Unlock instant, AI-driven research and patent intelligence for your innovation.

Feature extraction method based on power spectral density and cross-correlation entropy spectral density fusion

A power spectral density and feature extraction technology, applied in the field of feature extraction based on the fusion of power spectral density and cross-correlation entropy spectral density, can solve the problems of less robustness research and achieve good robustness

Active Publication Date: 2021-01-19
XI AN JIAOTONG UNIV
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

There are few studies on the robustness of frequency domain feature extraction methods, and traditional power spectral density is mainly used to extract frequency domain features

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
  • Feature extraction method based on power spectral density and cross-correlation entropy spectral density fusion
  • Feature extraction method based on power spectral density and cross-correlation entropy spectral density fusion
  • Feature extraction method based on power spectral density and cross-correlation entropy spectral density fusion

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0082] The process of extracting the eigenvectors of power spectral density and cross-correlation entropy spectral density fusion of a certain sample X is as follows:

[0083] 1) According to the existing training samples {{X 1 ,y 1}, {X 2 ,y 2},...,{X N ,y N}}, with the highest classification accuracy as the standard, use cross-validation to select the optimal kernel width σ o ;

[0084] 2) Calculate the power spectral density eigenvector of each sample according to formula (1):

[0085]

[0086] 3) Calculate the cross-correlation entropy spectral density eigenvector of each sample according to formula (2)-(6):

[0087]

[0088] 4) Merge the two eigenvectors to obtain a new eigenvector:

[0089]

[0090] simulation analysis

[0091] In order to demonstrate the advantages of the feature extraction method based on the fusion of power spectral density and cross-correlation entropy spectral density, two sets of experiments were carried out. The first set of simu...

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 feature extraction method based on the fusion of power spectral density and cross-correlation entropy spectral density. For the first time, the cross-correlation entropy spectral density is applied to EEG signal processing to extract frequency domain features, and the cross-correlation entropy spectral density and power spectrum are combined. Density fusion obtains a new feature. Compared with the traditional power spectral density and cross-correlation entropy spectral density, the fused feature extraction method can not only extract the frequency information in the signal well, but also suppress the influence of noise. Compared with power spectral density and cross-correlation entropy spectral density, the present invention is more suitable for signals with high signal-to-noise ratio and low signal-to-noise ratio. Compared with various interference signals including unknown characteristics in the environment, it provides a frequency domain feature extraction method with good performance. Therefore, the feature extraction method based on the fusion of power spectral density and cross-correlation entropy spectral density is easier to promote and use in the actual application of brain-computer interface.

Description

【Technical field】 [0001] The invention belongs to the technical field of signal processing and relates to a feature extraction method based on fusion of power spectral density and cross-correlation entropy spectral density. 【Background technique】 [0002] A brain-computer interaction system can be defined as a system that includes many external auxiliary devices that can be controlled by the user's thoughts, that is, a brain-computer interface is used to directly communicate with the nervous system and the outside world. The brain-computer interface can convert brain signals into control instructions, helping people interact directly with the outside world without using their own muscles. Because non-invasive electroencephalography (electroencephalography) signals are easy to collect, equipment is relatively cheap, and equipment is relatively portable, non-invasive electroencephalography is a signal acquisition technology that is widely used in brain-computer interaction. B...

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): G06K9/00G06K9/46G06K9/62
CPCG06V10/443G06F2218/02G06F2218/08G06F18/253
Inventor 陈霸东秦雪梅任鹏举袁泽剑郑南宁
Owner XI AN JIAOTONG UNIV