Research On Network Situation Prediction Model Based On GA-Elman Neural Network

In recent years,the rapid development of mobile Internet has brought great convenience to people's lives,but also brought serious network security issues.As people now work on a strong dependence on the network,the loss of network attacks are increasingly difficult to bear.Traditional network security devices are independent of their work,there is no comprehensive analysis of the information between the various devices that can not make an assessment of the overall network security situation.Network Security Situation Awareness(NSSA)has become a hot research area in the field of network security in order to master the security situation of the network and forecast the development trend of the future network situation.NSSA makes a quantitative assessment of the current state of the network based on the extraction of various factors of network security and uses the historical situation information to predict the future state of the network situation.NSSA can significantly improve the network management personnel to control the network.Based on the detailed study of NSSA model and related technology theory,aiming at the time series characteristics of network situation information,this paper improves the standard Elman neural network structure and establishes the situation forecasting model based on GA-Elman.Because the standard Elman neural network model only stores the historical information of the hidden layer neurons,this paper improves the structure of Elman neural network by adding feedback nodes so that it can store more historical information of neurons.In order to solve the problem that the error back propagation algorithm is easy to fall into the local optimum,a genetic algorithm with global search ability is used to train the network.Finally,the Elman network model based on genetic algorithm was used to predict the situation,and the experiment proved that the model could predict the future situation accurately.