Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Network attack recovery method of cascaded condition-based energy generative adversarial network

An energy generation and network attack technology, applied in the field of network information security, can solve problems such as difficult control, data recovery accuracy depends on physical system modeling accuracy, recovery accuracy and poor stability, etc., to achieve improved accuracy, high recovery accuracy and stability The effect of high reliability and high data recovery accuracy

Pending Publication Date: 2021-12-17
BEIJING UNIV OF TECH
View PDF0 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The technical problem to be solved by the present invention is to provide a network attack recovery method and system based on conditional energy generation against the network, which is used to solve the problem that the accuracy of data recovery depends on the accuracy of physical system modeling in the existing network attack recovery methods , and the existing network attack recovery technology based on deep learning, the problem of poor recovery accuracy and stability caused by the difficult control of the training process

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Network attack recovery method of cascaded condition-based energy generative adversarial network
  • Network attack recovery method of cascaded condition-based energy generative adversarial network
  • Network attack recovery method of cascaded condition-based energy generative adversarial network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0058] Such as figure 1 As shown, Embodiment 1 is a cascading condition-based energy generation confrontation network method, which consists of three steps: training a condition-based generation confrontation network model that can simulate the normal operation of the system, and establishing a condition-based energy generation confrontation network model. Generate an adversarial network model and implement an adaptive decision-making strategy, and evaluate the recovery performance of different models using the absolute error between the recovered measurement value and the real measurement value and the spatiotemporal loss caused by the model. The specific content is described as follows:

[0059] In step 1, the sensor measurement value expressed in the traffic data is used as input, the condition-based generative confrontation network is trained, and the trained model is used to generate feature data.

[0060] In step 2, the feature data generated by the condition-based gene...

Embodiment 2

[0093] The second embodiment is to use the recovery data obtained by the traditional generative confrontation network and the cascaded energy-based generative confrontation network with different numbers of training rounds as a reference, and then use the industrial control network traffic for experimental verification, and compare the results obtained by the method proposed in this paper. Recover the data, and compare the training time, storage space usage, and accuracy of the recovered data for different models.

[0094] Figure 5 and Figure 6 Five models are shown with the recovery accuracy over time for different register measurements. Analyzing the results in the table, it can be found that the model proposed in this paper is significantly better than the generative confrontation network and the cascaded energy-based generative confrontation network with different numbers of training rounds in terms of recovery accuracy and stability. The reason for this result is that...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a network attack recovery method for a cascaded condition-based energy generative adversarial network, which comprises the following steps of: representing a sensor measurement value in flow data as input, and generating feature data by using a trained model; taking feature data generated by the trained condition-based generative adversarial network as additional information and splicing with random noise to serve as input; calculating the absolute error between the recovery data and the real data, and the time required by different model training and the space required to be occupied, and measuring the performance difference between different models by using the indexes. The condition-based generative adversarial network model used in the method can quickly simulate behavior features during system schedule operation, the proposed condition-based energy generative adversarial network can effectively improve the recovery precision of the measured value, and the designed adaptive decision strategy can improve the recovery stability of the measured value. And the possibility of generating abnormal points is reduced.

Description

technical field [0001] The invention belongs to the technical field of network information security, relates to network attack recovery technology, and particularly relates to a cascade game-based deep learning model. Background technique [0002] With the continuous integration of industrialization and informatization, industrial control systems increasingly adopt standardized communication protocols, software and hardware, and realize remote control and operation through the Internet. Although this method significantly improves the efficiency of complex industrial systems, this connection breaks the closedness and proprietary nature of the original system, causing network security issues such as ransomware attacks and information leakage to spread rapidly to the industrial control field, directly affecting a large number of Industrial control related infrastructure security. Cyber-attacks against industrial control systems are intended to destroy control-related data in c...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06F21/55G06F21/56G06N3/02G06N3/08
CPCG06F21/55G06F21/568G06N3/02G06N3/08
Inventor 赖英旭田必涛王一鹏刘静孙墨童
Owner BEIJING UNIV OF TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products