关键词： Deep reinforcement learning   Internet of vehicles   Mobile edge computing   Scheduling optimization  

摘要： Driven by the construction of smart cities, networks and communication technologies are gradually infiltrating into the Internet of Things (IoT) applications in urban infrastructure, such as automatic driving. In the Internet of Vehicles (IoV) environment, intelligent vehicles will generate a lot of data. However, the limited computing power of in-vehicle terminals cannot meet the demand. To solve this problem, we first simulate the task offloading model of vehicle terminal in Mobile Edge Computing (MEC) environment. Secondly, according to the model, we design and implement a MEC server collaboration scheme considering both delay and energy consumption. Thirdly, based on the optimization theory, the system optimization solution is formulated with the goal of minimizing system cost. Because the problem to be resolved is a mixed binary nonlinear programming problem, we model the problem as a Markov Decision Process (MDP). The original resource allocation decision is turned into a Reinforcement Learning (RL) problem. In order to achieve the optimal solution, the Deep Reinforcement Learning (DRL) method is used. Finally, we propose a Deep Deterministic Policy Gradient (DDPG) algorithm to deal with task offloading and scheduling optimization in high-dimensional continuous action space, and the experience replay mechanism is used to accelerate the convergence and enhance the stability of the network. The simulation results show that our scheme has good performance optimization in terms of convergence, system delay, average task energy consumption and system cost. For example, compared with the comparison algorithm, the system cost performance has improved by 9.12% under different task sizes, which indicates that our scheme is more suitable for highly dynamic Internet of Vehicles environment.

关键词： Clustered scheduling   communication pipelining   federated learning   mobile edge computing  

摘要： This article proposes using communication pipelining to enhance the convergence speed of federated learning in mobile edge computing applications. Due to limited wireless subchannels, a subset of the total clients is scheduled in each iteration of federated learning algorithms. On the other hand, the scheduled clients wait for the slowest client to finish its computation. We propose to first cluster the clients based on the time they need per iteration to compute the local gradients of the federated learning model. Then, we schedule a mixture of clients from all clusters to send their local updates in a pipelined manner. In this way, instead of just waiting for the slower clients to finish their computations, more clients can participate in each iteration. While the time duration of a single iteration does not change, the proposed method can significantly reduce the number of required iterations to achieve a target accuracy. We provide a generic formulation for optimal client clustering under different settings, and we analytically derive an efficient algorithm for obtaining the optimal solution. We also provide numerical results to demonstrate the gains of the proposed method for different data sets and deep learning architectures.

关键词： UAV   Wireless networks   Aerial base station   Aerial relay   Mobile edge computing   Air-ground communications  

摘要： Due to their characteristics of mobility, flexibility and adaptive altitude, unmanned aerial vehicles (UAVs), also known as drones, have immense potential applications in wireless systems. In particular, UAVs can operate as flying mobile terminals within a cellular network, and such cellular-connected UAVs have enabled several applications ranging from real-time video streaming to item delivery. Furthermore, the deployment of UAVs in wireless networks as flying base stations (BSs), servers or relays is expected to enhance network performance in terms of coverage, capacity, reliability and energy-efficiency. In this paper, a comprehensive tutorial on the potential applications and benefits of UAVs is presented. Specifically, UAVs as aerial BSs collecting data from ground sensors, transmitting power to energy-constrained devices and broadcasting information to ground users are explored along with representative design objectives. Then, UAVs as relay nodes for coverage extension, pathloss compensation and interference mitigation are surveyed comprehensively. Besides, UAVs as mobile servers performing edge computing for ground users are introduced, and various aerial offloading modes are reviewed thoroughly. Furthermore, radio and computation resource optimization, as well as mathematical tools and developed algorithms, are described in UAV-assisted communicating, relaying, and computing systems. Finally, open problems and potential research directions pertaining UAV-assisted communications are presented. In a nutshell, this tutorial provides key guidelines on how to design, analyze and optimize UAV-assisted wireless communication systems.

关键词： Mobile edge computing   millimeter-wave communications   non-or thogonal multiple access   computation efficiency   resource allocation  

摘要： In this paper, the millimeter-wave (mmWave) communications and non-orthogonal multiple access (NOMA) are exploited for mobile edge computing (MEC) networks to improve the performance of task offloading. Aiming at improving the computation efficiency (CE) and ensuring the fairness among users, we study the CE optimization for mmWave-MEC with NOMA, where both the analog beamforming (ABF) and hybrid beamforming (HBF) architectures under the partial offloading mode are considered. First, according to the max-min fairness criterion, the CE optimization problem is formulated to jointly optimize the ABF at the base station and the local resource allocation of each user in mmWave-MEC with ABF. An efficient algorithm based on the penalized successive convex approximation is proposed to solve this non-convex problem. Then, the max-min CE optimization problem in mmWave-MEC with HBF is studied, where the joint design of the HBF at the BS and the local resource allocation of each user is carried out. By using the penalty function and the inexact block coordinate descent method, a feasible optimization algorithm is developed to tackle this challenging problem. Simulation results verify the convergence of the proposed algorithms and show that the proposed resource allocation schemes can improve the system CE effectively, and the mmWave-MEC with HBF scheme can obtain higher CE than that with ABF scheme. Besides, the NOMA scheme exhibits superior performance over the conventional orthogonal multiple access scheme in terms of CE.

关键词： Mobile edge computing   edge caching   task offloading   particle swarm optimization   genetic algorithm  

摘要： Computing-intensive and latency-sensitive user requests pose significant challenges to traditional cloud *** response to these challenges,mobile edge computing(MEC)has emerged as a new paradigm that extends the computational,caching,and communication capabilities of cloud *** caching certain services on edge nodes,computational support can be provided for requests that are offloaded to the ***,previous studies on task offloading have generally not considered the impact of caching mechanisms and the cache space occupied by *** oversight can lead to problems,such as high delays in task executions and invalidation of offloading *** optimize task response time and ensure the availability of task offloading decisions,we investigate a task offloading method that considers caching ***,we incorporate the cache information of MEC into the model of task offloading and reduce the task offloading problem as a mixed integer nonlinear programming(MINLP)***,we propose an integer particle swarm optimization and improved genetic algorithm(IPSO_IGA)to solve the ***_IGA exploits the evolutionary framework of particle swarm *** it uses a crossover operator to update the positions of particles and an improved mutation operator to maintain the diversity of ***,extensive simulation experiments are conducted to evaluate the performance of the proposed *** experimental results demonstrate that IPSO_IGA can save 20%to 82%of the task completion time,compared with state-of-theart and classical ***,IPSO_IGA is suitable for scenarios with complex network structures and computing-intensive tasks.

关键词： Mobile edge computing   Unmanned aerial vehicle   Online multi-agent proximal policy   optimization (O-MAPPO)   Energy efficiency  

摘要： The combination of mobile edge computing (MEC) systems with unmanned aerial vehicle (UAV) has gained a high profile in recent years due to its high flexibility and ability to handle intensive tasks. The computation offloading strategy and the resource allocation scheme affect the system performance significantly. Moreover, the system complexity increases with the numbers of users and servers exponentially. It is challenging to consider jointly the computation offloading and the resource allocation in MEC systems that have multiple users and multiple servers. This paper formulates the joint resources management in the UAV assisted MEC network as a partially observable markov decision process and proposes an online multi-agent proximal policy optimization (O-MAPPO) scheme to improve the energy efficiency while guaranteeing the requirements in task, power consumption, computation, and time. Specifically, users and servers are set as agents. All agents cooperatively make decisions of computation offloading and resource allocation to maximize the energy efficiency. Simulation results show that the O-MAPPO scheme significantly outperforms benchmark algorithms in robustness and stability.

关键词： Mobile edge computing   seagull optimization   deep belief network   resource management   parameter tuning  

摘要： Recently,computation offloading has become an effective method for overcoming the constraint of a mobile device(MD)using computationintensivemobile and offloading delay-sensitive application tasks to the remote cloud-based data *** city benefitted from offloading to edge *** a mobile edge computing(MEC)network in multiple *** comprise N MDs and many access points,in which everyMDhasM independent real-time *** study designs a new Task Offloading and Resource Allocation in IoT-based MEC using Deep Learning with Seagull Optimization(TORA-DLSGO)*** proposed TORA-DLSGO technique addresses the resource management issue in the MEC server,which enables an optimum offloading decision to minimize the system *** addition,an objective function is derived based on minimizing energy consumption subject to the latency requirements and restricted *** TORA-DLSGO technique uses the deep belief network(DBN)model for optimum offloading ***,the SGO algorithm is used for the parameter tuning of the DBN *** simulation results exemplify that the TORA-DLSGO technique outperformed the existing model in reducing client overhead in the MEC systems with a maximum reward of 0.8967.

关键词： Edge server placement   Dynamic movement   Game theory   Mobile edge computing   Query throughput  

摘要： Edge server placement plays a vital role in mobile edge computing to enable low-latency and high-throughput services by deploying edge servers at suitable geographical locations. Exiting work mainly focused on satisfying the QoS metrics of deployment while the impact of the dynamic movement of users is ignored. Dynamic movements result from different lifestyles and routines of mobile users whose points of interest thus changes over time which leads to significant unbalanced workload for edge servers. To address this challenge, we present a mobility-aware edge server placement method with two key features: (1) in the offline phase, we designed a fast heuristic algorithm to generate an edge server deployment strategy for massive data;(2) in the online phase, we designed a fine-tuning mechanism of deployment strategy based on cooperative game theory to adapt to users' dynamic movements. To demonstrate the utility of the proposed method, we have performed a comprehensive experimental evaluation on a large-scale dataset of over 1500 edge devices collected over a year. Experimental results show our method outperforms all baselines significantly in terms of query throughput and response time.

关键词： Mobile edge computing   mmWave communication system   delay minimisation   Red Deer Algorithm  

摘要： Mobile edge computing (MEC)-oriented solutions are essential for various 5G wireless communication systems. It is considered a key technology for future communication systems because of its capability for fulfilling a broad variety of necessities of the developing wireless terminals in the form of Intelligent Vehicles, augmented reality and virtual reality devices like huge computation, low latency and high data rate. Further, the resource package has made the research on mobile data offloading. One probable novel spectrum in the subsequent generation networks is the millimetre wave (mmWave) communication systems. It attains important attention because of its high rate. This research work, this area focuses on the delay of minimisation strategy in mmWave MEC by jointly optimising the hybrid beam-forming and also resource allocation. Here, the utilisation of a well-performing Red Deer Algorithm (RDA) is the ultimate aim of the suggested model that intends to optimise the analogue beam-forming vectors at the users, the analogue and digital beam-forming matrices at the Base Station (BS), the computation task offloading ratios and resource allocation at the MEC server. Here, the minimisation of the delay or latency is attained. The comparative analysis of the proposed model over the other models demonstrates the superiority of the proposed algorithm in assisting the mmWave MEC system.

关键词： Mobile edge computing   Collaborative decision-making   Deep reinforcement learning   Multi-agent deep deterministic policy gradient  

摘要： As is widely accepted, mobile edge computing (MEC) is a promising technology to enable wireless devices (WDs) to process computation-intensive tasks. Due to the mutual influence among different WDs, collaborative task offloading is needed in multi-agent environments. In this paper, a multiagent MEC network with delay-sensitive and non-partitionable tasks is considered, as well as the load on MEC servers. The joint optimization problem of offloading decision and resource allocation is formulated to minimize the average delay. To realize the collaborative decision-making, a multi-agent deep reinforcement learning based algorithm is proposed based on the framework of centralized training and decentralized execution. The centralized deep neural networks (DNN) learn from the past experience and the WDs learn policies from the evaluation of the actions from the centralized DNNs. Based on the learned policies, WDs can make offloading decisions with only local information. Simulation results show that the proposed algorithm achieves near-optimal performance and has the advantage of high stability in varying multi-agent environments. (c) 2023 Elsevier Inc. All rights reserved.