关键词： Mobile edge computing   Computation offloading   Resource allocation   Energy consumption   Algorithm design  

摘要： With lower network latency and powerful hardware, mobile edge computing (MEC) is effective for computation-intensive and delay-sensitive tasks. The rising energy and low-latency demands of mobile applications for MEC pose challenges to task allocation. In this work, we consider computation offloading in MEC that is composed of a set of mobile devices, each with multiple tasks to offload to a nearby MEC server, with both single and multiple access points (APs). With multiple APs deployed, each mobile device can directly communicate with one or more APs, and their tasks can offload to different MEC servers that are within direct communication range. By considering binary computation offloading mode and limited subchannels, we formulate the multitask allocation problem as an integer programming problem, with the objective of minimizing the total energy consumption of all mobile devices while meeting deadline requirements. To solve this complicated problem, we propose efficient algorithms for single and multiple APs, and analyze related properties, including the approximation ratio and complexity. Experiments show that the algorithm can find high-quality solutions in a short time.

关键词： mobile edge computing   offloading decision   task scheduling   wireless powered  

摘要： Efficient computing is an inevitable trend of mobile devices. In this paper, an algorithm of multi⁃user offloading decision and task scheduling is proposed for the wireless powered mobile⁃edge computing network. In the multi⁃user binary computing offload mode, an optimization problem to minimize the maximum execution delay of the system is established by jointly optimizing the wireless power supply time, the offloading decision of marginal users and the task scheduling. Since the optimization problem is a non⁃convex optimization problem for mixed integers and is difficult to solve, we design a two⁃layer alternate iterative solution. The first layer uses a greedy algorithm or an improved Johnson algorithm to solve the problem in a given wireless energy supply, and the second layer is based on the golden section method to solve the optimal wireless energy supply time. Simulation results show that the proposed algorithm has can effectively reduce the execution delay of the whole system with a fast convergence speed. © 2022 Journal of Nanjing Institute of Posts and Telecommunications. All rights reserved.

关键词： Mobile edge computing   Deep reinforcement learning   DNN partition   Game theory  

摘要： Currently, most edge-based Deep Reinforcement Learning (Deep-RL) applications have been deployed in the edge network, however, their mainstream studies are still short of adequate considerations on its limited compute and bandwidth resources. In this paper, we investigate the near on-policy of actions taking in distributed Deep-RL architecture, and propose a "hybrid near on-policy " Deep-RL framework, called Coknight, by leveraging a game-theory based DNN partition approach. We first formulate the partition problem into a variant of knapsack problem in device-edge setting, and then transform it into a potential game with a formal proof. Finally, we show the problem is NP-complete whereby an efficient distributed algorithm based on the potential game theory is developed from device perspective to achieve fast and dynamic partitioning. Coknight not only significantly improves the resource efficiency of the Deep-RL but also allows the inference to enforce the scalability of the actor policy. We prototype the framework with extensive experiments to validate our findings. The experimental results show that with the premise of a rapid convergence guarantee, Coknight, compared with Seed-RL, can reduce GPU utilization by 30% while providing large-scale scalability. (C) 2022 Elsevier Inc.

关键词： Task analysis   Autonomous aerial vehicles   Energy consumption   Digital twins   Trajectory   Servers   Resource management   Digital twin   unmanned aerial vehicle   mobile edge computing   user mobility   deep reinforcement learning  

摘要： Considering the user mobility and unpredictable mobile edge computing (MEC) environments, this paper studies the intelligent task offloading problem in unmanned aerial vehicle (UAV)-enabled MEC with the assistance of digital twin (DT). We aim at minimizing the energy consumption of the entire MEC system by jointly optimizing mobile terminal users (MTUs) association, UAV trajectory, transmission power distribution and computation capacity allocation while respecting the constraints of mission maximum processing delays. Specifically, double deep Q-network (DDQN) algorithm stemming from deep reinforcement learning is first proposed to effectively solve the problem of MTUs association and UAV trajectory. Then, the closed-form expression is employed to handle the problem of transmission power distribution and the computation capacity allocation problem is further addressed via an iterative algorithm. Numerical results show that our proposed scheme is able to converge and significantly reduce the total energy consumption of the MEC system compared to the benchmark schemes.

关键词： Task analysis   Servers   Autonomous aerial vehicles   Wireless communication   Resource management   Energy consumption   Performance evaluation   Mobile edge computing   resource allocation   service caching   task offloading   UAV communications  

摘要： Service caching refers to caching the necessary programs or/and the related databases for performing computational tasks at edge servers, which has been considered to save both computation and communication resources in mobile edge computing (MEC) systems. In this paper, we investigate computation service caching in a multi-unmanned aerial vehicle (UAV) enabled MEC system, where each UAV equipped with an edge server acts as an aerial computing platform to serve the ground devices. Furthermore, the UAVs can serve the devices cooperatively through the provided computing and caching resources. Aiming at minimizing the maximum task completion latency among all devices, we formulate a joint service caching, task offloading, communication and computation resource allocation, as well as UAV placement optimization problem, while guaranteeing the energy budget of all devices and UAVs. The problem is a mixed integer non-linear programming problem, and we decompose it into four sub-problems, and then propose an iterative algorithm based on block coordinate descent (BCD) and successive convex approximation (SCA) techniques to obtain near-optimal solution. Numerical results show that our proposed algorithm can achieve lower task completion latency than other baselines while guaranteeing better fairness among all devices.

关键词： Sensors   Radar   Performance evaluation   Wireless sensor networks   Servers   OFDM   Computational modeling   Radio resource allocation   integrated sensing and communication   mobile edge computing   network externality   matching theory  

摘要： Integrated sensing, communication, and computation (ISCC) will become a key enabler for automation applications. However, since the performance region of ISCC has a higher dimension than those of traditional wireless networks, existing schedulers typically fail to simultaneously meet the heterogeneous requests in ISCC. In this work, we propose a novel wireless scheduling architecture to explore the coordination gains of sensing, communication, and computation from a perspective of joint optimization. Specifically, we first construct an implementation framework of ISCC by combining the mobile edge computing paradigm with the integrated sensing and communication technology, where the inherent tradeoff between sensing, communication, and computation performance is characterized. Next, a joint device association and subchannel assignment problem is formulated to capture the network externalities induced by resource competition among mobile devices with multi-functional requirements. Due to its intractability, we then reformulate it in the matching theoretical manner. To obtain a mutually satisfactory solution under externalities, an iterative matching algorithm is developed by introducing pairwise stability and proved to be convergent and stable. The extensive simulations elucidate the significant superiority of our proposed scheme over those externality-unaware wireless schedulers.

关键词： Offloading optimization   Time allocation   Task offloading   Mobile edge computing  

摘要： This paper considers a wireless energy transfer based mobile edge computing system, where wireless devices can be charged by the radio-frequency signals broadcast by hybrid access point. With Mobile Edge Computing (MEC), wireless devices can execute their computation tasks locally or offload them to the MEC server by time division multiple access protocol. Based on this system, this paper studies the problem of system energy efficiency maximization by joint optimization of computing time allocation, energy consumption, capacity of local computing and task offloading. A Tabu search based system energy efficiency maximization algorithm is proposed for solving the optimization problems. Finally, the performance of the proposed algorithm is valued by extensive simulation experiments. Simulation results verify the effectiveness of the proposed algorithm.

关键词： Mobile blockchain   Mobile edge computing   5G   Mining-as-a-Service   Data offloading  

摘要： In this paper, we propose a mobile blockchain (MB) based mining-as-a-service (MaaS) scheme, MBMaaS for resource-constrained industrial internet-of-things (IIoT) environments. The scheme addresses the research gaps of fixed static allocation for miners to perform computationally intensive mining tasks through a multi-hop computational offloading (CO) scheme and addresses an auction mechanism for a fair bidding process among the miner nodes. The scheme operates in three phases. In the first phase, a multi-hop CO scheme with a fair incentive policy is formulated for miners. The CO schemes offer guaranteed offloading services to mobile devices from far-edge systems through a chain of neighbor nodes. Then, in the second phase, MaaS is proposed to leverage expensive mining tasks through 5Genabled pico/femtocells. Integration of 5G allows massive end-to-end device and service connectivity. As IIoT ecosystems have limited memory and compute requirements, MaaS assures that the proposed consensus has a responsive validation and mining time. To make the data exchange in the consensus process lightweight, and allow a large number of sensors to share the data in a lightweight manner, an effective consensus mechanism Lightweight Proof-of-Proximity (LPoP), is proposed that forms group validations instead of single block validation. The data is exchanged through javascript object notation (JSON) format, maintaining a steady transaction rate. MB-MaaS is compared against the existing scheme for parameters bid thresholds and request servicing times, and mining and consensus formation. For example, the request serving time at 12 requests is improved by 56.78%, and a significant improvement of 26.47% is observed for processed blocks;parsing time on average is improved by 7.89%. The comparative analysis suggests that the scheme is more efficient than other competing approaches. (c) 2022 Elsevier Inc. All rights reserved.

关键词： Mobile edge computing   Machine learning   Virtual machine placement   Bat Algorithm  

摘要： Mobile edge computing requires more and more high-performance servers, resulting in increasing energy consumption. As a well-established way to reduce energy consumption, virtual machine placement can be utilized to optimize the cost of wide-deployment of servers. However, traditional strategies tend to focus on single indicators, there are few existing research taking time delay limitation into account while solving energy consumption problems. In this paper, we propose a brand new method to settle the problems listed above, which is able to reduce the placement time of virtual machine and energy consumption. First, considering the excellent performance of bat swarm algorithm in NP-hard problem, we introduced the second order oscillation factor to avoid premature convergence, and combined the order exchange and migration local search technology. we proposed the OEMBA algorithm, which integrates underutilized servers to save energy. Subsequently, an improved Long Short-Term Memory model is utilized to fasten the placement of virtual machines and reduce latency based on historical data. Our results indicate that the improved learning model can save energy consumption and reduce placement latency.

关键词： Multi-period vehicular task offloading   Mobile edge computing   5G HetNets   Transmission method   Task completion rate   Timely rate of task completion  

摘要： Task offloading is one of the important research directions in Internet of Vehicles. Mobile edge computing and 5G heterogeneous network frameworks bring new opportunities for task offloading research. Most research works propose resource scheduling schemes and offloading strategies for task offloading within the same period. However, the allocation of resources for offloading tasks at different periods cannot be planned uniformly. The utilization and release of resources in previous periods affect the scheduling of resource allocations in later periods. To address the task offloading problem at different periods, this paper proposes the MPTO-MT algorithm. Specifically, the algorithm first determines whether the task can be offloaded in time at a certain period according to the number of candidate offloading nodes of the task in this period. Secondly, the algorithm selects the offloading node and the task transmission method. In order to select the appropriate offloading node and transmission method, we introduce the service utility and propose three heuristics of selecting the offloading node. Finally, the algorithm updates the resource information of nodes at the next period based on the utilization and release of node resources at the period, and the service time of the nodes. Comparative experiments with other baseline algorithms show that the proposed MPTO-MT algorithm improves the task completion rate by 4.55%, the timely rate of task completion by 3.57%, and shorter average service time.