The dramatic increase in throughput demands on transport systems has propelled the development of all-optical networks. These networks can provide tremendous capacity when they are designed with their own limitations in mind, such as coarse wavelength granularity and physical impairments. In this research we consider the holistic design of optical networks that include the interdependence of three network layers: the traffic grooming layer, the lightpath management layer, and the physical fiber layer.
The network is first viewed from the top-down, where sub-wavelength circuit requests arrive with specific quality of service requirements. Current traffic grooming approaches are altered to incorporate their dependence on the lightpath management and physical layer constraints. The system is then examined from the bottom-up, so that the quality of transmission and efficiency of resource utilization can be optimized as the higher layer protocols evolve. Total capacity is measured from an information theoretic view point and system optimization uses ideas from game theory.
The results of the research will be practical algorithms for improved capacity and survivability of future optical networks as well as providing a quantitative proof of their superiority. The enhancement of network capability will help satisfy our society?s ever-increasing need for information. It encourages the development of applications that require significant bandwidth. It also stimulates cross-fertilization of ideas from the two fields of networking and communications. The algorithms and software will be made publicly available via a web-site. The research enhances the education of the diverse group of graduate and undergraduate students participating in it.
