关键词： Mobile edge computing   task offloading   wireless power transfer   multi-agent   deep reinforcement learning  

摘要： Combined with wireless power transfer (WPT) technology, mobile edge computing can provide continuous energy supply and computing resources for mobile devices, and improve their battery life and business application scenarios. This article first designs the mobile edge computing (MEC) model of mobile devices with random mobility and hybrid access point (HAP) with data transmission and energy transmission. On this basis, the selection of target server and the amount of data offloading are taken as the learning objectives, and the task offloading strategy based on multi-agent deep reinforcement learning is constructed. Then combined with MADDPG algorithm and SAC algorithm, the problems of multi-agent environment instability and the difficulty of convergence are solved. The final experimental results show that the improved algorithm based on MADDPG and SAC has good stability and convergence. Compared with other algorithms, it has achieved good results in energy consumption, delay and task failure rate.

关键词： Machine type communication   mobile edge computing   computation offloading   deep reinforcement learning   energy efficiency  

摘要： Mobile edge computing (MEC) supports the internet of things (IoT) by leveraging computation offloading. It minimizes the delay and consequently reduces the energy consumption of the IoT devices. However, the consideration of static communication mode in most of the recent work, despite varying network dynamics and resource diversity, is the main limitation. An energy-efficient computation offloading method using deep reinforcement learning (DRL) is proposed. Both delay-tolerant and non-delay tolerant scenarios are considered using capillary machine type communication (MTC). Depending upon the type of service, an intelligent MTC edge server using DRL decides either process the incoming request at the MTC edge server or sends it to the cloud server. To control communication, we draft a markov decision problem (MDP). This minimizes the long-term power consumption of the system. The formulation of the optimization problem is considered under the constraint of computing power resources and delays. Simulation results delineate the significant performance gain of 12 & x0025;in computation offloading through the proposed DRL approach. The effectiveness and superiority of the proposed model are compared with other baselines and are demonstrated numerically.

关键词： Edge computing   Probability   Device-to-device communication   Throughput   Bipartite graph   Base stations   Internet   Mobile edge computing   capacity   time-varying   degree of files   transmission distance  

摘要： Capacity is a fundamental metric for mobile edge computing scenarios, where the system state plays an important role. Previous studies have mostly been based on the premise that the system state is stable. In reality, the network is dynamic and the system state changes with time. In this paper, we study the capacity of a mobile edge system in which users continuously join or leave the coverage of base station. We first change the problem of maximum network capacity into a minimum transmission distance problem. We observe that both the probability of the files being requested and the distance of the files transmission are related to the degree of files, i.e., the number of users who are interested in a file and request it with a certain probability. Then, we evaluate the degree of files in a time-varying situation, and calculate the probability of the files being requested and the transmission distance according to the degree of files. Finally, we calculate the capacity of the network under time-varying conditions. In the experimental section, we analyze the degree of files, the optimal copies number, and the change in network capacity over time. In addition, we compare the capacity in our system with classic studies. The experimental results verify the superiority of the proposed method.

关键词： V2X   data fusion   GPS   global positioning system   OBU   on board unit   RSU   road side unit   particle filter   positioning   vehicular network   MEC   mobile edge computing  

摘要： Precise positioning is a key issue for road vehicles in navigation, safety, and autonomous driving applications. While global positioning system (GPS) is widely accepted, it is still a challenge to achieve lane-level positioning. In this work, we consider the fusion of multi-sensory data using particle filter (PF), which is flexible in integrating different information in complex outdoor environments. We focus on three types of popular sensors: controller area network (CAN bus), GPS, and roadside camera. We propose a PF model that can adopt these types of sensory inputs for vehicle positioning. We show that in scenarios where vision sensory inputs are available, lane-level precision can be achieved. When there is no vision coverage, seamless localisation with reasonable precision can still be supported by GPS. Field trial results are presented to validate our model.

关键词： internet of things   IoT   power grid   renewable energy   optimisation strategy   mobile edge computing   MEC  

摘要： With deployment and application of the power grid internet of things (IoT), all the nodes need more and more resources which bring a great challenge to the power grid system. Mobile edge computing (MEC) and rational use of renewable energy in the grid are effective way to solve above problem of grid resource allocation. In this paper, we propose a MEC framework and an offloading policy that considers renewable energy. The model fully considers the stochasticity of renewable energy arrival randomness and task offloading. By analysing the renewable energy's dedicate character and adopt alternating direction method of multipliers (ADMM) algorithm, the method introduce a task computing offloading method based on renewable energy is proposed. The algorithm realises the optimisation strategy of the power grid internet node delay and grid energy consumption minimisation. The simulation results show that the proposed algorithm performance exceeds the two baseline computation and offloading strategies, which can effectively optimise the grid energy consumption of the nodes.

关键词： Internet of Vehicles   task offloading   heterogeneous wireless network architecture   mobile edge computing   combination auction   channel gain  

摘要： From a global perspective, an Internet of Vehicles task offloading solution based on mobile edge computing is proposed, which satisfies the application requirements (high reliability) strictly. The average time for completing a task can be minimized with the reasonable task offloading solution. Firstly, we model the wireless network, the transmission time and the movement of vehicles. Besides, heterogeneous wireless network architecture is adopted, data centers are deployed at Small-cell Base Stations, Macro-cell Base Stations and Internet. Then considering the limitedness, heterogeneity and task diversity of resources, we utilize matching model based on combination auction to design the offloading model. Furthermore, the multi-round sequential combination auction mechanism is proposed, which equals the matching problem to the multi-dimensional grouping knapsack problem and uses dynamic programming to get the optimal match. This solution is based on virtual machine technology and voltage scaling technology in the task execution time model. Moreover, the computing resources can be measured by CPU frequency. We propose an optimization problem for the shortest average task completion time with limited resources. Finally, the effects of these parameters (such as the number of tasks per unit time, the amount of data offloaded and the number of CPU cycles) on the task execution efficiency are analyzed and compared with other algorithms by simulation experiments. Compared to existing schemes, simulation results show that the proposed algorithm can reduce system overhead and shorten task execution time effectively.

关键词： Servers   Resource management   Task analysis   Delays   Computational modeling   Base stations   Edge computing   Vehicular networks   mobile edge computing   reinforcement learning   computation migration  

摘要： With the development of 5G technology, the requirements for data communication and computation in emerging 5G-enabled vehicular networks are becoming increasingly stringent. Computation-intensive or delay-sensitive tasks generated by vehicles need to be processed in real time. Mobile edge computing (MEC) is an appropriate solution. Wireless users or vehicles can offload computation tasks to the MEC server due to it has strong computation ability and is closer to the wireless users or vehicles. However, the communication and computation resources of the single MEC are not sufficient for executing the continuously generated computation-intensive or delay-sensitive tasks. We consider migrating computation tasks to other MEC servers to reduce the computation and communication pressure on current MEC server. In this article, we construct an MEC-based computation offloading framework for vehicular networks, which considers time-varying channel states and stochastically arriving computation tasks. To minimize the total cost of the proposed MEC framework, which consists of the delay cost, energy computation cost, and bandwidth cost, we propose a deep reinforcement learning-based computation migration and resource allocation (RLCMRA) scheme that requires no prior knowledge. The RLCMRA algorithm can obtain the optimal offloading and migration policy by adaptive learning to maximize the average cumulative reward (minimize the total cost). Extensive numerical results show that the proposed RLCMRA algorithm can adaptively learn the optimal policy and outperform four other baseline algorithms.

关键词： Mobile edge computing   massive MIMO   resource allocation   power control  

摘要： This paper designs a single-cell multi-user massive MIMO-MEC network. In order to ensure the fairness of users, a joint pilot transmission, data transmission and resource allocation model during the computation execution process with the goal of minimizing the maximum offload computing delay for all users is constructed. The resulted problem is non-convex and non-linear optimization, thus difficult to be solved optimally. To tackle this challenge, an improved fruit fly optimization algorithm (FOA) based on the external penalty function steepest descent algorithm (IFOA-PFSA) is proposed. The point obtained by the steepest descent algorithm based on the external penalty function has been employed as the initial point of the fruit fly optimization algorithm, which can greatly reduce the population size and the maximum number of iterations in the random search process of the traditional fruit fly optimization algorithm, reducing the algorithm complexity. Simulation results show that the proposed algorithm IFOA-PFSA has a smaller delay than the traditional FOA (TFOA) algorithm. The complexity of the proposed algorithm is also lower than the TFOA algorithm.

关键词： Edge intelligence   mobile edge computing   IoT   image distortion   latency and reliability   semi-definite relaxation  

摘要： The emerging time-critical Internet-of-Things (IoT) use cases, e.g., augmented reality (AR), virtual reality (VR), autonomous vehicle etc., on the one hand, involve computation intensive computer vision (CV) components in the services, on the other hand, the computation task should be completed within the stringent latency constraint, otherwise, the service reliability will deteriorate. The state-of-the-art work has shown that it is promising to tackle this challenge by offloading the computation tasks to the edge servers (ESs) using mobile edge computing (MEC). However, offloading tasks from local IoT devices to remote ESs could cause communication errors, thereby resulting in transmission failure even service timeout. The existing work mainly requires perfect transmission during task offloading at physical layer or transport layer. In fact, CV algorithms for, e.g., image classification and recognition, are able to tolerate certain level distortion of the input image to maintain the required inference accuracy. In this paper, we focus on the service reliability at application layer and study how feasible it is to improve the service reliability of the time-critical CV services in MEC system by allowing imperfect transmission. The service reliability is modeled by the transmission failure probability, service timeout probability and inference accuracy. The optimization goal is to maximize the service reliability, subject to the latency constraint. Due to the non-convexity, we solve this problem by the semi-definite relaxation based algorithm for a multi-user scenario. We evaluate the algorithm considering the practical scenarios, i.e., object detection with SSD and YOLOv2. The proposed algorithm achieves the performance closed to the exhaustive method but at a much lower complexity.

关键词： Hybrid energy harvesting   device-to-device communication   mobile edge computing   resource allocation  

摘要： Due to the limited battery capacity and computing capability of mobile users, the resource allocation strategy in device-to-device (D2D)-assisted edge computing system with hybrid energy harvesting is investigated in this paper. By employing magnetic induction-based wireless reverse charging technology, mobile user can supplement extra energy from nearby users when the energy harvested from ambient radio frequency sources is about to be exhausted. Moreover, mobile user can not only perform local computation, but also offload computing tasks to nearby users for auxiliary computation through D2D communication links or mobile edge computing (MEC) server under base station (BS) for edge computation. Due to the limited computing resources of MEC server, when the computing capability of the MEC server reaches the maximum value, an adjacent user under another nearby BS can be considered as a relay node. The computing tasks of the remaining users under the previous BS can be transferred to the MEC server with sufficient resources under another nearby BS by establishing D2D relay links. The objective of the resource allocation strategy is to maximize the energy efficiency under the constraints of computation delay and energy harvesting. The resource allocation problem is formulated as a mixed-integer nonlinear programming problem, which is not easy to obtain the optimal solution at low computational complexity. A suboptimal solution is obtained by adopting the quantum-behaved particle swarm optimization (QPSO) algorithm. Simulation results show that the performance of the proposed strategy is superior to other benchmark strategies, and QPSO algorithm can achieve higher energy efficiency than the standard particle swarm optimization algorithm.