Network security situation assessment model based on LAHP-IGFNN

Abstract

Provided is a network security situation assessment model based on LAHP-IGFNN; firstly, according to the method, linear programming and an analytic hierarchy process are combined, the linear programming is used for improving the consistency of judgment matrixes in the analytic hierarchy process, and the subjectivity problem of the analytic hierarchy process is reduced; then, the fuzzy neural network is optimized by using a gravitation search algorithm with an improved updating formula, so that the convergence speed of the algorithm is increased and the algorithm is prevented from falling intolocal optimum; simulation experiments show that the LAHP has reasonableness and better consistency, and the IGFNN has advantages in the aspects of convergence speed and fitness, so that the LAHP-has the advantages in the aspects of convergence speed and fitness. The IGFNN situation assessment method has better performance.

Description

technical field [0001] The invention relates to the technical field of network security situation situation deep neural network, in particular to a network security situation assessment model based on LAHP-IGFNN. Background technique [0002] The Internet has entered thousands of homes and has completely changed the traditional way of working and living. Although the arrival of the information age has greatly promoted people's life and work, while being ubiquitous, there are also a large number of security risks and network security. More and more network attacks start from terminals, and network security issues have become people's One of the important issues in the normal use of the Internet; for the security and stability of the entire Internet, no matter how perfect the security and comprehensive design of the legal system, computers, servers, software, hardware, etc., it will be difficult to eliminate due to the open environment of the Internet Unsafe risk factors; eli...

AI Technical Summary

Problems solved by technology

This method does not need to know the probability distribution of variables accurately, but it has a large amount of calculation, and there is a potential problem of combinatorial explosion; Ramaki proposed a risk assessment method based on Bayesian network, which has good convergence and fault tolerance; However, in a large-scale network application environment, it is necessary to obtain the corresponding parameters through complex training; Li Xiaoyan proposed a network security situation assessment method based on the hidden Markov model; this method can easily and intuitively track the dynamic characteristics of numerical fluctuations , but the establishment of the state transition matrix and the observation symbol matrix is ​​based on expert experience, and the quality of the model depends on the knowledge of the expert and the level of knowledge of the expert, which is highly subjective; Hu Jing proposed fuzzy analytic hierarchy process and BP neural network The combined situation assessment method can solve highly complex input, output and nonlinear mapping, and has good nonlinear fitting ability; however, the comparison matrix in the AHP method needs to be obtained through expert experience, which is too subjective, and the BP neural network It has the defects of slow convergence and can't solve the local minimum

Images