关键词： Mobile edge computing   UAV   radar sensing   radar sensing   data compression   data compression   multi-agent deep reinforcement learning   multi-agent deep reinforcement learning   multi-agent deep reinforcement learning  

摘要： In this paper, we propose a multi-unmanned aerial vehicle (UAV)-assisted integrated sensing, communication, and computation network. Specifically, the treble-functional UAVs are capable of offering communication and edge computing services to mobile users (MUs) in proximity, alongside their target sensing capabilities by using multi-input multi-output arrays. For the purpose of enhance the computation efficiency, we consider task compression, where each MU can partially compress their offloaded data prior to transmission to trim its size. The objective is to minimize the weighted energy consumption by jointly optimizing the transmit beamforming, the UAVs' trajectories, the compression and offloading partition, the computation resource allocation, while fulfilling the causal-effect correlation between communication and computation as well as adhering to the constraints on sensing quality. To tackle it, we first reformulate the original problem as a multi-agent Markov decision process (MDP), which involves heterogeneous agents to decompose the large state spaces and action spaces of MDP. Then, we propose a multi-agent proximal policy optimization algorithm with attention mechanism to handle the decision-making problem. Simulation results validate the significant effectiveness of the proposed method in reducing energy consumption. Moreover, it demonstrates superior performance compared to the baselines in relation to resource utilization and convergence speed.

关键词： Mobile edge computing   trajectory planning   task offoading   MIMO AAV   Mobile edge computing   trajectory planning   task offoading   MIMO AAV  

摘要： Edge computing is conducive to reducing service response time and improving service quality by pushing cloud functions to a network's edges. Most existing works in edge computing focus on utility maximization of task offloading on static edges with a single antenna. Besides, trajectory planning of mobile edges, e.g., autonomous aerial vehicles (AAVs) is also rarely discussed. In this paper, we are the first to jointly discuss the deadline-ware task offloading and AAV trajectory planning problem in a multi-input multi-output (MIMO) AAV-aided mobile edge computing system. Due to discrete variables and highly coupling nonconvex constraints, we equivalently convert the original problem into a more solvable form by introducing auxiliary variables. Next, a penalty dual decomposition-based algorithm is developed to achieve a global optimal solution to the problem. Besides, we proposed a profit-based fireworks algorithm in a relatively lower time to reduce the execution time for large-scale networks. Extensive evaluation results reveal that our proposed optimal algorithms could significantly outperform static offloading algorithms and other algorithms by 25% on average.

关键词： Mobile edge computing   blockchain   task- offloading   task- offloading   multi-armed bandits   multi-armed bandits  

摘要： As mobile devices proliferate and mobile applications diversify, Mobile Edge Computing (MEC) has become widely adopted to efficiently allocate computing resources at the network edge and alleviate network congestion. In the MEC initial phase, the absence of vital information presents challenges in devising task-offloading policies, and identifying malicious devices responsible for providing inaccurate feedback is complex. To fill in such gaps, we introduce a consortium blockchain-enabled Committee Voting based Task Offloading Model (CVTOM) to collaboratively formulate resource allocation policies and establish deterrence against malicious servers producing erroneous results intentionally. Different voting principle mechanisms of each committee member are first designed in a Blockchain-enabled system which helps to represent the system's resource status. Additionally, we propose a Multi-armed Bandits related Thompson Sampling based Adaptive Preference Optimization (TSAPO) algorithm for task-offloading policy, enhancing the timely identification of potent edge servers to improve computing resource utilization which first considers dynamic edge server space and parallel computing scenarios. The solid proof process greatly contributes to the theoretical analysis of the TSAPO. The simulation experiments demonstrate the delay and budget can be reduced by around 25% and 10% respectively, showcasing the superior performance of our approach.

关键词： Autonomous aerial vehicles   Trajectory   Trajectory planning   Games   Costs   Computational modeling   Resource management   Planning   Benchmark testing   Time-frequency analysis   Mobile edge computing   autonomous aerial vehicle   Stackelberg game   deep reinforcement learning  

摘要： Nowadays, Mobile Edge Computing (MEC) has been widely deployed to enhance the computational capabilities of mobile devices. However, the geographic location of MEC servers is usually fixed. In order to provide flexible edge computing services, some works have considered integrating Autonomous aerial vehicles (AAVs) into MEC networks. In the context of AAV-assisted edge computing, there usually exist multiple AAVs and users, and each AAV may aim to maximize its profit by providing computing services, while users will decide which AAVs to utilize based on their preferences. In this context, how AAVs and users effectively plan their trajectories becomes particularly important as it will affect the profitability of AAVs and the user experience. Since the trajectories of AAVs and users are affected by each other, we model the trajectories of AAVs and users as a Stackelberg game, and then design trajectory planning strategies for users and AAVs based on Independent Proximal Policy Optimization (IPPO) and Proximal Policy Optimization (PPO) respectively, aiming to maximize AAVs' profits while ensuring user acceptance of AAV services. Finally, we evaluate the proposed trajectory planning strategy against three typical benchmark strategies using synthetic and realistic datasets. The experimental results demonstrate that our strategy can outperform benchmark strategies in terms of AAV profit while guaranteeing users' service experience.

关键词： Servers   Streaming media   Data models   Training   Federated learning   Backhaul networks   Adaptation models   Mobile edge computing   federated learning   reinforcement learning   adaptive bitrate streaming   content caching  

摘要： The explosive growth of mobile data traffic generated primarily from resource-hungry video streaming sessions challenges the service providers to deliver a better Quality of Service to the heterogeneous end-users. Content caching in Edge Computing is a promising solution to cope with this exponential rise in video traffic. The most popular videos are typically stored in the local caches of edge servers to provide fast and continuous access to videos. However, various edge caching strategies fail to cope with the dynamic request patterns of the users. Most learning-based caching models are generally trained in a centralized way, which overconsumes the network resources during training and transmission of video requests. Therefore, we propose a Federated Learning-based Reinforcement Learning caching framework called FedCache in this work. In FedCache, the training is decentralised on the end-user devices with its local data. The trained parameters from the end users are aggregated at the central server. A comprehensive set of experiments were performed, and FedCache outperformed existing state-of-the-art caching strategies concerning cache hit rate, access delay and backhaul traffic significantly.

关键词： Ultra reliable low latency communication   Optimization   Minimization   Array signal processing   Reliability   Reconfigurable intelligent surfaces   Real-time systems   Metaverse   Wireless communication   Industries   Alternating optimization   digital twin   imperfect CSI   mobile edge computing   simultaneously transmitting and reflecting   ultra-reliable low-latency communication  

摘要： In the context of ultra-reliable low-latency communication (URLLC) in Internet-of-Things (IoT) networks, conventional half-space coverage limits the flexibility of reconfigurable intelligent surface (RIS) deployment. To overcome these constraints, this paper makes use of active simultaneously transmitting and reflecting RIS (STAR-RIS), which is seamlessly integrated into digital twin (DT) and mobile edge computing (MEC) frameworks. Our primary research objective is to achieve full-space coverage by enabling simultaneous transmission and reflection of the signals while improving uplink data transmission from IoT URLLC user nodes (UNs) to the base station (BS) with the assistance of active STAR-RIS, even in the presence of imperfect channel state information (CSI). We formulate the problem of minimizing total end-to-end (e2e) latency, computed using the alternating optimization (AO) algorithm. Subsequently, we have evaluated the performance of the AO algorithm against the stochastic gradient descent (SGD) algorithm, which serves as the benchmark solution. The simulation outcomes delineate a performance evaluation under perfect and imperfect CSI scenarios. The AO algorithm outperforms SGD with latency reductions of 19.7% at N=32 and 20.4% at N=64 . Increasing N from 32 to 64 results in a 39.3% latency reduction for AO, surpassing SGD's 38.8%. However, the SGD algorithm consistently exhibits lower computational complexity compared to the AO algorithm. Additionally, the energy splitting mode achieves the system's total e2e latency reductions of 28.4% over the mode switching mode and 11.04% over time switching mode. Furthermore, active STAR-RIS optimal beamforming (ARO) achieves approximate to 10 % latency reduction over the predictive optimal beamforming (PRO), which itself surpasses active STAR-RIS with random beamforming (ARR) by approximate to 9 %. This comparison considers key factors such as the power budget, the number of RIS elements, the caching capacity of the e

关键词： Servers   Delays   Costs   Quality of service   Heuristic algorithms   Optimization   Cloud computing   5G mobile communication   Resource management   Prediction algorithms   Mobile edge computing   batch learning   multi-armed bandits  

摘要： With the development of the fifth generation (5G) technology, Mobile Edge Computing (MEC) is becoming a useful architecture, which is envisioned as a cloud extension version. Users within the MEC system could save more time on data transmission, in which the tasks can be executed on the edge side. Multi-armed bandits (MAB) are powerful tools that help users offload tasks to their best servers in the MEC system. However, as the system scale grows, the traditional MAB algorithms are weak and the channel condition deteriorates rapidly, which will lead to an increasing time cost of edge tasks offloading and computing. Therefore, in this paper, we propose a Batch-based Multi-user Server Elimination (BMSE) to solve such a problem. BMSE contains two sub-algorithms, BMSE in User Level (BMSE-UL) and BMSE in System Level (BMSE-SL). BMSE-UL is applied among users for the initial stage which can help them discard servers with obvious bad performance. BMSE-SL is applied in the whole MEC system which can guide users offloading tasks collectively. Furthermore, we establish the optimality of the proposed algorithms by proving the sub-linearity convergence of their regrets and demonstrate the effectiveness through extensive experiments.

关键词： Training   Servers   Computational modeling   Accuracy   Data models   Convergence   Protocols   Uncertainty   Decision making   Costs   Federated learning   mobile edge computing   semi-asynchronous   timeGAN  

摘要： Federated learning (FL) has gained substantial attention as a promising solution to the need for client privacy in mobile edge computing (MEC). However, FL suffers from instability of accuracy because of the invalid clients who become stragglers caused by frequent fluctuation of available resources in MEC. To tackle this challenge, most of the frameworks of asynchronous FL allow the parameter server (PS) to schedule clients reasonably. This centralized decision-making paradigm makes it difficult to select all potential clients because client resources in MEC vary frequently. As another category of solutions, the semi-asynchronous FL frameworks follow the selection-after-training paradigm, which allows all potential clients to participate in FL but leads to a large waste of computation and communication resources. In this paper, we propose HiveFL, a new semi-asynchronous FL framework in which clients can proactively evaluate the changes in their resources, to improve global accuracy and reduce system overhead (i.e., the waste of resources induced by ineffective clients). HiveFL has the following notable properties. Firstly, HiveFL allows clients to perceive their resource status. Then, self-determining clients autonomously determine by themselves whether to participate in FL training according to the evaluation of their resource status. Secondly, comparing the experimental results with other baselines, HiveFL improves the average global test accuracy and the effective update ratio by 5.20%-22.21% and 20.3%-88.6%, respectively. Finally, HiveFL can reduce the average computation cost (measured by FLOPs) by 31.04%-81.45%. In addition, to address the problem brought by limited client resource status data, we adopt the time-series Generative Adversarial Networks (TimeGAN) method to provide more client data while training FL models. We prove the effectiveness of introducing the GAN-generated data in our experiments.

关键词： VANET   Mobile edge computing   Variable beacon power   Routing algorithms   Q-learning  

摘要： In traditional Vehicle-to-Vehicle-to-Infrastructure (V2V2I) routing algorithms, the beacon power of vehicles remains fixed, lacking adaptability to dynamic adjustments based on events or on-demand operational modes, leading to poor dynamic adaptability of vehicles and significant energy wastage. To enhance the flexibility of beacon power adjustment for vehicles in dynamic environments, while considering practical factors such as the signal coverage status, selfishness, and motion characteristics of routing vehicles, a self-adaptive V2V2I routing algorithm named Adaptive Beacon Power Routing algorithm(ABPR) is proposed. In the algorithm, in order to enhance the stability of the vehicle's connection to RSU, vehicles first collect neighbor beacons and analyze the current signal coverage status of the vehicle. Subsequently, they adaptively adjust the beacon power of the vehicle, and then analyze available neighbor vehicle nodes through improved Q-learning, searching for the optimal neighbor node and selecting it as the next-hop route. Simulation experiments conducted on the Cologne dataset demonstrate that compared to RSAR and GPSR algorithms, the ABPR algorithm achieves an average improvement of 15% in residual energy consumption, 13.79% in selfish vehicle count, 38.58% in routing hop count, and 51.37% in routing link lifetime proportion under +/- 30% standard beacon power conditions. This algorithm enhances the flexibility of beacon power adjustment in VANET environments while optimizing the overall data offloading link quality.

关键词： Servers   Wireless communication   Computational modeling   Artificial neural networks   Resource management   Energy consumption   Wireless sensor networks   Semantic communication   Performance evaluation   Optimization   Mobile edge computing   computation offloading   deep neural network   semantic communication  

摘要： In this paper, we focus on computation-aware offloading for executing deep neural network (DNN) inference tasks in a mobile edge computing (MEC) system. To cope with the challenges of insufficient wireless resources during task offloading, we resort to semantic communications (SCs), through which the users can offload the compressed task data to the edge server for remote execution. Specifically, we establish the relationship between the compression ratio and computation ratio for different DNN tasks. To achieve energy-efficient offloading, we formulate an optimization problem to minimize the energy consumption of all users by jointly optimizing the compression ratio, computation allocation, uploading time, and DNN layer selection. We first consider a special case with the preconfigured time scheduling and derive closed-form solutions to computation allocation and offloading time, which yield a threshold-based structure determined by users' channel conditions and local computation consumption. Inspired by the characteristics of these optimal solutions, a general low-complexity iterative algorithm is then designed to solve the original non-convex problem. Simulation results demonstrate that our proposed SC-based computation-aware offloading scheme can substantially reduce users' energy consumption compared to the conventional offloading and full offloading, especially with scarce wireless resources.