Intelligent Optical Network Layer Domain Routing And Survivability

Intelligent optical networks (ION) are the most promising candidate for the future transport networks, which not only meets the increasing needs of bandwidth, but also copes well with the bursty IP data traffic. In the intelligent optical networks, connections can be set up and released dynamically, the network resource can be assigned and reconfigured effectively, and differentiated bandwidth provisioning services are provided. Routing and survivability are two vital technologies which greatly impact the performance of optical networks. Meanwhile they are distinct and relative.In the support of China 863 high-tech program projects, China Nature Science Found projects the author has done a lot of work in the key areas of the ION. The main contribution and innovation of this dissertation are described as follows.1,The hierarchical routing infrastructure in the intelligent optical network is researched. Four algorithms which based on the fixed alternate hierarchical routing is proposed. The dissertation considers the resource balancing of multi-layer and route backtracking mechanism in these algorithms. The simulation results show that they can reduce the average block probability and acquire acceptable time performance simultaneously, and the network performance can be improved significantly by them.2,An inter-domain routing optimization scheme which based on the parallel ant colony algorithm is proposed. The dissertation also suggests an implementation model of the parallel ant colony algorithm in inter-domain environment. For the survivability technology in mesh optical networks, the dissertation proposes two new cycle-based protection methods. Compared with other cycle-based protection methods, methods proposed have many advantages such as simple configuration, independent of traffic model.3,Based on source routing, a dynamic inter-domain traffic grooming algorithm with two improved methods using the fixed alternate routing (FAR) is proposed. The average block probability can be reduced, the routing loops can be avoided efficiently and the network resources (wavelength and transceiver) can be saved by the proposed algorithms. For the combine survivability, the dissertation proposes a traffic grooming method in WDM optical network, including a solution method of SRLG restriction.4,An implementation method of the platform which supports dynamic hierarchical optical networking, structure adjustment (splitting, merging, etc.), routing area identifying, routing and survivability is proposed. The dissertation proposes a novel cost-effective capacity planning method of WDM networks to solve the planning problem for the increasing of traffic load, including a network cost model and an improved ant colony algorithm. The dissertation also proposes an implementation method of a network planning software.