【作者】 熊翱；

【导师】 孟洛明；

【作者基本信息】 北京邮电大学 ， 计算机科学与技术， 2013， 博士

【摘要】 传送网作为整个通信网的承载网络,其网络质量直接关系到通信网中各类业务网的质量,是通信网运营关注的重点和研究的热点。针对规模庞大、分布广泛、环境复杂的传送网,如何对其质量进行科学和全面的评价,从众多影响其网络质量的因素中找到关键因素,并采取有效措施进行优化,以提高传送网的整体质量,是通信网运营领域的一个难点。国内外研究人员在传送网网络质量评价和优化领域进行了大量的研究。由于缺乏对传送网网络结构的深入分析,传送网质量评价大都是从某些方面来评价传送网,对影响传送网网络质量的因素考虑不够全面,并且缺乏对传送网网络结构和承载业务的充分考虑,评价方法具有较大的片面性。在网络优化方面,主要关注的是网络设计方面,通过理论分析来优化网络结构,但是由于没有有效的质量评价模型,网络优化缺乏依据,并且没有有效的验证方法来验证其有效性。针对上述问题,论文对传送网质量评价模型和优化方法进行了研究,并对所提出的模型和算法进行了仿真验证。本文的主要研究内容和创新点如下：(1)论文详细分析了影响传送网运行质量的故障、电路服务质量、割接、性能等各个要素,给出了各要素的测量方法,在此基础上提出了传送网运行质量评价模型,全面反映传送网运行质量；(2)基于对传送网的深入分析和研究,论文指出传送网网络配置对网络质量具有很大影响。论文重点分析了影响网络质量的拓扑质量、传输系统质量、传输电路质量、资源利用率等主要配置要素,对各要素的测量方法进行了分析,并给出了配置质量评价模型。结合运行质量评价模型和配置质量评价模型,论文提出了能够全面、准确评价传送网网络质量的整体模型,并用一个完整的实例验证了模型的有效性；(3)基于传送网质量评价模型,论文对传送网网络优化的必要性、原则和方向进行了分析,提出了面向网络的宏观优化方法和面向电路的微观优化方法；(4)传送电路路由算法的优劣直接关系到传送网网络质量。论文提出采用蚁群算法实现电路路由选择,对蚁群系统工作模型和传送网电路路由选择原则进行了描述,指出传统蚁群系统在解决多约束路由选择问题中的不足,对传统蚁群系统进行了改进。论文采用改进后的蚁群系统对多约束电路路由选择算法过程进行了详细说明,并用实例验证了算法；(5)重要电路是传送网中需要重点保障的业务,对重要电路进行路由保护是提高传送网网络质量的重要方式。在对保护电路路由选择原则进行描述的基础上,论文指出通用路由算法无法实现保护路由的选择,提出采用多蚁群算法实现保护路由选择,对已有电路选择保护电路和同时选择主用及保护电路这两种不同情况下的保护路由选择算法分别进行了详细描述,并用实例验证了算法。更多还原

【Abstract】 As the supporting part of the whole communication network, transmission network is directly related to the QoS of all types of business network. The quality of transmission network becomes the focus of attention and research hotspot of communication network operation. In the large-scale, wide distribution and complex transmission network, it is the critical issue to evaluate quality scientifically and comprehensively, find the key influence factors and take effective measures to optimize and improve transmission network quality.A lot of researches have been done in the areas of network quality evaluation and optimization of transmission network. Due to the lack of in-depth analysis of the network structure of the transmission network, transmission network quality evaluation is focused on partial aspects of network, which is not comprehensive enough to consider the factors that affect the network quality, and also lack adequate consideration of the whole network structure of the transmission network and the bearer services on it. So these evaluation methods are incomprehensive. The main concern of network optimization is network design, using theoretical analysis to optimize the network structure. Because of the absence of an effective quality evaluation model, network optimization is lack of basis, and has no effective method to verify its validity.In response to these problems, this paper researches the quality assessment models and optimization methods for transmission network, and use some examples to verify the proposed model and algorithm. The main research contents and innovations of the paper are as follows:(1) This paper analyzed the elements which affect the running quality of transmission network, including fault, circuit service quality, cutover, performance, and gave the measurement of each elements. Based on these, we proposed the running quality evaluation model, which can fully reflect the quality of transmission network running;(2) Based on in-depth analysis and research of transmission network, the paper proposed that the network configuration affected the QoS greatly, analyzed the main influence factors such as the quality of the topology, the quality of the transmission system, the quality of the transmission circuit, the utilization of resource, and so on. Then analyzed their measuring method and proposed the evaluation model of configuration quality. Combined operating quality evaluation model and configuration quality evaluation model, the paper presented an overall model can evaluate network quality of the transmission network comprehensively and accurately, and then by using a complete instance, the validity of this model is verified;(3) Based on the quality Evaluation model of the transmission network, this paper analyzed necessity, principles and direction of the transmission network, then proposed network optimization network-oriented macro optimization method and circuit-oriented micro optimization method;(4) The transmission circuit routing algorithm is directly related to the network quality of the transmission network. The paper introduced the ant colony algorithm to realize circuit routing selection, described ant colony system model and the transmission network circuit routing selection principles, improving the lack of traditional ant colony system to solve the multi-constrained routing problem. By using the improved ant colony system, we describing the process of multi-constraint circuit routing selection algorithm, and using examples to verify this algorithm;(5) The important circuit routing protection is a key business need to focus on, and it also is an important way to improve the network quality of the transmission network. Based on the description of the protection circuit routing principle, the paper pointed out that universal routing algorithm cannot achieve the selection of protection routing, and presented using multi ant colony algorithm to achieve it. For two different circumstances:selecting protection routing of existing circuit and selecting the primary and protection circuit at the same time, we described circuit protection routing algorithms separately, and validated it by examples.更多还原