Photonic network design based on reference circuits

The objectives of this presentation are to clarify specific terms like transparency and transverse compatibility, and then to derive guidelines as a first approach to an engineered photonic network. These guidelines are applied to the planning of a core network with 8 and 16 wavelength channels per link and verified by first numerical results. Complementary to a layered network architecture, our methodology is based on the use of a specific reference configuration. Degradation effects like amplifier noise, chromatic and polarization-mode dispersion, nonlinear self-phase modulation are covered as well as node crosstalk and the impact of optical frequency misalignments. Based on ITU-T recommendations, a classification of ranges of bit-rates and other preliminary specifications, our method allows us to assemble a general photonic network from its elements in a bottom-up scheme. As a result, we show that photonic networks can exhibit transparent optical paths, ranging from 400 to several thousands of kilometres. A number of 16 wavelength channels at individual bit-rates of up to 10 Gbit/s traversing a couple of crossconnecting nodes can be implemented, taking into account present-day optical components like amplifiers, standard fibres, multiplexers and demultiplexers, fibre switches as well as dispersion-compensating techniques. The potential benefits of such networks are to be seen in their inherent high capacity and in a high degree of flexibility, supporting various applications. Considering the results obtained so far, it can be concluded that a country of the size of Germany could be covered by a transparent photonic network