Research On Network Security Situation Awareness In Enterprise Application

In the information age,a secure network environment is an important condition for the normal operation of society.Network security situational awareness,as a technical means with active defense capabilities in security big data,has broad application scenarios and great potential research value in network security.Network security situational awareness technology uses the acquired security data to understand and evaluate to predict attacks,and has active network security protection capabilities.However,the existing network security awareness technology is mainly used for large-scale network infrastructure protection,and is not suitable for the actual environment of small and medium-sized enterprises to apply situational awareness.Therefore,this paper conducts research on the application of network security situational awareness methods in small and medium-sized enterprises.The main work is as follows.First of all,in view of the complex situational awareness methods of existing large-scale multi-source data fusion,which cannot be migrated to enterprise security applications with a large number of applications,a relatively small scale,and strong business relevance,the research uses the Hidden Markov Model(HMM)to Enterprises use networksecurity data for security situation assessment;secondly,for the relevant parameters of the HMM model in the learning process of the traditional BW training algorithm,it is easy to fall into the local optimal solution,which leads to the problem of poor model evaluation effect.Research on the use of improved particle swarm based on elite strategy(PSO)algorithm optimizes the HMM evaluation model,and optimizes the transfer matrix of the HMM model before training.This will help improve the randomness of the training process and the actual evaluation ability of the model in enterprise applications.Set the non-linear update particle swarm acceleration coefficient to improve the global convergence ability of the model;finally,use the security data generated by the simulated attack on the enterprise web application,and combine the enterprise application security requirements to construct the HMM situation assessment model for experimental evaluation.The experimental results show that the PSO optimized HMM evaluation model based on the elite strategy can achieve convergence with fewer iterations,the results are more accurate,and it has a good use effect in enterprise application security situation evaluation.To meet the needs of small and medium-sized enterprises' situation prediction models for light weight and high efficiency,the RBF neural network with simple network structure and good fitting effect on non-linear data is used to evaluate the network security situation of enterprise applications.At the same time,the K-means clustering algorithm is used to evaluate the RBF.The neural network center is used to solve the problem,and the PSO algorithm is used to optimize the parameter base width and weight vector of the RBF neural network.Experiments show that the proposed method can effectively improve the shortcomings of the RBF gradient descent method that is easy to fall into local extremes,and has a better effect on the prediction of enterprise network situation.In summary,this article proposes a new situation assessment and situation prediction model for the insufficient application of situational awareness in small and medium-sized enterprises,which not only meets the specific needs of enterprise application situational awareness,but also takes into account the lightweight performance and practical application capabilities of the model.