关键词： Mobile edge computing  

摘要： Federated edge learning (FEEL) has drawn much attention as a privacy-preserving distributed learning framework for mobile edge networks. In this work, we investigate a novel semi-decentralized FEEL (SD-FEEL) architecture where multiple edge servers collaborate to incorporate more data from edge devices in training. Despite the low training latency enabled by fast edge aggregation, the device heterogeneity in computational resources deteriorates the efficiency. This paper proposes an asynchronous training algorithm for SD-FEEL to overcome this issue, where edge servers can independently set deadlines for the associated client nodes and trigger the model aggregation. To deal with different levels of staleness, we design a staleness-aware aggregation scheme and analyze its convergence performance. Simulation results demonstrate the effectiveness of our proposed algorithm in achieving faster convergence and better learning performance. Copyright © 2021, The Authors. All rights reserved.

关键词： Edge node cooperation   Forwarding group   Caching hit ratio   Mobile edge computing   ADMM  

摘要： In this study, we investigate the cooperation problem of edge nodes in mobile edge scenarios. We study two performance metrics: caching hit ratio and cooperation cost. We analyze the relationship of the two metrics for cooperative edge networks in which multiple edge nodes form a forwarding group to serve users. We construct a spanning tree to organize the forwarding group, and maximize the hit ratio subject to a budget constraint on cooperation cost. We first address the optimization problem using the method of Lagrangian multipliers and derive the optimal group parameter. Furthermore, a distributed optimization algorithm with alternating direction method of multipliers (ADMM) via constraint projection and variable splitting is proposed to achieve the desired goal. Finally, we build an edge computing simulation platform consisting of hundreds of edge nodes and users as a case study to verify the effectiveness and efficiency of our proposed method. Numerical results show that it significantly improves the caching hit ratio up to 60%, compared to the classic scheme.

关键词： Mobile Edge Computing   Popularity Prediction   Cooperative caching strategy   Branch and Bound algorithm  

摘要： Mobile edge computing effectively reduces the network delay of requests during data transmission, reduces the transmission burden of data traffic in the network, and improves the quality of mobile services. However, the introduction of mobile edge computing architecture increases the cost of the actual network infrastructure deployment, which increases the complexity of resource management. An edge network architecture based on software-defined network technology is proposed, in which the SDN controller with the entire network state can manage data transmission more intelligently to maximize the utilization of edge servers. In addition, to solve the problems of high request delay and high operating cost in current caching strategies based on video services, a video caching strategy for mobile edge computing is proposed. First, we analyze the user's request data and use the neural network model to predict the content of the user's subsequent time slice request and pre-cache the user's request. Then, improve the quality of user experience and choose the most appropriate edge node cache deployment plan. Finally, a video caching strategy for mobile edge computing with coordinated optimization of delay and energy consumption is proposed. The Branch and Bound algorithm is used to solve the optimization problem. Finally, we compared our algorithm with the LFU algorithm, PBC algorithm, COC algorithm, and R-LCCA algorithm. Experimental results show that the algorithm has a high cache hit rate, thereby reducing the cost of video providers and improving the quality of user experience. (c) 2021 Elsevier B.V. All rights reserved.

关键词： Mobile edge computing   Mobility   Edge service   Service recommendation   Quality of service(QoS)  

摘要： With the proliferation of smartphones and an increasing number of services provisioned by clouds, mobile edge computing (MEC) is emerging as a complementary technology of cloud computing. It could provide cloud resources and services by local mobile edge servers, which are normally nearby users. However, a significant challenge is aroused in MEC because of the mobility of users. User trajectory prediction technologies could be used to cope with this issue, which has already played important roles in service recommendation systems with MEC. Unfortunately, little attention and work have been given in service recommendation systems considering users mobility. Thus, in this paper, we propose a mobility-aware personalized service recommendation (MPSR) approach based on user trajectory and quality of service (QoS) predictions. In the proposed method, users trajectory is firstly discovered by a hybrid long-short memory network. Then, given users trajectories, service QoS is predicted, considering the similarity of different users and different edge servers. Finally, services are recommended by a center trajectory strategy through MPSR. Experimental results on a real dataset show that our proposed approach can outperform the traditional recommendation approaches in terms of accuracy in mobile edge computing.

关键词： Delays   Mobile handsets   Resource management   Energy consumption   Power demand   Task analysis   Cloud computing   Offloading   resource allocation   mobile cloud computing   mobile edge computing  

摘要： A promising technique to provide mobile applications with high computation resources is to offload the processing task to the cloud. Utilizing the abundant processing capabilities of the clouds, mobile edge computing enables mobile devices with limited batteries to run resource hungry applications and to save power. However, it is not always true that edge computing consumes less power compared to device computing. It may take more power for the mobile device to transmit a file to the cloud than running the task itself. This paper investigates the power minimization problem for the mobile devices by data offloading in multi-cell multi-user OFDMA mobile edge computing networks. We consider the maximum acceptable delay as QoS metric to be satisfied in our network. We formulate the problem as a mixed integer nonlinear problem which is converted into a convex form using D.C. approximation. To solve the converted optimization problem, we have proposed centralized and distributed algorithms for joint power allocation and channel assignment together with decision-making. Simulation results illustrate that by utilizing the proposed algorithms, considerable power savings can be achieved, e.g., about 60 percent for large bit stream size compared to local computing baseline.

关键词： IoT   LoraWAN   Mobile Edge Computing   Q-learning Internet of things   Reinforcement Learning   Simulation   Solar-powered   Task-offloading   Wireless power transmission  

摘要： Typical IoT(Internet of things) devices have a limited lifetime because they use a battery as an energy source. To fundamentally solve this problem, various environmental energy harvesting IoT is being researched around solar energy. Meanwhile, many studies are attempting to improve the quality of experience (QoE) in terms of the energy problem and delay time of IoT devices. Many of these studies on QoE improvement are approaching it as a cost minimization problem, especially an energy use minimization problem. However, such an energy cost minimization approach due to limited energy is not appropriate for energy harvesting IoT devices, which can continuously harvest energy. Since many energy-harvesting IoT devices can continuously harvest energy, we should approach in the direction of cost optimization that can utilize energy to the maximum rather than cost minimization approach. This study proposes a reinforcement learning-based dynamic task offloading scheme that can enhance user QoE while maximizing the utilization of harvested energy in an energy harvesting IoT edge computing environment. The proposed scheme modeled Q-learning using the information of dynamically changing amount of energy harvesting and the information of IoT devices and edge nodes. It can perform very simple, yet efficiently operates compared to complex models of the conventional minimization problem. © 2021, Korean Institute of Communications and Information Sciences. All rights reserved.

关键词： 5G   Internet of things(IoT)   Mobile edge computing   Feistel architecture   SCESW   GRC algorithm  

摘要： With the large-scale application of 5G in industrial production,the Internet of Things has become an important technology for various industries to achieve efficiency improvement and digital transformation with the help of the mobile edge *** the modern industry,the user often stores data collected by IoT devices in the cloud,but the data at the edge of the network involves a large of the sensitive information,which increases the risk of privacy *** order to address these two challenges,we propose a security strategy in the edge *** security strategy combines the Feistel architecture and short comparable encryption based on sliding window(SCESW).Compared to existing security strategies,our proposed security strategy guarantees its security while significantly reducing the computational *** our GRC algorithm can be successfully deployed on a hardware platform.

关键词： Mobile edge computing  

摘要： Over the past few years, we have witnessed an exponential increase in demand for seamless and continuous Internet access to online services - multimedia services, online business, gaming, scientific analysis, and so on. With the significant advances in computing and communication techniques of edge devices, the mobile edge computing (MEC) paradigm has emerged and has been demonstrated to be more feasible and efficient in providing wireless Internet access or computing in some scenarios compared to the traditional cloud computing platform. Furthermore, the promotion and popularization of unmanned aerial vehicles (UAVs) in recent years provides the promis of ubiquitous data sensing, leading to an era of aerial MEC. In this article, we first present four representative architectures for the UAV-based MEC system. Then we survey some recent research products in this area. In particular, we argue that the UAV can be regarded as a MEC unit and propose the paradigm of a multi-UAV-based MEC network. Furthermore, we employ software-defined networking to improve the network's scalability and controllability and present the framework of the software-defined UAV network. Based on the proposed framework, we introduce the concept of a 'smart function table' that can be applied to execute the network operation in the proposed framework. After evaluating our proposed framework, especially concerning the cooperative flying ability and control delay, several challenging issues and future research suggestions are presented to guide this direction. © 2018 IEEE.

关键词： Mobile edge computing   Task offloading   Optimization   Deep Q-learning  

摘要： In a mobile edge computing (MEC) network, mobile devices could selectively offload tasks to the edge server(s) to save time and energy. However, we should consider many dynamic factors in task offloading optimization, which increases the complexity of this problem. The traditional optimization approaches could require solving complex models to derive the optimal solution. This type of optimization problems are often NP-hard and will cause a considerable overhead on optimization. In contrast, a well-trained empirical model, such as an artificial neural network could be more efficient in decision making. In this research, considering the potential uneven spatial distribution of mobile devices in a MEC network with multiple wireless edge gateways, we allow an edge gateway to offload tasks to a nearby edge gateway further. We propose a deep Q-learning-based joint optimization approach for both device-level and edge-level task offloading. We also design a centralized mathematical programming solution for exploring the optimal trade-off performance. Simulation results show that the proposed approach achieves a satisfactory task delay performance and a better trade-off between the task delay and the energy consumption on tasks.

关键词： intelligent transportation system   short-term traffic state prediction   V2X communication   mobile edge computing   neural networks  

摘要： Real-time and reliable short-term traffic state prediction is one of the most critical technologies in intelligent transportation systems (ITS). However, the traffic state is generally perceived by single sensor in existing studies, which is difficult to satisfy the requirement of real-time prediction in complex traffic networks. In this paper, a short-term traffic prediction model based on complex neural network is proposed under the environment of vehicle-to-everything (V2X) communication systems. Firstly, a traffic perception system of multi-source sensors based on V2X communication is proposed and designed. A mobile edge computing (MEC)-assisted architecture is then introduced in a V2X network to facilitate perceptual and computational abilities of the system. Moreover, the graph convolutional network (GCN), the gated recurrent unit (GRU), and the soft-attention mechanism are combined to extract spatiotemporal features of traffic state and integrate them for future prediction. Finally, an intelligent roadside test platform is demonstrated for perception and computation of real-time traffic state. The comparison experiments show that the proposed method can significantly improve the prediction accuracy by comparing with the existing neural network models, which consider one of the spatiotemporal features. In particular, for comparison results of the traffic state prediction and the error value of root mean squared error (RMSE) is reduced by 39.53%, which is the greatest reduction in error occurrences by comparing with the GCN and GRU models in 5, 10, 15 and 30 min respectively.