Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Automatic model compression method for communication signal modulation identification

A communication signal and modulation recognition technology, which is applied in the fields of modulation type recognition, modulation carrier system, neural learning method, etc., can solve problems such as easy to be restricted and large number of parameters, achieve reduction in size, minimize reconstruction error, and meet light weight requirements. Effects of orders of magnitude deployable requirements

Pending Publication Date: 2021-10-22
PLA PEOPLES LIBERATION ARMY OF CHINA STRATEGIC SUPPORT FORCE AEROSPACE ENG UNIV
View PDF2 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The technical problem solved by the present invention is: in view of the current existing technology, the traditional residual network modulation identification model has a large amount of parameters, and the application on the portable device side is easily limited, and an automatic model compression for communication signal modulation identification is proposed method

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
  • Automatic model compression method for communication signal modulation identification
  • Automatic model compression method for communication signal modulation identification
  • Automatic model compression method for communication signal modulation identification

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0029] An automatic model compression method for communication signal modulation recognition, which is helpful for automatic compression of deep learning network models in the process of communication signal modulation recognition, replacing manual parameter adjustment, and can be used for deep learning network models for communication signal modulation recognition under portable devices The analysis of electromagnetic characteristics can ensure that the model still has the same or similar expressive ability, and reduce the volume of the model as much as possible, reduce the amount of calculation, and meet the requirements of lightweight deployment. The specific process of the compression method is as follows:

[0030] (1) Arranging the received communication signals into code element images in order according to the specified number of symbols, and converting the communication signals into modulation pattern texture images for identification;

[0031] (2) Build a residual netw...

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 automatic model compression method for communication signal modulation identification. The method comprises the following steps: starting from data, accumulating certain data to meet the requirement of model learning; mining representation modes of data features from enough data; and respectively calculating an optimal pruning proportion for each convolutional layer by adopting a depth deterministic strategy gradient algorithm. According to the invention, the reconstruction error can be minimized while the reasoning speed is increased, automatic compression of the deep learning network model in the communication signal modulation recognition process is facilitated, and manual parameter adjustment can be replaced.

Description

technical field [0001] The invention relates to an automatic model compression method for communication signal modulation recognition, which belongs to the technical field of communication signal modulation recognition. Background technique [0002] Communication signal modulation identification can provide a basis for selecting a suitable signal demodulator by performing a series of signal processing on the received signal to obtain the modulation mode of the signal. Automatic modulation identification has certain significance and value in both military and civilian fields. When the deep learning method is used to search the neural network model for communication signal modulation recognition, the residual network modulation recognition model can better complete the recognition task. However, due to the high computational complexity and large number of model parameters, it is limited in some portable devices. Deployment and application on the device side. In the actual app...

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): G06N3/04G06N3/08H04L27/00
CPCG06N3/082H04L27/0012G06N3/045
Inventor 方宇强张喜涛宋万均陈维高殷智勇
Owner PLA PEOPLES LIBERATION ARMY OF CHINA STRATEGIC SUPPORT FORCE AEROSPACE ENG UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap 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