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

Syndrome input RS code decoding method based on deep learning

A deep learning and syndrome technology, which is applied in the field of decoding algorithms with syndrome as input, and can solve problems such as analysis and research of deep learning translation of non-binary codewords.

Active Publication Date: 2020-01-24
TIANJIN UNIV
View PDF2 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, there is no example yet, and there is no analysis and research on deep learning to translate non-binary codewords

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
  • Syndrome input RS code decoding method based on deep learning
  • Syndrome input RS code decoding method based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0013] The RS code is a multi-ary code word obtained by hard judgment at the receiving end. Compared with the binary received code word, it will inevitably lose part of the channel noise information, resulting in uneven distribution of noise in the sample space and increasing the inaccuracy of neural network learning. Certainty. Multi-ary codewords mean more input possibilities than binary codewords. As the code length increases, the neural network structure and training process will be much more complicated, and it is not easy to expand to long code lengths. Therefore, it is not feasible to use the received multi-ary codeword as the input of the neural network for direct decoding. The present invention considers to use other data that is easier to find rules and whose length is as small as possible for input.

[0014] In the traditional decoding process of RS codes, a syndrome calculation step is involved. According to the analysis, the syndrome obtained in this step can we...

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 syndrome input RS code decoding method based on deep learning, comprising the following steps: generating a data set: calculating a syndrome S by using a receiving sequencey, and dividing the syndrome into a training set and a test set; calculating an error polynomial corresponding to the training set syndrome, and taking an error polynomial value as a deep learning training label; training a neural network: selecting a multi-layer perceptron (MLP) neural network structure, taking the syndrome S and the error polynomial E obtained in the step (1) as a data set of neural network training, and performing effective learning by using a back propagation algorithm; obtaining a neural network weight and bias with high decoding accuracy through training of a certain amount of data and algebra; when the number of the code word errors is within the decoding radius, performing inductive analysis on the neural network to obtain a correct corresponding relationship; andwhen the number of the code word errors is beyond the decoding radius, the neural network obtains the corresponding relation with the maximum possibility.

Description

[0001] Technical field [0002] The invention belongs to the field of error control coding in channel coding, and relates to a decoding algorithm based on deep learning and using a syndrome as an input. Background technique [0003] In the process of digital signal transmission, due to the interference and fading of the channel, errors may occur in the transmission signal, so the digital signal is usually encoded to enhance its ability to resist interference. Reed-Solomon (RS) code is a kind of error control code with strong error correction ability, which was constructed by Reed and Solomon in 1960, and can correct random errors and burst errors. At present, RS codes have been widely used in communication and data storage systems, involving many fields from deep space communication to high-density magnetic disks. [0004] In recent years, deep learning has attracted worldwide attention due to its powerful ability to solve complex tasks. Deep learning methods have been appli...

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): H03M13/15G06N3/04G06N3/08
CPCH03M13/1515G06N3/084G06N3/045
Inventor 梁煜安翔宇张为
Owner TIANJIN 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