LDPC Codeword Adjustment for Miscorrection Recovery

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current communications systems face challenges in accurately recovering original data words from encoded codewords due to burst errors and miscorrections during low-density parity-check (LDPC) decoding, particularly in noisy channels, where existing methods struggle to converge on the correct codeword and may result in incorrect data recovery.

Innovation Solution

The proposed solution involves a receiving path that implements a channel detection technique using Viterbi soft-output detection and interleaving schemes to process encoded codewords, followed by LDPC decoding and error-detection list-decoding methods to correct miscorrections, utilizing a list of candidate error vectors to adjust the codeword and ensure cyclic-redundancy check compliance, thereby improving data recovery accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If LDPC decoding is used to correct errors in noisy channels, then error correction capability is improved, but miscorrections occur reducing data recovery accuracy

Engineering Contradiction:
Improveerror correction capabilityVSAvoiddata recovery accuracy
Core Design Contradiction:
ReliabilityVSMeasurement precision

Solution Approach 1:

The decoding process is segmented into multiple stages: initial LDPC decoding, CRC verification, and conditional list decoding. By dividing the error correction process into distinct segments with different strategies, the system achieves both high error correction capability and accurate data recovery, avoiding miscorrections while maintaining reliability in noisy channels.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system implements feedback through CRC verification after LDPC decoding. When the CRC check fails, the system feedbacks to the list decoding stage to generate candidate codewords. This feedback mechanism ensures that miscorrections are detected and corrected, improving data recovery accuracy while maintaining the error correction benefits of LDPC decoding.

Inventive Principle:
Principle #23Feedback

2Measurement precision

If list decoding is implemented to correct miscorrections, then data recovery accuracy is improved, but processing complexity increases

Engineering Contradiction:
Improvedata recovery accuracyVSAvoidprocessing complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

Instead of implementing full list decoding for all received codewords, the system applies list decoding selectively only when CRC verification fails after initial LDPC decoding. This partial action approach maintains data recovery accuracy by correcting miscorrections when needed, while avoiding the excessive processing complexity of universal list decoding.

Inventive Principle:
Principle #16Partial or excessive action

Solution Approach 2:

The system applies different decoding strategies to different cases: standard LDPC decoding for most cases and list decoding only when miscorrections are detected via CRC failure. This local quality approach optimizes processing complexity by applying complex list decoding only where necessary, while maintaining high data recovery accuracy through targeted intervention.

Inventive Principle:
Principle #3Local quality

3Reliability

If multiple decoding iterations are performed to ensure correct codeword convergence, then data integrity is improved, but latency increases

Engineering Contradiction:
Improvedata integrityVSAvoiddecoding latency
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The system performs preliminary CRC verification immediately after initial LDPC decoding to determine whether further decoding iterations are needed. This preliminary action allows the system to quickly verify data integrity for most correctly decoded codewords, avoiding unnecessary additional iterations and reducing latency while maintaining high data integrity through targeted redecoding when needed.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS8898539B2Correcting errors in miscorrected codewords using list decoding
Publication Date: 2014.11.25 AVAGO TECHNOLOGIES INTERNATIONAL SALES PTE LTD
  • US8898539B2 patent drawing
  • US8898539B2 patent drawing
  • US8898539B2 patent drawing

AI summary

A receive path of a communications system comprises an error-correction decoder, an error-detection decoder, and a codeword adjuster. The error-correction decoder performs error-correction decoding on a received codeword to generate a valid codeword. The error-detection decoder performs error-detection decoding on the valid codeword to determine whether or not the valid codeword is the correct codeword that was transmitted. If the valid codeword is not the correct codeword, then the codeword adjuster generates an adjusted valid codeword by applying an error vector to the valid codeword. The error-detection decoder performs error-detection decoding on the adjusted valid codeword to determine whether or not the adjusted valid codeword is the correct codeword. When the error-correction decoder generates an incorrect valid codeword, adjusting the valid codeword enables the receive path to recover the correct codeword without retransmitting or re-detecting the codeword.