关键词： Autonomous aerial vehicles   Delays   Energy consumption   Resource management   Reinforcement learning   Trajectory   Uplink   Training   Rotors   Optimization   Mobile edge computing   AAV   personalized federated reinforcement learning   knowledge distillation  

摘要： In this letter, we propose an innovative autonomous aerial vehicle (AAV)-assisted computation offloading scheme, where multiple AAVs with heterogeneous computational capabilities collaborate to provide task offloading services for mobile users (MUs). Our objective is to minimize the weighted sum of the MUs' maximum task processing delay and the AAVs' average energy consumption by jointly optimizing the AAVs' service locations, velocities, user association and offloading policies. We model this problem as a fully decentralized partially observable Markov decision process (Dec-POMDP) and propose a personalized federated reinforcement learning (FRL) algorithm, named federated knowledge distillation based parameterized deep Q-network (FKD-PDQN), to collaboratively optimize the AAVs' policies in a decentralized and cost-efficient manner. In this approach, each AAV optimizes its local policy using PDQN, and updates its model by aggregating the distilled knowledge from the neighboring AAVs through an attention mechanism. Simulation results reveal that the proposed algorithm effectively reduces the average delay, energy consumption, and communication overhead, offering significant improvements over baseline algorithms such as Independent PDQN (IPDQN), Federated PDQN (FPDQN), and Krum.

关键词： Optimization   Computational modeling   Energy consumption   Delays   Collaboration   Genetics   Servers   Process control   Costs   Reliability   Mobile edge computing   task offloading   multi-objective optimization   task urgency   task criticality   Bayesian maximum entropy  

摘要： Mobile edge computing assisting vehicular networks to achieve collaborative computing is a significant research area, as it can effectively expand the on-board computational resources, thereby improving the responsiveness of vehicular applications and reducing energy consumption. However, the existing task offloading schemes still have various gaps and face challenges that should be addressed because task offloading problems are usually multi-objective optimization problems (MOOP) and NP-hard problems. In this respect, we formulate a new multi-objective model for vehicle-centered task offloading tacking into account the queuing and computing processes of tasks. The minimization of the average delay time, average energy consumption, and average payment cost for each vehicle is defined as a MOOP. We propose an efficient collaborative offloading approach based on task triage strategy and multi-objective optimization to address the model. Specifically, this approach can be decomposed into two stages. First, a task triage strategy is developed based on a comprehensive analysis of task urgency and task criticality to make offloading decisions for triaged tasks with extreme characteristics. Then, we propose a novel multi-objective optimization framework for the MOOP. Finally, we develop a multi-objective hybrid genetic algorithm that integrates the associative learning immediate memory strategy and genetic operations to solve the MOOP. Extensive experiments and performance evaluations based on real-world traces of taxis demonstrate the effectiveness of the proposed approach. The results indicate that the approach outperforms six other well-known multi-objective algorithms in solving the MOOP.

关键词： Mobile edge computing   Authentication   Session key   Scyther   Authenticated Encryption (AE)  

摘要： By bringing services closer to the network edge, mobile edge computing (MEC) improves traditional cloud computing and increases end-user accessibility. This represents a significant improvement in Internet of Things (IoT) applications. Ensuring strong security is becoming more and more important as more mobile users access MEC services across several domains. Keeping IoT applications safe from attacks is a major problem in MEC contexts. Moreover, it becomes considerably challenging to maintain robust security in circumstances where users often move between networks. This demands a trustworthy authentication framework that facilitates mobility for users while protecting MEC from illegal access. The framework also needs to make sure that insider attacks do not jeopardize the MEC environment's or its servers' security. Establishing a secure session key between MEC users and external edge servers is also crucial for maintaining safe communication in MEC environments. Another essential security requirement is maintaining user anonymity, since most users want to keep their personal information private. This work addresses these problems by presenting a mutual authentication framework developed especially for mobility-supported MEC contexts, that utilizes PUF, ECC, and AE. The presented framework incorporates three factors for user authentication and provides resilience against various attacks, and session key establishment in mobile MEC environments. It also presents flexibility for users to update their passwords when required. Security and performance evaluations of the framework demonstrate its practicality and effectiveness, exhibiting a reduction in communication cost by 61% to 71% percent and computational cost by 41.16%, and 75.44%.

关键词： Computation offloading   dynamic scheduling   DAG   DVFS   mobile edge computing  

摘要： Mobile edge computing (MEC), as a new computing paradigm, has become an effective solution to solve the contradiction between computing-intensive applications and resource-starved mobile user equipment (UE). UE enables it to offload tasks to small cell base stations (SBSs). However, as the demand for computation-intensive tasks continues to grow, the energy consumption of SBSs also continues to increase. This paper is to minimize the energy consumption of SBSs. We model an application with dependent subtasks as a directed acyclic graph (DAG), and propose an optimized energy-aware task offloading strategy (EAOC) to effectively allocate computing resources of edge servers to applications. EAOC makes decision for tasks by balancing the finish time and energy consumption. The simulation results show that the proposed EAOC has significant energy-saving compared to existing algorithms.

关键词： Internet of Vehicles   mobile edge computing   blockchain   offloading strategy   improved polar lights optimization algorithm  

摘要： The rapid growth of computationally intensive tasks in the Internet of Vehicles (IoV) poses a triple challenge to the efficiency, security, and stability of Mobile Edge Computing (MEC). Aiming at the problems that traditional optimization algorithms tend to fall into, where local optimum in task offloading and edge computing nodes are exposed to the risk of data tampering, this paper proposes a secure offloading strategy that integrates the Improved Polar Lights Optimization algorithm (IPLO) and blockchain. First, the truncation operation when a particle crosses the boundary is improved to dynamic rebound by introducing a rebound boundary processing mechanism, which enhances the global search capability of the algorithm;second, the blockchain framework based on the Delegated Byzantine Fault Tolerance (dBFT) consensus is designed to ensure data tampering and cross-node trustworthy sharing in the offloading process. Simulation results show that the strategy significantly reduces the average task processing latency (64.4%), the average system energy consumption (71.1%), and the average system overhead (75.2%), and at the same time effectively extends the vehicle's power range, improves the real-time performance of the emergency accident warning and dynamic path planning, and significantly reduces the cost of edge computing usage for small and medium-sized fleets, providing an efficient, secure, and stable collaborative computing solution for IoV.

关键词： Training   Convergence   Computational modeling   Adaptation models   Optimization   Servers   Accuracy   Data models   Analytical models   Federated learning   Distributed learning   split federated learning   client-side model aggregation   model splitting   mobile edge computing  

摘要： The increasing complexity of deep neural networks poses significant barriers to democratizing AI to resource-limited edge devices. To address this challenge, split federated learning (SFL) has emerged as a promising solution that enables device-server co-training through model splitting. However, although system optimization substantially influences the performance of SFL, the problem remains largely uncharted. In this paper, we first provide a unified convergence analysis of SFL, which quantifies the impact of model splitting (MS) and client-side model aggregation (MA) on its learning performance, laying a theoretical foundation for this field. Based on this convergence bound, we introduce AdaptSFL, an adaptive SFL framework to accelerate SFL under resource-constrained edge computing systems. Specifically, AdaptSFL adaptively controls MS and client-side MA to balance communication-computing latency and training convergence. Extensive simulations across various datasets validate that our proposed AdaptSFL framework takes considerably less time to achieve target accuracy than existing benchmarks.

关键词： UAV-assisted vehicular networks   mobile edge computing   multi-agent deep reinforcement learning  

摘要： Intelligent Connected Vehicles (ICVs) impose stringent requirements on real-time computational services. However, limited onboard resources and the high latency of remote cloud servers restrict traditional solutions. Unmanned Aerial Vehicle (UAV)-assisted Mobile Edge Computing (MEC), which deploys computing and storage resources at the network edge, offers a promising solution. In UAV-assisted vehicular networks, jointly optimizing content and service caching, computation offloading, and UAV trajectories to maximize system performance is a critical challenge. This requires balancing system energy consumption and resource allocation fairness while maximizing cache hit rate and minimizing task latency. To this end, we introduce system efficiency as a unified metric, aiming to maximize overall system performance through joint optimization. This metric comprehensively considers cache hit rate, task computation latency, system energy consumption, and resource allocation fairness. The problem involves discrete decisions (caching, offloading) and continuous variables (UAV trajectories), exhibiting high dynamism and non-convexity, making it challenging for traditional optimization methods. Concurrently, existing multi-agent deep reinforcement learning (MADRL) methods often encounter training instability and convergence issues in such dynamic and non-stationary environments. To address these challenges, this paper proposes a MADRL-based joint optimization approach. We precisely model the problem as a Decentralized Partially Observable Markov Decision Process (Dec-POMDP) and adopt the Multi-Agent Proximal Policy Optimization (MAPPO) algorithm, which follows the Centralized Training Decentralized Execution (CTDE) paradigm. Our method aims to maximize system efficiency by achieving a judicious balance among multiple performance metrics, such as cache hit rate, task delay, energy consumption, and fairness. Simulation results demonstrate that, compared to various representative b

关键词： UAV-assisted platooning   Dual-channel switching   Energy optimization   Reliability enhancement   Ground-air networks   Mobile edge computing  

摘要： Connected and Autonomous Vehicle (CAV) networks are rapidly advancing as a critical technology for enabling intelligent transportation systems. One promising approach to support such vehicular networks is the use of Unmanned Aerial Vehicles (UAVs) to enhance communication and computational capabilities through air-ground collaboration. However, existing UAV-assisted vehicular communication methods often face challenges, particularly in urban environments with frequent obstructions that degrade communication quality and reliability. Additionally, ensuring energy efficiency while maintaining high communication reliability remains an unsolved issue. In this paper, we investigate a UAV-assisted ground-air collaborative network for a platoon of vehicles navigating urban areas with frequent obstacles. Our objective is to maximize communication reliability while minimizing system energy consumption, specifically utilizing a dual-channel switching mechanism to adapt between Line-of-Sight (LoS) and Non-Line-of-Sight (NLoS) communication channels. We propose a joint optimization model that integrates UAV trajectory planning with communication and computation resource allocation for platooning vehicles. To solve the non-convex optimization problem, we propose a novel optimization algorithm that integrates the Broyden-Fletcher-Goldfarb-Shanno (BFGS) method with Sequential Quadratic Programming (SQP). This combined approach iteratively determines effective search directions, ensuring convergence towards a locally optimal solution. Furthermore, we theoretically analyze the convergence of the proposed joint optimization model. Extensive simulations are carried out to demonstrate the effectiveness of the proposed optimization model and algorithm, showing significant improvements in communication reliability and energy efficiency compared to conventional single-channel approaches.

关键词： Mobile edge computing  

摘要： Large language models (LLMs), which have shown remarkable capabilities, are revolutionizing artificial intelligence (AI) and shaping our society. However, the status quo cloud-based LLM deployment faces critical challenges such as long response time and privacy violation, while on-device LLM deployment is hindered by the limited capabilities of end devices. To address these issues, this article explores the transformative potential of deploying LLMs at the 6G edge. We first introduce killer applications to exemplify the urgent need for edge LLM deployment. Meanwhile, we identify the inherent limitations of on-device LLM deployment. We therefore argue that end-edge cooperation at the 6G edge is a promising solution for the dilemma. Towards this end, we elaborate on the 6G MEC architecture tailored for LLMs. Furthermore, we delve into edge training and edge inference for LLMs, with a focus on end-edge cooperation. In both aspects, we discuss a spectrum of cutting-edge techniques, including split learning/inference, parameter-efficient fine-tuning, parameter-sharing inference, and small-large language model cooperation. Finally, we investigate open problems in green and privacy-preserving edge LLM deployment. This work provides a comprehensive and forward-looking perspective and pathways for enabling LLM deployment at the network edge. Copyright © 2025, The Authors. All rights reserved.

关键词： Autonomous vehicles   Mobile edge computing   Deep neural networks   Refined pelican optimizer   LiDAR   Vehicle control   MobileNet   Microcontroller  

摘要： Autonomous vehicles and mobile edge computing's confluence have raised an innovative model for immediate decision-making and improved computational abilities. But, enhancing vehicle management systems to guarantee effective enactment remains an important challenge. Present approaches regularly depend on intricate algorithms and multiple sensors, that result in improved computational overhead and potential latency. The current study resolves the present gap by offering a new hybrid framework, which synergistically mixes optimization algorithms and deep neural networks through the advantages of mobile edge computing. Precisely, this research presents a hybrid model for autonomous vehicle management by integrating a refined version of the RPO or Pelican Optimizer with deep neural networks attuned to mobile edge computing environments. The chief contributions of the present study have been threefold: (1) the improvement of a particular autonomous driving method optimized for mobile edge computing platforms;(2) the arrangement of an optimized MobileNet method employing the RPO algorithm that uses LiDAR sensor data for effective object recognition and path design;and (3) the construction of an indoor vehicle prototype by mean of a microcontroller and LiDAR sensors, after a comprehensive performance evaluation of inference models, and analyzing the trade-offs between input size and computational effectiveness. Experimental outcomes show the efficiency and reliability of the suggested hybrid model, through improving autonomous vehicle management and decision-making abilities within the mobile edge computing paradigm. The current study contributes to the enhancement of autonomous vehicle research and provides an innovative solution for effective and precise vehicle control within edge computing environments.