通信网络的业务规划及保护关键技术研究
发布时间:2018-05-18 19:32
本文选题:路由规划与保护 + IP网络的快速重路由 ; 参考:《电子科技大学》2014年硕士论文
【摘要】:在近几年中,随着云计算技术和移动终端服务的快速发展,通信网络承载的业务流量迅速增多,通信网络在人们日常的生活工作中扮演的角色越来越重要,用户也对通信网络的承载能力、稳定性和健壮性提出了更高的要求。业务的路由规划和保护这一通信网络中的关键问题,对通信网络的承载能力、稳定性和健壮性有着极为重要的影响。对业务的路由规划直接影响到通信网络的业务接纳能力和资源利用状况,而对业务的路由提供保护则可以有效的提高通信网络的稳定性和健壮性。本文着眼于通信网络的业务路由规划和保护问题,重点关注了IP网络的业务路由保护问题和弹性光传输网络的业务规划问题。IP快速重路由是互联网工程任务组提出的快速响应IP网络的元件失效的框架标准。在网络发生失效后,IP快速重路由框架可以把业务恢复时间缩短至数十毫秒,从而保护业务不受失效的影响。在近十年内已出现了很多IP快速重路由机制,在这其中,无环保护路最简单最易于部署,但无法保证覆盖所有的失效场景。而另一方面,最大冗余树可以实现对单元件失效的100%保护,但是最大冗余树提供的保护路由更长,更容易引起流量集中和过载,引发网络拥塞。本文在分析对比了无环保护路和最大冗余树的优缺点后,提出了结合这两种机制的IP快速重路由机制MRT+LFA,以利用这两种机制的优点提高保护质量。仿真实验的结果显示,本文提出的IP快速重路由机制MRT+LFA,与无环保护路相比可以保护更多的业务不受失效影响,与最大冗余树相比可以有效的缩短保护路由、减少流量集中、避免网络拥塞。频谱弹性分切光路网络(SLICE networks)是一种可以实现灵活的频谱带宽分配的弹性光网络结构。与传统的波分复用(WDM)光网络相比,SLICE网络可以实现对频谱带宽的灵活分配,从而有效的提高频谱资源的利用率,增大网络容量,提高网络承载业务的能力。SLICE网络中的一个基础问题,是在业务请求接入网络时为业务计算路由和分配频谱资源,即路由与频谱分配(RSA)问题。本文关注SLICE网络在动态业务和静态业务两种场景下的路由与频谱分配问题,并提出相应的RSA算法。在解决动态RSA问题时,本文提出将动态RSA问题分解成路由问题与频谱分配问题两个子问题。本文提出一个多约束路由算法SFPS来解决路由问题,并证明了算法的正确性,分析了算法的时间复杂度。而在考虑频谱分配问题时,本文设计了两种基于频谱分段的频谱分配策略,固定分段(FS)和自适应分段(AS)频谱分配策略,以减少频谱碎片,提高资源利用率。仿真实验的结果显示,本文提出的动态RSA算法(SFPS+FS和SFPS+AS)与已有的动态RSA算法相比,可以有效的提高计算效率,降低网络阻断率,减少频谱碎片,提高资源利用率。在解决静态RSA问题时,本文提出将批量业务按照某种顺序逐个规划,从而将静态RSA问题转化成动态RSA问题加以解决。按照这种思路,本文提出静态RSA算法SA-mMRSA。该算法包括子算法SASP和子算法mMRSA。SASP算法采用模拟退火的搜索框架,搜索使得接纳业务最多的规划顺序。而mMRSA算法则在最大重用频谱分配算法框架中加入业务路由的计算,同时引入AS策略为业务分配频谱资源。仿真实验的结果显示,本文提出的静态RSA算法SA-mMRSA,在解决规模较小的静态RSA问题和业务频谱需求相同的静态RSA问题时,可以取得与整数规划模型相近的规划结果;在与已有的静态RSA算法相比时,可以有效的减少业务阻断,提高资源利用率。
[Abstract]:In recent years, with the rapid development of cloud computing technology and mobile terminal service, the traffic volume of communication network is increasing rapidly. Communication network plays a more and more important role in people's daily life work. Users also put forward higher requirements for the capacity, stability and robustness of communication network. The key problems in the communication network planning and protection have a very important influence on the capacity, stability and robustness of the communication network. The routing planning of the service directly affects the service acceptance and resource utilization of the communication network, while the protection of the service route can effectively improve the communication network. Stability and robustness. This paper focuses on the business routing planning and protection of communication networks, focusing on the service routing protection of the IP network and the business planning problem of the elastic optical transmission network (.IP fast rerouting) is the framework standard for the fast response of the IP network to the component failure of the Internet Engineering task force. After the failure, the IP fast rerouting framework can shorten the business recovery time to tens of milliseconds, thus protecting business from failure. In the last ten years, many IP fast rerouting mechanisms have emerged, in which the easiest and most easy to deploy, but not guaranteed to cover all failure scenarios. The large redundant tree can achieve 100% protection to the failure of the element, but the maximum redundant tree provides a longer protection route, which is easier to cause traffic concentration and overload, causing network congestion. In this paper, after analyzing and comparing the advantages and disadvantages of the acyclic protection path and the maximum redundant tree, the IP fast rerouting mechanism, MRT+LFA, which combines these mechanisms, is put forward. In order to improve the quality of protection by using the advantages of these two mechanisms, the simulation experiment shows that the IP fast rerouting mechanism proposed in this paper, MRT+LFA, can protect more services from the acyclic protection path. Compared with the maximum redundant tree, it can effectively shorten the protection route, reduce traffic concentration and avoid network congestion. SLICE networks is an elastic optical network structure which can realize flexible bandwidth allocation. Compared with the traditional WDM optical network, SLICE network can realize the flexible allocation of spectrum bandwidth, thus effectively improve the utilization of spectrum resources, increase the network capacity and improve the network bearing industry. A basic problem in.SLICE network is the routing and allocation of spectrum resources for Business Computing and allocation of spectrum resources, namely routing and spectrum allocation (RSA). This paper focuses on the routing and spectrum allocation problem of the SLICE network in two scenarios of dynamic service and static service, and proposes a corresponding RSA algorithm. In the dynamic RSA problem, this paper proposes to decompose the dynamic RSA problem into two sub problems of routing problem and spectrum allocation problem. In this paper, a multi constrained routing algorithm SFPS is proposed to solve the routing problem, and the correctness of the algorithm is proved and the time complexity of the algorithm is analyzed. In this paper, two kinds of algorithms are designed in this paper, which are based on the spectrum allocation problem. Spectrum allocation strategy, fixed segment (FS) and adaptive segmentation (AS) spectrum allocation strategy, to reduce spectrum fragmentation and improve resource utilization. The simulation experiment results show that the dynamic RSA algorithm (SFPS+FS and SFPS+AS) proposed in this paper can effectively improve the computing efficiency and reduce network blocking compared with the existing dynamic RSA algorithm. Rate, reduce spectrum debris and improve resource utilization. In solving static RSA problems, this paper proposes to plan batch business in a certain order to solve the static RSA problem into a dynamic RSA problem. According to this idea, the static RSA algorithm SA-mMRSA., which includes sub algorithm SASP and sub algorithm mMRSA.SASP, is proposed. The algorithm uses the search framework of simulated annealing to search the planning order that allows the most to accept business. While the mMRSA algorithm adds the computing of the business routing in the framework of the maximum reuse spectrum allocation algorithm, the AS strategy is introduced to allocate the spectrum resources for the service. The simulation experiment results show that the static RSA algorithm SA-mMRSA proposed in this paper is solved. When the small static RSA problem and the static RSA problem with the same business spectrum demand, the planning results similar to that of the integer programming model can be obtained. When compared with the existing static RSA algorithms, the business block can be reduced effectively and the utilization of resources can be improved.
【学位授予单位】:电子科技大学
【学位级别】:硕士
【学位授予年份】:2014
【分类号】:TN915.0
,
本文编号:1906930
本文链接:https://www.wllwen.com/kejilunwen/wltx/1906930.html