2D Data Dependent Noise Whitening Filters for Multi-Track Storage
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data storage devices face challenges in accurately demodulating read signals due to non-white, data-dependent noise in magnetic recording channels, which affects the performance of Viterbi data detectors, especially as the length of data sequences increases, requiring numerous noise whitening filters that double with each additional bit.
Innovation Solution
Implementing two-dimensional (2D) data dependent noise whitening (DDNW) filtering and equalization techniques to process read signals from multiple tracks, using 2D equalizers and DDNW filters to generate noise-whitened samples that can detect data sequences across multiple tracks, and computing branch metrics considering covariance matrices to improve demodulation accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the length of data sequence increases to improve noise whitening performance, then the accuracy of ML data detector improves, but the number of data dependent noise whitening filters doubles with each additional bit
Solution Approach 1:
The patent combines multiple data-dependent noise whitening filters into a single filter by merging their functionality. Instead of implementing separate filters for each possible data sequence (which would double in number with each additional bit), the invention integrates the noise whitening function to process multiple data sequences simultaneously, thereby maintaining detection accuracy while dramatically reducing the number of filters required.
Solution Approach 2:
The patent creates a universal noise whitening filter that can handle multiple data sequences with different optimal noise-whitening functions. This single filter is designed to be multi-functional, adapting its behavior based on the input data sequence while maintaining the noise whitening performance that would otherwise require multiple specialized filters.
2Measurement precision
If traditional noise whitening filters are used to address data-dependent noise, then demodulation accuracy improves, but the system cannot effectively process multi-track data with intersymbol interference
Solution Approach 1:
The patent transitions from one-dimensional noise whitening filtering to two-dimensional filtering that operates across multiple tracks simultaneously. This dimensional extension allows the filter to account for intersymbol interference between adjacent tracks while maintaining demodulation accuracy, thereby achieving both precise demodulation and effective multi-track processing.
Solution Approach 2:
The patent segments the filtering operation into track-specific components while maintaining overall coherence. By dividing the multi-track data processing into manageable segments that can be handled individually yet integrated collectively, the system achieves both high demodulation accuracy for individual tracks and effective handling of intersymbol interference across tracks.
Data Source
AI summary
A data storage device is disclosed wherein a first 2D data dependent noise whitening (DDNW) filter is configured to perform 2D DDNW of first and second 2D equalized samples to generate first 2D noise whitened samples. A second 2D DDNW filter is configured to perform 2D DDNW of the first and second 2D equalized samples to generate second 2D noise whitened samples. A 2D sequence detector is configured to detect a first data sequence recorded in a first data track from the first and second 2D noise whitened samples and to detect a second data sequence recorded in a second data track from the first and second 2D noise whitened samples.