关键词： Mobile edge computing   deep reinforcement learning   digital twin   digital twin edge network  

摘要： 6G is envisioned to empower wireless communication and computation through the digitalization and connectivity of everything, by establishing a digital representation of the real network environment. Mobile edge computing (MEC), as one of the key enabling factors, meets unprecedented challenges during mobile offloading due to the extremely complicated and unpredictable network environment in 6G. The existing works on offloading in MEC mainly ignore the effects of user mobility and the unpredictable MEC environment. In this paper, we present a new vision of Digital Twin Edge Networks (DITEN) where digital twins (DTs) of edge servers estimate edge servers' states and DT of the entire MEC system provides training data for offloading decision. A mobile offloading scheme is proposed in DITEN to minimize the offloading latency under the constraints of accumulated consumed service migration cost during user mobility. The Lyapunov optimization method is leveraged to simplify the long-term migration cost constraint to a multi-objective dynamic optimization problem, which is then solved by Actor-Critic deep reinforcement learning. Simulations results show that our proposed scheme effectively diminishes the average offloading latency, the offloading failure rate, and the service migration rate, as compared with benchmark schemes, while saving the system cost with DT assistance.

关键词： Full granularity   Partial offloading   Mixed overhead   Mobile edge computing  

摘要： Mobile edge computing (MEC) utilizes wireless access network to provide powerful computing resources for mobile users to improve the user experience, which mainly includes two aspects: time and energy consumption. Time refers to the latency consumed to process user tasks, while energy consumption refers to the total energy consumed in processing tasks. In this paper, the time and energy consumption in user experience are weighted as a mixed overhead and then optimized jointly. We formulate a mixed overhead of time and energy (MOTE) minimization problem, which is a nonlinear programming problem. In order to solve this problem, the block coordinate descent method to deal with each variable step by step is adopted. We further analyze the minimum value of delay parameters in the model, and examine two special cases: 1-offloading and 0-offloading. In 1-offloading, all the task data is offloaded to MEC server, and no data offloaded in 0-offloading. The necessary and sufficient conditions for the existence of two special cases are also deduced. Besides, the multi-user situation is also discussed. In the performance evaluation, we compare MOTE with other offloading schemes, such as exhaustive strategy and Monte Carlo simulation method-based strategy to evaluate the optimality. The simulation results show that MOTE always achieves the minimal overhead compared to other algorithms.

关键词： Deep reinforcement learning   Energy consumption   Mobile edge computing   Response time   Task offloading  

摘要： With the popularity of smart mobile equipment, the amount of data requested by users is growing rapidly. The traditional centralized processing method represented by the cloud computing model can no longer satisfy the effective processing of large amounts of data. Therefore, the mobile edge computing (MEC) is used as a new computing model to process the big growing data, which can better meet the service requirements. Similar to the task scheduling problem in cloud computing, an important issue in the MEC environment is task offloading and resource allocation. In this paper, we propose an adaptive task offloading and resource allocation algorithm in the MEC environment. The proposed algorithm uses the deep reinforcement learning (DRL) method to determine whether the task needs to be offloaded and allocates computing resources for the task. We simulate the generation of tasks in the form of Poisson distribution, and all tasks are submitted to be processed in the form of task flow. Besides, we consider the mobility of mobile user equipment (UE) between base stations (BSs), which is closer to the actual application environment. The DRL method is used to select the suitable computing node for each task according to the optimization objective, and the optimal strategy for solving the objective problem is learned in the algorithm training process. Compared with other comparison algorithms in different MEC environments, our proposed algorithm has the best performance in reducing the task average response time and the total system energy consumption, improving the system utility, which meets the profits of users and service providers. (C) 2020 Elsevier Inc. All rights reserved.

关键词： 3D connectivity   3D services   5G   6G   B5G   high-altitude platform stations   mobile edge computing   non-terrestrial communications   satellite   unmanned aerial vehicle  

摘要： Sixth generation will exploit satellite, aerial, and terrestrial platforms jointly to improve radio access capability and unlock the support of on-demand edge cloud services in three-dimensional (3D) space, by incorporating mobile edge computing (MEC) functionalities on aerial platforms and low-orbit satellites. This will extend the MEC support to devices and network elements in the sky and forge a space-borne MEC, enabling intelligent, personalized, and distributed on-demand services. End users will experience the impression of being surrounded by a distributed computer, fulfilling their requests with apparently zero latency. In this paper, we consider an architecture that provides communication, computation, and caching (C3) services on demand, anytime, and everywhere in 3D space, integrating conventional ground (terrestrial) base stations and flying (non-terrestrial) nodes. Given the complexity of the overall network, the C3 resources and management of aerial devices need to be jointly orchestrated via artificial intelligence-based algorithms, exploiting virtualized network functions dynamically deployed in a distributed manner across terrestrial and non-terrestrial nodes.

关键词： Mobile edge computing   Deep reinforcement learning   Computation offloading   Multi-user MIMO  

摘要： Mobile edge computing (MEC) emerges recently as a promising solution to relieve resource-limited mobile devices from computation-intensive tasks, which enables devices to offload workloads to nearby MEC servers and improve the quality of computation experience. In this paper, an MEC enabled multi-user multi-input multi-output (MIMO) system with stochastic wireless channels and task arrivals is considered. In order to minimize long-term average computation cost in terms of power consumption and buffering delay at each user, a deep reinforcement learning (DRL)-based dynamic computation offloading strategy is investigated to build a scalable system with limited feedback. Specifically, a continuous action space-based DRL approach named deep deterministic policy gradient (DDPG) is adopted to learn decentralized computation offloading policies at all users respectively, where local execution and task offloading powers will be adaptively allocated according to each user's local observation. Numerical results demonstrate that the proposed DDPG-based strategy can help each user learn an efficient dynamic offloading policy and also verify the superiority of its continuous power allocation capability to policies learned by conventional discrete action space-based reinforcement learning approaches like deep Q-network (DQN) as well as some other greedy strategies with reduced computation cost. Besides, power-delay tradeoff for computation offloading is also analyzed for both the DDPG-based and DQN-based strategies.

关键词： Mobile edge computing   Privacy of vehicle location   Computational collaborative scheduling strategy   Kalman filter   Double DQN algorithm  

摘要： This paper proposes a collaborative scheduling strategy for computing resources of the Internet of vehicles considering location privacy protection in the mobile edge computing environment. Firstly, a multi area multi-user multi MEC server system is designed, in which a MEC server is deployed in each area, and multiple vehicle user equipment in an area can offload computing tasks to MEC servers in different areas by a wireless channel. Then, considering the mobility of users in Internet of vehicles, a vehicle distance prediction based on Kalman filter is proposed to improve the accuracy of vehicle-to-vehicle distance. However, when the vehicle performs the task, it needs to submit the real location, which causes the problem of the location privacy disclosure of vehicle users. Finally, the total cost of communication delay, location privacy of vehicles and energy consumption of all users is formulated as the optimization goal, which take into account the system state, action strategy, reward and punishment function and other factors. Moreover, Double DQN algorithm is used to solve the optimal scheduling strategy for minimizing the total consumption cost of system. Simulation results show that proposed algorithm has the highest computing task completion rate and converges to about 80% after 8000 iterations, and its performance is more ideal compared with other algorithms in terms of system energy cost and task completion rate, which demonstrates the effectiveness of our proposed scheduling strategy.

关键词： Mobile edge computing  

摘要： In this paper, we study resource allocation algorithm design for multi-user orthogonal frequency division multiple access (OFDMA) ultra-reliable low latency communication (URLLC) in mobile edge computing (MEC) systems. To meet the stringent end-to-end delay and reliability requirements of URLLC MEC systems, we propose joint uplink-downlink resource allocation and finite blocklength transmission. Furthermore, we employ a partial time overlap between the uplink and downlink frames to minimize the end-to-end delay, which introduces a new time causality constraint. The proposed resource allocation algorithm is formulated as an optimization problem for minimization of the total weighted power consumption of the network under a constraint on the number of URLLC user bits computed within the maximum allowable computation time, i.e., the end-to-end delay of a computation task. Despite the non-convexity of the formulated optimization problem, we develop a globally optimal solution using a branch- and-bound approach based on discrete monotonic optimization theory. The branch-and-bound algorithm minimizes an upper bound on the total power consumption until convergence to the globally optimal value. Furthermore, to strike a balance between computational complexity and performance, we propose two efficient suboptimal algorithms based on successive convex approximation and second-order cone techniques. Our simulation results reveal that the proposed resource allocation algorithm design facilitates URLLC in MEC systems, and yields significant power savings compared to three baseline schemes. Moreover, our simulation results show that the proposed suboptimal algorithms offer different trade-offs between performance and complexity and attain a close-to-optimal performance at comparatively low complexity. Copyright © 2020, The Authors. All rights reserved.

关键词： game theory   autonomous aerial vehicles   mobile computing   relatively stable state   MEC   computing services   energy consumption   unmanned aerial vehicle   mobility   flexibility   edge server   instantaneous computing surge   resources costs   offloading model   mobile user   computing resources competition   UAV-enabled computation offloading algorithm   lower computing cost   cost-efficient computation   UAV-enabled edge computing   computationally intensive applications   mobile edge computing   increasing computing demands  

摘要： With the popularity of computationally intensive applications, more and more computing resources are required. Mobile edge computing (MEC) is widely applied as an effective method to meet the increasing computing demands. In a relatively stable state, MEC can provide computing services with low latency and energy consumption. However, in special cases such as communication traffic, the unmanned aerial vehicle (UAV), by taking advantage of its mobility and flexibility, can assist the edge server to cope with the challenge of instantaneous computing surge. In this study, the authors consider a UAV-enabled edge computing system. In addition to delay and energy consumption, the authors also consider computing resources costs in the offloading model. Besides, in order to minimise the computing cost of each mobile user (MU), they apply the non-cooperative game method to model the channel and computing resources competition among MUs. Then, the authors prove that the proposed game is an ordinal potential game and the existence of Nash equilibrium in the game. The authors propose the UAV-enabled computation offloading (UECO) algorithm to obtain the equilibrium strategy. Finally, the authors show that the UECO algorithm can quickly converge through iterative experiments, and it can achieve lower computing cost through comparative experiments.

关键词： cloud computing   integer programming   cellular radio   nonlinear programming   autonomous aerial vehicles   mobile computing   learning (artificial intelligence)   relay networks (telecommunication)   multi-robot systems   optimisation problem   complexity reduction   QCOA   quality of experience   MUs-Edge-Cloud three-layer network architecture   MU energy consumption   two-layer network architecture   traversal offloading   random offloading   efficient Q-learning based computation offloading algorithm   mixed-integer nonlinear programming problem   edge servers   remote cloud centre   computation resources   multiUAV-enabled MEC system   controllability   mobile edge computing   mobile users   relaying services   flying platform   UAV   unmanned aerial vehicles   multiUAV-enabled cloud-edge computing networks  

摘要： Unmanned aerial vehicles (UAVs) have been recently considered as a flying platform to provide wide coverage and relaying services for mobile users (MUs). Mobile edge computing (MEC) is developed as a new paradigm to improve quality of experience of MUs in future networks. Motivated by the high flexibility and controllability of UAVs, in this study, the authors study a multi-UAV-enabled MEC system, in which UAVs have computation resources to offer computation offloading opportunities for MUs, aiming to reduce MUs' total consumptions in terms of time and energy. Considering the rich computation resource in the remote cloud centre, they propose the MUs-Edge-Cloud three-layer network architecture, where UAVs play the role of flying edge servers. Based on this framework, they formulate the computation offloading issue as a mixed-integer non-linear programming problem, which is difficult to obtain an optimal solution in general. To address this, they propose an efficientQ-learning based computation offloading algorithm (QCOA) to reduce the complexity of optimisation problem. Numerical results show that the proposed QCOA outperforms benchmark offloading policies (e.g. random offloading, traversal offloading). Furthermore, the proposed three-layer network architecture achieves a 5% benefits compared with the traditional two-layer network architecture in terms of MUs' energy and time consumptions.

关键词： Unmanned aerial vehicle   Mobile edge computing   Non-orthogonal multiple access   Fair resource allocation design   Computation efficiency  

摘要： In this paper, we study the issue of fair resource optimization for an unmanned aerial vehicle (UAV)-enabled mobile edge computing (MEC) system with multi-carrier non-orthogonal multiple access (MC-NOMA). A computation efficiency (CE) optimization problem based on the max-min fairness principle under the partial offloading mode is formulated by optimizing the subchannel assignment, the local CPU frequency, and the transmission power jointly. The formulated problem belongs to the non-convex mixed integer nonlinear programming (MINLP), that is NP-hard to find the global optimal solution. Therefore, we design a polynomial-time algorithm based on the big-M reformulation, the penalized sequential convex programming, and the general Dinkelbach's method, which can choose an arbitrary point as the initial point and eventually converge to a feasible suboptimal solution. The proposed algorithm framework can be also applied to computation offloading only mode. Additionally, we derive the closed-form optimal solution under the local computing only mode. Simulation results validate the convergence performance of the proposed algorithm. Moreover, the proposed partial offloading mode with the CE maximization scheme outperforms that with the computation bits (CB) maximization scheme with respect to CE, and it can achieve higher CE than the benchmark computing modes. Furthermore, the proposed MC-NOMA scheme can attain better CE performance than the conventional OFDMA scheme.