Research On Virtual Resource Allocation Mechanism Of Intelligent Electric Communication Network

With the rapid development of the new generation of intelligent electric communication network(ECN),the number of network terminals and users of the electric communication network increases significantly.As a result,the quantity and types of services that need to be carried in ECN increase explosively,which causes the original communication network of the power distribution network cannot meet the requirements on differentiated transmission quality of services.Therefore,on the basis of the fiber-wireless network structure(FiWi)of ECN,the virtual network technology is used to shield the differences in the network topology,resource allocation and transmission protocol of the fiber-wireless network.Meanwhile,efficient and reasonable virtual network mapping algorithms that meet the reliability requirements of various services are designed to form virtual resource allocation algorithms for the ECN services.In addition,considering the transmission dynamics of the actual business in ECN,a dynamic allocation mechanism of virtual resources needs to be designed to guide the distribution of network resources in the distribution communication network more rationally and efficiently.Realize the optimal allocation of network resources in intelligent electric communication networkAiming at the differentiated demand for transmission reliability and bandwidth resource of various services of ECN,this paper studies the mapping scheme for the service reliability of ECN.The algorithm chooses a path protection scheme to ensure the reliability of high-priority services,and the main and auxiliary links adopt a two-stage mapping scheme.The algorithm proposed in this paper includes a primary link mapping algorithm based on reliability perception and risk balance,as well as a secondary link mapping algorithm based on bandwidth resource sharing.Then a business risk balance model is established.The main link solved by the algorithm not only meets the transmission quality requirements of different types of services,but also reduces the risk that a single point network failure affected more services,especially when there is a link with high failure rate.After the primary link is mapped,in order to reduce the network bandwidth resource overhead caused by path protection,this paper establishes a secondary link bandwidth resource sharing model.With the goal of minimizing network resource overhead,the restrictions are set as the primary link and the secondary link do not intersect as well as the main links whose secondary links share bandwidth resource do not intersect under certain conditions.Simulation experiments show that the algorithm proposed in this paper can improve the reliability of service transmission and reduce the blocking rate of service mapping at the same time.Considering that the dynamic changes in the demand for bandwidth resources of various services in ECN,this paper studies dynamic allocation algorithm of virtual resources based on RBF neural network.In this paper,when the service is initialized and mapped,the bandwidth resource is allocated according to the upper limit of the virtual service resource demand.In the case of limited network resources at the infrastructure layer and rapid business development,the fixed allocation of network resources is not conducive to the transmission and bearing of grid services.Therefore,combined with the dynamic and tidal characteristics of ECN,this paper establishes and trains the RBF neural network prediction model.By predicting the bandwidth requirements of the mapped virtual network services during the life cycle,resource occupation between services in the physical network is dynamically adjusted and managed.Resource occupation among various services in the physical network.Such scheme can ensure that when the network load of ECN is lower,the network resources of the infrastructure layer will not be idle,and when the network load is higher,the infrastructure layer can have more controllable and allocated resources.Through this algorithm,a more dynamic and flexible network service can be provided.By predicting the bandwidth demand of the mapped virtual network service in the next stage,the bandwidth resources between each service can be dynamically adjusted,and a resource adjustment mechanism that is allocated on demand can be provided.Simulation experiments show that the algorithm proposed in this paper can not only improve the acceptance rate of virtual network services,but also effectively improve the utilization of network resources while meeting the transmission requirements of different types of services.