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

A Modulation Intelligent Recognition Method Driven by Data and Knowledge

An intelligent recognition, dual-drive technology, applied in the field of communication, can solve problems such as poor performance and confusion, and achieve the effect of improving confusion, improving training speed, and improving recognition accuracy.

Active Publication Date: 2022-05-10
NANJING UNIV OF AERONAUTICS & ASTRONAUTICS
View PDF8 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] Aiming at the disadvantages of the existing modulation recognition technology in complex and dynamic actual scenes, such as poor performance, dependence on a large number of training samples, and serious confusion between high-order modulation modes, the present invention proposes a modulation mode intelligence driven by data and knowledge. The recognition method introduces semantic attributes as knowledge, and embeds attribute features into the visual feature space through the conversion model, which greatly improves the accuracy of modulation recognition under low signal-to-noise ratio, reduces the confusion between high-order modulation methods in the recognition process, and Ability to reduce dependence on the amount of training data

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
  • A Modulation Intelligent Recognition Method Driven by Data and Knowledge
  • A Modulation Intelligent Recognition Method Driven by Data and Knowledge
  • A Modulation Intelligent Recognition Method Driven by Data and Knowledge

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0050] Embodiments of the invention will be further described in detail below in conjunction with the accompanying drawings.

[0051] combined with figure 1 The specific steps of the method of the present invention are described as follows.

[0052] Step 1, spectrum data collection.

[0053] Modulation classification can be regarded as a K-type hypothesis testing problem, and the received signal under the k-th modulation hypothesis can be expressed as x k (n)=s k (n)+ω k (n). Among them, s k (n) represents the nth sampling point of the transmitted signal of the kth modulation mode, x k (n) represents the nth sampling point of the received signal of the kth modulation mode, ω k (n) means the mean is 0 and the variance is σ 2 additive white Gaussian noise. Express the received signal as a vector form composed of I / Q components, x k =I k +Q k , where x k means x k The vector form of (n), I k and Q k represent the in-phase and quadrature components of the signal, r...

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 modulation intelligent recognition method driven by data and knowledge, which mainly solves the problem of low classification accuracy of the existing modulation recognition method under low signal-to-noise ratio, dependence on a large number of training samples, and the difficulty of high-order modulation modes in the recognition process. confusion etc. The implementation steps are: spectral data collection; construction of corresponding attribute vector labels according to different modulation methods; construction and pre-training of attribute learning models according to attribute labels of different modulation methods; construction and pre-training of modulation recognition visual models; construction of feature space conversion Model, combined with the visual model and attribute learning model to build a data-knowledge dual-driven modulation mode intelligent recognition framework; transfer the parameters of the pre-trained visual model and pre-trained attribute learning model, retrain the conversion model; judge whether the network training is over, and output the classification results. The invention significantly improves the identification accuracy under low signal-to-noise ratio and reduces confusion among high-order modulation modes.

Description

technical field [0001] The invention belongs to the technical field of communication, and relates to a modulation intelligent identification method driven by data and knowledge. Background technique [0002] Intelligent identification of modulation mode is a crucial technology in wireless intelligent communication. It distinguishes different types of modulated signals by learning the unique characteristics of received signals. With the gradual popularization of 5G, research on 6G wireless communication networks has emerged. As a key component of 6G communication network, intelligent communication technology urgently needs in-depth research, among which the intelligent identification of modulation mode is one of the key technologies. Therefore, it is very important to carry out research on the intelligent identification of modulation modes. At the same time, modulation identification has been widely used in military and civilian fields. In military applications, modulation...

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): H04L27/00G06N3/04G06N3/08
CPCH04L27/0012G06N3/084G06N3/045G06N3/048G06N3/0464G06N3/09G06N3/08G06N5/02
Inventor 周福辉丁锐徐铭张浩袁璐吴启晖董超
Owner NANJING UNIV OF AERONAUTICS & ASTRONAUTICS
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