Next-Generation Optical WDM Mesh Networks

University Of California-Davis, Davis CA

Investigators

Abstract

This project aims at creating new knowledge on the design, analysis, and optimization of novel system architectures, network protocols, and software algorithms for building next generation optical Wavelength Division Multiplexing (WDM) backbone networks. Particular research focus will include hierarchical optical switching, optical multicasting, and optimal topology reconfiguration. For hierarchical optical switching, the generalized optical switch architecture with multiple levels will be developed where the advantages of both optics and electronics will be exploited. For example, a hybrid switch architecture will include wavebands (bands of wavelengths) being switched all-optically, while wavelength and sub-wavelength connections may be switched with optical-electrical-optical (OEO) switches. Results from this investigation will provide guidelines on where to strategically place resources such as hierarchical optical switches. This project will also design optical multicasting switch architectures and develop mathematical models and efficient algorithms for supporting light-trees in a network with hierarchical optical switching. The research team will develop new efficient models for network operators as well as will design new algorithms for protecting light-trees using various schemes such as link protection and shared protection. In particular, the signaling protocols for set up, maintenance, and termination of light-tree based connections will be designed and studied by simulation. Finally, in the part of the research related to the virtual topology design, the reconfigurations of WDM networks will be studied so that the bandwidth upgrades are taken into account, e.g. instead of fully taking down a lightly-loaded lightpath or adding a maximum-capacity lightpath, an existing lightpath can be upgraded when a fraction of its free capacity is needed. In particular, traffic fluctuations, such as within the global optical network over the time zones, can create another dynamism, i.e., traffic intensity peak rates. In this context, the research team will study different virtual topology-design methods to obtain a better understanding of this and similar problems related to dynamic traffic fluctuations.