关键词： Mobile edge computing   D2D underlay communication   Mean field game   Latency-critical applications   Dynamic system  

摘要： With the development of 5 G mobile users are increasing massively. Some mobile applications like healthcare are latency-critical and requires real-time data processing. A preference-based task offloading framework in mobile edge computing with a device-to-device offloading (MECD2D) system has been proposed to fulfill the latency demands of such applications for minimum energy consumption ensuring resiliency. The problem is formulated as a constraint-based non-linear optimization problem which is complex. The resources are allocated in two steps. In the first step, resources are allocated based on latency demand to ensure resiliency. In the second step, allocated resources are optimized using a non-cooperative mean field game for dynamic system. To ensure the performance of the system for dynamic network, the results are executed on a real-time Shanghai dataset. The computational results indicate that the proposed algorithm performs better in terms of energy consumption. Other parameters such as throughput, network utilization and task computation are also analysed. The results are verified by performing the proposed algorithm with existing Q learning and mean-field game algorithms. The results performed on the dataset indicate an improvement in energy consumption by 5-10 %, and 10-50 % as compared to Q learning and mean-field game respectively.

关键词： Servers   Computer architecture   Internet of Things   Task analysis   Energy consumption   Security   Optimization   Mobile edge computing   radio access networks   resource allocation   energy consumption   Q-learning algorithm  

摘要： As information technology rapidly advances, 5G technology, Radio Access Networks (RAN), and the Internet of Things (IoT) have emerged as the core elements of next-generation communication technology. There is an increasing demand for real-time communication and reduced latency in various applications. Therefore, this paper proposes a four-layer Mobile Edge Computing (MEC) architecture that connects user devices to the core network using RAN. Blockchain verification is used for data storage and access permission separation. The architecture aims to address the high latency, low flexibility, and security issues in cloud computing communication. We also propose a MEC server location algorithm to optimize communication distance, and a Q-learning algorithm for selection and resource allocation. Experimental results demonstrate significant energy savings compared to baseline algorithms.

关键词： Task analysis   Delays   Autonomous aerial vehicles   Optimization   Trajectory   Resource management   Energy consumption   Deep reinforcement learning   mobile edge computing   task offloading   unmanned aerial vehicle (UAV)  

摘要： Mobile edge computing (MEC) has emerged as a solution to address the demands of computation-intensive network services by providing computational capabilities at the network edge, thus reducing service delays. Due to the flexible deployment, wide coverage and reliable wireless communication, unmanned aerial vehicles (UAVs) have been employed to assist MEC. This paper investigates the task offloading problem in a UAV-assisted MEC system with collaboration of multiple UAVs, highlighting task priorities and binary offloading mode. We defined the system gain based on energy consumption and task delay. The joint optimization of UAVs' trajectory design, binary offloading decision, computation resources allocation, and communication resources management is formulated as a mixed integer programming problem with the goal of maximizing the long-term average system gain. Considering the discrete-continuous hybrid action space of this problem, we propose a novel deep reinforcement learning (DRL) algorithm based on the latent space to solve it. The evaluation results demonstrate that our proposed algorithm outperforms three state-of-the-art alternative solutions in terms of task delay and system gain.

关键词： Task analysis   Privacy   Topology   Delays   Costs   Vehicle dynamics   Processor scheduling   Mobile edge computing   directed acyclic graph   dependency-aware   computation offloading   deep reinforcement learning   vehicular edge computing networks  

摘要： The proliferation of computation-intensive and delay-sensitive applications in the Internet of Vehicles (IoV) poses great challenges to resource-constrained vehicles. To tackle this issue, Mobile Edge Computing (MEC) enabling offloading on-vehicle tasks to edge servers has emerged as a promising approach. MEC jointly augments network computing capabilities and alleviates resource utilization for IoV. Nevertheless, the efficacy of MEC depends heavily on the adopted offloading scheme, especially in the presence of complex subtask dependencies. Existing research has largely overlooked the crucial dependencies among subtasks, which significantly influences the decision making for offloading. This work attempts to schedule subtasks with guaranteed dependencies while minimizing system latency and energy costs in multi-vehicle scenarios. First, we introduce a subtask priority scheduling method based on the Directed Acyclic Graph (DAG) topological structure to ensure the priority order of subtasks, especially in the scenarios with complex interdependencies. Second, in light of privacy concerns and limited information sharing, we propose an Optimized Distributed Computation Offloading (ODCO) scheme based on deep reinforcement learning (DRL), alleviating the conventional requirement for extensive vehicle-specific information sharing to achieve optimal offloading performance. The adaptive k-step learning approach is further presented to enhance the robustness of the training process. Numerical experiments are presented to illustrate the advantages of the proposed scheme compared to the existing state-of-the-art offloading schemes. For instance, the ODCO achieved a system utility of approximately 0.80 within 300 episodes, obtaining utility gains of about 0.05 compared to the distributed earliest-finish time offloading (DEFO) algorithm with around 500 episodes.

关键词： Internet of vehicles   Mobile edge computing   Deep reinforcement learning   Federated learning   Task offloading   Resource allocation  

摘要： Mobile edge computing (MEC) enables computation intensive applications in the Internet of Vehicles (IoV) to no longer be limited by device resources. However, the lack of an effective task scheduling strategy will seriously affect users' quality of experience (QoE). In this paper, a task type-based task offloading and resource allocation strategy is proposed to reduce delay and energy consumption during task execution. First, we establish communication, computing, and system cost models based on task offloading schemes, and model the joint optimization problem of task offloading and resource allocation as a Markov decision process. The utility function is obtained based on the task completion rate and the system cost. Second, an algorithm framework based on multi-agent deep deterministic policy gradient (MADDPG) is designed to solve the difficulty that traditional single-agent reinforcement learning algorithms are difficult to converge in a dynamic environment. In distributed scenarios, the proposed framework can also reduce system costs while handling more tasks. Finally, federated learning is introduced in the training process to reduce the impact of non-IID data while protecting privacy. Simulation results show that the proposed algorithm can effectively improve system processing efficiency and reduce device energy consumption compared to the popular reinforcement learning algorithms.

关键词： Internet of Vehicles   mobile edge computing   iterative double auction   evolutionary game   cloud mining   Internet of Vehicles   mobile edge computing   iterative double auction   evolutionary game   cloud mining  

摘要： With the growing prominence of the Internet of Vehicles (IoV), there is an unprecedented generation of vehicular data, which harbors immense potential for diverse applications. Considering the convergence of blockchain technology and IoV, blockchain can be used as a distributed ledger to establish a decentralised and autonomous trading platform for IoV networks. However, the limitations in the power and computational capacities of IoV devices hinder their direct involvement in the Proof-of-Work (PoW) process. Addressing this challenge, offloading the computational demands of PoW to Mobile Edge Computing (MEC) servers emerges as a feasible solution. This scenario has prompted us to conceptualize the computational resource management in blockchain consensus as a two-layered resource game, aimed at optimizing both the profits of computing power providers and the utility for vehicles. In the Blockchain Market Layer subgame, we employ a game-theoretic framework to analyze the interactions between MEC servers and vehicles within a PoW-based blockchain network. To optimize the system utility, we introduce a centralized evolutionary game-based pool selection algorithm. Simultaneously, in the Edge Cloud Network Layer subgame, a reverse auction algorithm is implemented to align the computing power supply with the demand within the blockchain market. Here, the cloud data center functions as the seller of computing power, while the edge servers aggregate the computing demand from IoV devices. A broker is then tasked with the allocation and pricing of computing resources, in accordance with the bids, ensuring adherence to economic principles of authenticity and individual rationality. Simulation results demonstrate that our mechanism enhances system efficiency while ensuring revenue, thereby enabling an effective distribution of computing resources between vehicles and cloud data centers.

关键词： next-generation networks   mobile edge computing   decentralized mechanism  

摘要： With the rapid advancement of the Internet of Things and the emergence of 6G networks in smart city environments, a growth in the generation of data, commonly known as big data, is expected to consequently lead to higher latency. To mitigate this latency, mobile edge computing has been proposed to alleviate a portion of the workload from mobile devices by offloading it to nearby edge servers equipped with appropriate computational resources. However, existing solutions often exhibit poor performance when confronted with complex network topologies. Thus, this paper introduces a decentralized mechanism aimed at determining the locations of network edge nodes in such complex network topologies, characterized by lengthy execution times. Our proposal provides performance improvements and offers scalability and flexibility as networks become more complex. Experimental evaluations are conducted using the Shanghai Telecom dataset to validate our proposed approach.

关键词： 6 G Mobile Communication   Cloud Computing   Multi Access Edge Computing   5 G Mobile Communication   Computer Architecture   Servers   Security   Radio Access Networks   Internet Of Things   Edge Cloud   Infrastructure   Data Center   Cloud Computing   Open Research   Network Operators   Internet Of Things Devices   Computational Capabilities   5 G Networks   Edge Server   Mobile Edge Computing   Radio Access Network   Computation Offloading   Microservices   6 G Networks   Ultra Low Latency   High Degree Of Mobility   Future Of Humanity   Open Research Problem   Mobile Edge Computing Server   Local Edge   Distributed Energy Resources   Predictive Maintenance   Smart Manufacturing   Mobile Network   Intelligent Transportation Systems   Mobile Network Operators   Virtual Network Functions   Edge Caching  

摘要： Edge computing is one of the key driving forces to enable beyond 5G (B5G) and 6G networks. To cope with the unprecedented increase in traffic volumes and computation demands of future networks, multiaccess (or mobile) edge computing (MEC) is considered a promising solution that provides cloud computing capabilities within the radio access network. There exists a significant amount of research on MEC and its potential applications;however, very little has been said about the key factors of MEC deployment to meet the diverse demands of future applications. This article presents key considerations for edge deployments in B5G/6G networks, including edge architecture, server location, capacity, user density, security, etc. We further present state-of-the-art edge-centric services in future B5G/6G networks, including experimental evaluation of edge-based computing. Lastly, we present some interesting insights and open research problems in edge computing for 6G networks.

关键词： Federated learning   mobile edge computing   unmanned aerial vehicle   wireless powered communication network   Federated learning   mobile edge computing   unmanned aerial vehicle   wireless powered communication network  

摘要： In recent years, federated learning (FL) has been adopted in mobile edge computing (MEC) to protect user privacy. However, in some cases, in addition to performing FL to process private data, users also need to process other non-private data. Therefore, how to integrate private data and non-private data in a MEC system for comprehensive processing is an issue worth studying. In this paper, we propose an unmanned aerial vehicle (UAV)-enabled wireless powered communication network (WPCN) to process both FL tasks and offloadable MEC tasks of UEs, where the UAV charges UEs via wireless power transfer (WPT) technology, executes offloaded tasks, and aggregates the FL model parameters. To minimize energy consumption of the UAV, we formulate a problem to jointly optimize the hovering position of UAV, the WPT power, the proportion of UAV's computing resources, the percentage of offloaded tasks, and the time scheduling of FL under the energy harvesting constraint. The problem is divided into three subproblems with the aid of block coordinate descent (BCD) method. These subproblems are solved by Lagrange method and heuristic algorithm respectively. Numerical results show our algorithm can reduce the energy consumption of the UAV with low time complexity compared with several benchmarks.

关键词： Mobile edge computing   two-timescale   smart grid   resource allocation   Lyapunov optimization   Mobile edge computing   two-timescale   smart grid   resource allocation   Lyapunov optimization  

摘要： Powering mobile edge computing (MEC) with a hybrid supply of smart grid (SG) and renewable energy source (RES) offers an opportunity to utilize clean energy and cut down energy expenses under two-way energy transactions. We propose two-timescale online resource allocation and energy management (TSRE) for MEC with a hybrid power supply, adapting to dynamic task and RES arrivals, wireless channels, and energy prices. The TSRE minimizes the time-averaged cost of predictive energy planning and real-time energy trading of base stations (BSs), and the energy usage of mobile users. By generalizing the Lyapunov optimization and stochastic subgradient method, the energy planning (sub)problem is solved upon RES arrivals using only historical data. The real-time offloading schedules, energy trading decisions and CPU configurations are decoupled over time and (asymptotically) optimally made in a distributed manner. The feasibility and the asymptotic optimality of the TSRE are proved. Numerical results demonstrate that the TSRE saves system cost significantly by 33.7%, compared to its baselines.