Adjacent Track Interference Cancellation in Magnetic Recording
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In magnetic recording systems, reducing track pitch and eliminating guard bands leads to significant interference from adjacent tracks, causing low data reliability due to the read head picking up signals from neighboring tracks, especially in Shingle Write Recording and Two-Dimensional Magnetic Recording techniques.
Innovation Solution
The method involves using multiple read heads to read both the current track and adjacent tracks, processing signals from these read heads to decode and remove interference, employing techniques like least-mean-square analysis to isolate the current track signal, and using ITI cancellation filters to mitigate interference.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If track pitch is reduced to increase recording density, then more tracks can fit on the storage medium, but the read head picks up significant signals from adjacent tracks causing low data reliability
Solution Approach 1:
The read head is divided into multiple separate read heads, each responsible for reading specific tracks. This segmentation allows the system to read multiple tracks simultaneously without interference, as each read head focuses on a specific track portion, thereby maintaining data reliability while enabling higher recording density through parallel reading operations
Solution Approach 2:
An interference cancellation filter is introduced as an intermediary component between the read heads and the data decoding process. This filter processes the signals from multiple read heads and subtracts interference from adjacent tracks, allowing the system to achieve high recording density while maintaining data reliability by actively removing harmful interference signals
2Quantity of substance
If guard bands are removed to increase recording density, then more tracks can fit on the storage medium, but interference from adjacent tracks increases significantly
Solution Approach 1:
The interference cancellation filter converts the harmful interference signals from adjacent tracks into useful information by processing the signals from multiple read heads. The filter uses the interference patterns to calculate and subtract the adjacent track contributions, transforming the previously harmful interference into a mechanism for maintaining data integrity while achieving high recording density without guard bands
Solution Approach 2:
Multiple read heads create copies of the track signals, with each read head reading a specific track portion. These signal copies are then processed together through interference cancellation, allowing the system to reconstruct the original track data accurately while removing interference, thereby enabling high density recording without traditional guard bands
3Device complexity
If a single read head is used to read the current track, then the device complexity is low, but the read head picks up signals from adjacent tracks causing interference
Solution Approach 1:
The single read head is segmented into multiple read heads, each positioned to read specific track portions. This segmentation increases device complexity but enables the system to read multiple tracks simultaneously without mutual interference, thereby improving signal accuracy and data reliability through parallel reading operations
Solution Approach 2:
The system transitions from a single read head reading one track to multiple read heads reading multiple tracks simultaneously in parallel. This dimensional change from sequential to parallel reading operations increases device complexity but significantly improves signal accuracy by eliminating adjacent track interference through spatial separation of reading operations
Data Source
AI summary
A method for reading a current track of data from a storage device includes using a first read head to read the current track of data and at least a first portion of at least one adjacent track of data, using at least a second read head to read at least a portion of the current track of data and at least a second portion of the at least one adjacent track of data, and decoding the data read from the current track, including processing signals from the first read head and the at least a second read head, to at least partly remove, from the signal from the first read head, contributions from the first portion of the at least one adjacent track of data. The decoding may include deriving expressions relating contributions from the current and adjacent tracks, and solving the expressions using least-mean-square analysis.


