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

A Deep Learning Channel Decoding Method Based on Alternating Direction Multiplier Method

A technology of alternating direction multipliers and channel decoding, applied in the field of deep learning channel decoding, can solve problems such as low error correction performance and high computational complexity, and achieve the effect of easy training

Active Publication Date: 2021-06-08
ZHEJIANG UNIV
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, compared with the classical Belief Propagation (BP) decoder, the LP decoder has higher computational complexity and lower error correction performance in the low signal-to-noise ratio (SNR) region

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 Deep Learning Channel Decoding Method Based on Alternating Direction Multiplier Method
  • A Deep Learning Channel Decoding Method Based on Alternating Direction Multiplier Method
  • A Deep Learning Channel Decoding Method Based on Alternating Direction Multiplier Method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0076] Consider the transmission signal on the additive Gaussian channel, the considered code pattern is [96,48]MacKay96.33.964 LDPC code and [128,64]CCSDS LDPC code A deep learning channel decoding method based on the alternating direction multiplier method proposed for this system includes the following steps:

[0077] 1) Construct the maximum likelihood optimization problem based on channel decoding. Each codeword of the considered LDPC code is specified by an M×N parity-check matrix H. and Represent the variable node and check node of the codeword respectively, d j Indicates the degree corresponding to the jth check node. is the transmitted codeword, and y is the received signal. The maximum likelihood decoding problem is expressed in the following form:

[0078]

[0079] in,[·] 2 Indicates modulo 2 operation, For the log-likelihood ratio, each element of v is defined as:

[0080]

[0081] Among them, Pr(·) represents the conditional probability.

[0...

Embodiment 2

[0132] Consider the transmission signal on the additive Gaussian channel, the considered code pattern is [96,48]MacKay96.33.964 LDPC code and [128,64]CCSDS LDPC code A deep learning channel decoding method based on the alternating direction multiplier method proposed for this system includes the following steps:

[0133] 1) Construct the maximum likelihood optimization problem based on channel decoding. Each codeword of the considered LDPC code is specified by an M×N parity-check matrix H. and Represent the variable node and check node of the codeword respectively, d j Indicates the degree corresponding to the jth check node. is the transmitted codeword, and y is the received signal. The maximum likelihood decoding problem is expressed in the following form:

[0134]

[0135] in,[·] 2 Indicates modulo 2 operation, For the log-likelihood ratio, each element of v is defined as:

[0136]

[0137] Among them, Pr(·) represents the conditional probability.

[0...

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 provides a deep learning channel decoding method based on the alternating direction multiplier method, which is mainly oriented to binary linear codes. The method includes the following steps: 1) constructing a maximum likelihood optimization problem based on channel decoding; 2) converting the maximum likelihood optimization problem of channel decoding into a decoding optimization problem with penalty terms by transforming constraint conditions and introducing penalty terms; 3) Introduce the alternating direction multiplier method to solve the decoding optimization problem with penalty items, and obtain the alternating direction multiplier method channel decoder; 4) Construct the deep learning network LADN according to the iterative form of the alternating direction multiplier method, the preset Coefficients are converted into learning parameters; 5) Learning parameters are obtained through training; 6) The deep learning network is restored to an iterative alternating direction multiplier channel decoder, and learning parameters are loaded for decoding. The present invention uses the power of deep learning to learn the optimal parameters in the channel decoder of the alternating direction multiplier method to further improve the decoding performance.

Description

technical field [0001] The invention belongs to the field of wireless communication channel coding and decoding, and relates to a deep learning channel decoding method based on an alternating direction multiplier method. Background technique [0002] Channel decoding is how to judge the received symbol message. In a general information transmission system, the message received by the sink is not necessarily the same as the message sent by the source, and the sink needs to know which source message is sent by the source at this time, so the message received by the sink needs to be based on a certain A rule judges that it corresponds to one of the source symbol message sets. Linear programming (LP) decoders are based on a linear relaxation of the original maximum likelihood decoding problem and are a popular decoding technique for binary linear codes. Since linear program decoders have strong guarantees on decoding performance in theory, they have attracted extensive attenti...

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): H03M13/11H03M13/00
CPCH03M13/1105H03M13/1148H03M13/6522G06N3/088G06N3/063H04L1/0054G06N3/048G06N3/08
Inventor 韦逸赵明敏赵民建雷鸣
Owner ZHEJIANG UNIV