关键词： Task offloading   Mobile edge computing   Approximation algorithm  

摘要： With development well underway, 5G is envisioned as an enabler of lighting fast mobile services, such as virtual reality, augmented reality, live video analytics, and etc. In particular, multi-cell Mobile Edge Clouds (MEC) with 5G base stations endowed with computing capability are able to promote the Quality of Services (QoS) of mobile users by executing tasks in the edge cloud. Due to the varying 5G network conditions and limited computation capacity of each base station in the multi-cell MEC, as well as the stringent QoS requirements, a fundamental and challenging problem is how to offload user tasks to the edge cloud, such that the energy consumption of mobile devices is minimized. In this paper, we first formulate the offline and online location-aware mobile task offloading problems in a multi-cell MEC. For the offline location-aware mobile task offloading problem, we then develop an exact solution and an approximation algorithm with an approximation ratio. For the online problem, we thirdly propose a novel deep reinforcement learning-based offloading algorithm for mobile users to obtain the optimal offloading policy. We finally conduct extensive experiments by simulations to evaluate the proposed algorithms against existing benchmarks. The experimental results show that the proposed algorithms are promising and outperform the benchmark algorithms by significantly reducing energy cost of mobile devices and delays experienced by mobile users.

关键词： Mobile edge computing  

摘要： In this paper, a joint task, spectrum, and transmit power allocation problem is investigated for a wireless network in which the base stations (BSs) are equipped with mobile edge computing (MEC) servers to jointly provide computational and communication services to users. Each user can request one computational task from three types of computational tasks. Since the data size of each computational task is different, as the requested computational task varies, the BSs must adjust their resource (subcarrier and transmit power) and task allocation schemes to effectively serve the users. This problem is formulated as an optimization problem whose goal is to minimize the maximal computational and transmission delay among all users. A multi-stack reinforcement learning (RL) algorithm is developed to solve this problem. Using the proposed algorithm, each BS can record the historical resource allocation schemes and users’ information in its multiple stacks to avoid learning the same resource allocation scheme and users’ states, thus improving the convergence speed and learning efficiency. Simulation results illustrate that the proposed algorithm can reduce the number of iterations needed for convergence and the maximal delay among all users by up to 18% and 11.1% compared to the standard Q-learning algorithm. Copyright © 2020, The Authors. All rights reserved.

关键词： Mobile edge computing  

摘要： Mobile edge computing (MEC) enables to provide relatively rich computing resources in close proximity to mobile users, which enables resource-limited mobile devices to offload workloads to nearby edge servers, and thereby greatly reducing the processing delay of various mobile applications and the energy consumption of mobile devices. Despite its advantages, when a large number of mobile users simultaneously offloads their computation tasks to an edge server, due to the limited computation and communication resources of edge server, inefficiency resource allocation will not make full use of the limited resource and cause waste of resource, resulting in low system performance (the weighted sum of the number of processed tasks, the number of punished tasks, and the number of dropped tasks). Therefore, it is a challenging problem to effectively allocate the computing and communication resources to multiple mobile users. To cope with this problem, we propose a performance-aware resource allocation (PARA) scheme, the goal of which is to maximize the long-term system performance. More specifically, we first build the multiuser resource allocation architecture for computing workloads and transmitting result data to mobile devices. Then, we formulate the multiuser resource allocation problem as a Markova Decision Process (MDP). To achieve this problem, a performance-aware resource allocation (PARA) scheme based on a deep deterministic policy gradient (DDPG) is adopted to derive optimal resource allocation policy. Finally, extensive simulation experiments demonstrate the effectiveness of the PARA scheme.

关键词： Mobile edge computing   Intelligent resource allocation   Deep reinforcement learning   Federated learning  

摘要： With the emergence of a large number of computation-intensive and time-sensitive applications, smart terminal devices with limited resources can only run the model training part of most intelligent applications in the cloud, so a large amount of training data needs to be uploaded to the cloud. This is an important cause of core network communication congestion and poor Quality-of-Experience (QoE) of user. As an important extension and supplement of cloud computing, Mobile Edge Computing (MEC) sinks computing and storage resources from the cloud to the vicinity of User Mobile Devices (UMDs), greatly reducing service latency and alleviating the burden on core networks. However, due to the high cost of edge servers deployment and maintenance, MEC also has the problems of limited network resources and computing resources, and the edge network environment is complex and mutative. Therefore, how to reasonably allocate network resources and computing resources in a changeable MEC environment has become a great aporia. To combat this issue, this paper proposes an intelligent resource allocation model "DRL + FL". Based on this model, an intelligent resource allocation algorithm DDQN-RA based on the emerging DRL algorithm framework DDQN is designed to adaptively allocate network and computing resources. At the same time, the model integrates the FL framework with the mobile edge system to train DRL agents in a distributed way. This model can well solve the problems of uploading large amounts of training data via wireless channels, Non-IID and unbalance of training data when training DRL agents, restrictions on communication conditions, and data privacy. Experimental results show that the proposed "DRL + FL" model is superior to the traditional resource allocation algorithms SDR and LOBO and the intelligent resource allocation algorithm DRLRA in three aspects: minimizing the average energy consumption of the system, minimizing the average service delay, and balancing resource

关键词： Mobile edge computing   Non-orthogonal multiple access   Imperfect channel state information   Offloading outage probability  

摘要： In this paper, a non-orthogonal multiple access (NOMA)-based mobile edge computing (MEC) system is proposed, where paired users (themth user and thenth user) offload tasks to the MEC server under imperfect channel state information (ipCSI) condition. To evaluate the effect on performance under ipCSI condition in NOMA-MEC system, this study derives new exact and asymptotic expressions of the offloading outage probability for two users under ipCSI and perfect channel state information (pCSI) conditions. On the basis of the theoretical derivation, the diversity orders of thenth user under ipCSI and pCSI conditions arezeroandn, respectively, while those for the distant user are allzeros. In addition, we also investigate the system throughput and energy efficiency of NOMA-MEC in delay-limited transmission. Numerical results show that (1) the offloading outage behaviors of NOMA-MEC are better than those of time division multiple address (TDMA), (2) the offloading outage performance of paired users under ipCSI is worse than that under pCSI, and (3) the NOMA-MEC system achieves higher throughput and energy efficiency than does the TDMA-MEC system.

关键词： Mobile edge computing   Vocal music   System resources   Resource allocation   Nash equilibrium  

摘要： How to allocate vocal teaching system resources plays an important role in improving the performance of vocal teaching systems. With the development of mobile edge computing, the calculation and storage of resource data can be adapted to the operating needs of vocal teaching systems. This study proposes a method of system resource allocation based on power iteration and sets the throughput of the unloading process as an objective function and realizes the optimal allocation of normal power by iterative optimization. At the same time, in view of the low energy efficiency and resource utilization of edge servers, a heterogeneous network based on edge servers is proposed to find the optimal strategy based on ensuring the Nash equilibrium point. In addition, in order to verify the performance of the algorithm in this paper, a comparative experiment is performed by designing simulation experiments. The results show that the method proposed in this paper has certain effects and strong practicability, which can provide theoretical references for subsequent related research.

关键词： Flow classification   Graphics Processing Unit (GPU)   GPU cluster   Mobile edge computing   Software Defined Networking   Software Defined Mobile Edge Computing (SDMEC)  

摘要： Mobile Edge Computing (MEC) provides different storage and computing capabilities within the access range of mobile devices. This moderates the burden of offloading compute/storage-intensive processes of the mobile devices to the centralized cloud data centers. As a result, the network latency is reduced and the quality of service provided for the mobile end users is improved. Different applications benefit from the large-scale deployments of MEC servers. However, the considerable complexity of managing large scale deployments of the sheer number of applications for the millions of mobile devices is a challenge. Recently, Software Defined Networking (SDN) is leveraged to resolve the problem by providing unified and programmable interfaces for managing network devices. Most of the current SDN packet processing services are tightly dependent on the packet classification service. This primary service classifies any incoming packet based on matching a set of specific fields of its header against a flow table. Acceleration of this basic process considerably increases the performance of the SDN-based MEC. In this paper, the hierarchical tree algorithm, which is a packet classification method, is parallelized using popular platforms on a cluster of Graphics Processing Units (GPUs), a cluster of Central Processing Units (CPUs), and a hybrid cluster. The best scenario for the parallel implementation of this algorithm on the CPU cluster is that which combines OpenMP and MPI. In this case, the throughput of the classifier is 4.2 million packets per second (MPPS). On the GPU cluster, two different scenarios have been used. In the first scenario, the global memory is used to store the rules and the Hierarchical-trie of the classifier while in the second scenario we break the filter set in a way that the resulting Hierarchical-trie of each subset could be stored in the shared memory of GPU. According to the results, although the first GPU cluster scenario achieves a throughput of

关键词： Task analysis   Servers   Delays   Wireless communication   Resource management   Energy consumption   Optimization   Mobile edge computing   service caching   computation offloading   resource allocation  

摘要： In mobile edge computing (MEC) systems, edge service caching refers to pre-storing the necessary programs for executing computation tasks at MEC servers. Service caching effectively reduces the real-time delay/bandwidth cost on acquiring and initializing service applications when computation tasks are offloaded to the MEC servers. The limited caching space at resource-constrained edge servers calls for careful design of caching placement to determine which programs to cache over time. This is in general a complicated problem that highly correlates to the computation offloading decisions of computation tasks, i.e., whether or not to offload a task for edge execution. In this paper, we consider a single edge server that assists a mobile user (MU) in executing a sequence of computation tasks. In particular, the MU can upload and run its customized programs at the edge server, while the server can selectively cache the previously generated programs for future reuse. To minimize the computation delay and energy consumption of the MU, we formulate a mixed integer non-linear programming (MINLP) that jointly optimizes the service caching placement, computation offloading decisions, and system resource allocation (e.g., CPU processing frequency and transmit power of MU). To tackle the problem, we first derive the closed-form expressions of the optimal resource allocation solutions, and subsequently transform the MINLP into an equivalent pure 0-1 integer linear programming (ILP) that is much simpler to solve. To further reduce the complexity in solving the ILP, we exploit the underlying structures of caching causality and task dependency models, and accordingly devise a reduced-complexity alternating minimization technique to update the caching placement and offloading decision alternately. Extensive simulations show that the proposed joint optimization techniques achieve substantial resource savings of the MU compared to other representative benchmark methods considered.

关键词： 5G   edge network   deep learning   reinforcement learning   caching   task offloading   mobile computing   edge computing   mobile edge computing   cloud computing   network function virtualization   slicing   5G network standardization  

摘要： Mobile edge computing (MEC) within 5G networks brings the power of cloud computing, storage, and analysis closer to the end user. The increased speeds and reduced delay enable novel applications such as connected vehicles, large-scale IoT, video streaming, and industry robotics. Machine Learning (ML) is leveraged within mobile edge computing to predict changes in demand based on cultural events, natural disasters, or daily commute patterns, and it prepares the network by automatically scaling up network resources as needed. Together, mobile edge computing and ML enable seamless automation of network management to reduce operational costs and enhance user experience. In this paper, we discuss the state of the art for ML within mobile edge computing and the advances needed in automating adaptive resource allocation, mobility modeling, security, and energy efficiency for 5G networks.

关键词： Task analysis   Servers   Delays   Wireless communication   Computational modeling   Optimization   Edge computing   Internet of Things (IoT)   network association   event-triggered   Markov decision process (MDP)   Lyapunov optimization   distributed computing   machine learning   mobile edge computing  

摘要： Ultra-low latency communication for mobile intelligent machines, such as autonomous vehicles and robots, is a central technology in Internet of Things (IoT) to achieve system reliability. Proactive network association and communication has been suggested to achieve ultra-low latency under the assistance of mobile edge computing. Highly dynamic and stochastic nature of IoT mobile machines suggests applying machine learning methodology to effectively enhance the proactive network association. In this paper, an online proactive network association is proposed for this distributed computing and networking scenario, in order to minimize the average task delay subject to time-average energy consumption. We first formulate an event-triggered delay model for mobility-aware anticipatory network association mechanism that takes future possible handovers into account. Based on the Markov decision processes (MDP) and Lyapunov optimization, a two-stage online decision algorithm for proactive network association is innovated for individual mobile machine without the statistical knowledge of random events that may lack of enough prior data. Theoretical analysis proves that the delay performance of proposed algorithm attains asymptotic optimality within the bounded deviation. Furthermore, an asynchronous online distributed association decision algorithm based on the nonlinear problem transformation is proposed to support more general scenarios of multi-machine event-triggered associations. Simulations verify the effectiveness of the proposed methodology.