Eureka AIR delivers breakthrough ideas for toughest innovation challenges, trusted by R&D personnel around the world.

An intelligent radiation source identification method based on combined twin network

A twin network and recognition method technology, applied in the field of signal processing, can solve the problems of unknown target recognition and re-entry recognition, and achieve the effect of improving recognition accuracy

Active Publication Date: 2022-07-01
UNIV OF ELECTRONICS SCI & TECH OF CHINA
View PDF8 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] In view of the above-mentioned deficiencies in the prior art, an intelligent radiation source identification method based on a combined twin network provided by the present invention solves the problem that the prior art cannot effectively identify and re-entry the unknown target under the condition of small samples

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
  • An intelligent radiation source identification method based on combined twin network
  • An intelligent radiation source identification method based on combined twin network
  • An intelligent radiation source identification method based on combined twin network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0042] like figure 1 As shown, an intelligent radiation source identification method based on a combined twin network includes the following steps:

[0043] S1. Collect several samples and use the twin network for data enhancement;

[0044] S2. According to the enhanced data, the trained combined twin network is used to identify the unknown target;

[0045] S3, calibrate the unknown target, and update the combined twin network according to the calibrated unknown target;

[0046] S4, using the updated combined Siamese network to re-enter the unknown target.

[0047] The step S1 is specifically:

[0048] S1.1. Collect K-type radiation source samples, the number of each type of samples is n k , obtain the training sample pairs of a single Siamese network, the number of which is

[0049] S1.2. Through sample combination, the number of training samples of a single Siamese network is expanded to k*n k of (k*n k -1) / 2 times, complete data enhancement.

[0050] The step S2 i...

Embodiment 2

[0072] This embodiment provides an intelligent radiation source identification method based on a combined twin network.

[0073] Step 1: Use the Siamese network for data augmentation.

[0074] The input of the Siamese network is a pair of samples. The network identifies whether the sample pair is the same class by learning the similarity and difference of the sample pair. If the two input samples are of the same class, the network output is 1; otherwise, the input is 0. For k classification problems, the number of samples in each class is n k Under the condition of , through sample combination, the number of training samples of a single Siamese network is The number of samples is expanded to the original (k*n k -1) / 2 times for data augmentation.

[0075] Step 2: Under the condition of small samples, use the trained combined Siamese network to identify the unknown target.

[0076] On the basis of Step 1, use the Siamese network to train a binary classification network for ...

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 intelligent radiation source identification method based on a combined twin network, which specifically includes: S1, collecting a number of samples, and using a twin network for data enhancement; S2, using a trained combined twin network to detect unknown data according to the enhanced data Identify the target; S3, calibrate the unknown target, and update the combined twin network according to the calibrated unknown target; S4, use the updated combined twin network to re-enter the unknown target. On the basis of fully integrating the fingerprint information of traditional signal identification, the invention further excavates the hidden features of the radiation source signal through the deep learning network, and effectively solves the problem of low accuracy of the existing radiation source identification method.

Description

technical field [0001] The invention belongs to the field of signal processing, in particular to an intelligent radiation source identification method based on a combined twin network. Background technique [0002] In recent years, with the rapid development of electronic technology, the application of electromagnetic spectrum intelligence has shown an explosive growth, which provides convenience for people's life and provides the foundation for intelligent weapons. In fact, the electromagnetic data obtained in the complex battlefield environment has the problems of large data volume and few effective target samples. Small sample threat detection and individual identification technology is a difficult point in massive electromagnetic data processing. Because the signal characteristics of electromagnetic targets cannot be known in advance, the classifier design is complex and the recognition rate is low; in addition, it is difficult to obtain enough training samples for new e...

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 Patents(China)
IPC IPC(8): G06K9/00G06K9/62G06N3/04G06N3/08
CPCG06N3/08G06N3/045G06F2218/00G06F2218/12G06F18/2132G06F18/2415
Inventor 孙国敏邵怀宗王沙飞杨健潘晔林静然利强
Owner UNIV OF ELECTRONICS SCI & TECH OF CHINA
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
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