关键词： computation-communication tradeoff   Computational modeling   Data models   Distributed machine learning   distributed optimization   Learning systems   Machine learning   mobile edge computing   Performance evaluation   Servers   timely computing   Wireless communication  

摘要： We consider distributed machine learning at the wireless edge, where a parameter server builds a global model with the help of multiple wireless edge devices that perform computations on local dataset partitions. Edge devices transmit the result of their computations (updates of current global model) to the server using a fixed rate and orthogonal multiple access over an error prone wireless channel. In case of a transmission error, the undelivered packet is retransmitted until successfully decoded at the receiver. Leveraging on the fundamental tradeoff between computation and communication in distributed systems, our aim is to derive how many edge devices are needed to minimize the average completion time while guaranteeing convergence. We provide upper and lower bounds for the average completion and we find a necessary condition for adding edge devices in two asymptotic regimes, namely the large dataset and the high accuracy regime. Conducted experiments on real datasets and numerical results confirm our analysis and substantiate our claim that the number of edge devices should be carefully selected for timely distributed edge learning. IEEE

关键词： Mobile handsets   Task analysis   Servers   Edge computing   Resource management   Optimization   Cloud computing   Assisted mechanism   computation offloading scheme   mobile edge computing   resource allocation  

摘要： Mobile edge computing (MEC) is a promising paradigm for providing computing and storage capabilities in close proximity to mobile devices. To solve the scenario in which massive mobile devices have tasks to be processed at the same time, this paper proposes an assisted mechanism for the MEC system. When the primary MEC server is unable to meet the delay requirements of the mobile devices within its coverage area, a portion of the tasks can be offloaded to secondary MEC servers to obtain extra resources for processing. This MEC framework effectively reduces the computing and communication burden of the primary MEC server and improves the resource utilization of the secondary MEC servers. To maximize the system offloading utility in terms of latency, we formulated an optimization problem that jointly optimizes the task assignment, computing resource allocation and offloading decision of all mobile devices. Since the formulated problem is a mixed integer nonlinear problem, we use the decomposition method to convert the optimization problem into several subproblems. In addition, a heuristic algorithm based on the priorities of mobile devices and the MEC servers is proposed to obtain the suboptimal device offloading strategy. The numerical results show that the assisted mechanism can effectively reduce system latency and improve system reliability. In addition, the performance of our proposed algorithm is close to the optimal solution.

关键词： Task analysis   Neural networks   Approximation algorithms   Computational modeling   Energy consumption   Computer architecture   Servers   Mobile edge computing   multidimensional multiple knapsack problem   multi-pointer networks   reinforcement learning  

摘要： Computation offloading is an efficient approach to reduce the energy consumption of a mobile device (MD). In this paper, we consider the multi-user offloading problem for mobile edge computing (MEC) in a multi-server environment. Its aim is to minimize the total energy consumption of MDs. This problem has been proven to be NP-hard. We formulate the problem as a multidimensional multiple knapsack (MMKP) problem with constraints, and propose a neural network architecture called Multi-Pointer networks (Mptr-Net) to solve the problem. We train Mptr-Net based on the reinforcement learning method, and design an algorithm to search for feasible solutions that meet the constraints. The simulation results show that the probability of a Mptr-Net obtaining an optimal solution can exceed 98%, which is approximately 25% more than that of a baseline heuristic algorithm. Additionally, the time needed to solve the problem by our neural network is stable compared with that of a mathematical programming solver named or-tools.

关键词： Mobile edge computing   computation offloading   VANETs   link correlation   cost optimization  

摘要： Mobile edge computing (MEC) is a suitable solution to improve computational capabilities in vehicular environments, and the computation offloading plays a critical role in MEC. After the computation task generates, in traditional offloading strategies, vehicles offload data only when they have entered the communication range of the MEC server, which greatly increases the time cost and cannot satisfy the delay requirement of certain applications in many cases. To deploy a large number of MEC servers for full coverage of roads is not a realistic solution due to its high construction cost. In this paper, we make full use of the multi-hop vehicular ad hoc networks (VANETs) to assist in computation offloading of vehicles. In the real-time traffic environment, based on the link correlation theory in VANETs, the reliability model of multi-hop routing path is built. Meeting the delay constraints, the optimal offloading strategy with the lowest relaying and computing cost is executed by the binary search algorithm. The simulation results show that the proposed multi-hop VANETs-assisted offloading strategy (MHVA) shows better performance in offloading delay and cost than existing typical strategies under various environments.

关键词： Mobile edge computing   massive multiple input multiple output   jamming attack   eavesdropping   energy consumption  

摘要： Under the double threat of malicious user jamming attacks and eavesdropping, mobile edge computing (MEC) systems will have some information security issues, such as uplink offloading is hindered, user information leakage, and low system energy utilization. To address this problem, we propose to use physical layer security technology to prevent eavesdropping, and use massive multiple input multiple output (mMIMO) technology to reduce interference and improve system energy efficiency. In addition, under the conditions of limited delay, the user's total energy consumption is minimized by jointly optimizing the user's offloading data bits, transmission power, and offloading rate. For this non-convex multi-objective optimization problem, which can be decomposed into multiple sub-optimal sub-problems. First, the Lagrange multiplier method is used to obtain a closed expression of the user's offloading data bits, and then the iterative algorithm is used to obtain the user's optimal transmit power and maximum offloading rate. The simulation results show that, under the premise of ensuring information security, the system's total energy efficiency of antennas (M = 128) is improved 30% compared with antennas (M = 4) when the amount of data reaches 5 x 10(5) bits.

关键词： Cloud computing   Servers   Edge computing   Mobile handsets   5G mobile communication   Task analysis   Internet of Things   Mobile edge computing   cloudlet   scalability   communication   computation resource management   ML   artificial intelligence  

摘要： Mobile Edge Computing (MEC) is considered an essential future service for the implementation of 5G networks and the Internet of Things, as it is the best method of delivering computation and communication resources to mobile devices. It is based on the connection of the users to servers located on the edge of the network, which is especially relevant for real-time applications that demand minimal latency. In order to guarantee a resource-efficient MEC (which, for example, could mean improved Quality of Service for users or lower costs for service providers), it is important to consider certain aspects of the service model, such as where to offload the tasks generated by the devices, how many resources to allocate to each user (specially in the wired or wireless device-server communication) and how to handle inter-server communication. However, in the MEC scenarios with many and varied users, servers and applications, these problems are characterized by parameters with exceedingly high levels of dimensionality, resulting in too much data to be processed and complicating the task of finding efficient configurations. This will be particularly troublesome when 5G networks and Internet of Things roll out, with their massive amounts of devices. To address this concern, the best solution is to utilize Machine Learning (ML) algorithms, which enable the computer to draw conclusions and make predictions based on existing data without human supervision, leading to quick near-optimal solutions even in problems with high dimensionality. Indeed, in scenarios with too much data and too many parameters, ML algorithms are often the only feasible alternative. In this paper, a comprehensive survey on the use of ML in MEC systems is provided, offering an insight into the current progress of this research area. Furthermore, helpful guidance is supplied by pointing out which MEC challenges can be solved by ML solutions, what are the current trending algorithms in frontier ML research and

关键词： Task analysis   Servers   Delays   Cost function   Heuristic algorithms   Roads   V2X   vehicular networks   integer linear programming   intelligent transportation systems   mobile edge computing   multi-access edge computing   task offloading  

摘要： Intelligent Transportation Systems (ITS) will become an essential part of every city in the near future. They should support various vehicle-to-everything (V2X) applications that improve road safety or even enable autonomous driving. Recently, the European Telecommunications Standards Institute (ETSI) introduced a multi-access (mobile) edge computing concept as a promising solution to satisfy the V2X delay and computational requirements. Based on this concept, the tasks generated by V2X applications can be offloaded to servers at the edge of the radio access network (RAN). There is a need for a task offloading algorithm that minimizes the ITS operator expenses connected with the servers deployment and maintenance while satisfying the requirements of the V2X applications. Most of the existing papers in the literature do not pay much attention to queuing delays at servers. In this paper, the queuing delays are analyzed by considering a general-type task computational time distribution. A non-linear optimization problem is formulated to minimize the ITS operator expenses subject to delays and computational resources constraints. The flexibility is also improved by considering that a delay constraint is satisfied with a given probability. To solve this problem, a method for linearization of the problem is proposed, and consequently, an algorithm based on Integer Linear Programming (ILP) is designed. A heuristic algorithm called Cost-effective Heuristic Algorithm for Task offloading (CHAT) is also introduced that provides close to optimal results and has much lower computational complexity than the ILP algorithm. The efficiency of the CHAT algorithm is studied in several scenarios in terms of the computational time, delays, and the total server energy consumption as the cost function. The results show that the CHAT algorithm satisfies the requirements of the V2X applications in all the considered scenarios and reduces the ITS operator expenses over twice compared with ot

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： Mobile handsets   Resource management   Task analysis   Central Processing Unit   Performance evaluation   Edge computing   Blockchain   mobile edge computing   deep reinforcement learning  

摘要： Mobile edge computing (MEC) raises the issue of resisting selfish edge attackers that use less computation resources than promised to process offloading tasks or provide faked computation results. In this paper, we present a blockchain based trust mechanism to help MEC address selfish edge attacks and faked service record attacks. This mechanism evaluates the computational performance of the edge devices and broadcasts such information to the neighboring edge devices and mobile devices. By building a reputation assignment method for the edge devices, the edge reputation system chooses the miner of the blockchain, which applies the joint Proof-of-Work and Proof-of-Stake consensus protocol to append a block recording the new service reputations onto the MEC blockchain. We propose a reinforcement learning (RL) based edge central processing unit (CPU) allocation algorithm without knowing the mobile service generation model and the network model in the dynamic edge computing process and a deep RL version to further improve the computational performance. The security performance is analyzed and the performance bound of the edge utility is provided. Experimental results show that this framework suppresses the selfish edge attacks, decreases the response latency and saves the energy compared with a benchmark MEC scheme.

关键词： Mobile edge computing   cognitive radio   underlay spectrum sharing   physical-layer security   wireless power transfer  

摘要： Wireless power transfer (WPT) and cognitive radio (CR) are two promising techniques in designing mobile-edge computing (MEC) systems. In this paper, we study a robust secure wireless powered multiple-input single-output (MISO) cognitive MEC system, which integrates several techniques: physical-layer security, WPT, CR, underlay spectrum sharing and MEC. Three optimization problems are formulated to minimize the total transmission power (TTP) of the primary transmitter (PT) and the secondary base station (SBS) under perfect channel state information (CSI) model, bounded CSI error model and the probabilistic CSI error model, respectively. The formulated problems are nonconvex and hard to solve. Three two-phase iterative optimization algorithms combined with Lagrangian dual, semidefinite relaxation (SDR), S-Procedure and Bernstein-type inequalities are proposed to jointly optimize the beamforming vectors of the PT and the SBS, the central processing unit (CPU) frequency and the transmit power of the MD. Simulation results are provided to verify the effectiveness of the proposed algorithms.