Research On Risk Assessment Method Of Information System Security Based On Quantum Neural Network Theory

Information security risk assessment is a comprehensive evaluation process of uncertain and random potential risk.It can clarify the security status of the system and the main security risk so as to provide effective guarantee for information system security.There are many limitations in the existing assessment methods.On the one hand,the traditional assessment methods are deterministic algorithms and models which have difficulty in measuring the uncertain and random security risk.On the other hand,with the openness and complexity of information system business functions,the nonlinearity and complexity of assessment calculation are also increasing.Quantum information theory is a new interdisciplinary subject which integrates quantum physics and information theory.It can effectively solve uncertain problem by utilizing quantum probability and parallel computing advantages.The neural network has the intelligent characteristics of self-learning and self-adaptation which is suitable for dealing with non-linear problem.Therefore,this thesis has focus on exploring the application of quantum neural network algorithms in information security risk assessment.The research work can be summarized as follows:1.This thesis explores a method of information security risk assessment based on quantum gate circuit neural network.Firstly,on the basis of the analysis of information security risk characteristics and security system,aiming at strengthening the analysis of vulnerability factors in risk assessment,this thesis proposes the risk assessment index model based on information system assets.Then,a quantum neural network model is constructed by a group of quantum gate circuits.Quantum rotate gate is used to control the phase deflection and qubit flip,and the comprehensive risk value is obtained by the risk assessment calculation.Finally,the validity and reliability of the proposed method are verified by experimental simulation.Compared with the BP neural network,it is proved that quantum gate circuit neural network has advantage in convergence performance and risk prediction.2.This thesis explores a method of information security risk assessment based on quantum self-organizing feature map(QSOFM)network.Firstly,according to the risk assessment system and ALARP principle,the security risk is divided into tolerable risk and intolerable risk.Then,combining the quantum neuron model with the self-organizing feature map network,the QSOFM neural network model is constructed and applied to the risk assessment calculation so as to obtain the classification result of assessment samples.Finally,the validity and reliability of the proposed method are verified by experimental simulation.Compared with self-organizing feature map network,QSOFM neural network has advantage in the classification accuracy of assessment samples and running time.