Research On Network Security Situation Awareness And Prediction Method Based On Bayesian Network

Computer technology is growing,and with it,security issues are becoming serious.Most of the cyber attacks are highly contagious and destructive and have posed a considerable threat to the safe use of computers.From the time the Internet became widespread in every household,cyber security has taken on an increasingly elevated status.With the development of the Internet industry infrastructure and the emergence of new computer applications,networks are becoming larger and larger,their topologies are becoming increasingly complex,and network security management is becoming more and more difficult.As the scale of today’s network increases,the relationship between the Internet industry and traditional industries is getting closer.People’s lives are increasingly dependent on the Internet.However,the current cyber security environment is quite harsh,and cyber attacks are everywhere,causing more and more cyber losses to occur.Therefore,accurate network situational awareness is an urgent action in such a complex and volatile network environment nowadays.In this paper,we analyze the network security posture before the attack has occurred and the strength of computer systems that still have the ability to provide services after the attack has occurred to study the effectiveness of defensive behavior in response to network attacks.First,there are various types of attacks in blockchain networks,and this paper focuses on one of them,the eclipse attack.To address the problem of incomplete information quantification in the process of blockchain network attack and defense,this paper uses Bayesian theory to construct a Bayesian network model.According to the characteristics of the eclipse attack,the node types in the blockchain network are divided,and the state transfer probability between nodes is obtained according to the node attack value,node state transfer cost and node state transfer benefit of each node.And according to the Bayesian network model,the set of attack paths of different nodes can be obtained,and the probability of attack path selection is calculated,which improves the accuracy of the network situational awareness process.In addition,in the above set,a path with higher attack probability is selected from it,and a Markov multi-stage attack-defense game is conducted to calculate the maximum objective function value of both attackers and defenders,and the current network posture situation of the blockchain network is calculated.Through the analysis,the process of sensing the security posture of the whole blockchain network nodes is completed,and the purpose of predicting the future security situation and system maintenance is achieved.Finally,the effectiveness of the model proposed in this paper is verified by designing experiments.