Research On Waveband Grooming Algorithms For WDM Networks

With the rapid development of WDM (Wavelength Division Multiplexing) technology, the growing number of optical cross-connects not only increases the complexity of switching nodes, but also substantially increases control costs and related administrative costs of the node, these hinder the further development of WDM technology. So higher requirements are put forward for switching architecture and traffic grooming in WDM optical network and multi-granularity optical switching technology comes into being. Research on this technology is mainly about the architecture of Multi-Granularity OXC (MG-OXC) and waveband traffic grooming algorithms.Multi-granularity optical switching can reduce the number of wavelength switching ports for DWDM systems, but we have observed that too many switching ports between the waveband switching plane and the wavelength switching plane are needed in traditional MG-OXC, which adds complexity to switching architectures of MG-OXC. A new multi-granularity optical switching architecture is proposed in this paper. In the proposed architecture, the wavelength switching plane and the waveband switching plane are completely independent from each another, which simplifies the switching architecture and reduces switching ports of the wavelength switching plane efficiently. To improve the waveband utilization, traffic grooming algorithms for the proposed new architecture are presented. Simulation results show that presented algorithms decrease the traffic blocking possibility efficiently, and hence improve the performance of the network.In addition,based on the optimization of same end traffic convergence strategy, a new waveband grooming algorithm based on arma prediction is proposed. The simulation results show that the new strategies can effectively improve the wavebands utilization and the performance of traffic grooming stratery.