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

Potential graph learning method for sampling smooth graph signals

A learning method and signal technology, applied in the field of graph learning, can solve problems such as graph signal loss, data loss, pollution, etc., and achieve the effect of wide applicability and high guessing accuracy

Inactive Publication Date: 2021-06-25
GUILIN UNIV OF ELECTRONIC TECH
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] Although the existing graph learning methods have achieved good results, most graph learning methods assume that the complete graph signal data is known before finding its potential graph. However, this assumption is not always true.
In fact, many times the graph signal observed by people is missing or polluted. For example, in wireless sensor networks, sensor nodes are missing data due to energy constraints, natural disasters and other factors.

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
  • Potential graph learning method for sampling smooth graph signals
  • Potential graph learning method for sampling smooth graph signals
  • Potential graph learning method for sampling smooth graph signals

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0047] refer to figure 1 , a latent graph learning method for sampling smooth graph signals, comprising the following steps:

[0048]1) Data collection and preprocessing: graph signal data in real life often obey statistical laws, and many graph signals are statistically smooth, such as temperature data of wireless sensor networks. Generally, graphs can be used to describe the statistical correlation between data Relationship, where the weights of each side of the graph represent the degree of correlation between the data, this example assumes that the observed target graph signal data X=[x 1 x 2 ... x T ] satisfy the graph factor analysis model formula (1):

[0049] x=Uh+u+n (1),

[0050] Among them, U is the Fourier transform matrix of the graph, h is the spectrum of the graph signal x that is the latent variable, u is the average value of the graph signal, and n is the covariance matrix as σ 2 The zero-mean Gaussian white noise of I, σ represents the standard deviati...

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 potential graph learning method for sampling smooth graph signals, which is characterized by comprising the following steps of (1) collecting data and preprocessing, (2) estimating a covariance matrix, (3) constructing a convex optimization model, (4) solving an optimization problem and (5) solving a graph adjacency matrix. The method is used for the situation that the known graph signals are incomplete and has wider applicability.

Description

technical field [0001] The invention relates to a graph learning technology in graph signal processing, relates to a graph factor analysis model, in particular to a potential graph learning method for sampling smooth graph signals. Background technique [0002] In recent years, graph signal processing has received the attention of many scholars. In the graph signal processing framework, irregular networks such as wireless sensor networks, traffic networks, and social networks can be modeled as graphs with corresponding topological structures, and the data in the network is modeled as the signal of each vertex on the graph. As an extension of traditional signal processing theory, graph signal processing theory extends traditional concepts and analysis methods such as Fourier transform and filtering to graphs with more complex structures. Although researchers have made a lot of remarkable achievements in many aspects of the field of graph signal processing, such as the sampli...

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 Applications(China)
IPC IPC(8): G06F17/18G06F17/16G06F17/14
CPCG06F17/14G06F17/16G06F17/18
Inventor 蒋俊正池源冯海荣卢军志黄炟鑫
Owner GUILIN UNIV OF ELECTRONIC TECH
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

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

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com