Traffic grooming in WDM networks using genetic algorithm

The traffic-adaptive concept has become the key technology in WDM (optical wavelength division multiplexing) networks. A genetic algorithm (GA) is proposed for the wavelength assignment in WDM rings that minimize overall network cost. The network cost includes the cost of the transceivers required at the nodes as well as the number of wavelengths. By the design of a genetic structure, the evaluation of fitness, the selection of parents, and the operation of reproduction, simulation results demonstrate the superiority of the GA approach when comparing with a class of the existing heuristic strategies. To achieve better performance, the relationships both the cost distribution and the number of generation are also discussed.