Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

IEEE Consumer Communications and Networking Conference

Logistic Regression for LDPC Decoding Failure Prediction

Taehyun Kim, Joosung Park

Published Jan 9, 2021• 3 citations
Publisher

Abstract

In this paper, we propose a regression analysis for the early stopping of low-density parity-check (LDPC) decoder to reduce computational complexity and decoding latency. The proposed scheme predicts the decoding failure of the LDPC decoder using logistic regression. The data sets for the decodability classification consist of the index of iteration and the number of unsatisfied parity checks. Simulation results show that with the proposed scheme, the average number of iterations of the LDPC decoder can be significantly reduced without performance loss. In addition, since the proposed scheme determines the decoding failure based on the instantaneous value, it has advantages in terms of memory and computational complexity compared to the early stopping schemes based on the changes in parameters.

DOI
10.1109/ccnc49032.2021.9369532

Related papers

Stopping Criterion for LDPC Decoder based on PEXIT Chart Analysis

Taehyun Kim, Joosung Park, Jun Heo

In this paper, an early stopping scheme is proposed for low-density parity-check (LDPC) decoders. The proposed scheme predicts that the LDPC decoder will not be able to correct errors and then stops the iterative decoding process before reaching the maximum number of iterations. When the number of parity checks unsatisfied at both the l-th and the (l-T)-th decoding iterations is larger than a threshold, the decoder with the proposed scheme predicts that the decoding of the codeword is failed at the last iteration. Protograph-based extrinsic information transfer (PEXIT) chart analysis is used to determine the threshold. Simulation results show that the LDPC decoder can correct errors with the lower average number of iterations by using both the conventional syndrome test and proposed scheme.

IEEE Wireless Communications and Networking Conference•May 1, 2020•2 citations

A Stopping Criterion for Low-Density Parity-Check Codes

Donghyuk Shin, Kyoungwoo Heo, Sang-Bong Oh +1 more

Low-density parity-check (LDPC) codes have an inherent stopping criterion, parity-check constraints (equations). By testing the parity-check constraints, an LDPC decoder can detect successful decoding and stop their decoding, which is, however, not possible with turbo codes. In this paper, we propose a stopping criterion to predict decoding failure of LDPC codes, instead of detecting successful decoding. If the decoder predicts the decoding failure in advance, the receiver can more rapidly response to the transmitter and request for additional parity bits with an automatic repeat request (ARQ) protocol, which reduces overall system latency. The receiver can also save power consumption by avoiding unnecessary decoder iterations. The proposed stopping criterion makes use of the variations of the number of satisfied parity-check constraints in the belief-propagation (BP) decoding which is always tested in the conventional BP decoding to detect successful decoding. Thus, the proposed stopping criterion does not require any additional complexity. The counting of satisfied parity-check constraints shows behaviors of the BP decoding, which comes, otherwise, from the observations of changes of log-likelihood ratio (LLR) values in multi-bit resolution with additional complexity.

IEEE Vehicular Technology Conference•Apr 1, 2007•33 citations

An early stopping criterion for LDPC decoding based on average weighted reliability measure

Tso-Cho Chen

An early stopping criterion for low-density parity-check decoding is proposed to reduce the number of decoding iterations. The proposed early stopping criterion is based on the average weighted reliability measure of the hard output of decoder to detect undecodable blocks in an early stage of the decoding process. Simulation results show that the proposed algorithm can achieve about 0.2–0.4 dB coding gain improvement on average while reducing up to about 50%–85% iteration for decoding at low to medium Eb/N0, compared with the existing method.

Cross Strait Quad-Regional Radio Science and Wireless Technology Conference (CSQRWC)•Jul 1, 2012•2 citations

A stopping criterion for nonbinary LDPC codes over GF(q)

Xin Chen, Aidong Men, Wei Zhou

Nonbinary low-density parity-check (LDPC) codes over finite fields GF(q) have a better performance than those of the binary low-density parity-check codes, at short and medium block lengths, but the decoder of GF(q)-LDPC has more complexity. In this paper, a new stopping criterion is proposed. Based on the changes of the variance of variable node¿s extrinsic information, the criterion can predict a decoding failure and decide whether the iteration of the decoder should be stopped. The simulation results show that the stopping criterion can greatly reduces the average number of required iterations of GF(q)-LDPC decoder with negligible performance loss.

IEEE International Conference on Communication Systems (ICCS)•Nov 1, 2008•4 citations

A hybrid low complexity decoding of LDPC codes

Hua Wang, Guangrong Fan, Jingming Kuang

An efficient early stopping criterion based on the evolution of the number of reliable variable nodes (NRVN) is proposed for low density parity check (LDPC) codes in this paper. This criterion can significantly reduce the average number of iterations (ANI) at low to middle signal to noise ratios (SNRs) with minimal performance degradation. Furthermore, a hybrid decoding method is presented to reduce the decoding complexity of LDPC codes by combining the proposed stopping criterion and forced convergence (FC) decoding algorithm. Compared with the classic belief propagation (BP) decoding, significant computation complexity reduction can be achieved with small computation overhead.

IET International Conference on Wireless, Mobile and Multimedia Networks (ICWMMN)•Jan 1, 2010•1 citation

Low Complexity Stopping Criterion for LDPC Code Decoders

Frank Kienle, Norbert Wehn

Low density parity check (LDPC) codes are amongst the most powerful codes known today. They are decoded iteratively by a message-passing algorithm. For this, an inherent stopping criterion exists for decodable blocks based on parity checks. For undecodable blocks, typically, a fixed number of iterations are carried out before the decoding procedure is terminated, which is a waste of energy and time. We present a stopping criterion which detects undecodable blocks in an early stage of the decoding process. The novel stopping criterion works for floating and fixed point implementations. It greatly reduces the average number of required iterations without any loss in communication performance.

IEEE Vehicular Technology Conference•Dec 10, 2005•67 citations
In this paper, we propose a regression analysis for the early stopping of low-density parity-check (LDPC) decoder to reduce computational complexity and decoding latency. The proposed scheme predicts the decoding failure of the LDPC decoder using logistic regression. The data sets for the decodability classification consist of the index of iteration and the number of unsatisfied parity checks. Simulation results show that with the proposed scheme, the average number of iterations of the LDPC decoder can be significantly reduced without performance loss. In addition, since the proposed scheme determines the decoding failure based on the instantaneous value, it has advantages in terms of memory and computational complexity compared to the early stopping schemes based on the changes in parameters.
In this paper, an early stopping scheme is proposed for low-density parity-check (LDPC) decoders. The proposed scheme predicts that the LDPC decoder will not be able to correct errors and then stops the iterative decoding process before reaching the maximum number of iterations. When the number of parity checks unsatisfied at both the l-th and the (l-T)-th decoding iterations is larger than a threshold, the decoder with the proposed scheme predicts that the decoding of the codeword is failed at the last iteration. Protograph-based extrinsic information transfer (PEXIT) chart analysis is used to determine the threshold. Simulation results show that the LDPC decoder can correct errors with the lower average number of iterations by using both the conventional syndrome test and proposed scheme.
Low-density parity-check (LDPC) codes have an inherent stopping criterion, parity-check constraints (equations). By testing the parity-check constraints, an LDPC decoder can detect successful decoding and stop their decoding, which is, however, not possible with turbo codes. In this paper, we propose a stopping criterion to predict decoding failure of LDPC codes, instead of detecting successful decoding. If the decoder predicts the decoding failure in advance, the receiver can more rapidly response to the transmitter and request for additional parity bits with an automatic repeat request (ARQ) protocol, which reduces overall system latency. The receiver can also save power consumption by avoiding unnecessary decoder iterations. The proposed stopping criterion makes use of the variations of the number of satisfied parity-check constraints in the belief-propagation (BP) decoding which is always tested in the conventional BP decoding to detect successful decoding. Thus, the proposed stopping criterion does not require any additional complexity. The counting of satisfied parity-check constraints shows behaviors of the BP decoding, which comes, otherwise, from the observations of changes of log-likelihood ratio (LLR) values in multi-bit resolution with additional complexity.
An early stopping criterion for low-density parity-check decoding is proposed to reduce the number of decoding iterations. The proposed early stopping criterion is based on the average weighted reliability measure of the hard output of decoder to detect undecodable blocks in an early stage of the decoding process. Simulation results show that the proposed algorithm can achieve about 0.2–0.4 dB coding gain improvement on average while reducing up to about 50%–85% iteration for decoding at low to medium Eb/N0, compared with the existing method.
Nonbinary low-density parity-check (LDPC) codes over finite fields GF(q) have a better performance than those of the binary low-density parity-check codes, at short and medium block lengths, but the decoder of GF(q)-LDPC has more complexity. In this paper, a new stopping criterion is proposed. Based on the changes of the variance of variable node¿s extrinsic information, the criterion can predict a decoding failure and decide whether the iteration of the decoder should be stopped. The simulation results show that the stopping criterion can greatly reduces the average number of required iterations of GF(q)-LDPC decoder with negligible performance loss.
An efficient early stopping criterion based on the evolution of the number of reliable variable nodes (NRVN) is proposed for low density parity check (LDPC) codes in this paper. This criterion can significantly reduce the average number of iterations (ANI) at low to middle signal to noise ratios (SNRs) with minimal performance degradation. Furthermore, a hybrid decoding method is presented to reduce the decoding complexity of LDPC codes by combining the proposed stopping criterion and forced convergence (FC) decoding algorithm. Compared with the classic belief propagation (BP) decoding, significant computation complexity reduction can be achieved with small computation overhead.
Low density parity check (LDPC) codes are amongst the most powerful codes known today. They are decoded iteratively by a message-passing algorithm. For this, an inherent stopping criterion exists for decodable blocks based on parity checks. For undecodable blocks, typically, a fixed number of iterations are carried out before the decoding procedure is terminated, which is a waste of energy and time. We present a stopping criterion which detects undecodable blocks in an early stage of the decoding process. The novel stopping criterion works for floating and fixed point implementations. It greatly reduces the average number of required iterations without any loss in communication performance.