MEC系统中基于多智能体深度增强学习的协作式任务卸载和资源分配

移动边缘计算(MEC)通过将计算任务卸载到MEC服务器上,在缓解智能移动设备计算负载的同时,可以降低服务时延。然而目前在MEC系统中,关于任务卸载和资源分配仍然存在以下问题：1)边缘节点间缺乏协作;2)计算任务到达与实际环境中动态变化的特征不匹配;3)协作式任务卸载和资源分配动态联合优化问题。为解决上述问题,文章在协作式MEC架构的基础上,提出了一种基于多智能体的深度确定性策略梯度算法(MADDPG)的任务卸载和资源分配算法,最小化系统中所有用户的长期平均成本。仿真结果表明,该算法可以有效降低系统的时延及能耗。     

基于移动边缘计算的V2X任务卸载方案

移动边缘计算(MEC)通过在移动网络边缘提供IT服务环境和云计算能力带来高带宽、低时延优势,从而在下一代移动网络的研究中引起了广泛的关注。该文研究车载网络中车辆卸载请求任务时搜寻服务节点为其服务的匹配问题,构建一个基于MEC的卸载框架,任务既可以卸载到MEC服务器以车辆到基础设施(V2I)形式通信,也可以卸载到邻近车辆进行车辆到车辆(V2V)通信。考虑到资源有限性、异构性,任务多样性,建模该框架为组合拍卖模式,提出一种多轮顺序组合拍卖机制,由层次分析法(AHP)排序、任务投标、获胜者决策3个阶段组成。仿真结果表明,所提机制可以在时延和容量约束下,使请求车辆效益提高的同时最大化服务节点的效益。     

命名数据网络中基于内容请求相关性的协作缓存算法

针对命名数据网络(Named Data Networking, NDN)存储空间的有效利用和应答内容的高效缓存问题,该文采用差异化缓存的方式,提出一种依据内容请求序列相关性的协作缓存算法。在内容请求中,预先发送对于后续相关数据单元的并行预测请求,增大内容请求的就近响应概率;缓存决策时,提出联合空间存储位置与缓存驻留时间的2维差异化缓存策略。根据内容活跃度的变化趋势,空间维度上逐跳推进内容存储位置,时间维度上动态调整内容缓存时间,以渐进式的方式将真正流行的请求内容推送至网络边缘存储。该算法减小了内容请求时延和缓存冗余,提高了缓存命中率,仿真结果验证了其有效性。     

移动边缘网络中基于用户移动的服务功能链迁移策略

移动边缘计算(Mobile Edge Computing,MEC)通过在网络边缘部署服务器,提供计算和存储资源,可为用户提供超低时延和高带宽业务。网络功能虚拟化(Network Function Virtualization,NFV)与MEC技术相结合,可在MEC服务器上提供服务功能链(Service Function Chain,SFC),提升用户的业务体验。为了保证移动用户的服务质量,需要在用户跨基站移动时将SFC迁移到合适的边缘服务器上。主要以最小化用户服务的端到端时延和运行成本为目标,提出了MEC网络中具有资源容量约束的SFC迁移策略,以实现移动用户业务的无缝迁移。仿真结果表明,与现有方案相比,该策略具有更好的有效性和高效性。     

基于可重构路由器上缓存的流媒体协作分发策略

提出一种新颖的基于可重构路由器上缓存的协作分发策略来加速流媒体。通过网络存储即多个边缘路由器节点对热点视频数据进行合作缓存,就近为用户提供服务,从而使得流媒体服务器的性能要求尤其是带宽需求得到巨大的降低,骨干网传输的流量也明显减少,同时用户响应延迟也得到明显的改善。此外,实现了一个原型系统来评价基于路由器上缓存的流媒体协作分发策略的性能,结果表明该方案相比于现有的方案在改善网络性能以及用户体验方面取得很大的提升。     

面向5G的MEC系统关键技术

移动边缘计算(MEC)是未来5G移动通信系统提升服务应用能力的重要技术手段之一。通过在无线接入网络的边缘节点处部署具备计算、存储和通信能力的服务应用平台,MEC能够有效处理终端用户的高时效性业务需求,大幅度缩短端到端时延,并解决核心网络的数据流量瓶颈等相关问题。     

移动边缘计算中服务功能链的自适应优化部署策略

为解决移动边缘计算中面向用户的服务功能链（Service Function Chain,SFC）部署成本开销过大、时延过长问题,提出了针对SFC的支出成本与时延联合自适应优化的部署策略。首先,在虚拟网络功能（Virtualized Network Function,VNF）节点选取阶段,考虑路径损耗这一无线信道衰落问题,根据有线用户与无线用户的位置情况,选择当前最佳节点以降低SFC的响应时延。其次,在服务节点配置阶段,根据用户请求处理的数据内容的新鲜度记录,自适应动态增加和删减相应的缓存,利用资源感知算法在保证数据传递可靠性的同时,减少服务节点的配置个数,降低配置开销。最后,在SFC部署阶段,利用基于KSP（K-shortest Paths）的功耗感知算法确定最佳节点映射排序与通信链路,在减少通信链路重映射的同时还能保证部署的SFC的低成本与低时延。实验仿真结果表明,相比于已有方案,该方法能够有效降低部署成本与时延,并能对不同用户的SFC部署做到自适应优化,提高了SFC的部署成功率。     

内容中心网络中面向隐私保护的协作缓存策略

针对内容中心网络节点普遍缓存带来的隐私泄露问题,在兼顾内容分发性能的基础上,该文提出一种面向隐私保护的协作缓存策略。该策略从信息熵的角度提出隐私度量指标,以增大攻击者的不确定度为目标,首先对于缓存策略的合理性给予证明;其次,通过构建空间匿名区域,扩大用户匿名集合,增大缓存内容的归属不确定性。缓存决策时,针对垂直请求路径和水平匿名区域,分别提出沿途热点缓存和局域hash协同的存储策略,减小缓存冗余和隐私信息泄露。仿真结果表明,该策略可减小内容请求时延,提高缓存命中率,在提升内容分发效率的同时增强了用户隐私保护水平。     

基于匹配理论的电力物联网边缘服务器选择机制

移动边缘计算利用部署在用户附近基站或具有空闲资源的路侧单元、车辆和MEC服务器作为网络的边缘,为设备提供所需的服务以及云端计算能力,以减少网络操作和服务交付的时延。文章将移动设备和MEC服务器的任务分配问题描述为一对一的匹配博弈,解决了移动边缘计算中的任务卸载问题。文章提出的算法具有良好的扩展性,并且能够降低总体能耗,使任务卸载时延最小化。     

车辆网络多平台卸载智能资源分配算法

为了降低计算任务的时延和系统的成本,移动边缘计算(MEC)被用于车辆网络,以进一步改善车辆服务。该文在考虑计算资源的情况下对车辆网络时延问题进行研究,提出一种多平台卸载智能资源分配算法,对计算资源进行分配,以提高下一代车辆网络的性能。该算法首先使用K临近(KNN)算法对计算任务的卸载平台(云计算、移动边缘计算、本地计算)进行选择,然后在考虑非本地计算资源分配和系统复杂性的情况下,使用强化学习方法,以有效解决使用移动边缘计算的车辆网络中的资源分配问题。仿真结果表明,与任务全部卸载到本地或MEC服务器等基准算法相比,提出的多平台卸载智能资源分配算法实现了时延成本的显著降低,平均可节省系统总成本达80%。     

面向ICN切片的视频缓存与传输服务

随着视频流量爆炸式地增长,为了降低密集区域用户的视频检索时延和网络负载,利用网络功能虚拟化技术提出了一种将5G切片与信息中心网络(Information Center Network,ICN)相结合的内容分发服务系统集成的架构,并改进基于ICN的缓存策略.该缓存策略依据特定区域用户一个月与一周内的历史访问记录构建视频主...     

面向移动边缘计算中多应用服务的虚拟机部署算法

移动边缘计算(MEC)通过在用户近端以虚拟机(VM)形式部署应用服务,能有效降低服务响应延迟并减少核心网络数据流量。然而,当前MEC中虚拟机部署的大多数研究尚未具体考虑用户对多种应用服务的需求。因此,该文针对MEC中多应用服务的虚拟机部署问题,提出两种启发式算法,即基于适应度的启发式部署算法(FHPA)和基于分治的启发式部署算法(DCBHPA),通过在边缘网络中配置支持多种应用服务的虚拟机来最大限度地减少网络中的数据流量。FHPA和DCBHPA分别基于边缘服务器的网络连接特征和用户对应用请求的差异性,定义了不同的适应度计算模型。在此基础上,通过子问题划分机制实现VM配置。仿真结果表明,相比于基准算法,所提算法能更好地控制系统数据流量,有效地提高边缘网络服务资源的利用率。     

A blockchain-enabled security management framework for mobile edge computing

Mobile edge computing (MEC) integrates mobile and edge computing technologies to provide efficient computing services with low latency. It includes several Internet of Things (IoT) and edge devices that process the user data at the network's edge. The architectural characteristic of MEC supports many internet-based services, which attract more number of users, including attackers. The safety and privacy of the MEC environment, especially user information is a significant concern. A lightweight accessing and sharing protocol is required because edge devices are resource constraints. This paper addresses this issue by proposing a blockchain-enabled security management framework for MEC environments. This approach provides another level of security and includes blockchain security features like temper resistance, immutable, transparent, traceable, and distributed ledger in the MEC environment. The framework guarantees secure data storage in the MEC environment. The contributions of this paper are twofold: (1) We propose a blockchain-enabled security management framework for MEC environments that address the security and privacy concerns, and (2) we demonstrate through simulations that the framework has high performance and is suitable for resource-constrained MEC devices. In addition, a smart contract-based access and sharing mechanism is proposed. Our research uses a combination of theoretical analysis and simulation experiments to demonstrate that the proposed framework offers high security, low latency, legitimate access, high throughput, and low operations cost.     

Virtual network function deployment strategy based on improved genetic simulated annealing algorithm in MEC

In order to effectively improve the end-to-end service delay of the flow in multi-clusters coexisting mobile edge computing (MEC) network,a virtual network function deployment strategy based on improved genetic simulated annealing algorithm was proposed.The delay of mobile service flow was mathematically modeled through the open Jackson queuing network.After proving the NP attribute of this problem,a solution combining genetic algorithm and simulated annealing algorithm was proposed.In this strategy,the advance mapping mechanism avoids the possibility of network congestion,and the occurrence of local optima was avoided through using the methods of individual judgment and corrective genetic.Extensive simulation was set up to evaluate the effectiveness of the proposed strategy under different parameter settings,such as different volume of requests,different scale of service nodes,different number of MEC clusters,and logical link relationships between virtual network functions.Results show that this strategy can provide lower end-to-end services delay and better service experience for latency-sensitive mobile application.     

Unmanned aerial vehicles in collaboration with fog computing network for improving quality of service

In this paper, we study a UAV-based fog or edge computing network in which UAVs and fog/edge nodes work together intelligently to provide numerous benefits in reduced latency, data offloading, storage, coverage, high throughput, fast computation, and rapid responses. In an existing UAV-based computing network, the users send continuous requests to offload their data from the ground users to UAV–fog nodes and vice versa, which causes high congestion in the whole network. However, the UAV-based networks for real-time applications require low-latency networks during the offloading of large volumes of data. Thus, the QoS is compromised in such networks when communicating in real-time emergencies. To handle this problem, we aim to minimize the latency during offloading large amounts of data, take less computing time, and provide better throughput. First, this paper proposed the four-tier architecture of the UAVs–fog collaborative network in which local UAVs and UAV–fog nodes do smart task offloading with low latency. In this network, the UAVs act as a fog server to compute data with the collaboration of local UAVs and offload their data efficiently to the ground devices. Next, we considered the Q-learning Markov decision process (QLMDP) based on the optimal path to handle the massive data requests from ground devices and optimize the overall delay in the UAV-based fog computing network. The simulation results show that this proposed collaborative network achieves high throughput, reduces average latency up to 0.2, and takes less computing time compared with UAV-based networks and UAV-based MEC networks; thus, it can achieve high QoS.     

Users' experience matter: Delay sensitivity-aware computation offloading in mobile edge computing

As a promising computing paradigm, Mobile Edge Computing (MEC) provides communication and computing capability at the edge of the network to address the concerns of massive computation requirements, constrained battery capacity and limited bandwidth of the Internet of Things (IoT) systems. Most existing works on mobile edge task ignores the delay sensitivities, which may lead to the degraded utility of computation offloading and dissatisfied users. In this paper, we study the delay sensitivity-aware computation offloading by jointly considering both user's tolerance towards delay of task execution and the network status under computation and communication constraints. Specifically, we use a specific multi-user and multi-server MEC system to define the latency sensitivity of task offloading based on the analysis of delay distribution of task categories. Then, we propose a scoring mechanism to evaluate the sensitivity-dependent utility of task execution and devise a Centralized Iterative Redirection Offloading (CIRO) algorithm to collect all information in the MEC system. By starting with an initial offloading strategy, the CIRO algorithm enables IoT devices to cooperate and iteratively redirect task offloading decisions to optimize the offloading strategy until it converges. Extensive simulation results show that our method can significantly improve the utility of computation offloading in MEC systems and has lower time complexity than existing algorithms.     

Federated learning based intelligent edge computing technique for video surveillance

With the explosion of global data,centralized cloud computing cannot provide low-latency,high-efficiency video surveillance services.A distributed edge computing model was proposed,which directly processed video data at the edge node to reduce the transmission pressure of the network,eased the computational burden of the central cloud server,and reduced the processing delay of the video surveillance system.Combined with the federated learning algorithm,a lightweight neural network was used,which trained in different scenarios and deployed on edge devices with limited computing power.Experimental results show that,compared with the general neural network model,the detection accuracy of the proposed method is improved by 18%,and the model training time is reduced.     

Joint partial computation offloading and resource allocation inMEC-enable networks

The sudden surge of various applications poses great challenges to the computation capability of mobile devices. To address this issue, computation offloading to multi-access edge computing(MEC) was proposed as a promising paradigm. This paper studies partial computation offloading scenario by considering time delay and energy consumption, where the task can be splitted into several blocks and computed both in local devices and MEC, respectively. Since the formulated problem is a nonconvex probl...