Anti-attack and defense method based on LSTM (Long Short Term Memory) detector

A detector and detection rate technology, applied in the direction of neural learning methods, instruments, biological neural network models, etc., can solve the problems of poor detection effect, increased detection rate, low safety, etc., to achieve accurate measurement values ​​and detection results, increase The effect of detection rate, strong robustness

Inactive Publication Date: 2018-08-10
ZHEJIANG UNIV OF TECH
View PDF5 Cites 22 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0007] In order to overcome the disadvantages of poor detection effect and low security in the prior art, the present invention provides an LSTM detector-based adversarial attack defense method with high detection effect and high security, which combines the negative selection algorithm with the

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
  • Anti-attack and defense method based on LSTM (Long Short Term Memory) detector
  • Anti-attack and defense method based on LSTM (Long Short Term Memory) detector
  • Anti-attack and defense method based on LSTM (Long Short Term Memory) detector

Examples

Experimental program
Comparison scheme
Effect test

Example Embodiment

[0048] The present invention will be further described below in conjunction with the drawings.

[0049] Reference Figure 1 ~ Figure 3 , An LSTM detector-based adversarial attack defense method, including the following steps:

[0050] 1) Generate a set of adversarial sample detectors, the process is as follows:

[0051] 1.1) Use a sliding window of length l to intercept the time series data as the original training data set X=(x (1) ,x (2) ,...}, x (t) ={y (t) ,u (t) }, y (t) Represents the measured value at time t, u (t) Represents the control value at the moment, where the normal sample is denoted as X', and the adversarial sample is denoted as i represents the i-th type of adversarial samples, there are a total of w adversarial samples, let i=1; figure 1 The medium data set represents the original training data set, the circle represents the normal data, and the other shapes are the adversarial sample data obtained from different attacks;

[0052] 1.2) Since the time series data i...

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 an anti-attack and defense method based on an LSTM (Long Short Term Memory) detector. The method comprises the following steps that: 1) generating a candidate detector of an LSTM-FNN (Fuzzy Neural Network) structure, and asking the detector to detect abnormal samples as more as possible on a premise that the detector does not misjudge a normal sample; 2) storing the candidate detector into a register queue, detecting a training dataset by an abnormality detector taken out of the register queue, and deleting the detected abnormal sample to enable different abnormality detectors to cover different abnormal areas; and 3) in a detection stage, detecting a detected sample by all abnormality detectors, and comprehensively judging whether the sample is abnormal or not. Byuse of the method, a neural network is combined with a negative selection algorithm, each LSTM detector guarantees that the normal sample can not be misjudged, the abnormality detector set guaranteesthat the abnormal situation may be covered, and an algorithm detection effect is improved.

Description

technical field [0001] The invention belongs to the technical field of industrial control security, and in particular relates to an LSTM detector-based anti-attack defense method. Background technique [0002] Industrial Control System (ICS) is an integral part of the country's important infrastructure and is widely used in various fields, including chemical industry, pharmaceuticals, hydropower energy, natural gas, aerospace and other fields. However, with the combination of networking and automation with industrial control systems, the cyber attack surface of industrial control systems is gradually expanding, making it more important than ever to develop innovative solutions for ICS security. In view of the fact that the data acquired by the industrial control system through the sensor or the data transmitted in the network is mainly time-series data, the extraction of key information of the time-series data and the defense mechanism of hidden attacks have become the key t...

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
IPC IPC(8): G06F21/55G06N3/04G06N3/08
CPCG06F21/552G06N3/08G06N3/044G06N3/045
Inventor 陈晋音苏蒙蒙徐轩珩郑海斌林翔熊辉沈诗婧
Owner ZHEJIANG UNIV OF TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products