关键词： deep reinforcement learning   mobile edge computing   service caching   service migrating  

摘要： Emerging mobile edge computing(MEC)is considered a feasible solution for offloading the computation-intensive request tasks generated from mobile wireless equipment(MWE)with limited computational resources and *** to the homogeneity of request tasks from one MWE during a longterm time period,it is vital to predeploy the particular service cachings required by the request tasks at the MEC *** this paper,we model a service caching-assisted MEC framework that takes into account the constraint on the number of service cachings hosted by each edge server and the migration of request tasks from the current edge server to another edge server with service caching required by ***,we propose a multiagent deep reinforcement learning-based computation offloading and task migrating decision-making scheme(MBOMS)to minimize the long-term average weighted *** proposed MBOMS can learn the near-optimal offloading and migrating decision-making policy by centralized training and decentralized *** and comprehensive simulation results reveal that our proposed MBOMS can converge well after training and outperforms the other five baseline algorithms.

关键词： Mobile edge computing   fractional programming   active reconfigurable intelligent surface  

摘要： In this letter, we consider an active reconfigurable intelligent surface (ARIS)-assisted uplink mobile edge computing (MEC) system. We propose an ARIS assisted non-orthogonal multiple access (NOMA) scheme, denoted by ARIS-NOMA, in which multiple users process their local computing and simultaneously perform task offloading in NOMA manner to a base station (BS) with the help of a common ARIS. A total computation bits (TCB) maximization problem is formulated to improve the computing capability of proposed ARIS-NOMA scheme by jointly optimizing the BS's receive beamforming (RBF), ARIS's reflection coefficients, local computing frequency (LCF), and transmit power of each user. Specifically, we first use a fractional programming (FP) technique to reformulate the original objective function, and then utilize a block coordinate descent (BCD) method to separate the reformulated problem into several sub-problems and iteratively optimize one set of variables with others being fixed. In particular, we obtain closed-form solutions for BS RBF, ARIS reflection coefficients, LCF, and the transmit power of each user by jointly leveraging the Lagrange dual method and Karush-Kuhn-Tucker (KKT) conditions. Numerical results validate the advantages of the proposed ARIS-NOMA scheme compared to the conventional benchmarks regarding to the TCB.

关键词： Mobile edge computing   Classification   Trust   Task offloading   Entropy weight method  

摘要： To address the problems of how to resist internal attacks caused by offloading traffic surveillance video processing tasks to the edge cloud and how to reasonably allocate edge computing resources to reduce service latency and system energy consumption, this paper proposes a trust -based offloading scheme for edge computing tasks. Firstly, a novel trust scheme is presented, which establishes a trust feature value matrix and constructs a difference function based on linear discriminant analysis. To achieve the best classification performance, the maximization of the difference function is transformed into an optimization problem of finding the optimal weights. This scheme effectively solves the subjectivity issue in threshold determination in previous trust mechanisms. Finally, a technique for order preference by similarity to an ideal solution similarity ranking technique for multi -criteria decision making based on the entropy weight method (EW-TOM) is proposed to jointly optimize system latency and energy consumption. Simulation results show that the proposed scheme achieves more than 95% accuracy in filtering normal nodes, which is more than 20% better than other schemes. In addition, the latency and energy cost are better than other schemes, with an improvement of more than 6%.

关键词： Jamming   Task analysis   Computational modeling   Servers   Optimization   Data models   Resource management   Mobile edge computing   anti-jamming communications   hybrid hidden strategy   hierarchical reinforcement learning  

摘要： This letter investigates the scenario of resisting intelligent reactive jammer in mobile edge computing (MEC) networks. To solve the multi-dimensional offloading allocation problem while avoiding malicious jamming attacks, a hierarchical reinforcement learning (HRL) based hybrid hidden strategy is proposed to allocate multi-dimensional resources and reduce the probability of being perceived in the offloading process. Specifically, the HRL framework is divided into two layers for asynchronous update: the upper layer with the discrete policy network taking charge of outputting the channel access strategies to resist the dynamic jamming attack in the frequency domain, and the lower layer with the continuous policy network mainly offering hidden offloading strategies to avoid the detection of intelligent reactive jammers. By integrating the double deep Q-Network and twin delayed deep deterministic network, and exploiting their advantages in solving the hybrid strategy problems, the mobile device (MD) is able to execute effective hidden offloading strategies to reduce the computing cost as well as to significantly avoid intelligent reactive jamming in dynamic and unknown environments.

关键词： multiple UAVs   mobile edge computing   deep reinforcement learning   power inspection   offloading strategy   electric power IoT  

摘要： With the ongoing advancement of electric power Internet of Things (IoT), traditional power inspection methods face challenges such as low efficiency and high risk. Unmanned aerial vehicles (UAVs) have emerged as a more efficient solution for inspecting power facilities due to their high maneuverability, excellent line-of-sight communication capabilities, and strong adaptability. However, UAVs typically grapple with limited computational power and energy resources, which constrain their effectiveness in handling computationally intensive and latency-sensitive inspection tasks. In response to this issue, we propose a UAV task offloading strategy based on deep reinforcement learning (DRL), which is designed for power inspection scenarios consisting of mobile edge computing (MEC) servers and multiple UAVs. Firstly, we propose an innovative UAV-Edge server collaborative computing architecture to fully exploit the mobility of UAVs and the high-performance computing capabilities of MEC servers. Secondly, we established a computational model concerning energy consumption and task processing latency in the UAV power inspection system, enhancing our understanding of the trade-offs involved in UAV offloading strategies. Finally, we formalize the task offloading problem as a multi-objective optimization issue and simultaneously model it as a Markov Decision Process (MDP). Subsequently, we proposed a task offloading algorithm based on a Deep Deterministic Policy Gradient (OTDDPG) to obtain the optimal task offloading strategy for UAVs. The simulation results demonstrated that this approach outperforms baseline methods with significant improvements in task processing latency and energy consumption.

关键词： federated learning   mobile edge computing   dataset management  

摘要： Federated learning (FL) in mobile edge computing has emerged as a promising machine-learning paradigm in the Internet of Things, enabling distributed training without exposing private data. It allows multiple mobile devices (MDs) to collaboratively create a global model. FL not only addresses the issue of private data exposure but also alleviates the burden on a centralized server, which is common in conventional centralized learning. However, a critical issue in FL is the imposed computing for local training on multiple MDs, which often have limited computing capabilities. This limitation poses a challenge for MDs to actively contribute to the training process. To tackle this problem, this paper proposes an adaptive dataset management (ADM) scheme, aiming to reduce the burden of local training on MDs. Through an empirical study on the influence of dataset size on accuracy improvement over communication rounds, we confirm that the amount of dataset has a reduced impact on accuracy gain. Based on this finding, we introduce a discount factor that represents the reduced impact of the size of the dataset on the accuracy gain over communication rounds. To address the ADM problem, which involves determining how much the dataset should be reduced over classes while considering both the proposed discounting factor and Kullback-Leibler divergence (KLD), a theoretical framework is presented. The ADM problem is a non-convex optimization problem. To solve it, we propose a greedy-based heuristic algorithm that determines a suboptimal solution with low complexity. Simulation results demonstrate that our proposed scheme effectively alleviates the training burden on MDs while maintaining acceptable training accuracy.

关键词： Unmanned aerial vehicle   mobile edge computing   ultra-reliable and low-latency communications   computation latency  

摘要： In this paper, we consider an unmanned aerial vehicle (UAV)-enabled mobile edge computing (MEC) system, where multiple ground devices offload portions of their latency-sensitive and mission-critical computational tasks to a UAV-carried MEC server for remote computing and compute the remaining portions locally. To meet the low-latency requirements of the MEC, ultra-reliable and low-latency communication (URLLC) is used to offload tasks from the devices to the UAV. We minimize the maximum computation latency among all devices by jointly optimizing the computing times and CPU frequencies of the devices and the UAV, the offloading bandwidths of the devices, and the three-dimensional location of the UAV. We propose an algorithm that decomposes the joint optimization problem into three subproblems, which optimize the UAV's horizontal location, the UAV's altitude, and the offloading bandwidths and computing CPU frequencies, respectively. In solving the subproblems, the data rate expression of the devices' finite-blocklength offloading is accurately approximated by a tractable logarithmic function, and the successive convex approximation technique is applied to tackle the non-convex structure. Furthermore, a semi-closed-form solution to the subproblem that optimizes the bandwidths and CPU frequencies is derived to reduce the complexity. Simulation results show that the proposed algorithm can significantly reduce the system's computation latency compared to the benchmark schemes.

关键词： Computer architecture   Task analysis   Approximation algorithms   Mobile handsets   Servers   Mathematical models   Delays   Approximation algorithm   cloud radio access network   mobile edge computing   task offloading   user statisfaction  

摘要： With the continuous development of the communication service industry, users pay more attention to the quality of network service. Previous studies on offloading problems, especially in the Cloud Radio Access Network (C-RAN) architecture with Mobile Edge Computing (MEC), are primarily focused on the economic perspective, with little consideration given to user-oriented satisfaction problems. To fill this gap, this article proposes a mathematical model for maximizing user satisfaction in the C-RAN architecture with multi-layer MEC. The problem is divided into two stages for solution. The first stage addresses the optimal connection problem between users and Remote Radio Heads (RRHs). The second stage then schedules user tasks reasonably based on the solution obtained in the first stage. The two-stage problems are all proved to be NP-Hard. Two efficient approximation algorithms, namely User-to-RRH Association Algorithm (URAA) and Maximum Satisfaction Algorithm (MSA), are proposed to solve the problems in different stages. This article proves and analyzes the theoretical performance of the two algorithms. Finally, the performance of the proposed algorithms is verified by simulation experiments. The experimental results demonstrate that the two proposed algorithms can achieve reasonable solutions to the problems, and the user satisfaction level can be maintained at a high level.

关键词： Autonomous aerial vehicles   Sensors   Radar   Optimization   Precoding   Radar antennas   Energy consumption   Joint communication   sensing and computing   mobile edge computing   radar sensing   successive convex approximation   unmanned aerial vehicle  

摘要： In this paper, we propose a joint communication, sensing and computing (JCSC) framework for unmanned aerial vehicle (UAV) system that utilizes mobile edge computing (MEC) technology to offload radar sensing data to the ground for processing. To achieve optimal performance in sensing data processing, it is necessary to design the communication and radar precoding matrix of the UAV systems and optimize the location of the UAV system. The joint optimization model is established using the linear weighting method, which transforms the joint optimization problem into a single objective optimization problem. To solve this problem, the master problem is decomposed into two subproblems: one for precoding design in data offloading and the other for precoding design in radar sensing. Fractional programming and successive convex approximation (SCA) are used to transform each of these subproblems respectively into a convex problem, and two algorithms named "precoding design and location optimization for data offloading (PLDO)" and "precoding design and location optimization for radar sensing (PLRS)" are proposed to solve the optimization problem. The simulation results show that the proposed algorithm can greatly reduce the energy consumption of UAV compared to local processing, and it has a faster convergence speed.

关键词： Mobile edge computing   Deep reinforcement learning   Meta reinforcement learning   Virtual machine placement  

摘要： Mobile edge computing requires more and more high-performance servers, resulting in increased energy consumption. As an effective means to reduce energy consumption, virtual machine placement (VMP) has been widely studied. In the edge computing environment, as the number of terminal device requests continues to increase, the scale of VMP becomes larger and larger, and existing research algorithms may take a long time to converge. The reason is that as the number of VMs increases, the search space of the policy becomes larger and the agent needs to interact with the environment for a longer time to make the best decision. In addition, existing research methods only consider reducing energy consumption, rarely consider the response latency of virtual machines, and almost ignore the dynamic changes of the edge environment. To overcome these drawbacks, we propose a virtual machine placement algorithm based on meta-reinforcement learning, which consists of an inner and outer loop. The inner loop designs a deep reinforcement learning algorithm combined with the order exchange and migration mechanism to generate the best decision, and the outer loop provides meta-strategy parameters for the inner loop based on meta-learning to accelerate the convergence capability of the inner loop, thereby obtaining efficient virtual machine placement decisions quickly from a new environment. Through simulation experiments, we demonstrate that our approach effectively reduces the energy consumption of the edge server and the response latency of VMs at different problem sizes compared to the three baseline algorithms. At the same time, it quickly adapts to the new environment with only a small number of gradient updates.