Abstract
In the recent years many researchers have addressed the problem of energy efficient routing in the WDM optical networks. Typical strategies involve putting the network components into sleep mode or switching off the network components in low utilization periods. In this paper, we propose a strategy based on the principle of anycast routing, where it is possible to select the most suitable destination, from a set of available destination nodes to complete the work. A novel genetic algorithm is used for solving this problem for a set of scheduled traffic demands, where the start and end times of the demands are known in advance. The fitness function used in the genetic algorithm not only minimizes the power consumption of the network but also minimizes the overall cost of optical transceivers needed to implement the network.