Research On Network Security Situation Awareness Method Based On Quantum Neural Network

With the rapid development of Internet technology,the Internet has become a powerful driving force for social and economic development,the development of various fields is becoming more and more dependent on the network,and the importance of its security issues has gradually become increasingly obvious.Facing various threats existing in the network,compared to the traditional single point of protection,network security situational awareness can solve network security problems more efficiently,by extracting the security element indicators that can cause network situational changes within the network,completing the situational assessment,providing help for network security managers to make timely decisions.This paper mainly studies three aspects of network security situation index extraction,situation assessment,and situation prediction.With regard to the extraction of network security situation indicators,according to study the characteristics of network internal security index,classifies it,and gives the criterion for reasonably constructing the situation index set,so as to avoid blindly selecting the situation index and affect the result of situation assessment and prediction.For network security situation assessment,a bottom-up hierarchical situation assessment model is designed,and a fuzzy comprehensive evaluation method is selected to complete the assessment of the network security situation,when the fuzzy comprehensive evaluation method assigns weights to various indicators,scientific tomographic analysis method is used to determine the weight relationship among various indicators,effectively solve the situation assessment problem affected by multiple factors,and finally get the expected network security situation assessment results.For network security situation prediction,due to the many characteristics of network security situation element indicators and the characteristics of non-linear time series,combined with quantum computer system,this paper designs a new type of quantum neural network model,the input of the model is a discrete sequence.The hidden layer node is a quantum neuron,and the output layer is a common neuron,the quantum neuron includes type 0 and type 1 control which consists of a quantum revolving gate and a multi-bit controlled NOT gate,through a number of controlled not gate in target qubits to input,output feedback and thus the overall memory input sequence,using controlled not gate in the output of a number of quantum bits of entanglement for quantum neuron output,using L-M algorithm to train the model,this model can effectively make use of the characteristics of factor indexes to solve the shortcoming that the traditional prediction model is easy to lose the characteristics of factor indexes.Compared with the ordinary BP network,the convergence speed and approximation ability of this model are better than that of the ordinary BP network.This article builds a real network environment,extracts the measured data from the network,builds a network security situation index set,completes the network security situation assessment,and designs a network security situation prediction scheme based on the quantum neural network.Compared with the traditional network security situation prediction,the prediction model of quantum neural network based on sequence input designed in this paper has good prediction ability and prediction accuracy in solving the situation prediction problem,and can provide decision-making help for network security managers in a timely manner.