Eureka AIR delivers breakthrough ideas for toughest innovation challenges, trusted by R&D personnel around the world.

Signal demodulation method based on machine learning

A signal demodulation and machine learning technology, applied in receiver-specific devices, link quality-based transmission modification, multiple modulation transmitter/receiver arrangements, etc. Problems such as signal demodulation with unknown mode, to achieve the effect of high spectrum utilization and excellent performance

Active Publication Date: 2019-04-23
HARBIN INST OF TECH
View PDF6 Cites 11 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to solve the existing problem that different hardware circuits are required to demodulate signals of different modulation methods, and it is difficult to demodulate signals with unknown modulation methods, and propose a machine learning based signal demodulation 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
  • Signal demodulation method based on machine learning
  • Signal demodulation method based on machine learning
  • Signal demodulation method based on machine learning

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0029] Specific implementation mode 1: The specific process of a signal demodulation method based on machine learning in this implementation mode is as follows:

[0030] Step 1. Collect the original baseband signal, process the baseband signal to obtain the noise signal, build a denoising self-encoder, perform denoising processing on the noise signal, and obtain the signal after denoising processing; Figure 5 , Figure 6a , Figure 6b shown;

[0031] The denoising self-encoder consists of an encoder and a decoder;

[0032] The specific process is:

[0033] Step 11, constructing the encoder of the denoising self-encoder, and extracting the features of the modulated signal collected;

[0034] Step 1 and 2, constructing a decoder of the denoising self-encoder, denoising and restoring the signal of the extracted feature, and obtaining the signal after denoising processing;

[0035] Step 2, train the neural network in the encoder and decoder in step 1, optimize the parameters...

specific Embodiment approach 2

[0039] Specific embodiment two: the difference between this embodiment and specific embodiment one is that the original baseband signal is collected in the step one, and the baseband signal is processed to obtain the noise signal; the specific process is as follows:

[0040] Step 1. Collect the original baseband signal, carry out carrier modulation on the baseband signal, generate signals of different modulation modes, add Gaussian white noise to the collected carrier modulation signal according to the signal-to-noise ratio range of 0-15dB, with a step size of 1dB;

[0041] Different modulation methods refer to the fact that the original signal is a baseband signal, and then modulated onto the carrier. The modulation method of the signal is uncertain, it can be PAM modulation, or BPSK modulation and other different modulation methods;

[0042]Step 2. Sampling the communication signal after adding Gaussian white noise in step 1, sampling 128 points for each symbol, and 4 communi...

specific Embodiment approach 3

[0047] Specific embodiment three: the difference between this embodiment and specific embodiment one or two is that in the step one by one, the encoder of the denoising self-encoder is constructed, and the specific process is as follows:

[0048] Step 111. Construct the network structure of the encoder of the denoising self-encoder using a convolutional neural network. The network structure includes an input layer, a first convolutional layer, a first pooling layer, a second convolutional layer, and a second pooling layer. layer, output layer;

[0049] The first convolutional layer and the second convolutional layer use the RELU activation function in the form of g(z)=max{0,z}, that is, when the input is less than 0, the output is 0, and when the input is greater than zero, the output is equal to the input;

[0050] The length of the convolution kernel of the first convolution layer is set to 5*5, and the number of convolution kernels is set to 16;

[0051] The length of the ...

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 relates to a signal demodulation method based on machine learning, relates to demodulation of signals in a single modulation mode and signals in various hybrid modulation modes, and aimsat solving the problems that different hardware circuits are needed for demodulation when the signals in different modulation modes are demodulated at present, and the signals in various unknown modulation modes are difficult to demodulate. The method comprises the following steps of 1, acquiring an original baseband signal, processing the baseband signals to obtain a noise signal, constructing adenoising auto-encoder, and carrying out denoising processing on the noise signal to obtain a signal after denoising processing; 2, obtaining a trained denoising auto-encoder network and parameters;3, constructing a signal demodulation model based on a convolutional neural network; 4, obtaining a trained signal demodulation model; and 5, cascading the trained denoising auto-encoder network and the signal demodulation model together, and demodulating the signals in the single modulation mode and the signals in various hybrid modulation modes. The method is used for the field of signal demodulation under a Gaussian white noise channel.

Description

technical field [0001] The invention relates to the field of signal demodulation under a Gaussian white noise channel, and relates to the demodulation of a single modulation mode signal and multiple mixed modulation mode signals. Background technique [0002] With the continuous development of communication technology, people's demand for the Internet of Everything is becoming stronger and stronger, and communication spectrum resources and bandwidth are becoming more and more tight. Usually, people think of increasing the amount of information carried by the transmitted symbols by increasing the modulation order, so as to improve the utilization rate of spectrum and bandwidth. However, with the increase of the modulation order, it becomes more and more difficult for the receiver to demodulate. The rate is getting higher and higher. At this time, coherent demodulation is usually used to reduce the bit error rate. However, in an actual communication system, the conditions of ...

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): H04L27/00H04L1/00
CPCH04L1/0036H04L27/0008
Inventor 朱洪涛李德志王振永徐誉郭庆何辞
Owner HARBIN INST OF TECH
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
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