关键词： 5G core network   downtime   mobile edge computing   service migration  

摘要： Seamless cross-edge migration for real-time rendering applications is challenging. The strong interactive nature of real-time rendering applications demands a downtime lower than 15 ms to achieve an imperceptible migration. Existing methods based on virtual machine migration and container migration suffer from unpleasant downtime brought by dirty page retransmission-induced repeated memory data copy and the shared storage failure-induced extensive disk data copy. In this article, we propose Cloud-assisted Service Migration (CSM) which leverages cloud-edge collaboration to achieve seamless service migration for real-time rendering applications. CSM improves service migration user experience in three folds: First, it introduces a dual rendering mechanism to bypass the peer-to-peer data copy and compresses the freezing stage. Second, a user equipment-centric session switch mechanism is proposed to save time by well coordinating application session switches and 5G user plane session switches. Third, a smooth switching mechanism is leveraged to prevent unpleasant frame flickers during session switching. We implement CSM in edge-rendering multiplayer games and deploy it on a 5G test bed with a full-stack user plane protocol stack. The evaluation results show that CSM can reduce downtime to < 14 ms and the service migration process is user imperceptible.

关键词： Digital twin   continual learning   mobile edge computing   resource allocation   inference model accuracy   model retraining   approximation algorithm   online algorithm  

摘要： Propelled by recent advances in Mobile Edge Computing (MEC) and the Internet of Things (IoT), the digital twin technique has been envisioned as a de-facto driving force to bridge the virtual and physical worlds through creating digital portrayals of physical objects. In virtue of the flourishing of edge intelligence and abundant IoT data, data-driven modelling facilitates the implementation and maintenance of digital twins, where simulations of physical objects are usually performed based on Deep Neural Networks (DNNs). A significant advantage of adopting digital twins is to enable decisive prediction on the behaviours of objects in near future without waiting for that really happen. To provide accurate predictions, it is vital to keep each digital twin synchronized with its physical object in real-time. However, it is challenging to maintain the real-time synchronization between a digital twin and its physical object due to the dynamics of physical objects and sensing data drift over time, i.e., the live data from a physical object diverge from the model training data of its digital twin. To address this critical issue, continual learning is a promising solution to retrain models of digital twins incrementally. In this article, we investigate digital twin synchronization issues via continual learning in an MEC environment, with the aim to maximize the total utility gain, i.e., the enhanced model accuracy. We study two novel optimization problems: the static digital twin synchronization problem per time slot and the dynamic digital twin synchronization problem for a finite time horizon. We first formulate an Integer Linear Program (ILP) solution for the static digital twin synchronization problem when the problem size is small;otherwise, we develop a randomized approximation algorithm at the expense of bounded resource violations for it. We also devise a deterministic approximation algorithm with guaranteed performance for a special case of the static digital twin s

关键词： Federated learning   mobile edge computing   cloud radio access network   resource allocation   SDR method  

摘要： In this paper, a mobile edge computing (MEC) assisted multi-layer architecture is proposed to support the implementation of federated learning in Internet of Things (IoT) networks. In this architecture, when performing a federated learning based task, data samples can be partially offloaded to MEC servers and cloud server rather than only processing the task at the IoT devices. After collecting local model parameters from devices and MEC servers, cloud server makes an aggregation and broadcasts it back to all devices. An optimization problem is presented to minimize the total federated training latency by jointly optimizing decisions on data offloading ratio, computation resource allocation and bandwidth allocation. To solve the formulated NP hard problem, the optimization problem is converted into quadratically constrained quadratic program (QCQP) and an efficient algorithm is proposed based on semidefinite relaxation (SDR) method. Furthermore, the scenario with the constraint of indivisible tasks in devices is considered and an applicable algorithm is proposed to get effective offloading decisions. Simulation results show that the proposed solutions can get effective resource allocation strategy and the proposed multi-layer federated learning architecture outperforms the conventional federated learning scheme in terms of the learning latency performance.

关键词： Deep reinforcement learning   Mobile edge computing   Unmanned aerial vehicle   Counterfactual inference   Whale optimization algorithm  

摘要： In heterogeneous wireless networks, massive user equipments (UEs) generate computing tasks with timevarying heterogeneous requirements. To improve the service quality, this paper formulates a unmanned aerial vehicles (UAVs)-assisted mobile edge computing (MEC) framework for time -varying heterogeneous task requirements. In the framework, the task delay and the number of successfully executed tasks are optimized by jointly controlling the trajectories of multiple UAVs. To address the considered trajectory planning optimization problem, a collaborative multi -agent deep reinforcement learning (MADRL) algorithm is proposed, where each UAV is regarded as a learning agent. First, a counterfactual inference based personalized policy update mechanism is proposed to evaluate the independent policy of agents by comparing the policy with a designed counterfactual policy. Based on this idea, each agent updates a personalized policy from both group and individual interests to improve its cooperation ability in dynamic and complex environments. Then, a diversified experience sampling mechanism is proposed to enhance the efficiency of policy evaluation and update with rich experiences provided by the environment interaction and the modified whale optimization algorithm. Finally, evaluation results demonstrate the superiority and effectiveness of the proposed MADRL algorithm.

关键词： Streaming media   Backhaul networks   Satellites   Quality of experience   Satellite broadcasting   Bandwidth   Servers   Video streaming   mobile edge computing   super-resolution   satellite backhaul  

摘要： Using satellite communications as backhaul links facilitates extending network coverage to unconnected areas. However, providing high-quality video streaming service via satellite backhaul is not economical. This paper presents SatSR, a mobile edge computing (MEC)-based super-resolution (SR)-enhanced adaptive on-demand video streaming system for mobile networks with satellite backhauls. Particularly, SR-based video quality enhancement is integrated into the video streaming process, so that low-quality videos with small sizes can be transmitted by satellite links and then enhanced to be high-quality. Meanwhile, SatSR offloads computation-intensive SR processing from user equipment (UE) to the MEC server to relieve UEs' computation burden and speed up the SR processing. Specifically, the framework and the operation process of SatSR are designed first. Then, to mitigate the impact of SR processing delay, a pipelined mechanism is proposed, which can coordinate the video transmission and SR-based enhancement efficiently. Furthermore, an SR scale factor adaptation algorithm based on deep reinforcement learning is proposed to cope with the fluctuation of communication links. Finally, a system prototype and a chunk-level simulator of SatSR are built, respectively. The experiments results validate that SatSR outperforms baselines significantly, including both the UE-based SR-enhancement video streaming scheme and the traditional bitrate adaptation based video streaming scheme.

关键词： Computation offloading   Task dependency   Mobile edge computing   Multi-edge cloud computing   Multi-objective optimization Enhanced Whale   Optimization dynamic allocation  

摘要： In this paper, we introduce the Enhanced Whale Optimization Algorithm (EWA) to optimize dependent task offloading in a multi-edge cloud computing environment. Our proposed algorithm aims to identify the most suitable offloading scenario for dependent tasks, focusing on minimizing total processing latency, energy consumption, and associated costs. We operate within a system comprising many decentralized Mobile Edge Computing servers (MECs) and a centralized cloud server. Two novel improvement operations, namely Frame Shifting (FS) and Load Redistribution Strategy (LRS), are introduced to enhance the performance of the whale algorithm. Through simulation, our results demonstrate the superior performance of EWA. Specifically, compared to the Whale Optimization Algorithm (WOA), EWA achieves a remarkable reduction in latency by 22.84%, a substantial decrease in energy consumption by 78.28%, and a notable reduction in cost usage by 61.47%. These outcomes underscore the efficacy and practical significance of the proposed EWA in addressing the challenges posed by dependent task offloading in the multi-edge cloud computing landscape.

关键词： Aerial-vehicular networks   Aerial autonomous vehicles   Mobile edge computing   High-altitude platforms   Millimeter waves   Data-aware non-orthogonal multiple access  

摘要： In recent years, there has been a significant interest in ubiquitous coverage, high data rate connectivity, and mobile edge computing (MEC) as crucial services within the future sixth -generation (6G) wireless networks. These services are regarded as essential components, exemplifying the advancements anticipated in 6G technology. Nevertheless, the successful implementation of these services in MEC-enabled vehicular networks significantly relies on the availability of robust network coverage and telecommunication framework. Unfortunately, in far-flung and isolated regions, such framework is often lacking, posing significant challenges in achieving uninterrupted connectivity, comprehensive coverage, and efficient computation offloading. Targeting the aforementioned horizon, in this paper, an uplink non -orthogonal multiple access (NOMA)-MEC-enabled aerial -vehicular network operating at millimeter wave (mmWave) is proposed in which the ground vehicles are provided edge computing services by an aerial autonomous vehicle, i.e., a high -altitude platform (HAP). We present a system model in which a HAP equipped with MEC servers offers computation offloading capabilities to a vehicular network. Our main objective is to minimize the transaction time of a NOMA cluster offloading its data to MEC servers located at the HAP. We devise a dual -layer optimization scheme for optimizing the transmission power and computational resource allocation by using the Lagrange multipliers method and then attain convergence by implementing a sub -gradient approach. We further extend our work by proposing a dataaware NOMA clustering scheme. The simulation results demonstrate the efficacy of our proposed approach, showing a notable reduction in the transaction time in comparison to the baseline scheme. The optimal power allocation enhances the data rates which subsequently reduces the transmission time, and the optimal cores assignment effectively minimizes the computation time. Additionally,

关键词： Vehicular edge computing   latency   communication   storage   mobile edge computing  

摘要： The latency associated with accessing data stored on edge computing servers for vehicles encompasses both the communication between a vehicle and a server as well as a latency of a data storage system. To enable low-latency vehicular services, an efficient resource management should consider the communication as well as the storage I/O cache resource allocation along with a data access pattern and a priority of individual vehicular services. Therefore, we focus on a joint optimization of communication and storage I/O cache resource allocation for access to data of vehicular services hosted by the edge computing servers. The proposed framework determines the data placement for the services and allocates communication and storage I/O cache resources to each service. The objective is to minimize the overall latency experienced by vehicular services for access to data. The edge computing platforms share storage and communication resources among various vehicular services, each having distinct priorities and data access rates or patterns. Hence, to reflect different priorities of services in resource allocation, our objective metric takes into account the service priority, data access frequency, and latency. We propose a feasible solution using dual relaxation considering both communication and storage latencies. The proposed solution reduces the average latency of vehicular services by up to 1.8x compared to the state-of-the-art resource allocation method for vehicular edge computing. Even more notable improvement is observed for high priority vehicular services, where the proposal leads to 2.5x lower latency compared to the state-of-the-art storage I/O cache architecture for virtualized cloud services.

关键词： Task analysis   Servers   Resource management   Processor scheduling   Optimal scheduling   Computational efficiency   Energy harvesting   Mobile edge computing   asynchronous computing   user scheduling   wireless power transfer  

摘要： In this paper, the problem of joint user scheduling and computing resource allocation in asynchronous mobile edge computing (MEC) networks is studied. In such networks, edge devices will offload their computational tasks to an MEC server, using the energy they harvest from this server. To get their tasks processed on time using the harvested energy, edge devices will strategically schedule their task offloading, and compete for the computational resource at the MEC server. Then, the MEC server will execute these tasks asynchronously based on the arrival of the tasks. This joint user scheduling, time and computation resource allocation problem is posed as an optimization framework whose goal is to find the optimal scheduling and allocation strategy that minimizes the energy consumption of these mobile computing tasks. To solve this mixed-integer non-linear programming problem, the general benders decomposition method is adopted which decomposes the original problem into a primal problem and a master problem. Specifically, the primal problem is related to computation resource and time slot allocation, of which the optimal closed-form solution is obtained. The master problem regarding discrete user scheduling variables is constructed by adding optimality cuts or feasibility cuts according to whether the primal problem is feasible, which is a standard mixed-integer linear programming problem and can be efficiently solved. By iteratively solving the primal problem and master problem, the optimal scheduling and resource allocation scheme is obtained. Simulation results demonstrate that the proposed asynchronous computing framework reduces 87.17% energy consumption compared with conventional synchronous computing counterpart.

关键词： Mobile edge computing   Cyber-physical system   Workflow application   Task scheduling   Metaheuristic  

摘要： This paper proposes a cost-minimization scheduling algorithm for the joint optimization of task offloading and resource allocation for workflow applications in a mobile edge computing (MEC)-supported cyber-physical system. We model this scheduling problem as an integer program that minimizes the total communication and computation cost under resources and deadline constraints upon workflow applications. To cope with this complicated optimization problem, we develop an efficient heuristic method to dispatch the workflow tasks onto MEC servers with sufficient resources. By using the task dispatching heuristic to evaluate the quality of each candidate solution, we construct a novel gene-inspired metaheuristic algorithm (GIMA) that incorporates an offspring-production operator and a conditional insertion scheme into the improvement strategy to explore high-quality solutions. Specifically, the offspring-production operator can enhance the scheduling quality by generating improved offspring solutions based on existing solutions. The conditional insertion scheme is further incorporated to reduce the computational overhead involved in solution exploration. We perform extensive simulations to justify the performance of the proposed GIMA in solving the scheduling problem studied in this work. Experimental results on real-world and synthetic workflow applications show that GIMA outperforms other metaheuristics in terms of cost reduction and computational efficiency.