无线信道生成密钥

IF 4.8 3区计算机科学 Q1 COMPUTER SCIENCE, INFORMATION SYSTEMS

Ad Hoc Networks

Pub Date : 2025-10-06 DOI: 10.1016/j.adhoc.2025.104042

Syed Shafaq Ali Shah, Chunyi Chen, Ruiyue Liang

Physical layer key distribution is an emerging technology in which legitimate communicating parties can extract secret keys with low computational cost and high security by using inherent randomness in the wireless channel. Due to a large number of wireless communication devices, the wireless channel has become weakly correlated with substantial noise, making the channel imperfect, leading to unreliable extraction of secret keys and increased key disagreement rates. However, existing knowledge is working for specific environments where the training and testing conditions are the same, but in the real world the environment is frequently changing, which existing knowledge did not take into account. To address these problems, we present a novel physical layer key distribution scheme based on variational autoencoders and domain adversarial neural network (VAEDANN) that ensure secret key extraction between legitimate devices from weakly correlated wireless channels. The variational autoencoder extracts the latent feature and eliminates the noise in the data, domain adversarial learning makes the model domain invariant. We used a quantization method called channel quantization alternating to address high key disagreement rates. To control information leakage and correct the discrepancy, we used cascade protocol and the toeplitz algorithm. After extensive simulation experiments, the VAEDANN-based method demonstrates a lower MSE compared to other methods across environments, highlighting its superior capability in learning channel reciprocity. Additionally, it achieves a lower bit disagreement rate and key error rate, exhibits high randomness in the key, and maintains a low computational cost. These features make the VAEDANN method highly reliable for resource constrained devices.

物理层密钥分发是一种新兴的技术，合法通信方可以利用无线信道固有的随机性，以低计算成本和高安全性提取密钥。由于无线通信设备的大量存在，无线信道变得弱相关，存在大量的噪声，使得信道不完善，导致密钥提取不可靠，密钥不一致率增加。然而，现有的知识适用于训练和测试条件相同的特定环境，但在现实世界中，环境经常变化，现有的知识没有考虑到这一点。为了解决这些问题，我们提出了一种新的物理层密钥分发方案，该方案基于变分自编码器和域对抗神经网络（VAEDANN），确保从弱相关无线信道中合法设备之间提取密钥。变分自编码器提取潜在特征并消除数据中的噪声，领域对抗学习使模型域不变。我们使用了一种称为信道量化交替的量化方法来解决高密钥不一致率。为了控制信息泄漏和纠正误差，我们使用了级联协议和toeplitz算法。经过大量的仿真实验，与其他方法相比，基于vaedann的方法在不同环境下具有较低的MSE，突出了其在学习信道互易性方面的优越能力。此外，该算法具有较低的比特不一致率和密钥错误率，具有较高的密钥随机性，且计算成本较低。这些特点使得VAEDANN方法在资源受限的设备上具有很高的可靠性。

{"title":"Secret-key generation from wireless channels using variational autoencoders and domain adversarial neural networks","authors":"Syed Shafaq Ali Shah, Chunyi Chen, Ruiyue Liang","doi":"10.1016/j.adhoc.2025.104042","DOIUrl":"10.1016/j.adhoc.2025.104042","url":null,"abstract":"<div><div>Physical layer key distribution is an emerging technology in which legitimate communicating parties can extract secret keys with low computational cost and high security by using inherent randomness in the wireless channel. Due to a large number of wireless communication devices, the wireless channel has become weakly correlated with substantial noise, making the channel imperfect, leading to unreliable extraction of secret keys and increased key disagreement rates. However, existing knowledge is working for specific environments where the training and testing conditions are the same, but in the real world the environment is frequently changing, which existing knowledge did not take into account. To address these problems, we present a novel physical layer key distribution scheme based on variational autoencoders and domain adversarial neural network (VAEDANN) that ensure secret key extraction between legitimate devices from weakly correlated wireless channels. The variational autoencoder extracts the latent feature and eliminates the noise in the data, domain adversarial learning makes the model domain invariant. We used a quantization method called channel quantization alternating to address high key disagreement rates. To control information leakage and correct the discrepancy, we used cascade protocol and the toeplitz algorithm. After extensive simulation experiments, the VAEDANN-based method demonstrates a lower MSE compared to other methods across environments, highlighting its superior capability in learning channel reciprocity. Additionally, it achieves a lower bit disagreement rate and key error rate, exhibits high randomness in the key, and maintains a low computational cost. These features make the VAEDANN method highly reliable for resource constrained devices.</div></div>","PeriodicalId":55555,"journal":{"name":"Ad Hoc Networks","volume":"180 ","pages":"Article 104042"},"PeriodicalIF":4.8,"publicationDate":"2025-10-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145321998","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}

引用次数: 0

An energy-aware distributed federated soft actor-critic framework for intelligent task offloading in vehicular mobile edge computing networks 面向车辆移动边缘计算网络智能任务卸载的能量感知分布式联合软行为者评价框架

IF 4.8 3区计算机科学 Q1 COMPUTER SCIENCE, INFORMATION SYSTEMS

Ad Hoc Networks

Pub Date : 2025-10-05 DOI: 10.1016/j.adhoc.2025.104043

Komeil Moghaddasi , Raja Jurdak

Today, modern intelligent transportation systems increasingly depend on Mobile Edge Computing (MEC) to process real-time data for perception and cooperative driving. However, meeting strict low-latency requirements is challenging because Electric Vehicles (EVs) have limited battery capacity, and must carefully balance the energy used for computation and communication with preserving driving range. While each operation may consume only a small amount of power, their combined effect can substantially reduce how far an EV can travel on a single charge. In this study, we introduce a distributed Federated Soft Actor–Critic (Fed-SAC) framework designed specifically for intelligent, energy-aware, low-latency task offloading in dynamic vehicular MEC networks. Our approach enables each agent to learn privacy-preserving offloading policies in situ, jointly optimizing execution location, transmit power, and offloading ratio through a multi-dimensional discrete action space, addressing both scalability and adaptability under strict latency and energy constraints. Simulations of mobility, wireless channels, and deadline-sensitive workloads show Fed-SAC reduces mean task delay by 77 %, energy use by 89 %, and combined cost by 93 % compared to state-of-the-art methods.

如今，现代智能交通系统越来越依赖于移动边缘计算（MEC）来处理实时数据，以实现感知和协同驾驶。然而，满足严格的低延迟要求是具有挑战性的，因为电动汽车（ev）的电池容量有限，并且必须仔细平衡用于计算和通信的能量与保持行驶里程。虽然每个操作可能只消耗少量的功率，但它们的综合效应可以大大减少电动汽车一次充电的行驶距离。在本研究中，我们引入了一个分布式联邦软参与者-评论家（Fed-SAC）框架，专门为动态车辆MEC网络中的智能、能量感知、低延迟任务卸载而设计。我们的方法使每个agent能够就地学习保护隐私的卸载策略，通过多维离散动作空间共同优化执行位置、传输功率和卸载比例，在严格的延迟和能量约束下解决可扩展性和适应性问题。移动性、无线信道和对截止日期敏感的工作负载的模拟表明，与最先进的方法相比，Fed-SAC将平均任务延迟降低了77%，能源消耗降低了89%，综合成本降低了93%。

{"title":"An energy-aware distributed federated soft actor-critic framework for intelligent task offloading in vehicular mobile edge computing networks","authors":"Komeil Moghaddasi , Raja Jurdak","doi":"10.1016/j.adhoc.2025.104043","DOIUrl":"10.1016/j.adhoc.2025.104043","url":null,"abstract":"<div><div>Today, modern intelligent transportation systems increasingly depend on Mobile Edge Computing (MEC) to process real-time data for perception and cooperative driving. However, meeting strict low-latency requirements is challenging because Electric Vehicles (EVs) have limited battery capacity, and must carefully balance the energy used for computation and communication with preserving driving range. While each operation may consume only a small amount of power, their combined effect can substantially reduce how far an EV can travel on a single charge. In this study, we introduce a distributed Federated Soft Actor–Critic (Fed-SAC) framework designed specifically for intelligent, energy-aware, low-latency task offloading in dynamic vehicular MEC networks. Our approach enables each agent to learn privacy-preserving offloading policies in situ, jointly optimizing execution location, transmit power, and offloading ratio through a multi-dimensional discrete action space, addressing both scalability and adaptability under strict latency and energy constraints. Simulations of mobility, wireless channels, and deadline-sensitive workloads show Fed-SAC reduces mean task delay by 77 %, energy use by 89 %, and combined cost by 93 % compared to state-of-the-art methods.</div></div>","PeriodicalId":55555,"journal":{"name":"Ad Hoc Networks","volume":"180 ","pages":"Article 104043"},"PeriodicalIF":4.8,"publicationDate":"2025-10-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145269961","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}

引用次数: 0

Enhancing user fairness in UAV-assisted RSMA networks : A proximal policy optimization approach 增强无人机辅助RSMA网络中的用户公平性：一种近端策略优化方法

IF 4.8 3区计算机科学 Q1 COMPUTER SCIENCE, INFORMATION SYSTEMS

Ad Hoc Networks

Pub Date : 2025-10-04 DOI: 10.1016/j.adhoc.2025.104041

Donghyeon Hur , Donghyun Lee , Manh Cuong Ho , Wonjong Noh , Sungrae Cho

Rate-splitting multiple access (RSMA) and unmanned aerial vehicles (UAVs) are emerging as key technologies for enhancing connectivity and resource efficiency in future 6G networks. This study investigated a UAV-assisted RSMA downlink communication system and developed a novel framework that jointly optimizes the UAV’s trajectory, precoding matrix, and common rate to maximize the minimum achievable user rate. The proposed system was formulated as a Markov decision process (MDP) and solved using deep reinforcement learning (DRL), specifically using the proximal policy optimization (PPO) algorithm. This learning-based approach enables UAVs to adapt dynamically to the environment without relying on prior channel state information (CSI), allowing for efficient resource allocation and interference management in complex and dynamic wireless scenarios. Furthermore, a precoding design based on a uniform rectangular array (URA) was employed to enhance directional transmission and spatial multiplexing. In simulations, the proposed method significantly outperformed existing benchmarks, achieving an average minimum rate improvement of 23.31% over Deep Deterministic Policy Gradient (DDPG), 48.59% over Soft Actor–Critic (SAC), 50.87% over Trust Region Policy Optimization (TRPO), 63.36% over REINFORCE algorithm, 79.13% over Greedy algorithm, and 145.67% over Random strategies, respectively. These results confirm the potential of UAV-aided RSMA networks in next-generation wireless environments.

分频多址（RSMA）和无人机（uav）正在成为未来6G网络中增强连通性和资源效率的关键技术。研究了一种无人机辅助的RSMA下行通信系统，并开发了一种新的框架，该框架共同优化了无人机的轨迹、预编码矩阵和公共速率，以最大化最小可实现用户速率。该系统被表述为马尔可夫决策过程（MDP），并使用深度强化学习（DRL）求解，特别是使用近端策略优化（PPO）算法。这种基于学习的方法使无人机能够动态地适应环境，而不依赖于先前的信道状态信息（CSI），从而在复杂和动态的无线场景中实现有效的资源分配和干扰管理。此外，采用基于均匀矩形阵列（URA）的预编码设计来增强定向传输和空间复用。在仿真中，该方法显著优于现有基准，比Deep Deterministic Policy Gradient （DDPG）平均提高23.31%，比Soft Actor-Critic （SAC）平均提高48.59%，比Trust Region Policy Optimization （TRPO）平均提高50.87%，比reinforcement算法平均提高63.36%，比Greedy算法平均提高79.13%，比Random策略平均提高145.67%。这些结果证实了无人机辅助RSMA网络在下一代无线环境中的潜力。

{"title":"Enhancing user fairness in UAV-assisted RSMA networks : A proximal policy optimization approach","authors":"Donghyeon Hur , Donghyun Lee , Manh Cuong Ho , Wonjong Noh , Sungrae Cho","doi":"10.1016/j.adhoc.2025.104041","DOIUrl":"10.1016/j.adhoc.2025.104041","url":null,"abstract":"<div><div>Rate-splitting multiple access (RSMA) and unmanned aerial vehicles (UAVs) are emerging as key technologies for enhancing connectivity and resource efficiency in future 6G networks. This study investigated a UAV-assisted RSMA downlink communication system and developed a novel framework that jointly optimizes the UAV’s trajectory, precoding matrix, and common rate to maximize the minimum achievable user rate. The proposed system was formulated as a Markov decision process (MDP) and solved using deep reinforcement learning (DRL), specifically using the proximal policy optimization (PPO) algorithm. This learning-based approach enables UAVs to adapt dynamically to the environment without relying on prior channel state information (CSI), allowing for efficient resource allocation and interference management in complex and dynamic wireless scenarios. Furthermore, a precoding design based on a uniform rectangular array (URA) was employed to enhance directional transmission and spatial multiplexing. In simulations, the proposed method significantly outperformed existing benchmarks, achieving an average minimum rate improvement of 23.31% over Deep Deterministic Policy Gradient (DDPG), 48.59% over Soft Actor–Critic (SAC), 50.87% over Trust Region Policy Optimization (TRPO), 63.36% over REINFORCE algorithm, 79.13% over Greedy algorithm, and 145.67% over Random strategies, respectively. These results confirm the potential of UAV-aided RSMA networks in next-generation wireless environments.</div></div>","PeriodicalId":55555,"journal":{"name":"Ad Hoc Networks","volume":"180 ","pages":"Article 104041"},"PeriodicalIF":4.8,"publicationDate":"2025-10-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145269958","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}

引用次数: 0

A multi-AUV adaptive collaborative target coverage algorithm for unknown environment 未知环境下多auv自适应协同目标覆盖算法

IF 4.8 3区计算机科学 Q1 COMPUTER SCIENCE, INFORMATION SYSTEMS

Ad Hoc Networks

Pub Date : 2025-09-29 DOI: 10.1016/j.adhoc.2025.104033

Jie Li , Qing Chen

With the rapid development and popularisation of autonomous underwater vehicle (AUV) technology, AUVs are increasingly being used in target coverage. Due to the limited communication range and power of AUV, collaboration among AUV swarm is particularly critical, especially in unknown environment. To solve this problem, we propose a multi-AUV adaptive cooperative target coverage algorithm for unknown environment. First, we design a greedy-based multi-AUV adaptive cooperative detection mechanism, utilising a potential field-based AUV force model to achieve efficient detection. Second, a multi-AUV adaptive connectivity mechanism based on potential field is introduced to maintain the connectivity of the AUV network. Simulation results demonstrate the validity of the proposed algorithm. Compared to other methods, our algorithm achieves target coverage under unknown and known target scenarios, ensures energy balance within the AUV network throughout the coverage process, and extends the lifetime of the network.

随着自主水下航行器（AUV）技术的快速发展和普及，AUV越来越多地用于目标覆盖。由于水下机器人的通信范围和能力有限，水下机器人群之间的协作尤为关键，尤其是在未知环境下。针对这一问题，提出了一种未知环境下多auv自适应协同目标覆盖算法。首先，设计了一种基于贪婪的多水下机器人自适应协同检测机制，利用基于势场的水下机器人力模型实现高效检测；其次，引入基于势场的多AUV自适应连接机制，维持AUV网络的连通性；仿真结果验证了该算法的有效性。与其他方法相比，我们的算法实现了未知和已知目标场景下的目标覆盖，保证了AUV网络在整个覆盖过程中的能量平衡，延长了网络的生命周期。

引用次数: 0

M2PFSL: Multi-task and multi-modal IoMT healthcare system via personalized split federated learning M2PFSL：通过个性化分离联邦学习的多任务多模式IoMT医疗保健系统

IF 4.8 3区计算机科学 Q1 COMPUTER SCIENCE, INFORMATION SYSTEMS

Ad Hoc Networks

Pub Date : 2025-09-27 DOI: 10.1016/j.adhoc.2025.104032

Bohan Hu , Yuhang Li , Yinghao Zhang

This paper presents M

^{2}

PFSL, a novel Internet of Medical Things (IoMT) system that revolutionizes health monitoring through integrated multi-task multi-modal personalized processing. Unlike conventional approaches that address computational distribution, communication efficiency, and personalization as separate challenges, our unified framework leverages their synergistic relationships to achieve superior performance while dramatically reducing communication overhead. We develop a distributed processing paradigm where edge devices perform modality-level compression followed by cross-attention fusion to generate compact multi-modal features, while cloud servers execute intensive personalized multi-task analytics. Our systematic two-phase training methodology first establishes robust global models through iterative federated split learning across multiple communication rounds, then adapts these models to individual patient characteristics through GAN-based personalization in compressed feature spaces. This approach enables personalized model adaptation without compromising global knowledge or increasing communication costs, representing a fundamental advance in distributed healthcare system development. The strategic architecture achieves substantial communication overhead reduction while maintaining diagnostic quality, making continuous health monitoring feasible in bandwidth-constrained environments. Extensive experimental validation demonstrates consistent and substantial performance improvements across multiple datasets and compression ratios, with the discovered task-specific compression sensitivity hierarchy providing critical insights for adaptive IoMT system design.

M2PFSL是一种新型的医疗物联网（IoMT）系统，通过集成的多任务多模式个性化处理，彻底改变了健康监测。与将计算分布、通信效率和个性化作为单独挑战来解决的传统方法不同，我们的统一框架利用它们的协同关系来实现卓越的性能，同时显著降低通信开销。我们开发了一种分布式处理范例，其中边缘设备执行模态级压缩，然后进行交叉注意融合以生成紧凑的多模态特征，而云服务器执行密集的个性化多任务分析。我们系统的两阶段训练方法首先通过跨多个通信回合的迭代联合分裂学习建立鲁棒的全局模型，然后通过压缩特征空间中基于gan的个性化使这些模型适应个体患者特征。这种方法能够在不影响全局知识或增加通信成本的情况下实现个性化的模型适应，代表了分布式医疗保健系统开发的根本进步。该战略体系结构在保持诊断质量的同时大幅降低了通信开销，从而在带宽受限的环境中实现持续运行状况监控。广泛的实验验证表明，在多个数据集和压缩比中，性能得到了一致和实质性的改善，所发现的特定于任务的压缩灵敏度层次结构为自适应IoMT系统设计提供了关键见解。

{"title":"M2PFSL: Multi-task and multi-modal IoMT healthcare system via personalized split federated learning","authors":"Bohan Hu , Yuhang Li , Yinghao Zhang","doi":"10.1016/j.adhoc.2025.104032","DOIUrl":"10.1016/j.adhoc.2025.104032","url":null,"abstract":"<div><div>This paper presents M<span><math><msup><mrow></mrow><mrow><mn>2</mn></mrow></msup></math></span>PFSL, a novel Internet of Medical Things (IoMT) system that revolutionizes health monitoring through integrated multi-task multi-modal personalized processing. Unlike conventional approaches that address computational distribution, communication efficiency, and personalization as separate challenges, our unified framework leverages their synergistic relationships to achieve superior performance while dramatically reducing communication overhead. We develop a distributed processing paradigm where edge devices perform modality-level compression followed by cross-attention fusion to generate compact multi-modal features, while cloud servers execute intensive personalized multi-task analytics. Our systematic two-phase training methodology first establishes robust global models through iterative federated split learning across multiple communication rounds, then adapts these models to individual patient characteristics through GAN-based personalization in compressed feature spaces. This approach enables personalized model adaptation without compromising global knowledge or increasing communication costs, representing a fundamental advance in distributed healthcare system development. The strategic architecture achieves substantial communication overhead reduction while maintaining diagnostic quality, making continuous health monitoring feasible in bandwidth-constrained environments. Extensive experimental validation demonstrates consistent and substantial performance improvements across multiple datasets and compression ratios, with the discovered task-specific compression sensitivity hierarchy providing critical insights for adaptive IoMT system design.</div></div>","PeriodicalId":55555,"journal":{"name":"Ad Hoc Networks","volume":"180 ","pages":"Article 104032"},"PeriodicalIF":4.8,"publicationDate":"2025-09-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145269957","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}

引用次数: 0

Generative AI-based intrusion detection systems for intra-vehicle networks 基于生成式人工智能的车载网络入侵检测系统

IF 4.8 3区计算机科学 Q1 COMPUTER SCIENCE, INFORMATION SYSTEMS

Ad Hoc Networks

Pub Date : 2025-09-25 DOI: 10.1016/j.adhoc.2025.104031

Guettouche Asaouer, Djallel Eddine Boubiche

With the rise of connected and autonomous vehicles, securing Intra-Vehicle Networks against cyber threats has become a critical challenge. The Controller Area Network bus, a widely used communication protocol in modern vehicles, remains highly vulnerable to sophisticated intrusion attacks. Traditional Machine Learning and Deep Learning based Intrusion Detection Systems have demonstrated limitations in adaptability, real-time performance, and handling zero-day attacks. This survey explores the emerging role of Generative Artificial Intelligence in enhancing IVN security. It examines key GenAI—assessing their potential to address the shortcomings of conventional IDS techniques. A comprehensive review of recent literature is conducted, analyzing the effectiveness of generative approaches in intrusion detection compared to deterministic methods. Key aspects such as detection time, adaptability to unknown threats, and real-time processing constraints are evaluated. Additionally, this paper identifies existing research gaps, emphasizing the need for standardized datasets, federated learning strategies, and improved deployment techniques to ensure the practical viability of GenAI-based IDS in real-world vehicular environments. The insights presented aim to guide future research toward more robust and adaptive security solutions for IVNs.

随着联网和自动驾驶汽车的兴起，保护车载网络免受网络威胁已成为一项关键挑战。控制器区域网络总线是现代车辆中广泛使用的通信协议，它仍然极易受到复杂的入侵攻击。传统的基于机器学习和深度学习的入侵检测系统在适应性、实时性和处理零日攻击方面存在局限性。本调查探讨了生成式人工智能在增强IVN安全性方面的新兴作用。它审查了关键的基因分析，评估了它们解决传统IDS技术缺点的潜力。对最近的文献进行了全面的回顾，分析了与确定性方法相比，生成方法在入侵检测中的有效性。评估了检测时间、对未知威胁的适应性和实时处理约束等关键方面。此外，本文还指出了现有的研究差距，强调需要标准化的数据集、联邦学习策略和改进的部署技术，以确保基于genai的IDS在现实车辆环境中的实际可行性。提出的见解旨在指导未来对ivn更强大和自适应安全解决方案的研究。

{"title":"Generative AI-based intrusion detection systems for intra-vehicle networks","authors":"Guettouche Asaouer, Djallel Eddine Boubiche","doi":"10.1016/j.adhoc.2025.104031","DOIUrl":"10.1016/j.adhoc.2025.104031","url":null,"abstract":"<div><div>With the rise of connected and autonomous vehicles, securing Intra-Vehicle Networks against cyber threats has become a critical challenge. The Controller Area Network bus, a widely used communication protocol in modern vehicles, remains highly vulnerable to sophisticated intrusion attacks. Traditional Machine Learning and Deep Learning based Intrusion Detection Systems have demonstrated limitations in adaptability, real-time performance, and handling zero-day attacks. This survey explores the emerging role of Generative Artificial Intelligence in enhancing IVN security. It examines key GenAI—assessing their potential to address the shortcomings of conventional IDS techniques. A comprehensive review of recent literature is conducted, analyzing the effectiveness of generative approaches in intrusion detection compared to deterministic methods. Key aspects such as detection time, adaptability to unknown threats, and real-time processing constraints are evaluated. Additionally, this paper identifies existing research gaps, emphasizing the need for standardized datasets, federated learning strategies, and improved deployment techniques to ensure the practical viability of GenAI-based IDS in real-world vehicular environments. The insights presented aim to guide future research toward more robust and adaptive security solutions for IVNs.</div></div>","PeriodicalId":55555,"journal":{"name":"Ad Hoc Networks","volume":"180 ","pages":"Article 104031"},"PeriodicalIF":4.8,"publicationDate":"2025-09-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145160001","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}

引用次数: 0