LDPC Bit-Flipping Decoder Using Bit Confidence Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing LDPC bit-flipping decoders have limited error correction capabilities, leading to a high uncorrectable bit error rate (UBER) and reduced storage device lifespan due to their complexity and inefficiency in handling bit errors.
Innovation Solution
The method involves operating an LDPC bit-flipping decoder by receiving an LDPC codeword, classifying bits based on soft reliability information, and iteratively processing the codeword over multiple iterations to correct bit errors, while preventing high-confidence bits from exceeding 33% of total flipped bits in the initial iterations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional LDPC bit-flipping decoding is used, then the decoder complexity is low, but the error correction capability is limited leading to high UBER
Solution Approach 1:
The patent applies preliminary action by classifying bits into high-confidence and low-confidence categories before the decoding process begins, using soft reliability information from the storage medium. This pre-classification allows the decoder to prioritize which bits to flip during iterations, improving error correction capability without requiring complex algorithms. The classification is performed once at the start, and the same classification guide is used throughout all decoding iterations.
Solution Approach 2:
The patent applies local quality by treating different bits differently based on their confidence levels. Low-confidence bits are targeted for flipping while high-confidence bits are protected from being flipped. This localized approach means that only specific portions of the codeword (the low-confidence bits) undergo modification during decoding, rather than treating all bits uniformly. This selective processing improves error correction while maintaining simplicity.
2Reliability
If more iterations are performed to improve error correction, then the UBER decreases, but the decoding time and energy consumption increase
Solution Approach 1:
By performing bit classification before decoding and using this classification to guide the entire decoding process, the patent achieves better error correction in fewer iterations. The preliminary classification provides a roadmap that prevents wasted iterations on bits that don't need flipping, thus reducing total decoding time while maintaining low UBER.
Solution Approach 2:
The soft reliability information from the storage medium essentially performs part of the decoding work by identifying which bits are likely erroneous. This self-service approach means the storage medium provides reliability information that directly guides the decoder, reducing the computational burden and iteration count needed to achieve successful decoding.
3Reliability
If soft reliability information is used to classify bits, then the error correction improves, but the processing complexity increases
Solution Approach 1:
The patent extracts only the essential information needed for decoding - the soft reliability information that indicates bit confidence levels. Rather than processing all possible bit properties or using complex probabilistic models, the system extracts and uses only the confidence level classification. This extraction approach simplifies processing while maintaining improved error correction capability.
4Productivity
If high-confidence bits are allowed to be flipped freely, then the decoding may converge faster, but the error correction accuracy decreases
Solution Approach 1:
The patent applies local quality by imposing different rules on different bits based on their confidence levels. High-confidence bits are protected from flipping (local protection), while low-confidence bits are allowed to be flipped (local modification). This localized differential treatment ensures that bits likely to be correct are not mistakenly changed, maintaining error correction accuracy, while still allowing sufficient flexibility in the decoding process.
Data Source
AI summary
Systems and methods for operating a low-density parity-check (LDPC) bit-flipping decoder are disclosed herein. An LDPC codeword is received, and each bit in the LDPC codeword is classified as either a high-confidence bit or a low-confidence bit based on at least one criterion. The LDPC codeword is iteratively processed over a plurality of iterations based on parity check equations associated with each bit of the LDPC codeword to generate a processed LDPC codeword. For each iteration of the plurality of iterations, the iterative processing includes flipping at least one bit of the LDPC codeword, while preventing, for a first n number of the plurality of iterations, bits classified as high-confidence bits from comprising more than 33% of the total number of flipped bits. The processed LDPC codeword is decoded.