Multi-Agent System Network Intrusion Tolerance Evaluation Method Based on Multilayer Perceptron

A multi-agent system and multi-layer perceptron technology, applied in the field of multi-agent systems, can solve problems such as high algorithm complexity, many network nodes, and inability to evaluate effectively, and achieve matrix feature simplification, precision and precision Good results

CN110224862BActive Publication Date: 2021-12-21HANGZHOU DIANZI UNIV
0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Publication Date
2021-12-21

Smart Images

  • Figure 1
    Figure 1
  • Figure 2
    Figure 2
Patent Text Reader

Abstract

The invention discloses a multi-agent system network intrusion tolerance evaluation method based on a multi-layer perceptron. Traditional methods cannot be effectively evaluated due to the large number of network nodes and high algorithm complexity. The method of the invention firstly conducts matrix spectrum clustering according to the number of nodes for the multi-agent system network set, obtains the distribution number of the eigenvectors of the adjacency matrix in different clusters, and constructs a set of eigenvectors; The features extracted from the data set with the number of nodes are handed over to the multi-layer perceptron model for feature learning, and the weight matrix and bias vector are obtained; finally, the multi-layer perceptron model predicts the features obtained after the adjacency matrix preprocessing of the same data set to obtain the final Classification results. The present invention obtains the eigenvalues ​​from the adjacency matrix corresponding to the network topology of the multi-agent system, which can simplify the matrix features, is more conducive to the learning of the perceptron, and the accuracy and precision of the learned artificial neural network are improved. it is good.
Need to check novelty before this filing date? Find Prior Art

Description

Technical field

[0001] The present invention belongs to the field of multi-intelligent system, and in particular, the present invention relates to a multi-intelligent system system based multi-layer perception system network accusation capability assessment method. Background technique

[0002] With the development of robotics, computer, sensing, and communication technology, multi-agentsystems has caused a significant concern around the world, and has enormaborated social production and people's lives. Expertists at home and abroad have conducted deep research on the basic theories and key technologies of multi-intelligent system system in various aspects, and has achieved a large number of important results. Multi-intelligent system network topology features, system enabling capacity, safety level and survival capabilities, and corresponding preventive control measures are important in both theoretical and engineering. Therefore, it is necessary to analyze the network security ...

Examples

Embodiment Construction

[0025] The method of the present invention will be further described below with reference to the drawings.

[0026] like figure 1 As shown, multi-layer perception-based multi-intelligent system system network enabling ability assessment method, the specific steps are:

[0027] Step (1). For n multi-intelligent system system network set g = {g 1 , G 2 , ..., g N }, With its adjacency matrix collection a = {a 1 , A 2 , ..., A N } The intensive distribution statistics of each element (ie, mean, extreme value, number, median) are basically characterized, and based on the spectrum space of the adjacent matrix feature vector, the matrix spectrum cluster is obtained by the number of nodes to obtain an adjacency matrix. Feature vector in K a different cluster number, record as Count j , J = 1, 2, ..., k, parameter k to take the number of corresponding data set nodes; build feature vector set f = {f 1 , f 2 , ..., f N } Where f i Indicates that in the corresponding multi-intelligent system...