关键词： Collaboration   Prediction algorithms   Servers   Optimization   Predictive models   Base stations   Quality of experience   Proactive content caching   mobile edge computing   distributed learning   collaborative filtering   neural network  

摘要： Caching popular content at the network edge, such as roadside units (RSUs), is a promising solution that enhances the user's quality-of-experience (QoE) and reduces network traffic. In this regard, the most challenging issue is to correctly predict the future popularity of contents and effectively store them in the cache of edge nodes. Thus, in this paper, we propose a distributed proactive caching scheme at the edge to optimize the content retrieval cost and improve the QoE of the mobile users. This proactive content caching scheme, namely Distributed Collaborative Learning (DCoL), is a non-parametric content popularity prediction mechanism in a distributed setting. Next, we show the advantage of DCoL as two folds: (i) it leverages distributed content popularity information to develop local content caching strategy, and (ii) it exploits the regional database using the long short-term memory (LSTM)-based prediction model to capture the dependency between requested contents. Simulation results using real datasets demonstrate that our scheme yields 8.9% and 18% gains, respectively, in terms of the cache hit efficiency and content retrieval cost, compared with a competitive centralized baseline, and outperforms other traditional caching strategies.

关键词： Mobile edge computing  

摘要： Multi-access/mobile edge computing (MEC) is a structural design for enabling cloud computing platform at the edge of mobile network, so as to reduce network congestion, improves fast response, and optimization of mobile resources to compute complex applications. MEC provides a distributed computing environment for its applications. In MEC, all the computational reserves will be brought to the base location of the mobile networks. It has the ability to store and process the contents at physical proximity to the mobile subscribers to assist delay sensitivity and contextaware applications. Generally, the packet forwarding and filtering will be operated by the traditional edge network. At present, applications’ online computa-tions and storage are done at remote servers, and those servers are placed far away from the users. New technologies are emerged to shift all the available resources to a edge from a cloud. Radio access network (RAN) could able to do the computational off-loading to the edge of the mobile network. So, it is mandatory to have a load balancer at the edge to distribute all the job to all the available processors without any congestion or delay. A routing specifies how to route the traffic between each origin-destination pair across a network. The traffic sharing is applied in a routing and allocating process to enhance the survivability of a network. The experimental results show the proposed algorithm works better than the existing dynamic load balancing algorithms. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021.

关键词： mobile edge computing   user cooperation   wireless power transfer   energy consumption   resource management  

摘要： Mobile edge computing(MEC) meets the high-bandwidth and ultra-low-latency requirements of mobile networks and improves communication reliability by reducing the networks' load. In this study, we study a wireless powered MEC system, which includes an access point(AP) and two mobile devices. The endurance of mobile devices can be effectively enhanced using the wireless power transfer technology and user cooperation. However, during user cooperation, a closer user assists a farther user with forwarding tasks by sharing the same bandwidth, which causes interference. To overcome the interference issue, the closer user can instead assist other users in forwarding tasks by using a different bandwidth. We aim to minimize the power of the AP by joint power and bandwidth optimization while satisfying the delay and users' computational tasks. Simulation results show that the proposed user collaboration method can overcome user interference,reduce energy consumption, and improve user performance compared with the baseline policies.

关键词： Location awareness   Servers   Task analysis   Energy consumption   Computational modeling   Cloud computing   Batteries   Localization   computation offloading   game theory   latency   mobile edge computing  

摘要： Indoor localization has become one of the fundamental services that is required in a diverse set of applications these days, such as patient monitoring and smart parking. Highly accurate localization techniques impose high latency and high energy consumption on the underlying application system. Thus, for such indoor location-based application, offloading the computation of the localization process to a remote server with high resource capability has been recently introduced as an avenue to address such a challenge. In this paper, a computation offloading problem is formulated to find the optimal decision with regard to the operation of the localization process. This decision includes: a) Where to compute the localization task, either locally on the end device or on the edge server or on the cloud server, b) Which localization technique should be used, and finally, c) Which transmission technology is recommended to be chosen in combination with the localization technique. All these decisions are constrained by the device, and the servers resource capabilities load. They are also constrained by the fact that the localization algorithm has to satisfy a certain application QoS requirement. Within such context, three algorithms are proposed for task offload decision making. First, the Indoor Localization Latency Optimal Offloading algorithm, which finds the optimal offloading decision that minimizes the total latency of the system and is considered a benchmark for the other algorithms. Second, Indoor Localization Latency Centralized Offloading algorithm that finds a sub optimal solution with lower complexity. Third, Indoor Localization Latency Game-Theoretic Offloading decentralized algorithm that converges after finite improvement steps and achieves Nash equilibrium. Altogether, the paper finds the optimum localization strategy for all users with the minimum latency under mobile edge computing environment.

关键词： Task scheduling   mobile edge computing   cloud computing   QoS-aware  

摘要： Since the limitations of the user equipment, they do not have enough computing power to process large amounts of data. At the same time, it's unbearable for users to spend a long time uploading data to the remote cloud center. In order to solve these problems above, the concept of mobile edge computing(MEC) is proposed. The computing and storage resources are placed close to the end equipment, reducing the transmission delay. MEC can meet the high real-time requirements of the user equipment. In the real-time face recognition application scenario, a three-layer hierarchy MS-CE is proposed for the shortcoming of the traditional centralized cloud center. And the distributed MEC servers are utilized at the MEC layer to provide parallel computing capabilities. Aiming at the problem of how to perform task scheduling in a geographically distributed MEC server, we propose a task scheduling based queue algorithm(TSBQ), which considers the data transmission delay and server load, and carries out a reasonable task allocation policy. We evaluate the MS-CE and TSBQ through simulation experiments. We can find the MS-CE architecture are better than others and TSBQ is more effective than Corral and Greedy.

关键词： mobile blockchain   deep reinforcement learning   mobile edge computing   power allocation   bandwidth allocation  

摘要： In order to protect the privacy and data security of mobile devices during the transactions in the industrial Internet of Things(IIoT), we propose a mobile edge computing(MEC)-based mobile blockchain framework by considering the limited bandwidth and computing power of small base stations(SBSs). First,we formulate a joint bandwidth and computing resource allocation problem to maximize the long-term utility of all mobile devices, and take into account the mobility of devices as well as the blockchain throughput. We decompose the formulated problem into two subproblems to decrease the dimension of action space. Then,we propose a deep reinforcement learning additional particle swarm optimization(DRPO) algorithm to solve the two subproblems, in which a particle swarm optimization algorithm is leveraged to avoid the unnecessary search of a deep deterministic policy gradient approach. Simulation results demonstrate the effectiveness of our method from various aspects.

关键词： task scheduling   crowd sensing   MEC   mobile edge computing   task offloading   task shifting   Lyapunov optimisation  

摘要： Crowd sensing effectively solves the dilemma of massive data collection faced by most data-driven applications. Recently, mobile edge computing (MEC) is proposed to extend the frontier of cloud to the network edge so that it is quite suitable to integrate MEC with current crowd sensing systems. In this paper, we focus on the basic problem of task scheduling in such systems. The problem discussed here has some unique challenges, e.g., edge devices are not dedicated to perform sensing tasks, task scheduling on edge devices and edge servers are highly coupled, and it is hard to achieve long-term objectives. To this end, we first present a workflow framework that captures the unique execution logic of sensing tasks. Then we propose a staged scheme to decouple the original scheduling problem. Moreover, we leverage Lyapunov optimisation technique to achieve long-term objective. The experiment results verify the effectiveness and efficiency of our proposed algorithm.

关键词： Array signal processing   Computer architecture   deep learning   Deep learning   energy partition   Inference algorithms   Mobile edge computing   Optimization   receive beamforming   reconfigurable intelligent surface   Resource management   Task analysis  

摘要： In this paper, we explore optimization-based and data-driven solutions in a reconfigurable intelligent surface (RIS)-aided multi-user mobile edge computing (MEC) system, where the user equipment (UEs) can partially offload their computation tasks to the access point (AP). We aim at maximizing the total completed task-input bits (TCTB) of all UEs with limited energy budgets during a given time slot, through jointly optimizing the RIS reflecting coefficients, the AP’s receive beamforming vectors, and the UEs’ energy partition strategies for local computing and offloading. A three-step block coordinate descending (BCD) algorithm is first proposed to effectively solve the non-convex TCTB maximization problem with guaranteed convergence. In order to reduce the computational complexity and facilitate lightweight online implementation of the optimization algorithm, we further construct two deep learning architectures. The first one takes channel state information (CSI) as input, while the second one exploits the UEs’ locations only for online inference. The two data-driven approaches are trained using data samples generated by the BCD algorithm via supervised learning. Our simulation results reveal a close match between the performance of the optimization-based BCD algorithm and the low-complexity learning-based architectures, all with superior performance to existing schemes in both cases with perfect and imperfect input features. Importantly, the location-only deep learning method is shown to offer a particularly practical and robust solution alleviating the need for CSI estimation and feedback when line-of-sight (LoS) direct links exist between UEs and the AP. IEEE

关键词： green energy   mobile edge computing   MEC   solar panel placement   mobile cloud computing  

摘要： Mobile edge computing (MEC) enhances smart phone computational and storage capabilities by offloading tasks from mobile phones to the edge of the network where servers are placed near base stations. Since data centres consume electricity more and more each year, we consider locating renewable energy sources such as solar panels near each MEC. We tackle these challenges by strategically placing solar panels near MECs to minimise the energy costs. We formulate and solve the optimisation problem and present two algorithms to solve the optimisation problem. Moreover, we analyse the workload of each MEC during past years, and we present an algorithm to predict future workload of different MECs. We evaluate our solution using real Google traces collected over a 29-day period from a Google cluster. Simulation results show that with the help of our optimisation algorithms, we can effectively reduce the electricity costs and increase the green energy consumption.

关键词： Computational modeling   Training   Data models   Servers   Adaptation models   Data privacy   Collaborative work   Client selection   federated learning   mobile edge computing   reinforcement learning  

摘要： With data increasingly collected by end devices and the number of devices is growing rapidly in which data source mainly located outside the cloud today. To guarantee data privacy and remain data on client devices, federated learning (FL) has been proposed. In FL, end devices train a local model with their data and send the model parameters rather than raw data to server for aggregating a new global model. However, due to the limited wireless bandwidth and energy of mobile devices, it is not practical for FL to perform model updating and aggregation on all participating devices in parallel. And it is difficulty for FL server to select apposite clients to take part in model training which is important to save energy and reduce latency. In this paper, we establish a novel mobile edge computing (MEC) system for FL and propose an experience-driven control algorithm that adaptively chooses client devices to participate in each round of FL. Adaptive client selection mechanism in MEC can be modeled as a Markov Decision Process in which we do not need any prior knowledge of the environment. We then propose a client selection scheme based on reinforcement learning that learns to select a subset of devices in each communication round to minimize energy consumption and training delay that encourages the increase number of client devices to participate in model updating. The experimental results show that the unit of energy required in FL can be reduced by up to 50% and training delay required can be reduced by up to 20.70% compared to the other static algorithms. Finally, we demonstrate the scalability of MEC system with different tasks and the influence of different non independent and identically distributed (non-IID) settings.