Customized Branched Neural Network-Aided Shuffled Min-Sum Decoder for Protograph LDPC Codes

doi:10.1109/TVT.2024.3459692

Customized Branched Neural Network-Aided Shuffled Min-Sum Decoder for Protograph LDPC Codes

2025-01-01

0

PRE

AI

DOI：10.1109/TVT.2024.3459692

Original

Original request for help

Save

Abstract

En 中文

The paper designs a novel neural shuffled min-sum (NSMS) decoder with the model-driven deep learning method to achieve higher efficient and lower complexity decoding for protograph low-density parity-check (LDPC) codes. We propose a new type of customized branched neural network (CBNN) structure, which integrates shuffled min-sum (SMS) decoding algorithm and shuffled belief-propagation (SBP) decoding algorithm. In such a network structure, we can adjust layer arrangement and simplify parameter groups at a specific stage (i.e., training or inference stage) to reduce the unwarranted computational workload. Furthermore, we utilize the branched neuron mean difference (BNMD) to optimize the training targets of the proposed NSMS decoder, which significantly accelerates the convergence speed of the network. Analytical and simulation results show that the proposed NSMS decoder can achieve better performance than the state-of-the-art counterparts in terms of convergence speed, error rate and computational complexity.

Keywords:

Decoding

Codes

Iterative decoding

Training

Convergence

Iterative methods

Biological neural networks

Branched neuron mean difference (BNMD)

customized branched neural network (CBNN)

model-driven deep learning

neural shuffled min-sum decoder (NSMS)

protograph LDPC codes

Customized Branched Neural Network-Aided Shuffled Min-Sum Decoder for Protograph LDPC Codes

Abstract

Journal

Organization