Title:
SELF-LEARNING DECODING METHOD FOR PROTOGRAPH LOW DENSITY PARITY CHECK CODE AND RELATED DEVICE THEREOF
Document Type and Number:
WIPO Patent Application WO/2021/204163
Kind Code:
A1
Abstract:
The present disclosure provides a self-learning decoding method for a protograph low density parity check (LDPC) code. The method comprises: constructing a training sample based on a protograph LDPC code; constructing a self-learning neural network decoding model; training the self-learning neural network decoding model according to the training sample, to obtain a decoding mode for decoding the protograph LDPC code; and using the decoding model to decode the protograph LDPC code to be decoded. The present disclosure also provides a self-learning decoding apparatus for a protograph LDPC code, an electronic device and a non-transient computer readable storage medium corresponding to the method.
Inventors:
DAI JINCHENG (CN)
NIU KAI (CN)
TAN KAILIN (CN)
NIU KAI (CN)
TAN KAILIN (CN)
Application Number:
PCT/CN2021/085830
Publication Date:
October 14, 2021
Filing Date:
April 07, 2021
Export Citation:
Assignee:
UNIV BEIJING POSTS & TELECOMM (CN)
International Classes:
H03M13/11
Foreign References:
| CN111565051A | 2020-08-21 | |||
| CN109347601A | 2019-02-15 | |||
| CN109361404A | 2019-02-19 | |||
| CN106571831A | 2017-04-19 |
Attorney, Agent or Firm:
BEI JING FORESTSONG PATENT AGENCY CO., LTD. (CN)
Download PDF: