Dynamic network architectures

In the area of optical network architectures our research focuses on various aspects of wavelength routing and wavelength allocation in static and dynamic optical networks, including WRONs (wavelength-routed optical networks) and WR-OBS (wavelength-routed optically burst switched) networks.

In the quasi-static (traffic demand changes slowly) WRON architecture a wavelength is assigned to a source-destination pair and the question relates to the number of wavelength that is necessary to interconnect an arbitrary mesh network. We have made significant contributions in answering this question, and in particularly, in identifying the bounds on the minimum number of wavelengths required and helped explain why wavelength conversion does not reduce the number of wavelengths [1].

Whilst WRONs are relatively easy to analyse and design, one could argue that these networks are not able to respond to rapidly varying traffic demands and our current research is focused on the analysis and design of dynamic networks, including the WR-OBS architecture, proposed in [2], scheduling, service differentiation, burst aggregation and control.

[1] S. Baroni, P. Bayvel, "Wavelengths requirements in arbitrarily connected wavelength-routed optical networks", IEEE J. of Lightwave Techn., Vol. 15, No.2, 242-251 (1997).
[2] M. Dueser, P. Bayvel,"Analysis of a dynamically wavelength-routed optical burst-switched network architecture", IEEE J. of Lightwave Techn., Vol.20, No.4, 574-585 (2002).

Physical layer sub-systems

Here we focus on the development of burst mode subsystems for the implementation of dynamic optical networks and interoperabilty studies. In particular performance of physical layer subsystems that have been optimised for operation in burst mode networks. We have developed and experimental evaluated a number of subsystems including: wavelength agile transmitters, digital burst mode receiver and burst-compatible EDFAs. These subsytems have been evaluated using our extensive experimantal test facility in a realisitic dynamic network scenario that emulates the distortions and impairments that are expected to occur in dynamic optical networks.