关键词： Mobile edge computing   Task offloading   Average delay   Energy consumption level   Cost function   Lagrangian function   Karush-Kuhn-Tucker condition  

摘要： With the development for the technology of mobile edge computing (MEC) and the grave situation for the shortage of global energy, the problem of computation offloading in a cloud computing framework is getting more attention by network managers. In order to improve the experience quality of users and increase the energy efficiency of the system, we focus on the issue of task offloading strategy in MEC system. In this paper, we propose a task offloading strategy in the MEC system with a heterogeneous edge. By considering the execution and transmission of tasks under the task offloading strategy, we present an architecture for the MEC system. We establish a system model composed of M/M/1, M/M/c and M/M/infinity queues to capture the execution process of tasks in local mobile device (MD), MEC server and remote cloud servers, respectively. Moreover, by trading off the average delay of tasks, the energy consumption level of the MD and the offloading expend of the system, we construct a cost function for serving one task and formulate a joint optimization problem for the task offloading strategy accordingly. Furthermore, under the constraints of steady state and proportion scope, we use the Lagrangian function and the corresponding Karush-Kuhn-Tucker (KKT) condition to obtain the optimal task offloading strategy with the minimum system cost. Finally, we carry out numerical experiments on the MEC system to investigate the influence of system parameters on the task offloading strategy and to obtain the optimal results. The experiment results show that the task offloading strategy proposed in this paper can balance the average delay, the energy consumption level and the offloading expend with the optimal allocation ratio.

关键词： Edge computing   Mobile edge computing   Edge server   Placement   NSGA-II  

摘要： With the advent of mobile Internet and IoT era, various smart terminals generate a large amount of data at the edge of the network, and how to transmit and process these data at high speed poses a challenge to the traditional communication networks. Edge computing, as an emerging framework, can improve the communication capability and data processing capacity of traditional communication networks by improving their architecture. Edge server placement (ESP) technology is one of the key technologies of edge computing, which can effectively reduce data transmission delay and improve data processing efficiency by placing edge servers (ESs) with computing and data storage functions at base stations to sink some functions of the core network to the edge of the network. In this paper, we study the k edge servers placement problem (KESP problem) in smart cities. We first elaborate it as a multi-objective optimization problem for optimal workload balancing and system delay under constraints. Then a modified multi-objective non-dominated sorting genetic algorithm with elite policy (MNSGA-II) is proposed to optimize this problem. Finally, simulations are performed based on real network datasets. The simulation results show that MNSGA-II reduces the system overhead by about 38.4%, 40.6%, and 59.3% on average compared to Random, K-Means, and Top-K.

关键词： Task analysis   Unmanned aerial vehicles   Fading channels   Minimization   Resource management   Computational modeling   Approximation algorithms   Internet of Things   mobile edge computing   trajectory optimization   UAV communication  

摘要： The explosive computation demands in the Internet of Things (IoT) have triggered the research interests on unmanned aerial vehicle (UAV) assisted mobile-edge computing (MEC) systems even though there are still many challenges, such as computing delay requirement, multi-UAV cooperation, and resource management. This letter focuses on the computing delay issue in MEC systems assisted by multiple UAVs with the goal of task completion time minimization. In particular, both the partial offloading and binary offloading modes are considered by jointly optimizing time slot size, terminal devices scheduling, computation resource allocation, and UAVs' trajectories. Particularly, an non-LoS channel model is adopted for UAV-ground communication. To handle the formulated problems, we develop alternating optimization algorithms by invoking the successive convex approximation method, Karush-Kuhn-Tucker conditions and penalized method. Numerical results show that the completion time is significantly decreased by the proposed algorithms.

关键词： Task analysis   Unmanned aerial vehicles   Servers   Resource management   Decision making   Optimization   Edge computing   Mobile edge computing   5G   unmanned aerial vehicle   futures-based resource trading   power optimization  

摘要： Mobile edge computing (MEC) has emerged as one of the key technical aspects of the fifth-generation (5G) networks. The integration of MEC with resource-constrained unmanned aerial vehicles (UAVs) greatly enables flexible resource provisioning for supporting dynamic and computation-intensive UAV applications. Existing resource trading could facilitate this paradigm with proper incentives, which, however, may often incur unexpected negotiation latency and energy consumption, trading failures and unfair pricing, due to the unpredictable nature of the resource trading process. Motivated by these challenges, an efficient futures-enabled resource trading mechanism for edge computing-assisted UAV network is proposed, where a mutually beneficial and risk-tolerable forward contract is devised to promote resource trading between an MEC server (seller) and a UAV (buyer) with multiple tasks. Two key problems i.e. futures contract design before trading, and transmission power optimization during trading are studied. By analyzing historical statistics associated with future resource supply, demand, and air-to-ground communication quality, the contract design is formulated as a multi-objective optimization problem aiming to maximize both the seller's and the buyer's expected utilities, while estimating their acceptable risk tolerance. Accordingly, we propose an efficient bilateral negotiation scheme to help players reach a trading consensus on the amount of resources and the relevant price. For the power optimization problem, we develop a practical algorithm that enables the buyer to determine its optimal transmission power via convex optimization techniques. Comprehensive simulations demonstrate that the proposed mechanism offers mutually beneficial utilities to players, while achieving commendable performance on trading failures and fairness, negotiation latency and cost, comparing with baseline methods.

关键词： Unmanned aerial vehicles   Games   Task analysis   Energy consumption   Interference   Micromechanical devices   Edge computing   Unmanned aerial vehicle   mobile edge computing   hierarchical framework   Stackelberg game   exact potential game  

摘要： In this letter, we investigate the joint optimization of computation offloading, channel allocation and position deployment problem for coalition-based unmanned aerial vehicle (UAV) mobile edge computing networks. To characterize the hierarchical offloading feature in coalition-based UAV network, we propose the hierarchical framework and formulate a discrete multi-leader multi-follower energy minimization Stackelberg game. In addition, we prove that the multi-leader subgame and multi-follower subgame are exact potential games with Nash equilibrium. Then the existence of Stackelberg equilibrium (SE) is guaranteed. The spatial adaptive play-best response based hierarchical iterative learning (SAP-BRHIL) algorithm is proposed to achieve the SE. The simulation results show that the proposed SAP-BRHIL algorithm can achieve the lowest network energy consumption compared with some existing approaches.

关键词： mobile edge computing   NOMA   full offloading   partial offloading  

摘要： The development of mobile edge computing (MEC) is expected to offer better performance in mobile communications than the current cloud computing architecture. MEC involves offering the closest access to the data source or physical mobile network environment. The network services are able to respond faster, thus satisfying the demands of the mobile network industry when deploying various potential business applications in real-time. Since the harvested mobile data are transferred to the edge server to make calculations, data transfers and faults in the mobile network can be swiftly pinpointed and removed accurately. Nevertheless, there are still problems in the practical application of the systems, specifically in reducing delays and lessening energy consumption. Because of non-orthogonal multiple access (NOMA) superior spectrum efficiencies, it is best to combine NOMA with MEC for simultaneous support of multiple access for end users, thus reducing transmission latencies and lowering energy consumption. Combining MEC and NOMA would offer many advantages, including superior energy savings, reductions in latency, massive connectivity, and the potential of combining with additional transmission technologies, such as millimetre-wave (mmWave) and M-MIMO. In this paper, designing wireless resource allocation is crucial for an economically viable low-latency wireless network, which can be realised using the Karush-Kuhn-Tucker (KKT) approach to obtain the optimal solution for partial and full offloading network traffic scenarios to minimize the total latency of the MEC network. The convergence and performance for orthogonal multiple access (OMA), pure-NOMA (P-NOMA), and hybrid-NOMA (H-NOMA) are also compared under different network traffic offloading scenarios. The significant results from this study showed the convergence of the optimal resource allocation in the case of full and partial offloading. The results demonstrated that the P-NOMA reduces the total offloading dela

关键词： Array signal processing   Resource management   Optimization   NOMA   Three-dimensional displays   Energy harvesting   Wireless communication   Hybrid beamforming   mobile edge computing   non-orthogonal multiple access   unmanned aerial vehicle   wireless power transfer  

摘要： Beamforming and non-orthogonal multiple access (NOMA) serve as two potential solutions for achieving spectral efficient communication in the fifth generation and beyond wireless networks. In this paper, we jointly apply a hybrid beamforming and NOMA techniques to an unmanned aerial vehicle (UAV)-carried wireless-powered mobile edge computing (MEC) system, within which the UAV is equipped with a wireless power charger and the MEC platform delivers energy and computing services to Internet of Things (IoT) devices. Our aim is to maximize the sum computation rate at all IoT devices whilst satisfying the constraint of energy harvesting and coverage. The resultant optimization problem is non-convex involving joint optimization of the UAV's 3D placement and hybrid beamforming matrices as well as computation resource allocation in both partial and binary offloading patterns, and thus is quite difficult to tackle directly. By applying the polyhedral annexation method and the deep deterministic policy gradient (DDPG) algorithm, we develop an effective algorithm to derive the closed-form solution for the optimal 3D deployment of the UAV, and find the solution for the hybrid beamformer. Two resource allocation algorithms for partial and binary offloading patterns are thereby proposed. Simulation results verify that our designed algorithms achieve a significant computation performance enhancement as compared to the benchmark schemes.

关键词： mobile edge computing   joint resource allocation   hierarchical game   resource sharing   QoS  

摘要： In this paper, we propose a hierarchical Stackelberg game based resource allocation algorithm (HGRAA) to jointly allocate the wireless and computational resources of a mobile edge computing (MEC) system. The proposed HGRAA is composed of two levels: the lower-level evolutionary game (LEG) minimizes the cost of mobile terminals (MTs), and the upper-level exact potential game (UEPG) maximizes the utility of MEC servers. At the lower-level, the MTs are divided into delay-sensitive MTs (DSMTs) and non-delay-sensitive MTs (NDSMTs) according to their different quality of service (QoS) requirements. The competition among DSMTs and NDSMTs in different service areas to share the limited available wireless and computational resources is formulated as a dynamic evolutionary game. The dynamic replicator is applied to obtain the evolutionary equilibrium so as to minimize the costs imposed on MTs. At the upper level, the exact potential game is formulated to solve the resource sharing problem among MEC servers and the resource sharing problem is transferred to nonlinear complementarity. The existence of Nash equilibrium (NE) is proved and is obtained through the Karush-Kuhn-Tucker (KKT) condition. Simulations illustrate that substantial performance improvements such as average utility and the resource utilization of MEC servers can be achieved by applying the proposed HGRAA. Moreover, the cost of MTs is significantly lower than other existing algorithms with the increasing size of input data, and the QoS requirements of different kinds of MTs are well guaranteed in terms of average delay and transmission data rate.

关键词： Mobile Edge Computing   Offloading   Pricing   Resource allocation   Game theory  

摘要： Mobile Edge Computing is a new computing paradigm that offers cloud computation capabilities at the edge of the mobile network. Due to the proximity to Mobile Users (MUs), the edge cloud can be accessed with low latency by resource-limited mobile devices with the aim of extending their capabilities through computation offloading. Despite its unique features, the efficiency of offloading is highly influenced by the purchasing power of MUs as well as the competition arisen to access both limited communication and computation resources. However, since the MUs' demands for these resources change dynamically, the pricing approaches should be capable of adapting in real-world scenarios. To address the aforementioned challenges, we propose a game-based distributed scheme to jointly and dynamically allocate and price resources required for proper offloading in a two-tier NOMA-based mobile system. To this end, we consider the incentives of both the edge provider and MUs as well as conflicts arising from their interactions. Under the proposed scheme, MUs should pay not only based on the given resources but also based on their transmission power level as a punishment for their produced interference. We formulate the interactions among the monopolist edge provider and MUs using the Stackelberg game. Further, the power allocation in NOMA and the competition of MUs over computation resources are modeled using game theory and shown admitting the potential and the weakly coupled constraint game, respectively. Furthermore, we prove the existence and uniqueness of the equilibrium of both corresponding games and calculate it for both the edge provider and MUs. Simulation results demonstrate the efficiency of the proposed approach in terms of the utility of MUs and the edge provider, average allocated communication and computation resources, optimal pricing, and the number of rejected MUs under different purchasing power.

关键词： Cloud computing   Energy efficient resource allocation   Mobile edge computing   Service rate   User equipment   Utility function  

摘要： Limited battery and computing resources of mobile devices (MDs) induce performance limitations in mobile edge computing (MEC) networks. Computational offloading has the capability to provide computing and storage resources to MDs for resource-intensive tasks execution. Therefore, to minimize energy consumption and service delay, MDs offload the resource-intensive tasks to nearby mobile edge server (MES) for execution . However, due to time varying network conditions and limited computing resources at MES also, the offloading decision taken by MDs may not achieve the lowest cost. In this paper, we propose an energy efficient and faster deep learning based offloading technique (EFDOT) to minimize the overall cost of MDs. We formulate a cost function which considers the energy consumption and service delay of MDs, radio resources, energy consumption and delay due to task partitioning, and computing resources of the MDs and MESs. Due to high computational overhead of this comprehensive cost function, we generate a training dataset to train a deep neural network (DNN) in order to make the decision making process faster. The proposed work finds the optimal number of components, task partitioning, and fine-grained offloading policy simultaneously. We formulate the fine-grained offloading decision problem in MEC as multi-label classification problem and propose EFDOT to minimize the computation and offloading overhead. The simulation results show high accuracy of the DNN and high performance of EFDOT in terms of energy consumption, service delay, and battery life.