| Optical networks and Wavelength-Division Multiplexing (WDM) have been widely studied and utilized in recent years. By exploiting the huge bandwidth of optical networks, WDM appears to be one of the most promising technologies to meet the dramatically increased demand for bandwidth. Since optical resources in optical networks are very expensive, development of dynamic lightpath allocation strategies, which utilize network resource efficiently, is an important area of research. We assume that there is no optical wavelength conversion device in the network, and the wavelength-continuity constraint must be satisfied.; Exact optimization techniques are typically too time-consuming to be useful for practical-sized networks. In this thesis we present a tabu search based heuristic approach which is used to establish an optimal lightpath dynamically in response to a new communication request in a WDM network. As far as we know, this is the first investigation using tabu search techniques for dynamical lightpath allocation in WDM networks. We have tested our approach with networks having different sizes. And then we have compared our results with those obtained using the MILP approach. In the vast majority of cases, tabu search was able to quickly generate a solution that was optimal or near-optimal, indicating that tabu search is a promising approach for the dynamic lightpath allocation problem in WDM networks. |
