关键词： Cloud-edge collaboration   mobile edge computing   master-slave game model   calculate unload   computing resource allocation  

摘要： Aiming at the problem of long task calculation delay of traditional edge computing offloading strategy (COS), a joint caching and COS for power supply chain services under cloud edge collaborative computing environment is proposed by introducing the master-slave game model. First, for the network of information interaction between power grid system and electrical equipment suppliers, the corresponding overall model of cloud edge collaborative network is constructed by combining cloud computing, edge computing, supplier terminal communication and remote cloud. Then, the cloud-side collaborative network model is analyzed and solved by building the service function cache model, task unloading and data transmission model and power business data calculation model, respectively. Finally, by introducing the master-slave game model, the whole system is divided into three different interests of power service providers, communication providers and communication network terminal providers and the joint caching and COS for power supply chain services is given. Through simulation experiments, the proposed computing unloading strategy and the other three methods are compared and analyzed under the same experimental conditions. The results show that the service cache hit rate of the proposed method is the highest, reaching 96.35%. Compared with the other three comparison methods, it increased by 14.96%, 11.24% and 4.85%, respectively. In addition, its task computing delay is the least affected by agent computing power and agent computing resources, and its performance is better than the other three comparison methods.

关键词： Artificial intelligence   Class estimation   Data imbalance   Distributed machine learning   Federated learning   Edge devices   Mobile edge computing   Privacy preserving algorithms  

摘要： In Mobile Edge Computing, the framework of federated learning can enable collaborative learning models across edge nodes, without necessitating the direct exchange of data from edge nodes. It addresses significant challenges encompassing access rights, privacy, security, and the utilization of heterogeneous data sources over mobile edge computing. Edge devices generate and gather data, across the network, in non-IID (independent and identically distributed) manner leading to potential variations in the number of data samples among these edge networks. A method is proposed to work in federated learning under edge computing setting, which involves AI techniques such as data augmentation and class estimation and balancing during training process with minimized computational overhead. This is accomplished through the implementation of data augmentation techniques to refine data distribution. Additionally, we leveraged class estimation and employed linear regression for client-side model training. This strategic approach yields a reduction in computational costs. To validate the effectiveness of the proposed approach, it is applied to two distinct datasets. One dataset pertains to image data (FashionMNIST), while the other comprises numerical and textual data concerning stocks for predictive analysis of stock values. This approach demonstrates commendable performance across both dataset types and approaching more than 92% of accuracy in the paradigm of federated learning.

关键词： Mobile edge computing   Computation offloading   System-centric energy efficiency   Latency   Optimization  

摘要： Mobile edge computing (MEC) enables computation tasks of end users to be offloaded and processed at nearby edge servers, which is a promising technology that can effectively alleviate end users' computing pressure. However, how to balance two important metrics for MEC computation offloading, i.e., latency and energy efficiency, is still a key challenge. In this paper, we investigate joint computation offloading and resource allocation for latency-sensitive applications in order to maximize system-centric computation energy efficiency (SCEE) in MEC systems. An optimization problem is formulated to maximize the sum computation energy efficiency of all users which contains the energy consumption at both user and edge server sides. Meanwhile, task completion latency constraints for all users are considered. We decompose the original optimization problem into two tractable subproblems. The optimal computation offloading as well as communication and computing resource allocation solutions are obtained by solving these subproblems. Furthermore, a hierarchical system-centric energy efficient computation offloading and resource allocation (SEECR) algorithm is developed. Numerical results show that compared with the benchmark schemes, the proposed algorithm can effectively improve the computation energy efficiency performance while guaranteeing the latency requirements of all users.

关键词： computing resource allocation   mobile edge computing   satellite-terrestrial networks   task offloading decision  

摘要： Mobile edge computing(MEC)-enabled satellite-terrestrial networks(STNs)can provide Internet of Things(IoT)devices with global computing ***,the network state information is uncertain or *** deal with this situation,we investigate online learning-based offloading decision and resource allocation in MEC-enabled STNs in this *** problem of minimizing the average sum task completion delay of all IoT devices over all time periods is *** decompose this optimization problem into a task offloading decision problem and a computing resource allocation problem.A joint optimization scheme of offloading decision and resource allocation is then proposed,which consists of a task offloading decision algorithm based on the devices cooperation aided upper confidence bound(UCB)algorithm and a computing resource allocation algorithm based on the Lagrange multiplier *** results validate that the proposed scheme performs better than other baseline schemes.

关键词： Mobile edge computing   STAR-RISs   Multi-cell communication systems   Secure communications   MADRL   DDPG  

摘要： The poor propagation environment between user equipments (UEs) and mobile edge computing (MEC) servers may lead to UEs' high energy consumptions and task delay. Reconfigurable intelligent surface (RIS) can significantly improve the quality of wireless propagation environment by intelligently adjusting the phase-shifts and amplitudes of its reflecting elements. However, the passive or active RISs can only reflect signals, i.e., the transmitters and receivers must be located on the same side of RISs. Simultaneous transmitting and reflecting (STAR) RIS, which can transmit and reflect incident signals simultaneously, has been recognized a promising solution to solve the above problem. For STAR-RIS-aided multi-cell secure communications MEC systems, we first formulate an optimization problem to minimize the sum of weighted delay and energy consumptions of all UEs, by jointly optimizing the phase-shifts and amplitudes of STAR-RISs and system spectrum and computation resource allocations. Second, under the practical coupled phase-shift model of STAR-RIS, we develop a multi-agent deep reinforcement learning (MADRL) based scheme, where the access point (AP) in each cell acts an agent to decide phase-shifts, amplitudes, computation task offloading proportions, etc, for the STAR-RIS and UEs within its coverage based on deep deterministic policy gradient (DDPG), by taking into account interference among multiple cells and possible information leakage of UEs. Finally, we verify and evaluate the performance of the proposed scheme through extensive simulations. The simulation results show that our proposed MADRL-based scheme outperforms the existing benchmarks by employing multiple agents and considering co-channel interference among cells. Moreover, simulation results also show that the performance of our developed scheme can indeed be improved by using STAR-RISs.

关键词： Mobile edge computing  

摘要： With the development of communication and computation technologies, mobile edge computing (MEC) has become an emerging technology being deployed in 5G/B5G mobile networks to improve the mobile user experience. In this paper, we study the asymptotic performance of parallel computing in MEC networks with massive multi-input and multi-output (MIMO) in millimeter wave band. First, a massive MIMO enabled MEC network with a parallel-limited computing model is constructed, and the task completed probability is defined as a new network performance metric. In the proposed model, we derive the task offloading and computing probabilities via the probability generation functional of the Poisson point process, from which, the task completed probability is obtained. Then we propose a computing resource allocation optimization algorithm to maximize the task completed probability within a given delay constraint. The numerical simulation results prove the accuracy of the obtained theoretical results and verify the outperformance of the proposed algorithm. © 2024, The Authors. All rights reserved.

关键词： Mobile edge computing   Unmanned aerial vehicle   Multi-agent deep reinforcement learning   Trajectory optimization   Offloading strategy  

摘要： As a new network architecture, the air-ground cooperative network is a support for future 6G network to achieve ubiquitous connectivities. To effectively relieve the computational pressure of massive data in 6G wireless networks, Unmanned Aerial Vehicles (UAVs) equipped with Mobile Edge Computing (MEC) servers have become an emerging technology that provides computing resources for Mobile Devices (MDs). Due to limited on-board energy and computational capabilities, this paper investigates a multi-UAV collaborative assisted MEC architecture. The optimization problem of minimizing the total computational cost is constructed by jointly optimizing the UAVs trajectories and MDs offloading strategies scheduling. The coupling between the optimization variables and the non-convexity of the problem can make it difficult to solve directly. To address the above concerns, the non-convex optimization problem is converted into a Markov decision process. The UAVs-assisted Offloading Strategy based on Reinforcement Learning (UOS-RL) algorithm is proposed to address the convergence difficulties caused by the high-dimensional continuous action space. Furthermore, the experience data generated by agents interacting with the environment is highly differentiated due to the highly dynamic variation of the environment. Hence, a Priority Experience Replay (PER) mechanism is proposed to improve the training efficiency of the UOS-RL algorithm based on the priority of the experience data. Simulation results show that the proposed PER-UOS-RL algorithm outperforms the existing works in terms of the computational cost.

关键词： Deep Reinforcement Learning   Mobile Edge Computing   Multi-user   Collaborative inference   Hybrid action space  

摘要： This study tackles the problem of increasing efficiency and scalability in deep neural network (DNN) systems by employing collaborative inference, an approach that is gaining popularity because to its ability to maximize computational resources. It involves splitting a pre-trained DNN model into two parts and running them separately on user equipment (UE) and edge servers. This approach is advantageous because it results in faster and more energy-efficient inference, as computation can be offloaded to edge servers rather than relying solely on UEs. However, a significant challenge of collaborative belief is the dynamic coupling of DNN layers, which makes it difficult to separate and run the layers independently. To address this challenge, we proposed a novel approach to optimize collaborative inference in a multi-agent scenario where a single-edge server coordinates the assumption of multiple UEs. Our proposed method suggests using an autoencoder-based technique to reduce the size of intermediary features and constructing tasks using the deep policy inference Q-inference network's overhead (DPIQN). To optimize the collaborative inference, employ the Deep Recurrent Policy Inference Q-Network (DRPIQN) technique, which allows for a hybrid action space. The results of the tests demonstrate that this approach can significantly reduce inference latency by up to 56% and energy usage by up to 72% on various networks. Overall, this proposed approach provides an efficient and effective method for implementing collaborative inference in multi-agent scenarios, which could have significant implications for developing DNN systems.

关键词： mobile edge computing   UAV path planning   task offloading   communication   collision avoidance  

摘要： Due to CPU and memory limitations, mobile IoT devices face challenges in handling delay-sensitive and computationally intensive tasks. Mobile edge computing addresses this issue by offloading tasks to the wireless network edge, reducing latency and energy consumption. UAVs serve as auxiliary edge clouds, providing flexible deployment and reliable wireless communication. To minimize latency and energy consumption, considering the limited resources and computing capabilities of UAVs, a multi-UAV and multi-edge cloud system was deployed for task offloading and UAV trajectory optimization. A joint optimization model for computing task offloading and UAV trajectory was proposed. During model training, a UAV communication mechanism was introduced to address potential coverage issues for mobile user devices through multiple UAVs or complete coverage. Considering the fact that decisions made by UAVs during trajectory planning may lead to collisions, a MADDPG algorithm with an integrated safety layer was adopted to obtain the safest actions closest to the joint UAV actions under safety constraints, thereby avoiding collisions between UAVs. Numerical simulation results demonstrate that the optimization method based on safety reinforcement learning considering communication cost outperforms other optimization methods. Communication between UAVs effectively addresses the issue of redundant or incomplete coverage for mobile user devices, reducing computation latency and energy consumption for task offloading. Additionally, the introduction of safety reinforcement learning effectively avoids collisions between UAVs.

关键词： Computational Offloading   Internet of vehicles   Task management   Deep reinforcement learning   Decentralized Framework   Mobile edge computing  

摘要： This paper focuses on the problem of computation offloading in a high-mobility Internet of Vehicles (IoVs) environment. The goal is to address the challenges related to latency, energy consumption, and payment cost requirements. The approach considers both moving and parked vehicles as fog nodes, which can assist in offloading computational tasks. However, as the number of vehicles increases, the action space for each agent grows exponentially, posing a challenge for decentralised decision-making. The dynamic nature of vehicular mobility further complicates the network dynamics, requiring joint cooperative behaviour from the learning agents to achieve convergence. The traditional deep reinforcement learning (DRL) approach for offloading in IoVs treats each agent as an independent learner. It ignores the actions of other agents during the training process. This paper utilises a cooperative three-layer decentralised architecture called Vehicle-Assisted Multi-Access Edge Computing (VMEC) to overcome this limitation. The VMEC network consists of three layers: the fog, cloudlet, and cloud layers. In the fog layer, vehicles within associated Roadside Units (RSUs) and neighbouring RSUs participate as fog nodes. The middle layer comprises Mobile Edge Computing (MEC) servers, while the top layer represents the cloud infrastructure. To address the dynamic task offloading problem in VMEC, the paper proposes using a Decentralized Framework of Task and Computational Offloading (DFTCO), which utilises the strength of MADRL and NOMA techniques. This approach considers multiple agents making offloading decisions simultaneously and aims to find the optimal matching between tasks and available resources.