关键词： mobile edge computing   mobility   service migration   deep reinforcement learning   deep deterministic policy gradient   RESOURCE-ALLOCATION   SERVICE MIGRATION   OPTIMIZATION   ALGORITHM   NETWORKS  

摘要： Mobile Edge Computing (MEC) is a promising approach. Dynamic service migration is a key technology in MEC. In order to maintain the continuity of services in a dynamic environment, mobile users need to migrate tasks between multiple servers in real time. Due to the uncertainty of movement, frequent migration will increase delays and costs and non-migration will lead to service interruption. Therefore, it is very challenging to design an optimal migration strategy. In this paper, we investigate the multi-user task migration problem in a dynamic environment and minimizes the average service delay while meeting the migration cost. In order to optimize the service delay and migration cost, we propose an adaptive weight deep deterministic policy gradient (AWDDPG) algorithm. And distributed execution and centralized training are adopted to solve the high-dimensional problem. Experiments show that the proposed algorithm can greatly reduce the migration cost and service delay compared with the other related algorithms.

关键词： Mobile edge computing   Computation offloading   Multi-agent deep reinforcement learning   Energy harvesting   Resource allocation   Non-orthogonal multiple access  

摘要： Supporting latency-sensitive and computation-intensive applications is hardly possible for mobile devices (MDs) with limited battery capacity and low computing resources. Therefore, mobile edge computing (MEC) and wireless power transfer (WPT) have emerged as promising technologies that allow MDs to offload part or all of their workloads to MEC servers and harvest energy for prolonging their battery lifetime. However, the MEC server’s limited computing resources, available communication channel quality, and time-limited energy harvesting (EH) challenge the computation offloading. In this paper, we study the joint problem of decentralized computation offloading and resource allocation (JDCORA) in the environment of non-orthogonal multiple access (NOMA)-assisted MEC with multiple EH-enabled MDs. To learn decentralized offloading policies, we propose a multi-agent deep reinforcement learning (MADRL)-based scheme towards minimizing energy consumption and task completion time, which considers the cooperation between MDs to adjust their strategies. In particular, we improve the multi-agent deep deterministic policy gradient (MADDPG) by applying the features of double actors, double centralized critics, soft value estimation, critic regularization and proportional-based prioritized experience replay (pPER) and propose an algorithm called multi-agent twin actors regularized critics (MATARC). Simulation results demonstrate that the proposed MATARC has a better convergence performance compared to other baseline methods and also reduces the average energy consumption, task completion time and task drop rate.

关键词： Autonomous driving   Task offloading   Mobile edge computing  

摘要： With the increasing complexity of autonomous driving tasks, the computational demands on single vehicular computing units have escalated, more and more tasks need to be offloaded to the edge. These tasks vary in latency sensitivity: real-time tasks, critical for passenger safety, require strict deadline adherence, whereas the latency of standard tasks mainly affects the user experience and has more flexible constraints. Addressing the challenge of selecting suitable edge computing nodes to enhance the offloading success rate of real-time tasks amidst a vast and heterogeneous cluster becomes crucial. This paper introduces the adaptive priority-based hierarchical task offloading (APHTO) algorithm, which optimizes task offloading strategies by accounting for the diverse latency constraints of different task types. Experiments demonstrate that under optimal performance conditions, APHTO significantly outperforms existing algorithms such as Min–Min, Max–Min, CUS, and FMS in reducing task latency by 20.31%, increasing offloading success rates by 35.83%, and improving resource utilization by 30.21%, marking a substantial advancement in task offloading strategies for autonomous driving integrated with MEC.

关键词： Mobile edge computing   Dung beetle optimizer   Sequential game   Deep reinforcement learning   NOMA   ALLOCATION   NETWORKS  

摘要： The current technologies such as computer vision and artificial intelligence are continuously being applied to disaster rescue robots, enabling their adaptability to different environments. However, the challenge arising from this is that these robots will generate computationally intensive tasks, constrained by the size and energy supply of the robots. This limitation leads to a reduction in device endurance and an increase in processing latency, becoming a bottleneck for the application of artificial intelligence technologies. To address this concern, we propose a novel edge offloading system based on unmanned aerial vehicles (UAVs) and mobile edge service vehicles (MESVs) to assist with communication and computation tasks, aiming to minimize the overall computational burden of rescue robots in extreme scenarios where communication infrastructure is unavailable due to sudden natural disasters or accidents. Leveraging the high maneuverability of UAVs, we utilize an improved dung beetle optimization (DBO) algorithm to optimize the UAV's flight positions and assist robots in task processing and data relay. Moreover, we address the optimization problem of robot offloading strategies by employing a deep reinforcement learning-based Twin Delayed Deep Deterministic Policy Gradient (TD3) algorithm, which effectively reduces processing latency and energy consumption. Through simulation experiments and real -world simulation tests, our proposed method, compared to other UAV position optimization algorithms and computational strategy offloading methods, can effectively reduce the computational energy consumption of robots. In comparison with other methods presented in this paper, it can also reduce latency by nearly 40%.

关键词： Mobile edge computing   Trust management   Edge caching   Blockchain   Deep reinforcement learning  

摘要： Mobile edge computing (MEC) has introduced content edge caching to offload network backhaul and improve user quality of experience by transferring frequently requested content to edge nodes near end-users. However, the implementation of edge caching is fraught with significant security and confidentiality issues due to the lack of trust between the different entities involved in the caching process. Most existing solutions for managing trust in edge caching only consider the trust between caching edge servers and ignore the trust between edge servers and content providers. This paper aims to fill this gap by proposing a trust management-based edge caching system that exploits the potential of blockchain technology to secure the caching process by enabling trust between content providers and edge servers. The proposed trust management system establishes trust in both directions of the interaction, providing higher confidence between the caching actors. Our trust system is based on a direct trust score calculated using the subjective three value logic scheme (3VSL). This is complemented by a reputation-based indirect trust score managed by the blockchain. A deep reinforcement learning (DRL) algorithm is proposed to efficiently make secure and intelligent caching decisions. The simulation results show that the proposed blockchain-based trust management system is effective and guarantees a good level of security and confidentiality for edge caching. They also demonstrate that our DRL caching algorithm provides secure and intelligent caching decisions in MEC, outperforming the baseline edge caching schemes.

关键词： Mobile edge computing   5G   Edge server placement   Load balancing   Heterogeneous capacities   Knapsack problem   OPTIMIZATION   MAXIMIZATION   DEPLOYMENT  

摘要： The rapid proliferation of low-latency and high-bandwidth applications has brought edge computing to the forefront of mobile network architectures. However, the strategic placement of edge servers plays a vital role in balancing price-performance trade-offs significantly. Existing works addressing the Edge Server Placement Problem have assumed homogeneous computational capabilities across ESs, which is not a pragmatic assumption considering variations in user densities and workload fluctuations across typical cityscapes. This work proposes a solution to the Edge Server Placement Problem with heterogeneous ES capacities and introduces a novel scheme to evaluate the workload of ESs for 5G networks. Additionally, this paper also proposes a novel Knapsack-based Metaheuristic for allocating base stations to edge servers, turning the Edge Server Placement Problem into a 0-1 Knapsack problem. Experimental evaluation using popular 5G traffic demand datasets has found that the proposed approach improves workload balance by 40.79%, utilisation rates by 57.58%, and reduces energy consumption by 44.68% approximately vis-a-vis homogeneous counterparts.

关键词： Mobile edge computing   Service migration   Hybrid learning   Deep reinforcement learning   Multi-user decisions   TASK MIGRATION   5G  

摘要： Mobile edge computing, a new computing model, aims to reduce network latency and improve user experience by placing computing resources closer to the users. However, when users move away from the current service areas, the quality of services will decline. To guarantee the quality of services, service migration technique is proposed and studied. Due to the wireless resource contention among users, there is service migration decisionmaking coupling among multiple users. Meanwhile, making real-time service migration decisions for multiple users to minimize service cost with application deadlines remains an open challenge because of the huge state space. In this paper, we formalize the service migration problem with deadlines as a Markov decision process (MDP) and propose a hybrid learning based service migration strategy. The proposed approach integrates Monte Carlo method and DQN learning, which adopts online training without Q-values and neural network approximation to generate service migration decisions for multiple users. Experimental results demonstrate that the proposed hybrid learning based service migration strategy outperforms the existing migration strategies in terms of service success ratio and service cost.

关键词： Mobile edge computing   Non-orthogonal multiple access   Privacy outage probability   Resource allocation   Partial offload   NONORTHOGONAL MULTIPLE-ACCESS   EFFICIENCY   NETWORKS   COMPUTATION  

摘要： Mobile edge computing can use the wireless access network to provide the services required by telecom users and cloud computing functions nearby, thereby creating a service environment with high performance, low latency and high bandwidth. Computational task offloading is a core issue in mobile edge computing. In order to improve the security when computing tasks are partially offloaded in non-orthogonal multiple access-based mobile edge computing systems, this paper studies the physical layer security of MEC networks considering the presence of eavesdroppers. We employ the secrecy outage probability to measure the secrecy performance of computation offloading. Under the premise of comprehensively considering the joint constraints of transmit power, local task calculation and confidentiality outage probability, we introduce the energy consumption weight factor to balance the transmission energy consumption and calculation energy consumption, and finally achieve the minimum weighted sum of system energy consumption. On the premise of satisfying two user priorities, in order to reduce system overhead, we design a joint task offloading and resource allocation mechanism. The mechanism uses binary search-based iterative optimization algorithm to seek the optimal solution after problem transformation, and obtain optimal task offloading and power allocation. A large number of simulation results show that the algorithm proposed in this paper can effectively reduce system energy consumption.

关键词： Mobile edge computing   unmanned aerial vehicles   service caching and task offloading   online algorithm   machine learning  

摘要： Unmanned Aerial Vehicles (UAVs) have gained increasing attention by both academic and industrial communities, due to their flexible deployment and efficient line-of-sight communication. Recently, UAVs equipped with base stations have been envisioned as a key technology to provide 5G network services for mobile users. In this article, we provide timely services on the data streams of mobile users in a UAV-aided Mobile Edge Computing (MEC) network, in which each UAV is equipped with a 5G small-cell base station for communication and data processing. Specifically, we first formulate a flow-time minimization problem by jointly caching services and offloading tasks of mobile users to the UAV-aided MEC with the aim to minimize the flow time, where the flow time of a user request is referred to the time duration from the request issuing time point to its completion point, subject to resource and energy capacity on each UAV. We then propose a spatial-temporal learning optimization framework. We also devise an online algorithm with a competitive ratio for the problem based upon the framework, by leveraging the round-robin scheduling and dual fitting techniques. Finally, we evaluate the performance of the proposed algorithms through experimental simulation. The simulation results demonstrate that the proposed algorithms outperform their comparison counterparts, by reducing the flow time no less than 19% on average.

关键词： Mobile edge computing   integrity   data right confirmation   consensus   merkle tree