Research On Dynamic Routing And Wavelength Assignment Algorithms In Intelligent Optical Network

Intelligent Optical Network (ION), as the representative of Automatic Switching Optical Network (ASON) is the development trend of the next generation transport networks. Dynamic Routing and Wavelength Assignment (RWA) mechanism is one of the key technologies in IONs. Dynamic RWA algorithm can establish light path dynamically based on service connection requests which can compute and select a reasonable routing and assign suitable wavelength between source and destination nodes. This thesis presents a systematical and simulation study on dynamic RWA algorithms in IONs.Firstly, some fundamental principles and key technologies of ION are presented in this thesis. The static & dynamic RWA mechanisms and algorithms are analyzed in detail. Considering the wavelength allocation and network load balance conditions, an improved Ant Colony based RWA Algorithms, named Smart Ant Dynamic Routing Wavelength Assignment (SA-DRWA) algorithm is proposed. The link idle rate is introduced into SA-DRWA algorithm as new constraint along with the random perturbation to prevent searching converged into unexpected local optimum. Theoretical analysis shows that SA-DRWA algorithm can achieve network load balance in optical network effectively. SA-DRWA algorithm is both suitable to solve dynamic RWA problems and the allocation of wavelength converters in ION simultaneously. Finally, an optical network simulation platform is built to verify the performance of SA-DRWA algorithm.Simulations in different network topologies show that SA-DRWA algorithm has lower blocking rate and better resource utilization than traditional Dijkstra+FF algorithm. The best performance improvement is obtained in regular MESH network. Adding wavelength converter at some specified node can decrease the network blocking rate effectively especial for medium network load scenario.