Disk Read LLR Windowing for Defect-Aware Iterative Decoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing disk apparatuses face challenges in accurately detecting and mitigating the adverse effects of second medium defective portions on data sectors, which can lead to unreliable data reading due to the gradual decrease in signal amplitude, making it difficult to distinguish the boundary of defective areas and resulting in increased bit error rates.
Innovation Solution
A method is introduced where log likelihood ratios (LLRs) are multiplied by specific multipliers within set windows based on the amplitude of the read signal, allowing for effective control of the propagation of LLRs across different sections of the data sector, thereby reducing the impact of defective portions on normal data areas.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a scaling factor is used to reduce LLRs corresponding to medium defective portions, then the adverse effect of defective portions is suppressed, but it is difficult to accurately detect the boundary of second medium defective portions
Solution Approach 1:
The patent changes the parameter used for boundary detection from signal amplitude to LLR value distribution characteristics. By monitoring changes in LLR statistics (such as mean and variance) across data sectors, the system can accurately detect boundaries of second medium defective portions even when signal amplitude changes gradually, thereby resolving the contradiction between reliability improvement and measurement precision.
Solution Approach 2:
The patent replaces the traditional mechanical/amplitude-based detection method with a statistical analysis method based on LLR values. Instead of relying on physical signal amplitude measurements that fail to detect gradual degradation, the system uses probabilistic LLR analysis to identify defective portions, achieving both high reliability and precise boundary detection.
2Measurement precision
If LLRs are propagated through iterative decoding, then data decoding accuracy is improved, but defective portions affect other higher LLRs through probability propagation
Solution Approach 1:
The patent extracts and isolates the harmful LLR values corresponding to defective portions by identifying them through statistical analysis. Once identified, these problematic LLRs are excluded or given reduced weight in the iterative decoding process, preventing them from propagating adverse effects to other bits while maintaining the benefits of iterative decoding for reliable portions.
Solution Approach 2:
The patent applies different quality treatments to different portions of the data based on their reliability. LLRs from defective portions are treated differently (masked or weighted down) compared to LLRs from normal portions. This local differentiation allows the system to maintain high decoding accuracy for good portions while preventing error propagation from defective portions.
Data Source
AI summary
According to one embodiment, a method for operating log likelihood ratios in a disk apparatus is disclosed. Iterative decoding is applied to the disk apparatus. The method can set windows for a sequence of the log likelihood ratios output by a soft-decision most-likelihood decoder based on the sequence of the log likelihood ratios or an amplitude of a read signal acquired in response to read of data from a data sector on a disk carried out by a head. The method can multiply the log likelihood ratios contained in each of the windows, by a multiplier specific to each window. In addition, the method can transmit the sequence of the log likelihood ratios multiplied by the multiplier for each of the windows to a parity check decoder.