Ant colony optimisation for virtual-wavelength-path routing and wavelength allocation

Ant Colony Optimisation (ACO) is applied to the problem of routing and wavelength allocation in a multi-wavelength all-optical virtual-wavelength-path routed transport network. Three variants of our ACO algorithm are proposed: local update (LU), global update/distance (GU/D) and global update/occupancy (GU/O). All three extend the usual practice that ants are attracted by the pheromone trail of ants from their own colony: in our work, the artificial ants are also repelled by the pheromone of other colonies. Overall, the best ACO variant, GU/O, provides results that approach those of an earlier problem-specific heuristic on small- and medium-sized networks.