Wdm Optical Network Capacity Design Robust Research

With the explosive increase in network traffic and the emergence of high performance optical network devices, such as optical cross-connect(OXC) and optical add/drop multiplexer(OADM), wavelength division multiplexing(WDM) technology becomes the core technology of next generation backbone networks. As IP traffic, which is uncertain and unpredictable, drives the strong demand for optical networks to dynamically allocate bandwidth on demand to users. Thus, it requires reasonable configuration and ultimate of backbone network. At the same time, WDM technology provides the tremendous bandwidth, while poses the survivalbility issue urgent. In WDM network, the failure of a network component can lead to tremendous loss than in other traditional networks. Research on demand uncertainties and survivability in WDM networks has not only huge application value, but also deep theoretic value.Recently, a lot of novel methods for network design and configuration have been proposed. These design methods aim to make best use of the whole network resources by properly configuring routing schema and distributing bandwidth robust under traffic demand uncertainties or link-state uncertainties and thus are termed as"Robust Desigh"method. In this dissertation, we consider following problems: robust resource provisioning algorithms in WDM networks under traffic demand uncertainties, design of robust resilient in WDM network under link-state uncertainties, and linear programming design of several protection methods.Chapter 1: The dissertation's study background and the assignment of the structure would be introduced in this chapter.Chapter 2: The robust design in WDM networks under traffic demand uncertainties would be discussed in this chapter, and two novel optimization algorithms are proposed. In view of resource provisioning, first, an optimal Resource Reconfiguration (RR) algorithm is presented to minimize the average blocking rate. Second, considering the dynamic demand while initializing a network, Two-Phase Robust Provisioning (TPRP) algorithm is proposed. TPRP uses a regret value to estimate the total amount of over and/or underprovisioning in networks. Simulation result shows RR could effectively utilize the network resource and reduce the average blocking rate. And TPRP not only efficiently reduces the average blocking rate, but also proposed a robust method for uncertain traffic demands.Chapter 3: Design of robust resilient in WDM network under link-state uncertainties is studied in this chapter. In the view of restoration, three novel robust resilient algorithms are proposed, including Resilient Robust Optimization (RRO) algorithm, Stochastic Programming (SP) algorithm, and Worst Case Programming (WCP) algorithm. The main idea of these altorithms is to design a network which has good Quality of Serve (QoS) no matter which link sates appears. Simulation shows that each of these algorithms could produce an optimal design that is a robust design under link-state uncertainties, and RRO produces the best optimal design which has the best QoS.Chapter 4: Several protection methods would be discussed in this chapter, including shared path protection, shared link protection, shared segment protection and shared p-cycle protection. ILP formulations of these protection methods are presented in this chapter, and an advanced shared segment protection is also disscused in this chapter. Simulation shows and validates ILP model of the advanced shared segment protection is correct and reliable.Two main methods are used in our simulations: CPLEX10.0 and C++. The data used in simulations come from C++, and the prototypes of ILP models are designed on CPLEX10.0.And the last is summarizing of this dissertation.