Hybrid machine learning signal classification method based on PCA dimension reduction

A machine learning and signal classification technology, applied in the field of communication, can solve the problems of insufficient adaptability to unknown situations, low degree of automation, limited training samples, etc.

Active Publication Date: 2018-12-18
HOHAI UNIV
View PDF5 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Conventional intrapulse analysis methods use threshold judgment based on manual experience, which is not highly automated and has insuffici

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
  • Hybrid machine learning signal classification method based on PCA dimension reduction
  • Hybrid machine learning signal classification method based on PCA dimension reduction
  • Hybrid machine learning signal classification method based on PCA dimension reduction

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0137] MATLAB simulation is used to verify the classification and recognition performance of multi-class radar and communication signals, including two stages of training and testing.

[0138] In the training phase, the signal-to-noise ratio of the four signals is set to 20dB, and the symbol width is 10 -6 s, the number of sampling points is 5000, the sampling frequency is 100MHz, the carrier frequency is 20MHz, the chirp bandwidth is 10MHz, the time width is 50us, and the frequency of 2FSK is 10MHz and 20MHz respectively.

[0139] In the first level of training, the four signals are trained separately based on the instantaneous autocorrelation method. The instantaneous autocorrelation processing results of the four signals are as follows: Figure 4 , 5 , 6, and 7. After the instantaneous autocorrelation cycle is processed, the zero-crossing points of the four signals are cyclically extracted 50 times, and the results are as follows Figure 8 As shown, it can be seen that the number...

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 hybrid machine learning signal classification method based on PCA dimension reduction, includes the following steps: 1 according to instantaneous autocorrelation classification method, setting zero-crossing threshold and standard deviation threshold for LFM signal, BPSK signal, binary frequency shift keying 2FSK signal and quadrature phase shift keying QPSK signal, separating LFM signal and QPSK signal from other signals; 2, adopting normalized amplitude duty cycle, normalized central instantaneous phase absolute value variance and normalized central instantaneous frequency absolute value variance for residual signal BPSK signal and 2FSK signal in two-stage classification, and adopting principal component analysis PCA algorithm to realize feature dimension reduction;3, the SVM classifier being used to obtain the optimal classification objective function, and the BPSK and 2FSK signals are distinguished. The invention adopts machine learning technology to realize classification, which has high automation degree and good classification effect.

Description

Technical field [0001] The present invention relates to the field of communication technology, in particular to a hybrid machine learning signal classification method based on PCA dimensionality reduction. Background technique [0002] Radar and electronic warfare systems need to automatically learn to recognize the electromagnetic signals obtained by reconnaissance. The intra-pulse characteristics of radar signals are important parameters for electromagnetic signal sorting and identification in radar and electronic warfare systems. Therefore, to reliably sort and identify radar signals, it is necessary to analyze the intra-pulse characteristics of radar signals. Conventional intrapulse analysis methods use threshold judgments based on manual experience, which are not highly automated and have insufficient adaptability to unknown situations, and due to limited training samples, manually set experience thresholds may have poor classification effects. Summary of the invention [00...

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): G06K9/00G06K9/62
CPCG06F2218/12G06F18/2135
Inventor 王峰黄珊珊
Owner HOHAI UNIV
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

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap