Soft-Input Decoding Margining for Defective Sector Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Hard disk drive systems face challenges in determining the margin of error correction, especially with LDPC codes, as they are not bounded distance decodable, making it difficult to assess when the decoder is close to failure and thus hard to ensure disk quality over time due to degradation.
Innovation Solution
A figure-of-merit is introduced to identify defective sectors by analyzing the number of incorrect log-likelihood ratios (LLRs) with magnitudes greater than a configurable threshold, which helps in determining the sector's defectiveness and the decoder's likelihood of failure, allowing for effective margining and quality assessment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If LDPC codes are used for error correction, then error correction capability is improved, but the ability to determine decoder proximity to failure deteriorates
Solution Approach 1:
The patent introduces log-likelihood ratios (LLRs) as an intermediary metric to assess decoder proximity to failure. LLRs provide soft information about the reliability of decoded bits, enabling the system to determine when the decoder is approaching failure without compromising the error correction capability of LDPC codes. This mediator allows measurement of decoder state while maintaining the advanced error correction properties of LDPC.
Solution Approach 2:
The patent replaces traditional bounded-distance decoding metrics with a statistical approach using LLRs. Instead of relying on hard decision metrics that fail to provide proximity information, the system uses soft probabilistic metrics that continuously indicate decoder confidence, enabling detection of failure proximity while maintaining superior error correction performance.
2Reliability
If product qualification testing is performed beyond intended use, then product quality assurance is improved, but test time and resource consumption increase
Solution Approach 1:
The patent performs margining assessment during the manufacturing qualification process by analyzing LLR distributions from qualification tests. This preliminary action identifies defective sectors before product shipment, ensuring that only products meeting quality thresholds are released. By conducting this assessment during manufacturing rather than requiring extensive field testing, the system reduces overall time and resource consumption while maintaining high quality assurance.
Solution Approach 2:
The patent replaces extensive physical stress testing with a computational analysis of LLR distributions from normal qualification tests. Instead of performing additional beyond-intended-use testing, the system uses statistical analysis of existing test data to assess product margin, significantly reducing test time while maintaining rigorous quality assurance through probabilistic evaluation of decoder reliability.
Data Source
AI summary
Determining a parameter associated with whether a portion of a storage device is defective is disclosed. Determining comprises: obtaining known data associated with the portion; reading back from the portion to produce a read-back waveform; decoding the read-back waveform, including producing statistical information; and determining a parameter associated with whether the portion is defective based at least in part on the statistical information.