2026· IEEE Transactions on Cognitive Communications and Networking· Vol 12, pp. 9176-9187· 0 citations· 35 references
Computer Science
Abstract
Federated learning (FL) has emerged as a promising paradigm for enabling distributed model training in vehicular networks while keeping raw data local. However, the dynamic mobility of vehicles and the limited spectrum resources create critical challenges for efficient FL execution. In particular, due to the high mobility of vehicles, the candidate set of participating vehicles keeps changing, which makes fixed-threshold selection strategies difficult to be applied effectively. Moreover, vehicle mobility also causes time-varying channel status, and if resource allocation is performed only once at the beginning of each training round, it might not match the varying channels, resulting in imprecise resource allocation and accordingly low resource utilization. To address these issues, we propose a dynamic vehicular FL framework where the long time duration is discretized into fine-grained time slots. A long time sequence two-timescale optimization problem is then formulated to jointly conduct vehicle selection and slot-level communication bandwidth allocation. To solve it, we design a hierarchical Markov Decision Process (H-MDP) framework, and then develop a hierarchical Proximal Policy Optimization-based vehicle selection and bandwidth allocation (HPPO-VSBAFL) strategy, consisting of two cooperative agents: a vehicle selection agent (VSA) for round-level participant selection, and a bandwidth allocation agent (BAA) for slot-level spectrum allocation. Extensive experimental results based on CIFAR-10 with ResNet-18 demonstrate that the proposed HPPO-VSBAFL framework significantly improves FL accuracy compared to baseline schemes, and can effectively adapt to the highly dynamic vehicular environments.
The results demonstrate that the proposed PP-SAPF is suitable for real-time deployment in intelligent transportation systems (ITS) and autonomous vehicles where low latency, reliable connectivity, and adaptive resource management is significant.
Irshad Khan, Neetha Papanna Umalakshmi, Somshekhar Durgaiah et al.· Bulletin of Electrical Engin...· 0 citations
The rapid advancements of next-generation vehicular networks require intelligent, low-latency, and efficient resource management to support heterogeneous services. In this work, we propose a Traffic-aware Dynamic Resource Allocation (TADRA) architecture for UAV-assisted vehicular O-RAN to address the challenges of dynamic traffic conditions, infrastructure failures, and stringent quality of service (QoS) requirements. Due to the dynamic mobility and flexible deployment characteristics, UAV Open Radio Units (O-RUs) in the TADRA architecture support the terrestrial infrastructure under overload or failure conditions, dynamically extending coverage, balancing traffic loads, and restoring service to maintain uninterrupted QoS across diverse and heterogeneous traffic demands. Unlike existing static or single-layer solutions, our proposed TADRA integrates RAN Intelligent Controllers (RICs) with a Hierarchical Traffic-Aware Multi-Agent Twin-Delayed (TMT) algorithm to optimize the allocation of computation and radio resources. This joint optimization problem is NP-hard, highly dynamic, and coupled across agents, making TMT a tractable and adaptive alternative. This hierarchical framework performs traffic prioritization at the upper (application) layer and resource allocation at the lower (MAC) layer, facilitating adaptive decision-making under diverse vehicular traffic patterns. Numerical results demonstrate that our solution provides substantial gains over MATD3, MADDPG, and GA, achieving 17% lower latency, 10% higher throughput, 14% lower energy consumption, and 6.5% higher reliability.
Hayla Nahom Abishu, Ahmed Badawy, Amr Mohamed et al.· IEEE Transactions on Network...· 0 citations
: In dense multi-drone FANET environments, routing efficiency is often reduced due to constant competition for a wireless channel, frequent changes in network topology, and uneven energy consumption between UAV nodes. These factors lead to a decrease in the packet delivery ratio, an increase in delays on the busiest or most unstable routes, and a reduction in the total network lifespan. Although many existing routing methods utilize mobility forecasting or reinforcement learning, most are based on independent decision-making by individual agents. As a result, such approaches may create a significant service load and insufficiently account for competition for the communication channel, as well as the fair distribution of energy consumption between network nodes. To overcome these integrated challenges, this paper proposes FL-CARE, a Federated Learning-based Contention-Aware and Energy-balanced Routing protocol. In FL-CARE, each UAV acts as a learning agent using a multidimensional state (predicted link stability, channel contention, queue occupancy, residual energy) and a tail-latency-sensitive reward function. A lightweight federated learning mechanism enables collaborative, swarm-level intelligence with bounded communication overhead. Extensive MATLAB simulations demonstrate that FL-CARE outperforms state-of-the-art protocols MP-QGRD and RL-MPEAOLSR, improving PDR by 12-18%, reducing average delay by 20 – 25% and tail latency (p99) by up to 30%. Additionally, FL-CARE yields a 15-22% energy consumption per bit delivered, and control overhead reduction of 35-45%, and advances network lifetime by approx. 25% in dense network deployments. The proposed framework maintains low computational and communication overhead through lightweight local learning and compact federated model updates, while adaptive aggregation intervals facilitate efficient collaborative learning and scalable operation under varying network conditions. These simulation results validate the holistic integration with federated multi-agent learning, with emphasis on contention awareness and energy balancing is essential for scalable and efficient routing in next-generation dense FANETs.
H. Khujamatov, Elyanora Jolimbetova, Khaleel Ahmad et al.· Journal of Computer Science· 0 citations
With the evolution of smart networked vehicles towards smart driving, smart cabin, and ecology, the modern Internet of Vehicles (IoV) has become a typical application of new quality productivity in the automotive industry. In IoV resource allocation, intelligent decisions require Reinforcement Learning (RL). However, existing RL research methods are constrained by limited power supply capacity, and vehicles will face significant challenges in controlling energy consumption during decision training, which seriously hinders their ability to acquire sufficient data to support efficient decision optimization. This paper investigates the problems of energy consumption control and personalized spectrum resource allocation in a Federated Learning system comprising multiple IoV cells and edge clouds, with the objectives of reducing the energy consumption of end-side vehicles in the IoV system and improving the effectiveness of spectrum resource allocation decisions. Aiming at the training scale limitation and energy consumption control problems faced by end-side vehicles during spectrum resource allocation policy learning, this paper designs Digital Twin-Based Low-Energy Reinforcement Learning for Multi-Cell IoV (DT-LERL) distributed collaborative training architecture in cellular-based IoV scenarios, which allows the twin to replace the end-side vehicular entities by introducing digital twins to carry out scenario interactions and model training. It combines the decision frequency of digital twins with decision experience to dynamically balance digital twin error against model decision quality. By reducing the dimensionality of the locally trained network, the client can perform multiple local updates in each communication round, facilitating the learning of personalized local decisions. Experimental simulations show that, compared with other mainstream personalized FL algorithms, the proposed joint FL and digital twin scheme achieves better performance for the personalized spectrum resource allocation decision problem.
A. Alamoudi, Abdullah S. Almansouri· Journal of Big Data· 0 citations
Unmanned aerial vehicle (UAV)-assisted mobile edge computing (MEC) is a key enabler for meeting the stringent low-latency and energy-efficiency requirements of emerging low-altitude economy applications. However, achieving these objectives remains challenging due to dynamic environments, limited communication and computation resources, and the heterogeneity of network entities. This paper investigates the long-term joint optimization framework that minimizes system-wide latency and energy consumption simultaneously by coordinating UAV association, subchannel selection, uplink/downlink power allocation, and computational resource distribution. This sequential decision-making process is formulated into a partially observable Markov decision process (POMDP) to account for localized observations and dynamic channel states. To solve it, we propose a heterogeneous multi-agent proximal policy optimization (MAPPO)-based framework where both user devices (UDs) and UAVs act as heterogeneous agents. This architecture utilizes a centralized training and decentralized execution (CTDE) paradigm to enable collaborative strategies between computing requesters and providers. Numerical results demonstrate that the proposed scheme effectively navigates the high-dimensional action space and achieves superior convergence and cost reduction compared to benchmarks, including PPO, independent PPO (iPPO), Q-learning multi-agent extension (QMIX), value decomposition networks (VDN), independent deep Q-network (iDQN), and genetic algorithm (GA).
Ming Cheng, Canlin Zhu, Jiang-Hang Tang et al.· Journal of King Saud Univers...· 0 citations
The rapid growth of Internet of Vehicles (IoV) applications has imposed strict requirements on low-latency and energy-efficient computing services. This letter investigates a multi-Uncrewed Aerial Vehicle (UAV)-assisted IoV system, where multiple Mobile Edge Computing (MEC)-enabled UAVs (MUs) collaboratively provide computing services for vehicular terminals (VTs). To improve service capability, we propose an energy-efficient task offloading and load balancing scheme that jointly considers vehicle mobility, task offloading and migration, and computing resource allocation to formulate an optimization problem. To solve this problem, a collective learning (CL)-enabled multi-agent reinforcement learning (CL-MARL) algorithm is proposed, where each agent learns optimal policies through centralized training and collective cooperative learning. Simulation results demonstrate that the proposed scheme outperforms benchmark strategies in terms of energy efficiency, task completion rate, and load balancing.
Yongbin Wang, Peng Lin, Yan Liu et al.· IEEE Wireless Communications...· 0 citations