Research And Application Of Multi-dimensional Network Model For Digital Twin Networks

The digital twin network includes the physical target network and its corresponding virtual twin,and the state of the two is synchronized in real time and information can be bidirectionally exchanged in real time.Data acquisition is performed on the target network to construct a target network model,and a twin network is created based on the data and the model.The twin network can be used to efficiently simulate,control,analyze and diagnose the target network.At present,the related research on digital twin network technology is still in its infancy,and there are still a series of technical difficulties and challenges in the digital twin reproduction of various heterogeneous networks.Accurately modeling the target network and comprehensively describing the information at all levels of its network is one of the main problems faced in the research of Siamese network construction technology.For the technical requirements of the network model in the construction and synchronization process of the digital twin network,the model construction problem for the target network is studied.The all-round and high-precision modeling of various main attributes of the target network from multiple dimensions are explored,and a unified converged multidimensional network model architecture is built in this thesis.The research object of this thesis is computer communication network,which has the characteristics of large number of nodes and links,complex network information and dynamic topology change.Through the network simulation technology based on the multi-dimensional network model architecture,the twin network construction and synchronization of the computer communication network are studied.Firstly,in-depth research and analysis on the requirements at all levels of the twin network construction and synchronization process are conducted in this thesis.Secondly,various network elements required for the twin network construction and synchronization process,abstracts and models the four dimensions of network entities,network rules,network services,and network behavior,and innovatively proposes a multi-dimensional network model architecture are summarized in this thesis.Various models are described using XML language,and the overall realization of the multi-dimensional network model architecture is simplified by using the strategy pattern.By analyzing the relationship between the four dimensions of the above model framework,the four-dimensional model is integrated and fused from the structure and function to form a multi-dimensional network model architecture.Then,in order to evaluate the multi-dimensional network model architecture,a digital twin network model evaluation index system is proposed in this thesis,and the traditional network model and the multi-dimensional network model proposed are compared and analyzed in this thesis in terms of accuracy,efficiency and flexibility.The analysis results show that,compared with the traditional network model,the multi-dimensional network model has a 22.7% increase in the accuracy of the parameters created by the digital twin network model,and an 11.2% increase in the accuracy of the synchronization parameters of the digital twin network model.In order to demonstrate the practical application of the multi-dimensional network model proposed by the research in the construction of twin networks and state synchronization,a target network of 5000 nodes is built,information of various dimensions of the target network is collected,and network virtualization on the infrastructure based on the virtualization platform is used in this thesis.It uses technologies such as network virtualization and host virtualization to create a twin network that is consisted of the target network through multi-dimensional network model drivers.The experimental results show that the multi-dimensional network model can not only fully and accurately reproduce the initial topology of the target network,but also synchronize the state of the target network.