本文提出一种新颖的动态全光网络监控分布式算法，该算法可估计光网络中光纤链路上的误码率，在不需要额外光监控元件的情况下同时监控，检测和定位多处光纤链路损坏。 / 在光网络传输过程中，各个终端结点的接受机可以时时地估计出收到光流的误码率，这些误码率信息可以通过扩展OSPF-TE协议在全网共享。基于这些共享的误码率信息，我们将光纤损坏检测问题抽像成一个线性编程（LP）算法，其中每一个误码率信息代表一个限制条件。我们之后运用一些算法优化技巧将这个问题的维度和复杂度大大地降低，以便可以直接嵌入到每个网络结点可能自带的微处理器单元中进行实时计算运用。本文提出的算法同时适用于没有光波长转换器的光网络和配备光波长转换器的光网络。 通过沿用OSPF协议的分层多域思想，大规模网络可以分化成小的域和连接各域的主干网络，从而可以将一个复杂的大规模网络检错问题转化成一系列简单小网络检错问题。通过将该算法在一个由408 节点组成，支持40波长的大规模GMPLS 网络仿真平台上仿真，算法的有效性得到了验证。 / 为了保证用于仿真的网络流量模型合理且符合实际，本文也对动态全光网络流量模型做了一定研究。在自相似网络流量模型下，我们发现长短光流的不公平性问题可以给动态全光网络带来很大问题，会大大地降低网络的吞吐率。我们运用一种截短长光流的方法可以将这个问题很有效地解决。 / 据我们所知，这是目前唯一的一个能运用于现实中超大规模光网络的低成本可实现且可以作到波长级监控和同时监控多个链路错误的算法。该方法可以不用额外添加昂贵的光监控元件就可实现对动态全光网络的监控，并且该方法同时适用于透明，半透明及配置波长转换器的光网络。 / A new and efficient distributed algorithm for estimating the bit-error-rate (BER) of links in dynamic optical networks is proposed. The method can be used to monitor, detect and localize multiple soft link-failures without incurring any additional optical monitoring equipment. During the transmission of each optical flow the end node’s receiver can estimate the digital BER information, and the BER information can be shared among the network by extending the Open Shortest Path First-Traffic Engineering Extension (OSPF-TE) protocol easily. We model the faults localization problem as a linear programming (LP) algorithm, where each BER information measured from a flow serves as a constraint. Optimization techniques are applied to significantly simplify the complexity of the LP algorithm in order to make it solvable in real time by an integrated processor attached to the network node. The proposed algorithm is capable of monitoring networks with or without wavelength converters. A large scale network can be divided into several layers according to the OSPF protocol, thus the algorithm can be applied to large networks in the real world similar to OSPF. The monitoring algorithm is demonstrated by network simulations over a 408-node, 40-wavelength network test-bed where up to twenty faulty links are identified. / To make sure the traffic generator model is reasonable, the traffic model for dynamic all-optical network is also studied in this work. Under self-similar traffic, we found that the dynamic optical networks suffer from the long flow short flow unfairness problem, which would reduce the throughput as well. So a segmentation strategy is proposed to solve this problem. / To the best of our knowledge, this is the first realistic and low-cost framework which can monitor channel level BER changes to identify multi-link-failures efficiently for large scale dynamic all-optical WDM networks, without using expensive optical monitors or additional supervisory channels. The approach proposed is applicable to transparent, translucent and wavelength-converted optical networks. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Li, Huadong. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2012. / Includes bibliographical references (leaves 64-66). / Abstracts also in Chinese. / Chapter Chapter 1 --- Introduction --- p.1 / Chapter Chapter 2 --- Backgrounds --- p.11 / Chapter 2.1 --- ROADMs, Dynamic networks --- p.11 / Chapter 2.1 --- Types of failures considered: --- p.13 / Chapter 2.2 --- Brief review of OSPF routing protocol --- p.15 / Chapter Chapter 3 --- Traffic model used --- p.16 / Chapter 3.1 --- Introduction --- p.16 / Chapter 3.2 --- LFSF unfairness problem --- p.19 / Chapter 3.3 --- Flow segmentation strategy --- p.23 / Chapter 3.4 --- Simulation results --- p.24 / Chapter 3.5 --- Summary and Conclusion --- p.29 / Chapter Chapter 4 --- Estimated digital BER monitoring and faults diagnosis algorithm --- p.31 / Chapter 4.1 --- Intra-domain faults diagnosis algorithm --- p.31 / Chapter 4.2 --- Hierarchically layering scheme for inter-domain network monitoring --- p.37 / Chapter Chapter 5 --- Simulation results and analysis --- p.40 / Chapter 5.1 --- Simulation set up --- p.40 / Chapter 5.1.1 --- 100Gbps simulation set up --- p.40 / Chapter 5.1.2 --- 10Gbps simulation set up --- p.42 / Chapter 5.2 --- Simulation results --- p.44 / Chapter 5.2.1 --- 100Gbps simulation results: --- p.44 / Chapter 5.2.2 --- 10Gbps simulation: --- p.51 / Chapter 5.3 --- Conclusion --- p.61 / Chapter Chapter 6 --- Conclusion --- p.62 / Reference --- p.64
| Identifer | oai:union.ndltd.org:cuhk.edu.hk/oai:cuhk-dr:cuhk_328710 |
| Date | January 2012 |
| Contributors | Li, Huadong., Chinese University of Hong Kong Graduate School. Division of Information Engineering. |
| Source Sets | The Chinese University of Hong Kong |
| Language | English, Chinese |
| Detected Language | English |
| Type | Text, bibliography |
| Format | electronic resource, electronic resource, remote, 1 online resource (5, 66 leaves) : ill. (some col.) |
| Rights | Use of this resource is governed by the terms and conditions of the Creative Commons “Attribution-NonCommercial-NoDerivatives 4.0 International” License (http://creativecommons.org/licenses/by-nc-nd/4.0/) |
Page generated in 0.002 seconds