Dynamic assignment of wavelengths in agile photonic networks

Abstract

In an automatically switched optical network, the wavelengths are assigned to optical path based on their intrinsic physical performance and on the current network operating parameters. The wavelength performance information is organized in binning tables, based primarily on the wavelength reach capabilities. A network topology database provides the distance between the nodes of the network, which is used to determine the length of the optical path. Other network operating parameters needed for wavelength selection are also available in this database. Once a bin corresponding to the path length is identified in the binning table, the wavelength for that path is selected based on length only, or based on the length and one or more additional parameters. The optical path performance is estimated for the selected wavelength, and the search continues if the estimated path performance is not satisfactory. Several available wavelengths are searched and of those, the wavelength that is most used along the optical path in consideration or alternatively network-wide is selected and assigned. This method helps minimize wavelength fragmentation. The binning tables may have various granularities, and may be organized by reach, or by reach, wavelength spacing, the load on the respective optical path, the fiber type, etc.