关键词： Task analysis   Servers   Protocols   Uplink   Downlink   Energy consumption   Time-frequency analysis   Mobile edge computing   task offloading   energy efficient   resource allocation   convex optimization  

摘要： This paper investigates an asynchronous offloading protocol for mobile edge computing (MEC) systems where the computational tasks of mobile users are partially offloaded to a MEC server at a base station (BS). In the proposed asynchronous approach, upload, computation, and download of the tasks are performed in an asymmetric and non-orthogonal manner. A joint optimization problem of the transmit power, the offloading size, and the time-frequency resources is addressed to minimize the energy consumption of the overall system. By applying convex optimization techniques, the optimal solution for the asynchronous MEC offloading problem can be obtained. Numerical results demonstrate the effectiveness of the proposed asynchronous protocol over the conventional synchronous approach.

关键词： Mobile edge computing   5G   Network management   Network topology   Network inventory  

摘要： The Fifth-Generation (5G) mobile networks leverage virtualisation and softwarisation to reduce both capital and operating expenditures whilst benefiting from network programmability and flexibility for various use cases. Meanwhile, virtualisation and softwarisation introduce unprecedented complexity to network infrastructure topologies, which poses substantial challenges to network topology management. This paper proposes a novel architecture to enable network administrators to achieve real-time network discovery and spatial representation of the software data path, containers, virtual machines, and geographically distributed edges to help network troubleshooting and management tasks. Another important innovation is the flexibility to support a significant number of business roles by means of a novel method to perform data model alignment between different business roles. The architecture has been designed, prototyped and validated, and experiment results have demonstrated promising scalability of the proposed solution. (C) 2020 Elsevier B.V. All rights reserved.

关键词： energy consumption   mobile edge computing   response time   smart cities   task offloading  

摘要： The Mobile Edge Computing (MEC) technology can greatly facilitate the development of smart cities, as a result, MEC-based smart cities have attracted much attention from some global research communities. In such filed, there is always a non-negligible challenge, that is, efficient task offloading from mobile devices to edge servers. To this end, this letter proposes a task offloading optimization strategy to make a tradeoff between energy consumption and response time, including the modeling of two sub-objectives and a heuristic bi-objective optimization algorithm. The proposed strategy is simulated and the experimental results demonstrate that it is more efficient compared to the the-state-of-the-art strategy.

关键词： Mobile Edge Computing   energy-harvesting   load balancing   energy consumption   queue redundancy  

摘要： With the development of cloud computing, the Mobile Edge Computing has emerged and attracted widespread attentions. In this paper, we focus on the load balancing in MEC with energy harvesting. We first introduce the load balancing in MEC as a problem of minimizing both the energy consumption and queue redundancy. Thereafter, we adapt such a optimization problem to the Lyapunov algorithm and solve this optimization problem. Finally, extensive simulation results validate that the obtained strategy improves the capabilities of MEC systems.

关键词： Future internet architecture   Mobile edge computing   Cloud computing   Analytics   Adaptation  

摘要： Information-Centric Networking (ICN) enables in-network data management and communication between multiple parties by replicating data and activating interactions between decoupled senders and receivers. Existing data management and offloading schemes in ICNs primarily use the transport layer hence it becomes inefficient to actively develop and update the ICN standards because of continuously evolving heterogeneous future internet architectures such as mobile edge cloud computing (MECC) architectures. In this paper, we present an adaptive execution model for mobile data stream mining (MDSM) applications in MECC environments to enable device-centric adaptive data management and offloading. We designed the proposed execution model considering multiple factors of complexity such as volume and velocity of continuously streaming data, the selection of data fusion and data preprocessing methods, the choice of learning models, learning rates, learning modes, mobility, limited computational and memory resources in mobile devices, the high coupling between application components, and dependency over Internet connections. We integrated the proposed execution model with multiple MDSM applications mapping to a real-word use-case for activity detection using MECC as a future network architecture. We thoroughly evaluated the proposed execution model in terms of battery power consumption, memory utilization, makespan, accuracy, and the amount of data reduced during in-network communication. The comparison showed that our proposed adaptive execution model outperformed the static and dynamic execution models which were deployed in the same ICN architecture. (C) 2020 Elsevier B.V. All rights reserved.

关键词： mobile edge computing   vehicle-to-everything   deep reinforcement learning   double deep q network  

摘要： With the rapid development of vehicular networks, vehicle-to-everything (V2X) communications have huge number of tasks to be calculated, which brings challenges to the scarce network resources. Cloud servers can alleviate the terrible situation regarding the lack of computing abilities of vehicular user equipment (VUE), but the limited resources, the dynamic environment of vehicles, and the long distances between the cloud servers and VUE induce some potential issues, such as extra communication delay and energy consumption. Fortunately, mobile edge computing (MEC), a promising computing paradigm, can ameliorate the above problems by enhancing the computing abilities of VUE through allocating the computational resources to VUE. In this paper, we propose a joint optimization algorithm based on a deep reinforcement learning algorithm named the double deep Q network (double DQN) to minimize the cost constituted of energy consumption, the latency of computation, and communication with the proper policy. The proposed algorithm is more suitable for dynamic scenarios and requires low-latency vehicular scenarios in the real world. Compared with other reinforcement learning algorithms, the algorithm we proposed algorithm improve the performance in terms of convergence, defined cost, and speed by around 30%, 15%, and 17%.

关键词： Task analysis   Mobile handsets   Edge computing   Energy consumption   Resource management   Servers   Computational modeling   Unmanned aerial vehicles (UAVs)   mobile edge computing   tasks assignment   block successive upper-bound minimization (BSUM)  

摘要： Unmanned aerial vehicles (UAVs) have been deployed to enhance the network capacity and provide services to mobile users with or without infrastructure coverage. At the same time, we have observed the exponential growth in Internet of Things (IoTs) devices and applications. However, as IoT devices have limited computation capacity and battery lifetime, it is challenging to process data locally on the devices. To this end, in this letter, a UAV-aided mobile edge computing system is proposed. The problem to jointly minimize the energy consumption at the IoT devices and the UAVs during task execution is studied by optimizing the task offloading decision, resource allocation mechanism and UAV's trajectory while considering the communication and computation latency requirements. A non-convex structure of the formulated problem is revealed and shown to be challenging to solve. To address this challenge, a block successive upper-bound minimization (BSUM) algorithm is introduced. Finally, simulation results are provided to show the efficiency of our proposed algorithm.

关键词： partition offloading   deep neural networks   mobile edge computing   mobility management  

摘要： The latest results in Deep Neural Networks (DNNs) have greatly improved the accuracy and performance of a variety of intelligent applications. However, running such computation-intensive DNN-based applications on resource-constrained mobile devices definitely leads to long latency and huge energy consumption. The traditional way is performing DNNs in the central cloud, but it requires significant amounts of data to be transferred to the cloud over the wireless network and also results in long latency. To solve this problem, offloading partial DNN computation to edge clouds has been proposed, to realize the collaborative execution between mobile devices and edge clouds. In addition, the mobility of mobile devices is easily to cause the computation offloading failure. In this paper, we develop a mobility-included DNN partition offloading algorithm (MDPO) to adapt to user's mobility. The objective of MDPO is minimizing the total latency of completing a DNN job when the mobile user is moving. The MDPO algorithm is suitable for both DNNs with chain topology and graphic topology. We evaluate the performance of our proposed MDPO compared to local-only execution and edge-only execution, experiments show that MDPO significantly reduces the total latency and improves the performance of DNN, and MDPO can adjust well to different network conditions.

关键词： Delay deadline   Mobile edge computing   Mobility   Quality of service   Task migration  

摘要： Mobile edge computing (MEC) is envisioned to integrate cloud-like capabilities into the edge of networks for improving quality of service (QoS). This makes it possible for users with resource-limited devices to execute computation-intensive tasks by offloading them to MEC nodes. Extensive works have been done for MEC. However, few of them involve user mobility. Whether to migrate task dynamically cannot be ignored when taking QoS into account. In this paper, we try to optimize task migration with user mobility consideration, in which deadlines of tasks are also involved. The problem is proved to be NP-hard. To solve it, we analyze three variants of this problem and devise a group migration (GM) algorithm with known trajectories of users. Our goal is to maximize the number of tasks whose deadlines are guaranteed. Extensive experiments are carried out, and the results confirm that GM algorithm can achieve up 35%-75% performance improvement compared three other common heuristics.

关键词： Mobile edge computing   Computation offloading   Resource competition   Dynamic programming   Energy-efficient  

摘要： In mobile edge computing systems, the energy consumption and execution delay can be reduced dramatically by mobile edge computation offloading (MECO) . However, due to the limited computing capacity of edge cloud, an energy-efficient offloading strategy plays a significant role. In this paper, the offloading decision problem for multi-device edge computing systems based on time-division multiple access is studied. The scheduling of offloading devices at the edge cloud is considered when modelling the edge computing system. Then, the offloading decision problem is formulated as an energy consumption minimization problem with the constraint of latency tolerance. It is a mixed integer programming problem of NP-hardness. To address the problem, a Dynamic Programming-based Energy Saving Offloading (DPESO) algorithm is designed to obtain the offloading strategy including the offloading option, offloading sequence and transmission power. First, the MECO with infinite edge cloud capacity is solved by device classification and transmission power decision. Then, we sort and adjust the offloading devices to meet the latency tolerance for the MECO with finite edge cloud capacity. Finally, simulation results demonstrate that the DPESO algorithm achieves better energy efficiency than the baseline strategies and has good scalability.