Optimization of Wavelengths, Grooming and Conversion Resources in All Optical WDM Networks

Abstract

Wavelength division multiplexing (WDM) in all optical networks is a popular optical networking technology which involves routing and wavelength assignment (RWA) problem as a key aspect to be dealt with. The RWA problem is known to be NP-complete integer linear programming (ILP) problem. Most of the previous work in the field focuses RWA problem in optical networks with static connection requests and wavelength conversion. In this paper our study attempts to provide better utilization of wavelength and devices for wavelength conversion and grooming, thereby leading to minimization of number of wavelengths and cost of wavelength conversion and grooming. The proposed genetic algorithm (GA) heuristic with single objective and multi objective fitness functions provide better results for the RWA adopting first fit wavelength assignment and most used wavelength assignment methods. The simulation studies have been carried out over 14-node NSFNET network with the objective of minimizing blocking probability. We observe that multi objective heuristic adopting most used wavelength assignment outperforms the single objective heuristic.