基于最长有效功能序列的服务功能链部署算法

服务功能链的服务性能取决于功能的部署位置和数据传输路径的选择。针对资源有限的网络中的服务功能链部署问题,该文设计了一种基于最长有效功能序列(LEFS)的服务功能链部署算法,以功能复用和带宽需求联合优化为目标,在控制路径长度的同时根据LEFS逐步搜索中继节点,直至满足服务请求。仿真结果表明,该算法能够均衡网络资源,同时优化网络的功能部署率和带宽利用率,与其他算法相比,网络资源利用率降低了10%,可以支持更多的服务请求,且算法计算复杂度低,可以实现对服务请求的快速响应。     

面向软件定义网络的服务功能链优化部署算法研究

针对网络功能虚拟化(NFV)环境下,现有服务功能链部署方法无法在优化映射代价的同时保证服务路径时延的问题,该文提出一种基于IQGA-Viterbi学习算法的服务功能链优化部署方法。在隐马尔可夫模型参数训练过程中,针对传统Baum-Welch算法训练网络参数容易陷入局部最优的缺陷,改进量子遗传算法对模型参数进行训练优化,在每一迭代周期内通过等比例复制适应度最佳种群的方式,保持可行解多样性和扩大空间搜索范围,进一步提高模型参数的精确度。在隐马尔科夫链求解过程中,针对隐含序列无法直接观测这一难点,利用Viterbi算法能精确求解隐含序列的优势,解决有向图网络中服务路径的优化选择问题。仿真实验结果表明,与其它部署算法相比,所提IQGA-Viterbi学习算法能有效降低网络时延和映射代价的同时,提高了网络服务的请求接受率。     

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。     

基于改进深度强化学习的虚拟网络功能部署优化算法

针对网络功能虚拟化/软件定义网络(NFV/SDN)架构下,网络服务请求动态到达引起的服务功能链(SFC)部署优化问题,该文提出一种基于改进深度强化学习的虚拟网络功能(VNF)部署优化算法.首先,建立了马尔科夫决策过程(MDP)的随机优化模型,完成SFC的在线部署以及资源的动态分配,该模型联合优化SFC部署成本和时延成本,同时受限于SFC的时延以及物理资源约束.其次,在VNF部署和资源分配的过程中,存在状态和动作空间过大,以及状态转移概率未知等问题,该文提出了一种基于深度强化学习的VNF智能部署算法,从而得到近似最优的VNF部署策略和资源分配策略.最后,针对深度强化学习代理通过ε贪婪策略进行动作探索和利用,造成算法收敛速度慢等问题,提出了一种基于值函数差异的动作探索和利用方法,并进一步采用双重经验回放池,解决经验样本利用率低的问题.仿真结果表示,该算法能够加快神经网络收敛速度,并且可以同时优化SFC部署成本和SFC端到端时延.     

基于联合备份的服务功能链可靠性保障的部署方法

在网络功能虚拟化(NFV)环境中,针对服务功能链(SFC)部署时的可靠性问题,该文提出对备份虚拟网络功能选择、备份实例放置和服务功能链部署的联合优化方法。首先,定义一个单位开销可靠性提高值的虚拟网络功能衡量标准,改进备份虚拟网络功能选择方法;其次,采用联合备份的方式调整相邻备份实例之间的放置策略,以降低带宽资源开销;最后,将整个服务功能链可靠性保障的部署问题构建成整数线性规划模型,并提出一种基于最短路径的启发式算法,克服整数线性规划求解的复杂性。仿真结果表明,该方法在优先满足网络服务可靠性需求的同时,优化资源配置,提高了请求接受率。     

基于微服务架构的粒度可变服务功能链映射算法

针对5G环境下服务功能链（SFC）端到端时延无法满足时延敏感型应用需求的问题,将传统虚拟网络功能（VNF）拆分成粒度不一的映射单元,提出了基于微服务架构的粒度可变服务功能链映射（VG-SFCM）算法。首先将传统粗粒度的VNF解耦成细粒度的微服务单元,随后通过SFC内冗余微服务单元的合并及SFC间微服务单元的复用,减少微服务单元的实例化,降低SFC的处理时间。仿真结果表明,所提算法在降低平均部署网络成本的同时,其SFC端到端时延相较于传统的映射算法降低了14.81%。     

一种采用NFV架构的服务功能链动态编排与部署

随着虚拟化技术的出现,网络功能虚拟化(Network Function Virtualization,NFV) 和软件定义网络(Software Defined Networking,SDN) 技术使网络功能从硬件中抽象出来并可运行在虚拟机上,将虚拟网络功能映射部署到物理网络上,为用户提供定制化服务。服务功能链(Service Function Chaining,SFC) 由一组虚拟网络功能组成,目前其部署的一个关键问题是如何在保证为用户提供相应服务的同时降低网络资源消耗和负载压力。为此,首先介绍了一种可用于服务功能链编排的NFV 架构,并对基于SRv6 技术的SFC 编排过程进行说明;然后提出了一种有效的SFC动态编排算法——第一步使用广度优先搜索遍历网络并找到部署服务链的最短物理路径,第二步使用蚁群优化算法生成最优部署方案。仿真结果表明,该方案能够有效减轻网络负载,同时平均降低端到端延迟22%,减少带宽资源消耗18%,优化部署成功率23%。     

基于改进深度强化学习的虚拟网络功能部署优化算法

针对网络功能虚拟化/软件定义网络 (NFV/SDN)架构下,网络服务请求动态到达引起的服务功能链(SFC)部署优化问题,该文提出一种基于改进深度强化学习的虚拟网络功能(VNF)部署优化算法。首先,建立了马尔科夫决策过程 (MDP)的随机优化模型,完成SFC的在线部署以及资源的动态分配,该模型联合优化SFC部署成本和时延成本,同时受限于SFC的时延以及物理资源约束。其次,在VNF部署和资源分配的过程中,存在状态和动作空间过大,以及状态转移概率未知等问题,该文提出了一种基于深度强化学习的VNF智能部署算法,从而得到近似最优的VNF部署策略和资源分配策略。最后,针对深度强化学习代理通过ε贪婪策略进行动作探索和利用,造成算法收敛速度慢等问题,提出了一种基于值函数差异的动作探索和利用方法,并进一步采用双重经验回放池,解决经验样本利用率低的问题。仿真结果表示,该算法能够加快神经网络收敛速度,并且可以同时优化SFC部署成本和SFC端到端时延。     

基于深度Q网络的在线服务功能链部署方法

针对5G网络资源状态动态变化和网络模型高维度下服务功能链部署的复杂性问题,该文提出一种基于深度Q网络的在线服务功能链部署方法(DeePSCD)。首先,为描述网络资源动态变化的特征,将服务功能链部署建模成马尔可夫决策过程,然后,针对系统资源模型的高维度问题采用深度Q网络的方法进行在线服务功能链部署策略求解。该方法可以有效描述网络资源状态的动态变化,特别是深度Q网络能有效克服求解复杂度,优化服务功能链的部署开销。仿真结果表明,所提方法在满足服务时延约束条件下降低了服务功能链的部署开销,提高了运营商网络的服务请求接受率。     

基于改进混合灰狼优化算法的无人机三维路径规划

针对传统灰狼优化(Grey Wolf Optimization, GWO)算法求解无人机三维路径规划问题时会出现收敛速度慢、容易陷入局部最优等问题,提出一种改进混合灰狼优化算法——CLGWO。基于Cat混沌映射和反向学习策略初始化灰狼种群,为算法全局搜索过程中丰富种群多样性奠定基础;提出新型非线性收敛因子的改进策略,提高算法全局搜索能力。在灰狼位置更新中提出引入狮群优化(Lion Swarm Optimization, LSO)算法的扰动因子和动态权重,使灰狼具有主动的搜索能力,避免因灰狼失去种群多样性而陷入局部最优。为验证改进算法的有效性,进行了8个国际通用的标准测试函数收敛性对比实验和无人机三维路径规划仿真实验。实验结果表明,CLGWO算法在单峰、多峰函数上均有较好的收敛性、较高的寻优精度;三维路径仿真环境下,CLGWO算法的平均路径长度、平均迭代次数、平均运行时间相比于GWO算法分别优化了33%、31%、52%,且路径转折少,能较好地得到全局最优值,验证了CLGWO算法的有效性。     

Resource optimization algorithm of combination of NFV and SDN for application of multiple services

Various services of internet of things (IoT) require flexible network deployment to guarantee different quality of service (QoS).Aiming at the problem of IoT service function chain deployment,network function virtualization (NFV) and software defined networking (SDN) were combined to optimize resources.Considering forwarding cost and traffic load balance,a joint optimization model of virtual network function placement and service function chain routing was given and was proved to be NP-Hard.In order to solve this model,two heuristic algorithms were proposed.One was the service chain deployment algorithm of first routing then placing (FRTP) and the other was the placing followed by routing (PFBR) based on node priority.Simulation results demonstrate that FRTP and PFBR algorithm can significantly balance network traffic load while alleviating congestion and improving the acceptance ratio of the chain requests compared with other algorithms.     

Service chain mapping algorithm based on reinforcement learning

A service chain resource scheduling architecture of multi-agent based on artificial intelligence technology was proposed.Meanwhile,a service chain mapping algorithm based on reinforcement learning was designed.Through the Q-learning mechanism,the location of each virtual network element in the service chain was determined according to the system status and the reward and punishment feedback after the deployment.The experimental results show that compared with the classical algorithms,the algorithm effectively reduces the average transmission delay of the service and improves the load balance of the system.     

Intelligent Service Function Chain Mapping Framework for Cloud-and-Edge-Collaborative IoT

With the rapid development of Internet of thing (IoT) technology, it has become a challenge to deal with the increasing number and diverse requirements of IoT services. By combining burgeoning network function virtualization ( NFV) technology with cloud computing and mobile edge computing ( MEC), an NFV-enabled cloud-and-edge-collaborative IoT (CECIoT) architecture can efficiently provide flexible service for IoT traffic in the form of a service function chain (SFC) by jointly utilizing edge and cloud resources. In this promising architecture, a difficult issue is how to balance the consumption of resource and energy in SFC mapping. To overcome this challenge, an intelligent energy-and-resource-balanced SFC mapping scheme is designed in this paper. It takes the comprehensive deployment consumption as the optimization goal, and applies a deep Q-learning(DQL)-based SFC mapping (DQLBM) algorithm as well as an energy-based topology adjustment (EBTA) strategy to make efficient use of the limited network resources, while satisfying the delay requirement of users. Simulation results show that the proposed scheme can decrease service delay, as well as energy and resource consumption.     

Service function chain deployment algorithm based on optimal weighted graph matching

Service function chain can support flexible network service requirement by linking virtual network functions.Aiming at the problem of service function chain deployment in a resource-constrained network,an algorithm for service function chain deployment based on optimal weighted graph matching was proposed.The service function chains was composed into graphs of functional topography,and the optimal matching results between graphs of functional topology and physical topology was obtained using eigendecomposition approach,and furtherly the matching results by hill-climbing method was optimized.Simulation results show that,the proposed algorithm can reduce the required bandwidth to deploy service function chains,balance the load of nodes and bandwidth of links,and support more service requests.What is more,the algorithm has a lower computation complexity and higher time efficience.     

基于改进灰狼优化算法的DV-Hop定位算法

针对无线传感器网络节点DV-Hop定位算法由于节点分布不均,距离估计不准确,导致定位精度较低的问题,提出了一种基于改进灰狼优化算法的DV-Hop定位算法,采用先进的灰狼优化算法以寻找最优值的方式得到未知节点、坐标。同时,为进一步提高优化算法的寻优能力,克服可能出现局部最优的情况,将优化算法与免疫算法相结合,提高优化算法中灰狼种群的多样性,进而提高对最优解的搜索能力,达到提高定位精度的目的。实验结果表明,相对于普通的DV-Hop定位算法和普通的灰狼优化算法,改进之后的定位算法精度更高。     

Hybrid gray wolf optimization-cuckoo search algorithm for RFID network planning

In recent years, with the rapid development of Internet of things (IoT) technology, radio frequency identification (RFID) technology as the core of IoT technology has been paid more and more attention, and RFID network planning(RNP) has become the primary concern. Compared with the traditional methods, meta-heuristic method is widely used in RNP. Aiming at the target requirements of RFID, such as fewer readers, covering more tags, reducing the interference between readers and saving costs, this paper proposes a hybrid gray wolf optimization-cuckoo search (GWO-CS) algorithm. This method uses the input representation based on random gray wolf search and evaluates the tag density and location to determine the combination performance of the reader's propagation area. Compared with particle swarm optimization ( PSO) algorithm, cuckoo search( CS) algorithm and gray wolf optimization ( GWO) algorithm under the same experimental conditions, the coverage of GWO-CS is 9.306% higher than that of PSO algorithm, 6.963% higher than that of CS algorithm, and 3.488% higher than that of GWO algorithm. The results show that the GWO-CS algorithm cannot only improve the global search range, but also improve the local search depth.     

Virtual network function–forwarding graph embedding: A genetic algorithm approach

Network function virtualization (NFV) provides a simple and effective mean to deploy and manage network and telecommunications' services. A typical service can be expressed in the form of a virtual network function–forwarding graph (VNF‐FG). Allocating a VNF‐FG is equivalent to place VNFs and virtual links onto a given substrate network considering resources and quality‐of‐service (QoS) constraints. The deployment of VNF‐FGs in large‐scale networks, such that QoS measures and deployment cost are optimized, is an emerging challenge. Single‐objective VNF‐FGs allocation has been addressed in existing literature; however, there is still a lack of studies considering multiobjective VNF‐FGs allocation. In addition, it is not trivial to obtain optimal VNF‐FGs allocation due to its high computational complexity even in case of single‐objective VNF‐FGs allocation. Genetic algorithms (GAs) have been proved its ability in coping with multiobjective optimization problems; thus, we propose a GA‐based scheme to solve multiobjective VNF‐FGs allocation problem in this paper. The numerical results confirm that the proposed scheme can provide near Pareto‐optimal solutions within a short execution time.     

面向SDN的动态网络策略部署与实现

软件定义网络（SDN）为未来网络业务的管理要求提供了一种新的解决方案。以动态服务功能链作为服务功能编排模式、以高级网络编程语言作为服务功能部署工具,并结合虚拟网络映射设计了业务到策略、从控制器到网络节点的部署方案。基于SDN三层结构,提出动态策略管理系统。动态策略管理系统可以根据反馈的网络状态信息调整网络策略,并对同一节点上的网络策略进行冲突检测,根据不同的冲突类型选择组合方式,有效避免了网络策略冲突,实现网络服务的自适应部署。最后,通过实验验证了动态策略管理系统从业务到策略的完整部署过程。