Research On Spectrum Resource Allocation Mechanism For Space Division Multiplexing Elastic Optical Networks

Nowadays,Multi-type services with high-bandwidth requirements are constantly emerging in the network.The network traffic is increasing exponentially,which puts forward higher requirements on the carrying capacity of optical transmission network.Space division multiplexing elastic optical network(SDM-EONs)with multi-core fiber(MCF)introduces spatial dimension,provides higher optical network transmission capacity and more flexible service carrying capacity.It has become an effective method to solve the bottleneck problem of transmission capacity in the network,which has received widespread attention.The introduction of the spatial dimension in EONs not only expands the transmission capacity of optical network,but also increases the complexity of spectrum resource management and allocation.In addition,the inter-core crosstalk and the inherent linear and nonlinear impairment in the core together produce a more complex deterioration effect on the quality of transmission(Qo T)of optical signals.Therefore,how to give full play to the advantages of SDM optical network transmission capacity,while eliminating the limitation of physical layer impairment constraints to ensure the Qo T of connection requests has become a concern in the research field.In this context,based on the mechanism of physical layer impairment constraint in spectrum,the degree of dynamic influence of spectrum state change on lightpath is explored.Further,form the spectrum resource optimization research in SDM-EONs and improve the Qo T and spectrum utilization efficiency.The main research works are as follows:First,aiming at the degradation of Qo T for lightpaths caused by inter and intra cores impairment.According to the characteristics of inter-core crosstalk affecting the lightpaths,we design the strategy of core grouping and preset spectrum allocation direction,in order to reduce the overlap of lightpaths in adjacent cores to the greatest extent.Moreover,considering the influence of dynamic spectrum condition on nonlinear impairment of lightpaths,a quantitative model of nonlinear impairment influence is established.On this basis,a spectrum resource allocation algorithm with minimal impairment is proposed.This algorithm selects the spectrum resource with the least influence for connection requests and reduce the lightpaths that cannot be transmitted normally due to the Qo T constraint.The simulation results show that the proposed algorithm can reduce blocking probability and improve the average Qo T of lightpahts in COST239 and NSFNET network topologies.Second,aiming at the problem of accumulated impairment to network caused by the establishment new lightpaths.Based on the analysis of the constraining effect of the joint impairment within and between cores on spectrum allocation,a quantitative model of impairment considering the dynamic spectrum state was established.Then,a spectrum resource allocation algorithm based on dynamic joint impairment perception was proposed.This algorithm preferentially selects the spectral position with the least influence on the existing lightpaths to minimize the accumulated impairment in the network.The simulation results show that the proposed algorithm has lower blocking probability and higher average Qo T of lightpaths in N6S9 and NSFNET network topologies.