Interleaved Servo Codewords for More Accurate Trellis Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data storage devices face challenges in accurately positioning the head over data tracks due to limitations in trellis sequence detection, particularly in decoding codewords with varying code rates and sizes, which affects the reliability of servo bursts and position error signals.
Innovation Solution
The implementation of a data storage device that interleaves first and second codewords with different code rates and sizes, using non-uniform and uniform interleaving techniques to generate an interleaved codeword, which improves the accuracy of trellis detectors by increasing redundancy and facilitating successful decoding of the first codeword, thereby enhancing the recovery of the second codeword.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional trellis sequence detection is used for decoding codewords with varying code rates and sizes, then the decoding process can be simplified, but the accuracy of head positioning and data recovery deteriorates
Solution Approach 1:
The codeword is divided into multiple segments (first codeword and second codeword) with different code rates. The first codeword has a lower code rate and is decoded first to provide reliable metrics, while the second codeword has a higher code rate and is decoded subsequently. This segmentation allows the trellis detector to process information in manageable stages, improving positioning accuracy without overwhelming computational complexity.
Solution Approach 2:
The first codeword is decoded before the second codeword in a preliminary action. The successful decoding of the first codeword provides reliable metrics that are used to prune the trellis branches during subsequent decoding of the second codeword. This preliminary decoding action reduces the search space and improves the accuracy of head positioning while managing the overall decoding complexity.
2Reliability
If codewords with different code rates are decoded simultaneously, then the decoding process is faster, but the reliability of position error signals deteriorates
Solution Approach 1:
The decoding process is segmented into multiple iterations, with each iteration focusing on specific codewords. In the first iteration, the first codeword is decoded to generate reliable metrics. In subsequent iterations, these metrics are used to aid the decoding of the second codeword. This segmented approach ensures that position error signals are generated with high reliability from the first codeword before being used for the second codeword.
Solution Approach 2:
The decoding process uses feedback from the first codeword decoding to improve the second codeword decoding. The metrics obtained from decoding the first codeword are fed back into the trellis detector to prune branches during the decoding of the second codeword. This feedback mechanism enhances the reliability of position error signals while maintaining efficient decoding throughput through iterative refinement.
3Measurement precision
If the trellis detector processes all codeword branches equally, then the decoding is simpler, but the accuracy of data recovery deteriorates
Solution Approach 1:
The trellis detector applies different processing quality to different branches based on their reliability. Branches corresponding to the first codeword, which has a lower code rate and higher reliability, are processed with higher accuracy and used to prune the trellis. Branches for the second codeword are processed subsequently with the benefit of the pruned search space. This local quality differentiation improves data recovery accuracy without uniformly increasing complexity across all branches.
Solution Approach 2:
The trellis detector performs partial processing in the first iteration by focusing on decoding the first codeword and generating metrics for it. This partial action is sufficient to prune the trellis branches for the second codeword in subsequent iterations. The excessive action of fully decoding all codewords in a single pass is avoided, instead using multiple passes with progressively refined information to achieve high accuracy data recovery.
Data Source
AI summary
A data storage device is disclosed comprising a storage medium. First data is encoded into a first codeword comprising a plurality of i-bit symbols, and second data is encoded into a second codeword comprising a plurality of j-bit symbols, wherein i is different than j and a first code rate of the first codeword is less than a second code rate of the second codeword. The first codeword and the second codeword are symbol interleaved to generate an interleaved codeword, and the interleaved codeword is written to the storage medium.