Data-driven Dynamic Mining of Vulnerable Nodes in Power Optical Fiber Communication Network
When vulnerable nodes in power fiber optic communication networks are affected by external factors or system interference,it can lead to the successive failures of line components,reducing the service life of the network.Therefore,a data-driven dynamic mining method for vulnerable nodes in power fiber optic communication networks is proposed.Firstly,in the big data environment,redundant nodes in the network are eliminated,and node information is exchanged based on the elimination results.After determining the location and connection of vulnerable nodes in the communication network,the abnormal range of the communication network is obtained.Then,the vulnerable performance characteristics of nodes in the area based on the divided communication anomaly range are extracted,and a vulnerable node mining model is established.Finally,based on this model,vulnerable nodes in the communication network are excavated,and dynamic mining of vulnerable nodes in the power fiber optic communication network is achieved.The experimental results show that when using this method to mine vulnerable nodes in communication networks,the average accuracy of node mining is 98%,the minimum network transmission delay is 19 ms,the consumption time of node mining is 0.42 s,and the node coverage is 96.88%.The mining performance of vulnerable nodes in communication networks is good.
big data environmentpower fiber optic communication networkvulnerable nodesdynamic mining methodmining model
万方数据
维普
