关键词： opportunistic social network   federated learning   community restructuring   Internet of Things   deep reinforcement learning   mobile edge computing  

摘要： With the advent of the 5G era, the number of Internet of Things (IoT) devices has surged, and the population's demand for information and bandwidth is increasing. The mobile device networks in IoT can be regarded as independent "social nodes", and a large number of social nodes are combined to form a new "opportunistic social network". In this network, a large amount of data will be transmitted and the efficiency of data transmission is low. At the same time, the existence of "malicious nodes" in the opportunistic social network will cause problems of unstable data transmission and leakage of user privacy. In the information society, these problems will have a great impact on data transmission and data security;therefore, in order to solve the above problems, this paper first divides the nodes into "community divisions", and then proposes a more effective node selection algorithm, i.e., the FL node selection algorithm based on Distributed Proximal Policy Optimization in IoT (FABD) algorithm, based on Federated Learning (FL). The algorithm is mainly divided into two processes: multi-threaded interaction and a global network update. The device node selection problem in federated learning is constructed as a Markov decision process. It takes into account the training quality and efficiency of heterogeneous nodes and optimizes it according to the distributed near-end strategy. At the same time, malicious nodes are screened to ensure the reliability of data, prevent data loss, and alleviate the problem of user privacy leakage. Through experimental simulation, compared with other algorithms, the FABD algorithm has a higher delivery rate and lower data transmission delay and significantly improves the reliability of data transmission.

关键词： Task analysis   Resource management   Pricing   Cloud computing   Computational modeling   Mobile applications   Edge computing   Pricing mechanisms   resource allocation   market equilibrium   microeconomic theory   mobile edge computing  

摘要： Mobile edge computing (MEC) can significantly improve the performance of mobile applications by leveraging nearby servers as the edge cloud to provide task offloading execution service for a smart mobile device (SMD) through wireless access points (APs). However, the edge cloud and AP will not provide free services. Their radio frequency resources and computing resource are limited but the service requests from various mobile devices could be massive. The goal of this article is to provide a pricing mechanism to efficiently allocate limited resources in the MEC system according to the budget of SMDs. To this end, we first present a market model of MEC resources that can give a real insight into the incentives for resource sharing at network edges. In the model, computation, and radio resources can be traded between resource suppliers (AP and edge cloud) and buyers (SMDs). Furthermore, we employ the microeconomic theory to get an optimal budget allocation strategy for the SMD to maximize its utility within a limited budget. Moreover, we propose an Equilibrium Price Finding (EPF) algorithm to find the equilibrium price of the MEC system, maximizing the whole system utility and leading to optimal resource allocation. Finally, simulation results show that, compared with state-of-the-art resource allocation methods, our optimal budget allocation algorithm can find budget allocation strategy more effectively and our equilibrium price finding algorithm can achieve market equilibrium to optimally allocate computation and radio resources in the MEC system.

关键词： Broadcast scheduling   Mobile edge computing   Mobile P2P streaming network   News broadcast applications  

摘要： The enhanced mobile broadband (eMBB) of 5G drives the News-Broadcast-on-Demand (NBoD) service to be delivered to mobile users in high quality video streams. In NBoD, each user receives a cascade of news clips chosen from a set of prestored news video clips according to the user's preference or subscription policy. While in some dense population areas, the high volume of video traffic generated may congest the data path from the backbone news servers to the mobile users connected to the radio access networks (RAN) under the same vicinity. As a result, users may experience poor viewing quality. To solve this problem, the mobile peer-to-peer streaming (MPPS) overlay network provides a promising solution. It conducts stream sharing in the RAN to reduce the backbone traffic. Unfortunately, the sharing efficiency of MPPS is deteriorated by the sparely distributed requests and mobile users' limited cache buffer space. In this regard, we introduce a novel mobileedge-computing (MEC) server at the mobile cloud edge, which purposely clusters requests of the same video clip to exploit the stream sharing opportunity in the RAN. The MEC server consists of two parts: a news directory manager (NDM) and a broadcast-channel scheduling manager (BSM). NDM takes the users' requests and compiling customized news playlists. BSM schedules the news-clip play times with the objective of maximizing the number of peers participating stream sharing. For BSM, two scheduling algorithms are investigated, termed the clip-wise scheduling (CS) for the short-term-stream-sharing optimization with a lower computational cost and the session-wise scheduling (SS) for the long-term-stream-sharing optimization with a higher computational cost. Through packet-level simulation, we evaluate these two schemes with respect to the parameters of user cache buffer size, startup delay tolerance and system load. The results reveal that with a small (9-s) cache buffer, CS and SS can achieve 70.1% similar to 76% and 74%

关键词： Training   Servers   Computational modeling   Analytical models   Wireless communication   Resource management   Energy consumption   Federated learning   energy efficiency   mobile edge computing  

摘要： In this letter, a novel joint dataset and computation management (DCM) scheme for energy-efficient federated learning (FL) in mobile edge computing (MEC) is proposed. For this purpose, with respect to the amount of dataset and computation resources, we rigorously formulated analytical models for i) learning efficiency, which considers the estimated global accuracy tendency according to the amount of dataset and service latency, and ii) the overall energy consumption of FL participants, including local training and model parameter transmission. To consider the trade-off between these two factors in the FL procedure with MEC, a theoretical framework for the DCM problem that jointly optimizes the amount of dataset and the computation resources used for local training over multiple FL clients was designed. Additionally, the extensive simulation-based performance evaluations validate the superior performance of the proposed DCM;compared to the various benchmarks in terms of the proposed cost function and test accuracy on the MNIST dataset with independent identically distributed (IID) / non-IID settings.

关键词： Mobile edge computing   synthetic theories   mobile device   suboptimal   security strategies   quality of service   deep learning  

摘要： The implementation of a variety of complex and energy-intensive mobile applications by resource-limited mobile devices (MDs) is a huge challenge. Fortunately, mobile edge computing (MEC) as a new computing paragon can offer rich resources to perform all or part of the MD's task, which greatly reduces the energy consumption of the MD and improves the quality of service (QoS) for applications. However, offloading tasks to the edge server is vulnerable to attacks such as tampering and snooping, resulting in a deep learning (DL) security feature developed by major cloud service providers. An effective security strategy method to minimize ongoing attacks in the MEC setting is proposed. The algorithm is based on the synthetic principle of a special set of strategies, and it can quickly construct suboptimal solutions even if the number of targets achieves hundreds of millions. In addition, for a given structure and a given number of patrollers, the upper bound of the protection level can be obtained, and the lower bound required for a given protection level can also be inferred. These bounds apply to universal strategies. By comparing with the previous three basic experiments, it can be proved that our algorithm is better than the previous ones in terms of security and running time.

关键词： Mobile edge computing   Machine learning   Deep learning   Big data analytics   Bio-inspired computing   Deep forest   Ant colony optimization   Ant lion optimization  

摘要： Today, incorporating advanced machine learning techniques into intrusion detection systems (IDSs) plays a crucial role in securing mobile edge computing systems. However, the mobility demands of our modern society require more advanced IDSs to make a good trade-off between coping with the rapid growth of traffic data and responding to attacks. Thus, in this paper, we propose a lightweight distributed IDS that exploits the advantages of centralized platforms to train and learn from large amounts of data. We investigate the benefits of two promising bioinspired optimization algorithms, namely Ant Lion Optimization and Ant Colony Optimization, to find the optimal data representation for the classification process. We use Deep Forest as a classifier to detect intrusive actions more robustly and generate as few false positives as possible. The experiment results show that the proposed approach can enhance the reliability of lightweight intrusion detection systems in terms of accuracy and execution time.

关键词： Mobile edge computing  

摘要： Recently, Collaborative Edge Computing (CEC) has been proposed as an effective method to improve the performance of Mobile Edge Computing (MEC) systems by offloading workload from busy MEC servers to idle ones. Although extensive researches have been conducted to study the system design and scheduling algorithms in CEC, the incentive mechanism between end-users and collaborative servers yet receives much less attention. To fill the gap, in this paper, we propose an online auction-based incentive mechanism where the task requests arrive stochastically and MEC servers need to decide whether to accept them without future information. To guarantee the applicability of our mechanism, we consider a comprehensive model where both the valuation of tasks and operation costs of edge servers are represented by general functions. Through theoretical analysis, we formally proved our mechanism achieves desirable properties, such as truthfulness and polynomial time complexity. Based on the primal-dual optimization framework, we further demonstrate the competitive ratio of our mechanism with respect to the optimal offline social welfare. Extensive simulations are conducted to verify our theoretical analysis and validate the effectiveness of our mechanism. © 2022, The Authors. All rights reserved.

关键词： Mobile edge computing   Multicarrier non-orthogonal multiple access   Resource block assignment   Deep reinforcement learning  

摘要： Future network services are emerging with an inevitable need for high wireless capacity along with strong computational capabilities, stringent latency and reduced energy consumption Two technologies are promising, showing potential to support these requirements: multi-access (or mobile) edge computing (MEC) and non orthogonal multiple access (NOMA). While MEC allows users to access the abundant computing resources at the edge of the network, NOMA technology enables an increase in the density of a cellular network. However, integrating NOMA technology into MEC systems faces challenges in terms of joint offloading decisions (remote or local computation) and inter-user interference management. In this paper, with the objective of maximizing the system-wide sum computation rate under latency and energy consumption constraints, we propose a two stage deep reinforcement learning algorithm to solve the joint problem in a multicarrier NOMA-based MEC system, in which the first-stage agent handles offloading decisions while the second-stage agent considers the offloading decisions to determine the resource block assignments for users. Simulation results show that compared with other benchmark algorithms, the proposed algorithm improves the sum computation rate while meeting the latency and energy consumption requirements, and it outperforms the approach in which a single agent handles both offloading decisions and resource block assignments due to faster convergence performance.

关键词： Mobile edge computing  

摘要： Mobile edge computing (MEC) systems (in which intensive computation and data storage tasks are performed locally, due to absence of communication infrastructure for connectivity to cloud) are currently being developed for disaster response applications and for tactical environments. MEC applications for these scenarios generate and process significant mission critical and personal data that require resilient and secure storage and sharing. In this paper we present the design, implementation and evaluation of R-Drive, a resilient data storage and sharing framework for disaster response and tactical MEC applications. R-Drive employs erasure coding and data encryption, ensuring resilient and secure data storage against device failure. R-Drive adaptively chooses erasure coding parameters to ensure highest data availability with minimal storage cost. R-Drive's distributed directory service provides a resilient and secure namespace for files with rigorous access control management. R-Drive leverages opportunistic networking, allowing data storage and sharing in mobile and loosely connected edge computing environments. We implemented R-Drive on Android, integrated it with existing MEC applications. Performance evaluation results show that R-Drive enables resilient and secure data storage and sharing. © 2022, CC BY.

关键词： Mobile edge computing   Calculation task allocation   Unmanned aerial vehicle communications  

摘要： Owing to the high economic benefits, flexible deployment, and controllable maneuverability, unmanned aerial vehicles (UAVs) have been envisioned as promising and potential technologies for dispensing wireless communication services. This paper investigates a mobile edge computing (MEC) system assisted by multiple access points (APs) and an UAV, in which APs may not be able to straightly establish wireless communications with terrestrial Internet of Thing devices (IoT) due to ground signal blockage. Consequently, an UAV is dispatched as a mobile AP to serve a group of users and render the air-to-ground channel. In this scenario, we contemplate dividing the computing tasks of IoTDs into three parts: either be calculated locally, or offloaded to the UAV for processing, or accomplished on AP through relaying. This work attempts to minimize the weighted sum of communication consumption, calculation consumption, and the UAV's flight consumption over a finite UAV mission duration by jointly optimizing calculation task allocation ratio, power distribution as well as the UAV's trajectory. However, the resulting problem we put forward is demonstrated to be highly non-convex and challenging to solve. To tackle this issue, we decompose the original problem into two sub-problems hinging on the block coordinate descent (BCD) method. We settle the two sub-problems iteratively through the Lagrangian duality method and succession convex approximation (SCA) technique. The simulation results further reveal that the proposed approach is superior to other comparison baselines.