Font Size: a A A

Design And Simulated Implementation Of Game Theory Based Intelligence Optimization Multicast Routing Algorithm In Multi-domain Optical Networks

Posted on:2011-08-25Degree:MasterType:Thesis
Country:ChinaCandidate:C S WangFull Text:PDF
GTID:2248330395958071Subject:Communication and Information System
Abstract/Summary:PDF Full Text Request
With the network size getting larger gradually, the entire network will be divided to many routing areas according to geographical span, technical requirements, management, flexibility and scalability requirements, where the networks between different network operators and different networks within every network operator are all different routing areas. To ensure the network scalability and security, the detailed information of the connection and bandwidth allocation in one domain will be limited only to itself, and only some aggregated information will be exposed to the external domains. As a result, no nodes in multi-domains have the global information about the whole multi-domain network. In such circumstance, how to provide excellent routing to establish connections is a new problem which is brought by the separation of optical networks.This thesis focuses on the multicast routing technology of multi-domain optical networks. Thus, a multicast routing algorithm in multi-domain optical networks is proposed. The algorithm considers the uncertain information in the networks. In order to support flexible QoS, fuzzy method is adopted to consider both the network service providers and users comprehensively, and also the microeconomics and game theory are used to achieve the "double win" between them.Due to the NP-complete nature of multicast routing problem in multi-domain optical networks, based on the established Logical Topology, multi-agent evolutionary algorithm is used along with the routing in Physical Topology together such that the multicast routing problem in multi-domain optical networks can be addressed through jointing the two layered routes in logical topology and physical topology. Meanwhile, the related wavelength assignment algorithm is embedded so that the problem of routing and wavelength assignment can be solved integratedly.To evaluate the performance of the proposed routing algorithm above, simulation software has been designed and developed based on Microsoft VC++. Performance analysis and tests have been done, and the simulation results have shown that the proposed routing algorithm in this thesis can solve the multicast routing problem in multi-domain optical networks efficiently and effectively.
Keywords/Search Tags:Optical networks, Multi-domain, Game theory, Quality of Service(QoS), Multicast, Intelligence optimization, Multi-agent evolutionary algorithm
PDF Full Text Request
Related items