Abstract:
In Wavelength Routed WDM Networks a routing and wavelength assignment (RWA) algorithm is responsible to make a routing and wavelength assignment decision upon the arrival of a lightpath connection request in the network using a predefined performance metric such as Blocking Probability. Both routing and wavelength assignment decisions have a profound effect on the Blocking Probability of the network or how much traffic a network can accommodate under a given traffic load. Our thesis work involves performance evaluation of some currently available existing routing algorithms (FAR, LLR and WLCR) using simulation techniques in sparse and full wavelength conversion scenarios to demonstrate their performance shortcomings under Wavelength Convertible WDM Networks. Than we propose a new routing algorithm Enhanced WLCR (EWLCR) that is dynamic in nature and circumvents the weaknesses of the previously available routing algorithms in Wavelength Convertible Networks. EWLCR also considers the criticality weight of a route when making a routing decision along with other performance parameters. Our simulation results demonstrate that a more consistent performance is possible under a diverse set of network topologies and wavelength conversion scenarios in terms of overall network blocking probability-using EWLCR. An additional benefit is that performance advantage comes without incurring any extra overhead of global network state information as compared to other dynamic routing algorithms such as LLR and WLCR.
