关键词： Mobile edge computing   Edge server selection   Resource allocation   Deep Q-learning  

摘要： This paper discusses the challenges of mobile edge computing with low latency and energy consumption caused by the explosive growth of communication traffic and data generated by mobile devices. To address the issue of selecting edge servers, a multi-user-oriented edge server selection strategy is proposed. The strategy minimizes the weighted sum of network latency and total energy consumption and develops an improved SPEA2-based algorithm to choose the most suitable edge server. The resource allocation problem is also considered, and a resource allocation strategy is proposed that maximizes the energy efficiency ratio. A DQN-based resource allocation strategy is devised to find the best resource allocation strategy. Experimental results demonstrate that the proposed server selection strategy reduces system overhead in terms of energy consumption and latency while improving resource utilization. The resource allocation strategy improves the computational efficiency of edge servers while reducing latency and total energy consumption.

关键词： Mobile edge computing   Cost minimization   Markov decision process   Deep reinforcement learning   Power allocation  

摘要： Mobile edge computing (MEC) can dispatch its powerful servers close by to assist with the computation workloads that intelligent wireless terminals have offloaded. The MEC server's physical location is closer to the intelligent wireless terminals, which can satisfy the low latency and high reliability demands. In this paper, we formulate an MEC framework with multiple vehicles and service devices that considers the priority and randomness of arriving workloads from roadside units (RSUs), cameras, laser radars (Lidar) and the time-varying channel state between the service device and MEC server (MEC-S). To minimize the long-term weighted average cost of the proposed MEC system, we transit this issue (cost minimization problem) into the Markov decision process (MDP). Furthermore, considering the difficulty realizing the state transition probability matrix, the dimensional complexity of the state space, and the continuity of the action space, we propose a deterministic policy gradient (MADDPG)-based bandwidth partition and power allocation optimization policy. The proposed MADDPG-based policy is a model-free deep reinforcement learning (DRL) method, which can effectively deal with continuous action space and further guide multi-agent to execute decision-making. The comprehensive results verify that the proposed MADDPG-based optimization scheme has fine convergence and performance that is better than that of the other four baseline algorithms.

关键词： Mobile edge computing   Deep reinforcement learning   Computation offloading   Resource allocation   Multi-branch architecture  

摘要： Mobile Edge Computing (MEC) is a promising computing paradigm in the context of 5G networks, as it enables the migration of workloads from User Equipments (UEs) to nearby MEC servers, thereby providing additional computing resources to UEs. In this paper, we propose a joint optimization approach to offloading decisions and resource allocation in a multi-user and multi-server MEC system, which operates in a time-varying environment. Our objective is to minimize the average task latency and discard rate under the constraints of latency and limited computing resources of the server. While traditional optimization methods have been used to solve computational offloading problems in static environments, these methods are not suitable for time-varying systems. Deep reinforcement learning can be an effective method for solving optimization problems in time-varying environments, as it can be used to adjust strategies in real-time in response to changes in the environment. However, the increased number of UEs in the system leads to a combinatorial increase in the number of possible actions, making it difficult for the algorithm to learn. To address this issue, we propose a multi-branch network based Deep Q Network (DQN) algorithm called Branch Deep Q Network (BDQN), which modifies the action generation network into a multi-branch network structure, each branch generates one-dimensional action. This modification makes the number of network outputs increase linearly. Numerical results show that the BDQN algorithm outperforms other baseline algorithms in terms of average task latency and discard rate.

关键词： Sensors   Internet of Things   Servers   Task analysis   Optimization   Monitoring   Process control   Age of information   mobile edge computing   computation offloading   status update  

摘要： IoT devices have been widely utilized to detect state transition in the surrounding environment and transmit status updates to the base station for system operations. To guarantee the accuracy of system control, age of information (AoI) is introduced to quantify the freshness of the sensory data and meet the stringent timeliness requirement. Due to the limited computing resources, the status update can be offloaded to the mobile edge computing (MEC) server for execution. Since status updates generated by insufficient sensing operations may be invalid and lead to additional processing time, a joint data sensing and processing optimization problem needs to be considered. Therefore, this work formulates an NP-hard problem that considers the freshness of the status updates and energy consumption of the IoT devices. Subsequently, the problem is decomposed into sampling, sensing, and computation offloading optimization problems. To optimize the system overhead, a multi-variable iterative system cost minimization algorithm is proposed. Simulation results illustrate the efficacy of our method in decreasing the system cost, and indicate the influence of sensing and processing under different scenarios.

关键词： Servers   Resource management   Task analysis   Costs   Cloud computing   Delays   Minimization   Deep learning   Q-factor   Edge computing   Computation offloading   double deep Q-Network   mobile edge computing   resource allocation   service caching  

摘要： Mobile Edge Computing (MEC) provides cloud-like computing functionalities for mobile devices (MDs) by deploying servers at the edge of future 6G network. The cooperation among multi-servers and remote cloud offers low-latency services for MDs, meeting the demand of numerous delay-sensitive computation tasks. However, the limited storage capacity of edge servers makes it difficult to cache all kinds of services from the cloud simultaneously. Furthermore, the available bandwidth of communication links and the computation capacity of edge servers are insufficient for all MDs. In this paper, we consider a multi-server MEC scenario, where exists the competitions of multiple MDs for communication and computation resource, and various services for storage capacity. We investigate task offloading and resource allocation strategy in the three-tier MEC system, and explore the service caching strategy of the edge server. We aim at minimizing the cost of delay and energy consumption, and formulate the joint optimization problem as Markov Decision Process (MDP). Based on Double Deep Q-Network (DDQN) algorithm, the joint service caching, resource allocation and computation offloading scheme (SCRACO) is proposed to solve the problem. The simulation results demonstrate that the proposed scheme can effectively reduce total cost of the system compared with other three benchmarks.

关键词： Mobile edge computing   Resource allocation   Non-orthogonal multiple access   Unmanned aerial vehicle   Task offloading  

摘要： To address the challenges in the era of artificial intelligence, this paper investigates how computing offloading techniques can be utilized to improve the system performance in cellular connected unmanned aerial vehicle (UAV) networks with limited resources using non-orthogonal multiple access (NOMA) and mobile edge computing techniques. Energy consumed by the UAV in performing tasks and the time required to complete tasks are minimized by joint optimal task offloading decision, sub-channel allocation, and computing resource allocation while satisfying the user's quality of service requirements. We consider this problem as a nonlinear programming problem with integer and non-integer variables and solve it by iteratively updating the resource allocation and offloading decision. Specifically, given a fixed offloading decision, a many-to-one matching model is first applied for sub-channel allocation, then a detailed allocation of computing resource is performed by optimizing the total system overhead, and are solved by utilizing the matching algorithm and the Lagrange dual method, respectively. Then, according to the scenarios we have been given for resource allocation, we design a multi-objective task offloading algorithm to update the offloading decision. Simulation experiments demonstrate the feasibility of the scheme, which shows a significant reduction in both energy consumption and latency compared to the benchmark scheme.

关键词： Mobile edge computing   task offloading   high speed railway   fluctuation of transmission and processing   potential game   stochastic network calculus  

摘要： The advanced 5G and mobile edge computing promote the development of the intelligent high-speed railway and enable computing-intensive and latency-sensitive tasks. Mobile edge computing can effectively release the pressure of tasks on on-board computing resources by migrating communication, computing, and storage to edge servers. However, the fluctuation of data transmission and task processing duration exists in practical projects. The current work assumes that the traditional communication model is ideal and that the latency can be derived as a fixed value, which is the most conservative bound. Obviously, there will be some system performance loss because the transmission latency is usually much shorter than this conservative bound. Therefore, an offloading strategy is proposed to maximize the task completion rate, considering the fluctuation of transmission latency and processing time in this paper. First, the stochastic network calculus is adopted to evaluate the fluctuation and probability of the transmission latency. Additionally, the task processing duration is regarded as an exponential distribution that is affected by the computing resource. Then, the task offloading model is generated based on the potential game to maximize the task completion rate. Moreover, the Nash Equilibrium and best response are derived. A reinforcement learning algorithm combined with a game is proposed to reach the Nash Equilibrium and obtain the task offloading strategy. Finally, extensive theoretical analysis and simulations are illustrated to prove the effectiveness of the model.

关键词： deep reinforcement learning   advanced asynchronous advantage actor-critic (A3C)   multi-agent system   mobile edge computing   cloud computing   computational offloading   energy efficiency  

摘要： Mobile edge computing networks (MECNs) based on hierarchical cloud computing have the ability to provide abundant resources to support the next-generation internet of things (IoT) network, which relies on artificial intelligence (AI). To address the instantaneous service and computation demands of IoT entities, AI-based solutions, particularly the deep reinforcement learning (DRL) strategy, have been intensively studied in both the academic and industrial fields. However, there are still many open challenges, namely, the lengthening convergence phenomena of the agent, network dynamics, resource diversity, and mode selection, which need to be tackled. A mixed integer non-linear fractional programming (MINLFP) problem is formulated to maximize computing and radio resources while maintaining quality of service (QoS) for every user's equipment. We adopt the advanced asynchronous advantage actor-critic (A3C) approach to take full advantage of distributed multi-agent-based solutions for achieving energy efficiency in MECNs. The proposed approach, which employs A3C for computing offloading and resource allocation, is shown through numerical results to significantly reduce energy consumption and improve energy efficiency. This method's effectiveness is further shown by comparing it to other benchmarks.

关键词： Mobile edge computing   Distributed computing   DNN offloading   Interleaved offloading   Edge computing   Computational offloading  

摘要： Deep Neural Networks (DNNs) are increasingly used on the edge in mobile and other resource-constrained devices for inference tasks ranging from object detection and image recognition to video processing. Many of these tasks have low latency requirements that cannot be satisfied if they are processed locally due to their high computational complexity. Offloading computation to the edge and cloud offers a way to alleviate this computational latency. Doing so introduces communication delays, which makes offloading a balancing act between the benefits of reduced processing time and the communication delays incurred. Existing algorithms for DNN offloading based on DNN partitioning are optimised for handling successive tasks on a single remote server, and perform sub-optimally when tasks are interleaved, or with multiple servers, or when privacy concerns require local processing of certain DNN layers. A viable alternative are generic computational offloading algorithms (GOAs), which can break down DNN tasks into their components and perform fine-grained offloading. We perform a simulation-based comparison of traditional GOAs with various levels of proactivity and offloading constraints and a naive DNN partitioning approach. We identify key requirements for offloading algorithms in the ability to create processing chains between several remote servers in order to reduce communication overheads, and the ability to prioritise already running tasks. The results confirm the expectations about the shortcomings of DNN partitioning, and show that GOA algorithms can provide a significant performance improvement.

关键词： 6G edge networks   mobile edge computing   survivability guarantee   node-overload protection   traffic prediction   task offloading  

摘要： With the development of sixth-generation (6G) mobile networks, the rise of emerging intelligent services has led to a huge increase in traffic. As an important technology to support the development of 6G, mobile edge computing (MEC) effectively meets the ultra-low latency requirements of most emerging services. However, due to the limited processing capacity of edge nodes, overload on any edge node will cause service degradation, interruption, and even node failure, weakening the advantages of MEC and reducing the survivability of the whole network. In this paper, we propose a task offloading scheme based on traffic prediction for node-overload protection to ensure the survivability of the 6G edge networks. We transformed the network survivability guarantee problem into a task offloading problem under the constraint of future available resources based on traffic prediction and developed a particle swarm optimization algorithm based on policy gradient (PSO-PG) to jointly optimize offloading decisions, routing, and computing resource allocation. Simulations verify the effectiveness of our proposed scheme and guarantee the survivability of 6G edge networks. Meanwhile, evaluation in multiple scenarios with different node scales has verified the wide applicability of our work.