关键词： Computation offloading   Mobile edge computing   Processing time optimization   Simulated annealing  

摘要： In recent years, the importance of the mobile edge computing (MEC) paradigm along with the 5G, the Internet of Things (IoT) and virtualization of network functions is well noticed. Besides, the implementation of computation-intensive applications at the mobile device level is limited by battery capacity, processing capabalities and execution time. To increase the batteries life and improve the quality of experience for computationally intensive and latency-sensitive applications, offloading some parts of these applications to the MEC is proposed. This paper presents a solution for a hard decision problem that jointly optimizes the processing time and computing resources in a mobile edge-computing node. Hence, we consider a mobile device with an offloadable list of heavy tasks and we jointly optimize the offloading decisions and the allocation of IT resources to reduce the latency of tasks’ processing. Thus, we developped a heuristic solution based on the simulated annealing algorithm, which can improve the offloading rate and reduce the total task latency while meeting short decision time. We performed a series of experiments to show its efficiency. Finally, the obtained results in terms of full-time treatrement are very encouraging. In addition, our solution makes offloading decisions within acceptable and achievable deadlines. Copyright © 2020 Institute of Advanced Engineering and Science. All rights reserved.

关键词： Road-Side Unit   Software Defined Network   Reinforcement Learning   Mobile Edge Computing  

摘要： The growing number of IoT devices and the different computing and communication capabilities of IoT devices require an efficient offloading scheme. This offloading scheme need to consider the mobility of IoT devices and helps to intelligently select the optimal server for offloading. An efficient offloading scheme need to take in consideration important factors such as mobility of the IoT user device, speed and direction of the IoT user device as well as the computational capabilities of the user mobile device and the load of nearby servers. Unbalanced load of data or task offloading lead to high latency and poor services. An optimal selection of offloading server will clearly improve latency and QoS. Some new architecture of cellular network suggest the deployment of small-cell base stations (SBS) [1], [2] with a certain computing capabilities which can help offloading task of IoT mobile device or of their nearby SBS. In smart city environment, the mobile IoT device user needs to choose an SBS from several available SBSs within the its communication proximity. In this paper, we propose a Smart Ranking based Task Offloading approach for selecting an SBS and to improve the Quality of Service. This approach uses Q-Learning for SBS selection which will be modelled in Software Defined Network controller to deal with the problem of choosing the SBS in an intelligent way for Task offloading.

关键词： Road-Side Unit   Software Defined Network   Reinforcement Learning   Mobile Edge Computing  

摘要： The evolution of wide variety of applications that are used by vehicular users includes a lot of data hungry applications. This increases the workload on the cellular networks, thereby delivering poor service to the users. We can overcome this problem by sharing this workload with open wireless networks. As such, this improves the Quality of Service provided by cellular networks. Road-Side Units (RSU) are a wireless network which plays a major role in data offloading. Our approach discusses switching the communication network from cellular to RSU whenever there is an opportunity for a vehicle to offload vehicles data. Busy roads/urban traffic consists of several RSUs with many users. In urban environment, the vehicular user needs to choose an RSU from several available RSUs within the vehicle communication proximity. For seamless connectivity, the delay in network communication because of selecting the best RSU and frequent switching of connection between vehicles and RUSs must be minimized. In this paper, we propose a Smart Ranking based Data Offloading (SRDO) algorithm for selecting an RSU and to improve the Quality of Service. In SRDO algorithm, Q-Learning is utilized for RSU selection. This algorithm is modelled in Software Defined Network controller to deal with the problem of choosing the RSU in an intelligent way for data offloading.

关键词： Mobile social networks   mobile edge computing   caching   deep reinforcement learning  

摘要： Social networks have continuously been expanding and trying to be innovative. The recent advances of computing, caching, and communication (3C) can have significant impacts on mobile social networks (MSNs). MSNs can leverage these new paradigms to provide a new mechanism for users to share resources (e.g., information, computation-based services). In this paper, we exploit the intrinsic nature of social networks, i.e., the trust formed through social relationships among users, to enable users to share resources under the framework of 3C. Specifically, we consider the mobile edge computing (MEC), in-network caching and device-to-device (D2D) communications. When considering the trust-based MSNs with MEC, caching and D2D, we apply a novel deep reinforcement learning approach to automatically make a decision for optimally allocating the network resources. The decision is made purely through observing the network's states, rather than any handcrafted or explicit control rules, which makes it adaptive to variable network conditions. Google TensorFlow is used to implement the proposed deep Q-learning approach. Simulation results with different network parameters are presented to show the effectiveness of the proposed scheme.

关键词： Internet of Things (IoT)   joint communication and computation resource allocations   mobile edge computing   non-orthogonal multiple access  

摘要： The past decades have witnessed an explosive growth of the Internet of Things (IoT) services requiring intensive computation resources. The conventional IoT devices, however, are usually equipped with very limited computation resources, which results in degraded quality of experience when executing the resource-hungry applications. Mobile edge computing (MEC), which enables smart terminals (STs) to offload parts of their computation workloads to the edge servers located at cellular base stations (BSs), has provided a promising approach to address this issue. In this article, we investigate the nonorthogonal multiple access (NOMA)-enabled multiaccess MEC. Specifically, by exploiting the advanced NOMA, an ST can simultaneously offload its computation workloads to different edge servers (ESs), which thus reduces the overall delay in completing the ST's computation workloads. To study this problem, we formulate a joint optimization of the computation resource allocations at the ESs, the ST's offloaded workloads and its radio resource allocations for NOMA transmission, with the objective of minimizing a system wise cost that accounts for the overall delay in finishing the ST's total computation workload and the total computation resource usage cost at the ESs. Despite the nonconvexity of the joint optimization problem, we exploit its layered structure and propose an efficient layered algorithm to find the optimal solution. By exploiting the optimal offloading solution of a single ST, we further investigate the scenario of multiple STs and propose two algorithms to determine the optimal grouping among different ESs for serving the STs, with one algorithm aiming at minimizing the total cost of all STs and the other algorithm aiming at determining the Nash stable grouping for the ESs. Numerical results are presented to validate the effectiveness of our proposed algorithms and show the performance gain of our proposed NOMA-enabled multiaccess computation offloading.

关键词： Vehicular model   Mobile edge computing   Vehicular service cloud   Vehicular cloud computing   5G  

摘要： With 5G network services around the corner, vehicular cloud networks providing computation capabilities have taken precedence over traditional costly cloud solutions. However, with vehicular cloud computing, a variety of new challenges have grown. In this paper we propose an infra-vehicle resource sharing model to provide a range of cloud services such as on-demand entertainment and speech recognition for driver assistance. The proposed solution forms nearby low-latency Vehicular Service Clouds (VSC) on-the-fly as per the needs of vehicular users. Vehicles in parking lots or moving on the road collaborate and share their computation and storage resources to complete different vehicular service requests. We develop an incentive-based model that uses edge-based Road Side Units (RSU) to compose heterogeneous node resources and produce a usable resource that satisfies users' requests with minimal delays. Through proof of concept simulations, we compare our solution against traditional cloud solutions to showcase the effectiveness of adopting our proposed framework.

关键词： Eavesdropping   Jamming   Security   Unmanned aerial vehicles   Task analysis   Edge computing   Base stations   Physical layer security   mobile edge computing   UAV communication   secrecy capacity  

摘要： In this paper, we propose a secure unmanned aerial vehicle (UAV) mobile edge computing (MEC) system where multiple ground users offload large computing tasks to a nearby legitimate UAV in the presence of multiple eavesdropping UAVs with imperfect locations. To enhance security, jamming signals are transmitted from both the full-duplex legitimate UAV and non-offloading ground users. For this system, we design a low-complexity iterative algorithm to maximize the minimum secrecy capacity subject to latency, minimum offloading and total power constraints. Specifically, we jointly optimize the UAV location, users' transmit power, UAV jamming power, offloading ratio, UAV computing capacity, and offloading user association. Numerical results show that our proposed algorithm significantly outperforms baseline strategies over a wide range of UAV self-interference (SI) efficiencies, locations and packet sizes of ground users. Furthermore, we show that there exists a fundamental tradeoff between the security and latency of UAV-enabled MEC systems which depends on the UAV SI efficiency and total UAV power constraints.

关键词： Mobile edge computing   Task offloading   Deep reinforcement learning   LSTM network   Candidate network  

摘要： With the maturity of 5G technology and the popularity of intelligent terminal devices, the traditional cloud computing service model cannot deal with the explosive growth of business data quickly. Therefore, the purpose of mobile edge computing (MEC) is to effectively solve problems such as latency and network load. In this paper, deep reinforcement learning (DRL) is first proposed to solve the offloading problem of multiple service nodes for the cluster and multiple dependencies for mobile tasks in large-scale heterogeneous MEC. Then the paper uses the LSTM network layer and the candidate network set to improve the DQN algorithm in combination with the actual environment of the MEC. Finally, the task offloading problem is simulated by using iFogSim and Google Cluster Trace. The simulation results show that the offloading strategy based on the improved IDRQN algorithm has better performance in energy consumption, load balancing, latency and average execution time than other algorithms. (C) 2019 Elsevier B.V. All rights reserved.

关键词： Mobile edge computing   user-adaptive offloading   computation-intensive offloading   per-task delay   tasks satisfaction optimization  

摘要： Mobile-edge computing (MEC) is a new paradigm with a great potential to extend mobile users capabilities because-of its proximity. It can contribute efficiently to optimize the energy consumption to preserve privacy, and reduce the bottlenecks of the network traffic. In addition, intensive-computation offloading is an active research area that can lessen latencies and energy consumption. Nevertheless, within multi-user networks with a multi-task scenario, select the tasks to offload is complex and critical. Actually, these selections and the resources' allocation have to be carefully considered as they affect the resulting energies and delays. In this work, we study a scenario considering a user-adaptive offloading where each user runs a list of heavy computation-tasks. Every task has to be processed in its associated MEC server within a fixed deadline. Hence, the proposed optimization problem target the minimization of a weighted-sum normalized function depending on three metrics. The first is energy consumption, the second is the total processing delays, and the third is the unsatisfied processing workload. The solution of the general problem is obtained using the solutions of two sub-problems. Also, all solutions are evaluated using a set of simulation experiments. Finally, the execution times are very encouraging for moderate sizes, and the proposed heuristic solutions give satisfactory results in terms of users cost function in pseudo-polynomial times.

关键词： Task analysis   Wireless communication   Sensors   Resource management   Wireless sensor networks   Servers   Energy consumption   Mobile edge computing   binary offloading   optimization algorithms   resource allocation  

摘要： Mobile-edge computing (MEC) has recently emerged as a cost-effective paradigm to enhance the computing capability of hardware-constrained wireless devices (WDs). In this paper, we first consider a two-user MEC network, where each WD has a sequence of tasks to execute. In particular, we consider task dependency between the two WDs, where the input of a task at one WD requires the final task output at the other WD. Under the considered task-dependency model, we study the optimal task offloading policy and resource allocation (e.g., on offloading transmit power and local CPU frequencies) that minimize the weighted sum of the WDs' energy consumption and task execution time. The problem is challenging due to the combinatorial nature of the offloading decisions among all tasks and the strong coupling with resource allocation. To tackle this problem, we first assume that the offloading decisions are given and derive the closed-form expressions of the optimal offloading transmit power and local CPU frequencies. Then, an efficient bi-section search method is proposed to obtain the optimal solutions. Furthermore, we prove that the optimal offloading decisions follow an one-climb policy, based on which a reduced-complexity Gibbs Sampling algorithm is proposed to obtain the optimal offloading decisions. We then extend the investigation to a general multi-user scenario, where the input of a task at one WD requires the final task outputs from multiple other WDs. Numerical results show that the proposed method can significantly outperform the other representative benchmarks and efficiently achieve low complexity with respect to the call graph size.