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

A Neural Network Based Polar Code bp Decoding Method

A neural network and polar code technology, which is applied in the field of polar code BP decoding based on neural network, can solve the problems of unfavorable URLLC high reliability implementation, increased performance degradation, and decreased decoding performance, and achieves improved BP. Decoding performance, reduced estimation error, significant performance gain

Active Publication Date: 2022-04-12
BEIJING UNIV OF POSTS & TELECOMM
View PDF9 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, in an actual communication system, this ideal assumption is not always accurate, and the channel will have noise correlation, making it in a non-ideal channel environment, affecting the received signal
Therefore, under non-ideal channel conditions, BP decoding will encounter difficulties in dealing with the correlation in channel noise. When the number of iterations is similar, the decoding performance of the algorithm will decrease, that is, the reliability will decrease, and as the noise correlation Enhanced performance and increased performance degradation, which is not conducive to the high reliability of URLLC

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 Neural Network Based Polar Code bp Decoding Method
  • A Neural Network Based Polar Code bp Decoding Method
  • A Neural Network Based Polar Code bp Decoding Method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0023] The present invention proposes a neural network-based polar code BP decoding method. The present invention will be further described in detail below in conjunction with the accompanying drawings.

[0024] Step 1: Transformer noise estimation model of the present invention such as figure 1 shown. Suppose the input data is Dimension size is N×1. In the present invention, the method of using the Transformer model is similar to the method of sequence modeling in natural language processing, but the difference is that the dimension of the noise vector in the network of the present invention is 1-D, which is the same as the high-level vector used in the word embedding technology. The dimension vectors are not consistent. Considering that there is no recursive or convolutional structure in the Transformer model, the position information between the sequences needs to be added through the position code. The dimension size of the position code (PE) is consistent with the inpu...

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 neural network-based polar code BP decoding method, which belongs to the technical field of wireless communication. In the non-ideal channel environment, the deep learning technology is introduced into the existing polar code BP decoding to optimize the design, and the parallel computing and global feature extraction capabilities of the Transformer model are fully utilized to implement the rough estimation noise obtained by BP decoding. Further accurate estimation, then feed back the updated LLR value to the connected BP decoding to continue iteration, thereby reducing the impact of noise correlation on BP decoding performance, and achieving the purpose of improving the reliability of BP decoding; meanwhile, the present invention The proposed method can reduce the complexity of decoding while achieving similar decoding performance.

Description

technical field [0001] The invention belongs to the technical field of wireless communication, and relates to a neural network-based polar code BP decoding method. Background technique [0002] With the explosive growth of wireless traffic, 5G is facing increasingly diverse demand drivers. As one of the three major application scenarios of 5G, ultra-reliable low-latency communication (URLLC) has strict requirements on latency and reliability. Polar codes are a suitable choice for URLLC channel coding schemes, and their bit error rate performance depends largely on the decoding algorithm used, so polar code decoding methods have attracted attention in URLLC scenarios. [0003] The BP decoding algorithm is a commonly used polar code decoding algorithm, and the existing research mainly evaluates its performance under the ideal AWGN channel. However, in an actual communication system, this ideal assumption is not always accurate, and the channel will have noise correlation, mak...

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): H03M13/13
CPCH03M13/13
Inventor 刘芳芳曾志民张瑞颐
Owner BEIJING UNIV OF POSTS & TELECOMM
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