Skip to Main Content
The emergence of translucency in WDM networks as a trade-off between the low cost of full transparency and high signal quality provided by full opacity has led to immense interest in translucent network design by the research community in recent years. Efficient translucent network design necessitates the judicious choice of regeneration sites and number of regenerators in order to maintain a predefined quality of transmission and to reduce the overall network cost. In this paper, we explicate a framework that corroborates the offline version of Physical Layer Impairment Aware Routing and Wavelength Assignment (PLI-RWA) problem in translucent networks. The originality of the proposed PLI-Signal Quality Aware RWA (PLI-SQARWA) algorithm lies in finding the routes which use fewest amounts of regenerators and constituting an efficient Regenerator Placement (RP) algorithm to place regenerators before wavelengths are assigned leading to maximal use of regenerators as wavelength converters. Subsequently, in the WA phase, for any further wavelength contention resolutions; we resort to optical wavelength converters. A relevant difference between existing studies and the proposed algorithm is that PLI-SQARWA provisions both, regeneration and wavelength conversion which guarantees zero signal and wavelength blocking. We proceed to the performance comparison of PLI-SQARWA with a heuristic for RWA and RP called COR2P (Cross-Optimization for RWA and RP) and the comparison results reveal that PLI-SQARWA outperforms COR2P in terms of number of network regenerators and time delay while demonstrating similar blocking performance to COR2P at various traffic loads. The results also suggest that to minimize time delay due to Optical-Electrical-Optical (OEO) conversions, using optical wavelength converters for only wavelength contention resolution is a judicious choice rather than resorting to regenerators. Finally, it can be inferred from the study that there exists a trade-off between- the overall network cost and the maximum tolerable delay.