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

Intrusion detection method and device based on ADASYN algorithm and improved convolutional neural network

A convolutional neural network and intrusion detection technology, applied in neural learning methods, biological neural network models, neural architectures, etc., can solve problems such as data imbalance, feature redundancy, and low detection accuracy, and improve intrusion detection and recognition. rate, improve learning and recognition ability, and solve the effect of feature redundancy between channels

Pending Publication Date: 2021-01-29
ELECTRIC POWER SCI RES INST OF STATE GRID XINJIANG ELECTRIC POWER +1
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The present invention provides an intrusion detection method and device based on the ADASYN algorithm and an improved convolutional neural network, which overcomes the above-mentioned deficiencies in the prior art, and can effectively solve the incompatibility of existing intrusion detection algorithms based on the convolutional neural network. The characteristics of data imbalance and feature redundancy lead to low detection accuracy and high false alarm rate.

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
  • Intrusion detection method and device based on ADASYN algorithm and improved convolutional neural network
  • Intrusion detection method and device based on ADASYN algorithm and improved convolutional neural network
  • Intrusion detection method and device based on ADASYN algorithm and improved convolutional neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0035] Embodiment 1: as attached figure 1 As shown, this embodiment discloses an intrusion detection method based on ADASYN algorithm and improved convolutional neural network, including:

[0036] S101. Obtain a number of original data, preprocess the original data, and use the ADASYN algorithm to perform enhanced processing on small samples in the number of original data;

[0037] S102. Divide some original data into a training sample set and a test sample set, and use the training sample set to train the preset model, and use the test sample set to test and evaluate the trained preset model; The establishment of the product neural network SPC-CNN algorithm;

[0038] S103, using the preset model with the best evaluation result as the intrusion detection model, and using the intrusion detection model to perform intrusion detection on the acquired network data.

[0039] as attached figure 2 As shown, step S101 of the above technical solution acquires the original data, prep...

Embodiment 2

[0048] Embodiment 2: as attached image 3 As shown, this embodiment discloses an intrusion detection method based on ADASYN algorithm and improved convolutional neural network, wherein the preset model is trained using the training sample set, wherein the preset model is improved through the improved convolutional neural network SPC-CNN algorithm , the training process further includes:

[0049] S201, establish a preset model by improving the convolutional neural network SPC-CNN algorithm, the preset model includes an input layer, a convolutional layer, 2 SPConv modules, a fully connected layer, a Softmax layer and an output layer, and the SPConv module includes a channel splitting module , 2 convolution modules and feature fusion modules;

[0050] S202, using the vector convolution module in the convolution layer to perform convolution processing on the training samples, and output L feature maps;

[0051] S203, using the channel splitting module in the SPConv module to div...

Embodiment 3

[0057] Embodiment 3: as attached Figure 4 As shown, this embodiment discloses an intrusion detection device based on ADASYN algorithm and improved convolutional neural network, including:

[0058] The preprocessing unit acquires some raw data, preprocesses the raw data, and uses the ADASYN algorithm to enhance the small samples in the several raw data;

[0059] The model output unit divides some original data into a training sample set and a test sample set, and uses the training sample set to train the preset model, and uses the test sample set to test and evaluate the trained preset model; wherein the preset model passes Improved convolutional neural network SPC-CNN algorithm establishment;

[0060] The intrusion detection unit uses the preset model with the best evaluation result as the intrusion detection model, and uses the intrusion detection model to perform intrusion detection on the acquired network data.

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 relates to the technical field of intrusion detection, in particular to an intrusion detection method and device based on an ADASYN algorithm and an improved convolutional neural network, and the method comprises the steps of obtaining a plurality of pieces of original data, carrying out the preprocessing of the original data, and carrying out the enhancement of small samples in theplurality of pieces of original data through the ADASYN algorithm; dividing the plurality of original data into a training sample set and a test sample set, and training a preset model by using the training sample set; and taking the preset model with the best evaluation result as an intrusion detection model, and performing intrusion detection on the acquired network data through the intrusion detection model. According to the invention, the ADASYN data enhancement algorithm can be utilized to enhance the small samples, the learning and recognition capability of the model for the small samples is improved, the multi-scale features of the data extracted by the convolutional neural network SPCCNN algorithm model are improved, the problem of inter-channel feature redundancy is effectively solved, and thus the intrusion detection recognition rate of the model is improved.

Description

technical field [0001] The invention relates to the technical field of intrusion detection, and relates to an intrusion detection method and device based on ADASYN algorithm and improved convolutional neural network. Background technique [0002] With the rapid popularization of 5G technology, wireless network has become an inevitable development trend. Due to the distribution and openness of the wireless network, it becomes an ideal attack target, so the security of the wireless network is paid more and more attention by people. The mainstream network security technology includes firewall technology, access authentication technology, etc., but when these security measures fail, the intrusion detection system (IDS) is undoubtedly one of the most effective network security measures. [0003] At present, machine learning intrusion detection methods are mostly used for intrusion detection. Machine learning algorithms include Naive Bayesian, Decision Trees (DTs), Support Vector...

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): G06F21/55G06K9/62G06N3/04G06N3/08
CPCG06F21/55G06N3/084G06F2221/034G06N3/045G06F18/253G06F18/214
Inventor 郭学让李峰张强何玲郭庆瑞李亚平张志军解鹏马林樊树铭张庚
Owner ELECTRIC POWER SCI RES INST OF STATE GRID XINJIANG ELECTRIC POWER
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

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com