.png){kind=icon}
.png){kind=icon}
Return
Efficient Active Deep Decoding of Linear Codes Using Importance Sampling
.png){kind=icon}
PRE
AI
DOI：10.1109/LCOMM.2024.3514493.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
The quality and quantity of training data significantly affect deep learning model performance. In error correction, generating high-quality samples with minimal noise is crucial. This letter presents a method that combines a modified Importance Sampling (IS) distribution with active learning to generate high-quality samples. The suggested IS distribution generates samples iteratively from shells with error probabilities within a specific range. This approach enhances the performance of BCH(63,36) and BCH(63,45) codes with cycle-reduced parity-check matrices. The proposed IS-based-active Weight Belief Propagation (WBP) decoder improves the error-floor region by up to 1.9dB on the BER curve compared to the conventional WBP decoder.
Keywords:
Decoding
Training
Codes
Signal to noise ratio
Iterative decoding
Error probability
Belief propagation
Training data
Noise measurement
Monte Carlo methods
Importance sampling distribution
deep learning
error-correction codes
active learning
belief propagation
Journal
IF:
4.4
Papers:
1.3W
Citations:
2.2W