The coexistence of vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) communication in 6G-enabled vehicular networks introduces complex challenges in spectrum sharing, semantic prioritization, and real-time coordination. To address these issues, we propose G-FEDMAP, a graph-based federated multi-agent deep reinforcement learning framework that supports semantic- and intent-aware resource allocation in distributed vehicle-to-everything (V2X) environments. G-FEDMAP integrates GraphSAGE-based spatiotemporal embeddings, shared actor–centralized critic multi-agent proximal policy optimization (MAPPO) training under a centralized training and decentralized execution (CTDE) paradigm, and event-adaptive reward shaping guided by user intent profiles. To preserve data locality and promote scalable collaboration, federated policy coordination is introduced across geographically partitioned vehicular domains. The system dynamically rebalances semantic priorities using intent reprioritization weights, enabling responsiveness to mission-critical context. We evaluate G-FEDMAP in a federated urban vehicular network comprising multiple regions with high agent density, dynamic event intents, and shared spectrum constraints. The proposed framework is tested under diverse communication and traffic conditions, and compared against MAPPO, graph neural network-MAPPO (GNN-MAPPO), and federated PPO variants. G-FEDMAP demonstrates improved V2V delivery success, higher semantic retention, greater intent satisfaction, and better coexistence of V2V and V2I links through graph-based scheduling and federated policy adaptation. These results position G-FEDMAP as a reliable and trustworthy AI-driven solution for future 6G-internet of things (6G-IoT) vehicular networks.

In this paper, we propose a novel semantic communication framework (SCF6) tailored for 6G-enabled vehicular networks, targeting ultra-reliable low-latency communication (URLLC) scenarios. By integrating semantic encoding/decoding with traditional channel processing, our framework optimizes data transmission, focusing on the meaning of the data rather than raw information. To quantify the integrity of the transmitted messages, we employ advanced natural language processing techniques, such as BERT (bidirectional encoder representations from transformers), ensuring semantic similarity between the sent and received information. We formulate an optimization problem that maximizes semantic spectrum efficiency evaluation (SSEE) and success rate (SR) for vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) communications under stringent URLLC constraints. The optimization solutions are solved by the multi-agent hierarchical attention-based semantic deep reinforcement learning (MAHAS-DRL) framework, which coordinates resource allocation and spectrum sharing among multiple vehicles. By incorporating hierarchical attention mechanisms at both semantic and channel levels, MAHAS-DRL enhances decision-making, optimizes transmission power, and reduces interference. Extensive simulation results demonstrate that our proposed framework outperforms traditional DRL approaches in terms of spectrum efficiency and reliability while significantly reducing transmission delays, making it ideal for dynamic urban vehicular networks.

In vehicle-to-everything (V2X) networks, real-time telemetry is essential for enabling predictive analytics and fault detection in intelligent transportation systems. However, frequent wireless disruptions due to interference, mobility, and congestion lead to telemetry gaps that degrade downstream decision-making. To address this challenge, we propose a framework that enhances wireless telemetry robustness using large language models (LLMs) guided by digital twin-based context. Our system combines retrieval-augmented generation with environmental priors to recover high-dimensional, time-correlated telemetry streams lost during communication outages. We also integrate federated continual learning to maintain fault classification performance across non-i.i.d. V2X conditions without centralized data exchange. Extensive evaluations on real-world driving datasets with simulated wireless impairments show that our method significantly improves reconstruction fidelity, reduces degradation from multi-step gaps, and sustains long-term classifier stability. This work demonstrates how AI-driven semantic recovery mechanisms can improve the functional reliability of wireless V2X telemetry under dynamic and lossy network conditions.

This study presents a dynamic UAV swarm framework to support ground networks in disaster zones. The framework leverages Generative AI (GenAI) for real-time hover point generation to guide waypoint-based UAV navigation and realistic task modeling, integrated with graph neural networks (GNN) for safe navigation and obstacle avoidance. A multi-agent graph reinforcement learning (MAGRL) mechanism optimizes UAV coordination, enhancing energy efficiency, task completion, and load balancing in response to environmental changes. The framework's graph attention mechanism further improves inter-UAV communication, enabling adaptive task allocation and efficient coverage of high-risk zones. Extensive simulations show that the integrated GenAI-GNN and MAGRL approach achieves superior performance in task completion, energy savings, and system utility, outperforming benchmarks including MADDPG, GCRL, PSO, and Greedy strategies in dynamic disaster scenarios.

In this study, we propose SCF6, a novel semantic communication framework for 6 G -enabled vehicular networks tailored to ultra-reliable low-latency communication (URLLC) scenarios. SCF6 integrates semantic encoding/decoding with conventional channel processing, optimizing transmission by focusing on data meaning. Leveraging BERT (bidirectional encoder representations from transformers)-based natural language processing, it ensures high semantic similarity between transmitted and received messages. To maximize semantic spectrum efficiency (SSEE) and success rate (SR) for vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) communications under strict URLLC constraints, we design a multi-agent hierarchical attention-based semantic deep reinforcement learning (MAHAS-DRL) framework. MAHASDRL coordinates resource allocation and spectrum sharing, embedding hierarchical attention at both semantic and channel levels to enhance decision-making, optimize power control, and reduce interference. Simulations demonstrate SCF6's superiority over traditional DRL methods in spectrum efficiency, reliability, and latency, proving effective for dynamic urban vehicular networks.

Effective maritime search and rescue (SAR) requires fast, coordinated action from Internet of Maritime Things (IoMT) nodes operating under extreme communication, energy, and environmental constraints. Existing solutions treat semantic sensing, digital twin modeling, and decentralized control in isolation, limiting their responsiveness and scalability. We propose SEMADT-RL, a unified framework that integrates semantic-driven communication, predictive digital twin forecasting, and decentralized multi-agent deep reinforcement learning with graph attention networks (MADRL-GNN). The semantic layer enables lightweight, anomaly-triggered updates, significantly reducing bandwidth while preserving critical detection cues. The digital twin assimilates these updates using an extended Kalman filter to forecast survivor drift and node dynamics. These forecasts guide decentralized agents that collaboratively optimize mobility, processing, and transmission policies under dynamic and constrained maritime conditions. Simulation results demonstrate that SEMADT-RL achieves faster survivor detection, lower communication overhead, and higher energy efficiency than state-of-the-art baselines, providing a scalable solution for next-generation IoMT-assisted SAR operations.

In this study, we introduce a digital twin (DT)-assisted IoV framework that combines a semi-synchronous adaptive federated learning (AdFL) method with multi-agent deep reinforcement learning, enhanced by generative artificial intelligence (GenAI) techniques, specifically conditional variational autoencoders (CVAE). This framework optimizes partial task offloading across distributed mobile edge computing (MEC) servers, ensuring scalable and efficient decision-making in diverse vehicular networks. By continuously reflecting the real-time conditions of vehicles and roadside units (RSUs), the DT framework ensures precise resource distribution and adaptive task handling. To handle the complexity of dynamic environments, we develop a global model that includes transformer layers in the federated learning (FL) process, which captures long-range dependencies. A semi-synchronous aggregation mechanism is introduced to maintain a balance between timely updates and model quality. The adaptive federated multi-agent reinforcement learning (AF-MARL) algorithm enables decentralized, collaborative learning among vehicles and RSUs, optimizing overall cost and energy use, reducing delays, and improving task completion rates. Comprehensive simulations show the framework's effectiveness compared to existing methods, emphasizing its potential to revolutionize real-time decision-making in IoV networks.

Uncrewed aerial vehicles (UAVs) are essential for providing communication and computation services in disaster recovery scenarios where traditional infrastructure is compromised. However, challenges related to energy efficiency, real-time adaptability, coverage, load balancing, and safe navigation persist, particularly in dynamic disaster environments. In this study, we propose a comprehensive framework that integrates generative AI (GenAI) with graph neural networks (GNNs) to dynamically generate hover points for waypoint-based UAV navigation and realistic task generation based on environmental conditions. The GNN-based collision avoidance mechanism further ensures safe navigation by allowing UAVs to avoid obstacles and no-fly zones while coordinating with neighboring UAVs in real time. To optimize UAV swarm operations, we introduce a multiagent graph reinforcement learning (MAGRL) framework, enabling UAVs to maximize overall system utility by refining hover point selection, task allocation, and load balancing in response to environmental changes. A graph attention mechanism enhances UAV coordination, improving communication efficiency and decision-making. Extensive simulations show that the proposed GenAI-GNN and MAGRL framework significantly outperforms existing methods in task completion, energy efficiency, and overall system utility in disaster recovery scenarios.

Achieving real-time decision-making and efficient resource management in dynamic, large-scale Internet-of-Vehicles (IoV) networks is a significant challenge due to their inherent complexity and scale. To address this, we propose a digital twin (DT)-assisted IoV framework that integrates a novel semi-synchronous adaptive federated learning (AdFL) approach with multiagent deep reinforcement learning, enhanced by generative artificial intelligence (GenAI) techniques, specifically conditional variational autoencoders (CVAEs). The framework optimizes partial task offloading across distributed mobile-edge computing (MEC) servers, ensuring scalable, efficient, and accurate decision-making in heterogeneous vehicular networks. By continuously mirroring the real-time states of vehicles and roadside units (RSUs), the DT framework enables precise resource allocation and adaptive task management. To tackle the complexities of dynamic environments, we design a global model with transformer layers embedded in the federated learning (FL) process, capturing long-range dependencies. A novel semi-synchronous aggregation mechanism is introduced to balance timely updates with model quality. The proposed adaptive federated multiagent reinforcement learning (AF-MARL) algorithm facilitates decentralized, collaborative learning among vehicles and RSUs, optimizing overall cost, and energy efficiency, reducing delay, and improving task completion rates. Extensive simulations demonstrate the effectiveness of the proposed framework against other existing approaches, highlighting its potential to transform real-time decision-making in IoV networks.

In the rapidly evolving domain of vehicular metaverse, this study introduces a cutting-edge quantum-based decentralized and heterogeneity-aware federated learning framework for vehicular metaverse named QV-FEDCOM, which stands as a testament to the innovative fusion of quantum computing principles with federated learning (FL). This framework is ingeniously tailored to address the challenges in a vehicular metaverse, offering a cost-efficient and adaptive solution for the dynamic vehicular landscape. QV-FEDCOM is strengthened by key components like quantum sequential-training-program, with reinforcement learning-based dynamic mode switching to reduce communication costs and manage vehicle states adaptively, and the quantum vehicle-context-grouping utilizing hierarchical clustering and simulated annealing for effective vehicle grouping based on contextual data similarity, addressing the complexities of data heterogeneity. Additionally, the integration of quantum-inspired principal component analysis (Q-PCA) enhances memory efficiency, further optimizing the framework. These elements converge in the QV-FEDCOM algorithm, establishing a decentralized, efficient, and context-aware quantum federated learning (QFL) process that redefines learning dynamics in the vehicular metaverse. Our study also introduces an innovative quantum trajectory loss (QTL) function, specifically designed for trajectory prediction tasks, which combines the Huber loss with an angular deviation penalty to robustly handle errors and penalize large deviations in the predicted trajectory angle. The effectiveness of the QV-FEDCOM framework is rigorously validated through comprehensive simulations, with its performance meticulously compared against various adaptations, showcasing its transformative capabilities within the vehicular metaverse ecosystem.

Vehicular crowdsensing (VCS) encounters challenges within social Internet of Vehicles networks, including interdependent behaviors and the necessity for long-term sensing strategies that balance energy efficiency and delay tolerance in dynamic settings. To tackle these obstacles, this study explores a VCS model tailored for social IoV networks, considering dynamic environmental parameters. We further develop a utility model that seamlessly integrates data-quality aware functional and social incentives for each vehicle, ensuring optimal task payoff, efficient energy usage, and minimized processing time within the dynamic social IoV environment. Additionally, we introduce a non-cooperative game between vehicles and propose a multi-agent deep reinforcement learning (DRL)-based solution for the dynamic VCS framework. This enables vehicles to autonomously adjust sensing levels, maximizing both individual and collective utility. Finally, through comparative simulations, we demonstrate the effectiveness of our approach in comparison to baseline methods.

In this study, we present a hybrid deep reinforcement learning (DRL) algorithm, trained using vehicular federated learning (VFL), specifically tailored for dynamic vehicular networks with historical data. Our approach utilizes VFL-based DRL to refine the caching scheme in these networks, focusing on predicting and storing the most effective content nearby to enhance cache efficiency and reduce content request delays. We propose a modified proximal policy optimization (mPPO) based approach for the DRL-based decision-making for caching management, which combines the advantages of proximal policy optimization (PPO) and double deep Q-network (DDQN). Our study encompasses a vehicular framework that includes a central edge node (CEN), roadside units (RSUs), unmanned aerial vehicles (UAVs), and vehicles equipped with historical data. We tackle the challenges posed by varying vehicle density and mobility, non-uniform RSU coverage, and constrained caching capacity. Through comprehensive simulations, we demonstrate that mPPO outperforms conventional DRL methods like PPO and DDQN, as well as heuristic approaches. These results underscore the efficacy of the VFL-based mPPO in dynamic vehicular networks, confirming its potential for real-world applications.

In the rapidly evolving domain of vehicular metaverse, this study introduces a cutting-edge quantum-based decentralized and heterogeneity-aware federated learning framework for vehicular metaverse named QV-MetaFL, which stands as a testament to the innovative fusion of quantum computing principles with federated learning (FL). This framework is ingeniously tailored to address the challenges in a vehicular metaverse, offering a cost-efficient and adaptive solution for the dynamic vehicular landscape. QV-MetaFL is strengthened by the quantum sequential-training-program (Q-STP) algorithm, a quantum-based sequential training program that transforms model training, reducing communication costs and adeptly managing vehicle states. Complementing this, the quantum vehicle-context-grouping (Q-VCG) mechanism groups vehicles based on contextual data similarity, effectively tackling the complexities of data heterogeneity. The synergy of Q-STP and Q-VCG culminates in the QV-MetaFL algorithm, a decentralized, efficient, and context-aware quantum federated learning (QFL) process that redefines learning dynamics in the vehicular metaverse. Additionally, our research introduces an innovative composite loss function that amalgamates classical loss metrics with quantum parameter regularization, deftly addressing quantum sensitivity to noise. The effectiveness of the QV-MetaFL framework is rigorously validated through comprehensive simulations, with its performance meticulously compared against various adaptations, showcasing its transformative capabilities within the vehicular metaverse ecosystem.

Vehicular crowdsensing (VCS) within the social Internet of Vehicles (IoV) significantly advances urban transportation management by enhancing road safety, traffic efficiency, and the overall driving experience. This article presents an intelligent multiagent deep reinforcement learning (DRL) framework for augmented dynamic task prioritization in a multi-incentive VCS system. Our framework, named intelligent multiagent reinforcement learning (IMARL), leverages augmented intelligence to integrate human-like decision-making processes with autonomous vehicle operations, ensuring more adaptive and robust task management. The proposed IMARL framework offers several key advantages: it dynamically adjusts the sensing levels of each vehicle, ensuring efficient energy usage and minimized processing times, and employs a data-quality aware multi-incentive utility model to capture both functional and social incentives. Additionally, our framework incorporates a layered server architecture, enhancing system resilience and scalability. Simulation results demonstrate the superiority of our approach. IMARL achieves significant improvements in task completion rates, energy consumption, and processing delays compared to other DRL and non-DRL benchmark methods. Furthermore, our approach exhibits strong adaptability to changing environmental conditions, maintaining high performance even in high-density traffic scenarios. These quantified results validate the effectiveness of the proposed framework, highlighting its potential to significantly enhance VCS systems in real-world applications.

In this study, we present a hybrid deep reinforcement learning (DRL) algorithm, trained using vehicular federated learning (VFL), specifically tailored for dynamic vehicular networks with historical data. Our approach utilizes VFL-based DRL to refine the caching scheme in these networks, focusing on predicting and storing the most effective content nearby to enhance cache efficiency and reduce content request delays. We propose a modified proximal policy optimization (mPPO)-based approach for the DRL-based decision making for caching management, which combines the advantages of proximal policy optimization (PPO) and double deep Q-network (DDQN). Our study encompasses a vehicular framework that includes a central edge node (CEN), roadside units (RSUs), unmanned aerial vehicles (UAVs), and vehicles equipped with historical data. We tackle the challenges posed by varying vehicle density and mobility, nonuniform RSU coverage, and constrained caching capacity. Through comprehensive simulations, we demonstrate that the mPPO outperforms the conventional DRL methods like PPO and DDQN, as well as heuristic approaches. These results underscore the efficacy of the VFL-based mPPO in dynamic vehicular networks, confirming its potential as a viable solution for real-world applications.

In this study, we investigate the performance of a hybrid reconfigurable intelligent surface (RIS)-assisted non-orthogonal multiple access (NOMA) network, augmented with backscattering capabilities, designed to facilitate integrated sensing and communication (ISAC). Our primary objective is two-fold: first, to enhance the overall communication throughput, and second, to strengthen the sensing power for target detection. To achieve these goals, we introduce two novel dynamic user clustering algorithms namely composite distance and angle-based (CDA) and channel-oriented adaptive (COA) algorithm for grouping users into clusters with fixed base station and RIS positions, where the successive interference cancellation (SIC) is employed for effective communication within each pair. Moreover, we present a comprehensive optimization problem that jointly maximizes the sum rate and sensing power. This problem involves optimizing the transmit beamformer at the base station, the power allocation factors within each cluster, and the phase shifts at the RIS. This methodology not only adheres to strict power constraints and quality of service requirements at each receiving node but also ensures equitable resource allocation among the targets and enforces unit modulus phase shifts at each RIS element. To tackle the complex interdependencies and non-convex nature of the optimization problem, we introduce an advanced iterative algorithm based on alternative optimization (AO). This state-of-the-art technique employs successive convex approximation (SCA) to systematically address this multifaceted problem. Finally, the simulation results empirically validate the proposed algorithm’s effectiveness, considering the number of RIS elements, maximum power budget, number of targets, and imperfect channel state information (CSI) while showing the trade-off between communication and sensing performance.

In this paper, we propose a novel approach for optimal resource management in ultra-reliable low-latency communication (URLLC)-enabled Internet of Vehicles (IoV) networks. The framework includes mobile edge computing (MEC) servers integrated into roadside units (RSUs), unmanned aerial vehicles (UAVs), and base stations (BSs) for hybrid vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) communication. We utilize asynchronous federated learning (AFL) approach to enhance the accuracy of the global model by considering the mobility characteristics of vehicles. The problem of optimal resource allocation is formulated to achieve the best allocation of frequency, computation, and caching resources while complying with the delay restrictions. To solve the non-convex problem, a multi-agent actor-critic type deep reinforcement learning algorithm called D-MAAC algorithm is introduced. Extensive simulations show the effectiveness of the proposed framework and algorithms compared to existing schemes.

Digital twin (DT) has emerged as a promising technology for improving resource allocation decisions in Internet of Vehicles (IoV) networks. In this paper, we consider an IoV network where mobile edge computing (MEC) servers are deployed at the roadside units (RSUs). The IoV network provides ubiquitous connections even in areas uncovered by RSUs with the assistance of unmanned aerial vehicles (UAVs) which can act as a relay between RSUs and task vehicles. A virtual representation of the IoV network is established in the aerial network as DT which captures the dynamics of the entities of the physical network in real-time in order to perform efficient resource allocation for delay-intolerant tasks. We investigate an intelligent delay-sensitive task offloading scheme for the dynamic vehicular environment which provides computation resources via local execution, vehicle-to-vehicle (V2V), and vehicle-to-roadside-unit (V2I) offloading modes based on the energy consumption of the system. Moreover, we also propose a multi-network deep reinforcement learning (DRL)-based resource allocation algorithm (RADiT) in the DT-assisted network for maximizing the utility of the IoV network while optimizing the task offloading strategy. Further, we compare the performance of the proposed algorithm with and without the presence of V2V computation mode. RADiT is further evaluated by comparing it with another benchmark DRL algorithm called soft actor-critic (SAC) and a non-DRL approach called greedy. Finally, simulations are performed to demonstrate that the utility of the proposed RADiT algorithm is higher under every condition compared to its respective conditions in SAC and greedy approach. Consequently, the proposed framework jointly improves energy efficiency and reduces the overall delay of the network. The proposed algorithm with UAV relay further increases the efficiency of the network by increasing the task completion rate.

Reconfigurable Intelligent Surface (RIS) is a transformative technology which can enhance the performance of the ubiquitous wireless networks and achieve better signal quality with the aide of multiple reflecting surfaces. In this work, an analytical framework of a RIS-aided full duplex (FD) communication network consisting of a FD-access point (AP) that communicates with an uplink and a down-link users simultaneously is provided. In particular, we analyze the performance of the considered system by deriving analytical expressions of outage probability for both uplink and downlink transmissions. Further, the accuracy of the derived expressions is validated using simulation results. Finally, from the comparative analysis, it is shown that the RIS outperforms the system without RIS providing remarkable improvement in the outage probability.

Due to the dynamic nature of a vehicular fog computing environment, efficient real-time resource allocation in an Internet of Vehicles (IoV) network without affecting the quality of service of any of the onboard vehicles can be challenging. This article proposes a priority-sensitive task offloading and resource allocation scheme in an IoV network, where vehicles periodically exchange beacon messages to inquire about available services and other important information necessary for making the offloading decisions. In the proposed methodology, the vehicles are stimulated to share their idle computation resources with the task vehicles, whereby a deep reinforcement learning algorithm based on soft actor–critic is designed to classify the tasks based on priority and computation size of each task for optimally allocating the power. Furthermore, we also design deep deterministic policy gradient (DDPG) and twin delayed DDPG (TD3) algorithms for the considered framework. In particular, the algorithms work toward achieving the optimal policy for task offloading by maximizing the mean utility of the considered network. Extensive numerical results under different network conditions, along with comparison among the three algorithms, are presented to validate the feasibility of distributed reinforcement learning for task offloading in future IoV networks.