Research On Virtual Mapping Algorithm In Space Division Multiplexing Elastic Optical Network Environment

The explosive growth of network traffic has brought challenges to optical networks with limited spectrum resources.The capacity of single core optical fibers approaches the physical limit in the frequency domain,while space division multiplexing elastic optical networks can expand the network transmission capacity from the spatial dimension.Meanwhile,network virtualization is a technology that abstracts physical network resources into multiple logical networks,which can improve the utilization and flexibility of network resources.Therefore,the combination of virtualization technology and space division multiplexing elastic optical networks can further improve resource utilization.Due to inter core crosstalk,which can reduce the transmission quality of services,crosstalk perception is the key to the research of virtual network mapping in space division multiplexing elastic optical networks.In addition,with the continuous increase in energy consumption,a large amount of greenhouse gases are emitted into the atmosphere,causing irreversible damage to the global ecological environment.Therefore,in this context,this thesis focuses on the virtual network mapping problem in the space division multiplexing elastic optical network environment.Firstly,considering the existence of inter core crosstalk,a virtual network mapping based on crosstalk perception is proposed;Secondly,in-depth research has been conducted on the energy consumption issue in the virtual network mapping of space division multiplexing elastic optical networks.The main research content of the thesis is as follows:Firstly,in order to address the crosstalk issue in space division multiplexing elastic optical networks,this thesis designs a crosstalk impact assessment method to dynamically evaluate the crosstalk impact between adjacent fiber cores.On this basis,a virtual network mapping algorithm for crosstalk perception in space division multiplexing elastic optical networks is proposed.In the node mapping stage,this article uses the virtual node proximity sorting method to sort virtual nodes,with the aim of strengthening the correlation between virtual nodes and virtual links.The physical node priority evaluation method considers inter core crosstalk on adjacent links of nodes to achieve balanced inter core crosstalk effect.In the virtual link mapping stage,design a spectrum partitioning allocation method to reduce inter core crosstalk.The simulation results show that the proposed algorithm improves the request acceptance rate and crosstalk improvement rate of the virtual optical network when the load is 100-900 Erlang.Secondly,in order to address the energy consumption issue in virtual network mapping in space division multiplexing elastic optical networks,this thesis establishes an objective function for virtual network mapping that minimizes the spectral resources consumed by virtual links,minimizes the maximum occupied frequency slots,and minimizes energy consumption.Based on this,this article proposes an adaptive discrete slime mold differential evolution virtual network mapping algorithm to optimize and solve the established objective function.Through simulation experiments,it can be found that the proposed algorithm optimizes network performance and reduces energy consumption by an average of 13.2% in NSFNET networks.