关键词： Task analysis   Heuristic algorithms   System dynamics   Measurement   Data processing   Minimization   Servers   Age of information   mobile edge computing   post-decision state   deep reinforcement learning   RESOURCE-ALLOCATION   STATUS UPDATE   PEAK AGE   INFORMATION   COMPUTATION   OPTIMIZATION   NETWORKS   MANAGEMENT   TRADEOFF   QUEUE  

摘要： With the rapid development of Mobile Edge Computing (MEC), various real-time applications have been deployed to benefit people's daily lives. The performance of these applications relies heavily on the freshness of collected environmental information, which can be quantified by its Age of Information (AoI). In the traditional definition of AoI, it is assumed that the status information can be actively sampled and directly used. However, for many MEC-enabled applications, the desired status information is updated in an event-driven manner and necessitates data processing. To better serve these applications, we propose a new definition of AoI and, based on the redefined AoI, we formulate an online AoI minimization problem for MEC systems. Notably, the problem can be interpreted as a Markov Decision Process (MDP), thus enabling its solution through Reinforcement Learning (RL) algorithms. Nevertheless, the traditional RL algorithms are designed for MDPs with completely unknown system dynamics and hence usually suffer long convergence times. To accelerate the learning process, we introduce Post-Decision States (PDSs) to exploit the partial knowledge of the system's dynamics. We also combine PDSs with deep RL to further improve the algorithm's applicability, scalability, and robustness. Numerical results demonstrate that our algorithm outperforms the benchmarks under various scenarios.

关键词： Task analysis   Processor scheduling   Scheduling   System performance   Probability density function   Probability   Power system reliability   Mobile edge computing   user scheduling   computational access point selection   system outage probability   NOMA  

摘要： The advent of mobile edge computing (MEC) has effectively resolved the challenges associated with computation-intensive applications. In this letter, we investigate a multi-user MEC system with multiple computational access points (CAPs) in the presence of an eavesdropper. In contrast to existing works that only considered user scheduling or CAP selection, we propose a joint user scheduling and CAP selection policy, and derive the system outage probability under the proposed policy in closed form. Simulations are provided to validate the correctness of the theoretical analysis and demonstrate the effectiveness of the proposed policy. The results confirm that the proposed joint user scheduling and CAP selection policy outperforms various existing policies that independently consider user scheduling and CAP selection.

关键词： Computational modeling   Training   Data models   Costs   Privacy   Mobile computing   Degradation   Differential privacy   federated learning   lottery ticket hypothesis   mobile edge computing   zero -concentrated DP  

摘要： Federated learning (FL) can train collaboratively with several mobile terminals (MTs), which faces critical challenges in communication, resource, and privacy. Existing privacy-preserving methods usually adopt instance-level differential privacy (DP), which provides a rigorous privacy guarantee but with several bottlenecks: performance degradation, transmission overhead, and resource constraints. Therefore, we propose Fed-LTP, an efficient and privacy-enhanced FL framework with Lottery Ticket Hypothesis (LTH) and zero-concentrated DP (zCDP). It generates a pruned global model on the server side and conducts sparse-to-sparse training from scratch with zCDP on the client side. On the server side, two pruning schemes are proposed: (i) the weight-based pruning (LTH) determines the pruned global model structure; (ii) the iterative pruning further shrinks the size of the pruned model. Meanwhile, the performance of Fed-LTP is boosted via model validation based on the Laplace mechanism. On the client side, we use sparse-to-sparse training to solve the resource-constraints issue and provide tighter privacy analysis to reduce the privacy budget. We evaluate the effectiveness of Fed-LTP on several real-world datasets in both independent and identically distributed (IID) and non-IID settings. The results confirm the superiority of Fed-LTP over state-of-the-art (SOTA) methods in communication, computation, and memory efficiencies while realizing a better utility-privacy trade-off.

关键词： 5G/6G vehicular networks   Graph neural network   Mobile edge computing  

摘要： Mobile Edge Computing (MEC) is a key technology that emerged to address the increasing computational demands and communication requirements of vehicular networks. It is a form of edge computing that brings cloud computing capabilities closer to end-users, specifically within the context of vehicular networks, which are part of the broader Internet of Vehicles (IoV) ecosystem. However, the dynamic nature of traffic flows in MEC in 5G/6G vehicular networks poses challenges for accurate prediction and resource allocation when aiming to provide edge service for mobile vehicles. In this paper, we present a novel approach to predict the traffic flow of MEC in 5G/6G vehicular networks using graph-based learning. In our framework, MEC servers in vehicular networks are construed as nodes to construct a dynamic similarity graph and a dynamic transition graph over a duration of multiple days. We utilize Graph Attention Networks (GAT) to learn and fuse the node embeddings of these dynamic graphs. A transformer model is subsequently employed to predict the vehicle frequency accessing the edge computing services for the next day. Our experimental results have shown that the model achieves high accuracy in predicting edge service access volumes with low error metrics.

关键词： Mobile edge computing   Bi-level optimization   Nash equilibrium   Game theory   RESOURCE-ALLOCATION   COMPUTATION   ALGORITHM   OPTIMIZATION   GREEDY  

摘要： In the mobile edge computing system subject to wireless interference, the Edge Server Provider (ESP) aims to offer profitable computing resources to Device Managers (DMs), who make optimal strategies based on the provided prices. However, the existence of mixed variables typically constitutes an NP -hard problem, posing a significant challenge for optimization. In response to address this issue, we formulate a bi-level optimization problem, where the upper level is devoted to optimizing the pricing of computing resources. The lower level optimizes DM migration strategies and resource allocation at specified prices. Leveraging game theory, we achieve distributed and efficient computation offloading by formulating the distributed computing offloading strategy among lower -level DMs as a task offloading game. Our analysis identifies Nash equilibrium and finite improvement characteristics within the game. Based on these insights, we present a bi-level distributed computation offloading algorithm capable of reaching Nash equilibrium, thus optimizing the profit for both DMs and ESP. The experiments have demonstrated the algorithm's effectiveness in reducing system costs and maximizing the profits of ESP with DMs across diverse scenarios.

关键词： Delays   Multitasking   Internet of Things   Artificial neural networks   Task analysis   Computational modeling   Inference algorithms   Asynchronous advantage actor-critic   distributed DNN inference   mobile edge computing   model partitioning   multi-task learning   RESOURCE-ALLOCATION   AWARE   ACCELERATION  

摘要： Model partitioning is a promising technique for improving the efficiency of distributed inference by executing partial deep neural network (DNN) models on edge servers (ESs) or Internet-of-Things (IoT) devices. However, due to heterogeneous resources of ESs and IoT devices in mobile edge computing (MEC) networks, it is non-trivial to guarantee the DNN inference speed to satisfy specific delay constraints. Meanwhile, many existing DNN models have a deep and complex architecture with numerous DNN blocks, which leads to a huge search space for fine-grained model partitioning. To address these challenges, we investigate distributed DNN inference with fine-grained model partitioning, with collaborations between ESs and IoT devices. We formulate the problem and propose a multi-task learning based asynchronous advantage actor-critic approach to find a competitive model partitioning policy that reduces DNN inference delay. Specifically, we combine the shared layers of actor-network and critic-network via soft parameter sharing, and expand the output layer into multiple branches to determine the model partitioning policy for each DNN block individually. Experiment results demonstrate that the proposed approach outperforms state-of-the-art approaches by reducing total inference delay, edge inference delay and local inference delay by an average of 4.76%, 10.04% and 8.03% in the considered MEC networks.

关键词： Resource management   Optimization   Task analysis   NOMA   Vehicle dynamics   Servers   Energy consumption   Mobile edge computing   non-orthogonal multiple access   resource allocation   Lyapunov optimization   vehicular networks   RESOURCE-ALLOCATION   POWER ALLOCATION   MANAGEMENT   FRAMEWORK  

摘要： Mobile Edge Computing (MEC) has envisioned to be a promising technology to provide more efficient services for computation-intensive but delay-sensitive onboard mobile services. In this paper, the Non-Orthogonal Multiple Access (NOMA) technology is applied in a vehicular edge computing network, in which vehicular users (VUs) can offload partial computation tasks to MEC servers over wireless channels for remote execution. In this network, an optimization problem for the long-term energy consumption of the system is presented and aims to minimize it by jointly optimizing the Successive Interference Cancellation (SIC) ordering of NOMA, the VUs' transmit power for computation offloading, and computation resource allocation of the MEC server. To deal with the intractable long-term optimization problem, we first transform it into an equivalent instantaneous form based on the Lyapunov optimization theory. Since the transformed problem is still highly non-convex, we further decompose it into the interactive resource allocation and SIC ordering sub-problems. For the resource allocation sub-problem, we exploit its convexity through the transformation and reparameterization, and derive the optimal solution in accordance with the Karush-Kuhn-Tucker (KKT) conditions and the gradient descent algorithm. After that, we propose a low-complexity algorithm by leveraging the Tabu search to obtain the sub-optimal SIC ordering. Simulation results validate the effectiveness of the proposed algorithm and the superiority of NOMA compared to Frequency Division Multiple Access (FDMA).

关键词： Mobile edge computing  

摘要： Edge computing places the computational services and resources closer to the user proximity, to reduce latency, and ensure the quality of service and experience. Low latency, context awareness and mobility support are the major contributors to edge-enabled smart systems. Such systems require handling new situations and change on the fly and ensuring the quality of service while only having access to constrained computation and communication resources and operating in mobile, dynamic and ever-changing environments. Hence, adaptation and self-organisation are crucial for such systems to maintain their performance, and operability while accommodating new changes in their *** article reviews the current literature in the field of adaptive edge computing systems. We use a widely accepted taxonomy, which describes the important aspects of adaptive behaviour implementation in computing systems. This taxonomy discusses aspects such as adaptation reasons, the various levels an adaptation strategy can be implemented, the time of reaction to a change, categories of adaptation technique and control of the adaptive behaviour. In this article, we discuss how these aspects are addressed in the literature and identify the open research challenges and future direction in adaptive edge computing *** results of our analysis show that most of the identified approaches target adaptation at the application level, and only a few focus on middleware, communication infrastructure and context. Adaptations that are required to address the changes in the context, changes caused by users or in the system itself are also less explored. Furthermore, most of the literature has opted for reactive adaptation, although proactive adaptation is essential to maintain the edge computing systems' performance and interoperability by anticipating the required adaptations on the fly. Additionally, most approaches apply a centralised adaptation control, which does not perfectly fit the mo

关键词： Mobile edge computing  

摘要： As mobile devices increasingly become focal points for advanced applications, edge computing presents a viable solution to their inherent computational limitations, particularly in deploying large language models (LLMs). However, despite the advancements in edge computing, significant challenges remain in efficient training and deploying LLMs due to the computational demands and data privacy concerns associated with these models. This paper explores a collaborative training framework that integrates mobile users with edge servers to optimize resource allocation, thereby enhancing both performance and efficiency. Our approach leverages parameter-efficient fine-tuning (PEFT) methods, allowing mobile users to adjust the initial layers of the LLM while edge servers handle the more demanding latter layers. Specifically, we formulate a multi-objective optimization problem to minimize the total energy consumption and delay during training. We also address the common issue of instability in model performance by incorporating stability enhancements into our objective function. Through novel fractional programming technique, we achieve a stationary point for the formulated problem. Simulations demonstrate that our method reduces the energy consumption as well as the latency, and increases the reliability of LLMs across various mobile settings. Copyright © 2024, The Authors. All rights reserved.

关键词： game theory   mobile edge computing   Nash equilibrium   space-air-ground integrated network   task offloading   TERRESTRIAL NETWORKS   RESOURCE-ALLOCATION   SATELLITE   ENERGY   OPTIMIZATION   DESIGN   QOS  

摘要： Due to geographical factors, it is impossible to build large-scale communication network infrastructure in remote areas, which leads to poor network communication quality in these areas and a series of delay-sensitive tasks cannot be timely processed and responded. Aiming at the problem of limited coverage in remote areas, the space-air-ground integrated networks (SAGIN) combined with mobile edge computing (MEC) can provide low latency and high reliability transmission for offloading delay-sensitive tasks for users in remote areas. Considering the strong limitation of satellite resources in the space-ground integrated network and insufficient energy of local user equipment, firstly, a satellite-UAV cluster-ground three-layer edge computing network architecture is proposed in this paper. Under the condition that the delay requirements of various ground tasks are met, the task offloading problem is transformed into a Stackelberg game between ground user equipment and edge servers. In addition, it is proved that the existence of Nash equilibrium in non-cooperative game between ground user equipment by using potential game. Finally, a Nash equilibrium iterative offloading algorithm based on Stackelberg game (NEIO-SG) is proposed to find the optimal offloading strategy for tasks to minimize the system offloading cost and the optimal forwarding percentage strategy for offloading tasks to maximize the utility function of the edge server. Simulation results show that compared to other baseline algorithms, NEIO-SG reduces the total system latency during task offloading by about 13%$$ \% $$ and the energy consumption of the edge server by about 35%$$ \% $$.