Routing And Spectrum Allocation In Elastic Optical Network

With the rapid development of broadband network and Internet as well as the popularity of intelligent terminals, the demand of network bandwidth is growing exponentially. All-optical network with large capacity and dynamic flexibility is the trend of future network development. The traditional WDM network whose wavelength assignment is fixed, can’t flexibly meet business needs, causing the waste of source. Elastic optical networks divide the network spectrum resource into small frequency gap by using fine points. According to business needs, elastic optical networks allocate appropriate spectrum resources to each vocational work and achieve intelligent management and allocation of network, improving the using efficiency of network bandwidth resources.This thesis mainly introduces the research on the routing spectrum allocation and spectrum fragmentation reconstruction of elastic optical network.Firstly, according to source model of elastic optical network, a new description method for network state of spectrum adjacency and the spectral window is proposed. Two core issues of elastic optical network are studied based on the description method.Secondly, for routing allocation, KSP algorithm is used; for spectrum allocation, a new spectrum flexible allocation algorithm is designed. It is named maximum spectrum adjacency. The function of the algorithm is realized by using network simulation tools on elastic optical network s imulation platform.Thirdly, for spectrum fragmentation reconstruction, the research points of it is described in detail. A reconstruction algorithm based on alternate path is put forward and simulated. This algorithm is compared with shortest path algorithm.The simulation results show that:the MSA algorithm can significantly improve performance of the network blocking probability and increase the using rate; the alternate path can effectively improve the performance of the network according to shortest path algorithm.