Wavelet Defect Scanning for Storage Media
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for identifying defects in storage device media, such as hard disc drives and solid-state drives, are inadequate as they either fail to provide clear information about defect duration, location, and type, especially when defects manifest in frequency changes, and are prone to erroneous decisions due to spurious noise.
Innovation Solution
The implementation of a wavelet-based defect scanning system that applies a wavelet transform to Analog to Digital Converter (ADC) samples from a readback signal, allowing for simultaneous localization of defects in both the time and frequency domains, using parameters determined by an ADC samples analyzer to categorize and map defects accurately.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If time domain analysis is used to detect defects, then defect location and duration can be identified, but frequency changes of defects cannot be clearly detected
Solution Approach 1:
The patent transitions from single-domain analysis (time or frequency) to dual-domain analysis by applying wavelet transform. This mathematical transform decomposes the readback signal into time-frequency representations, allowing simultaneous observation of both temporal location and frequency characteristics of defects. The wavelet coefficients provide a multi-dimensional view that captures both when a defect occurs and what frequency changes are associated with it.
2Measurement precision
If frequency domain analysis is used to detect defects, then frequency changes can be identified, but defect duration information is lost
Solution Approach 1:
The wavelet transform adds a time dimension to the frequency domain analysis. By representing the signal in the time-frequency plane through wavelet coefficients, the system can simultaneously determine both the frequency content and the temporal location/duration of defects. This resolves the information loss problem inherent in traditional Fourier-based frequency domain analysis.
3Ease of manufacture
If traditional defect scanning methods are used, then simple defect detection is possible, but accurate categorization of defect types and locations is inadequate
Solution Approach 1:
The patent segments the defect detection process into multiple analytical stages using wavelet decomposition. The wavelet transform breaks down the readback signal into different frequency bands and time intervals, creating a hierarchical structure of wavelet coefficients. This segmentation allows systematic analysis of defects at multiple resolutions, enabling accurate categorization of defect types (e.g., drop-out, drop-in, erasure) and precise location identification while maintaining a structured approach that builds upon simpler detection methods.
Data Source
AI summary
A method and system for providing simultaneous localization of defects in both the time and frequency domain. A high frequency repeating pattern is written on media, and the pattern is read to generate a readback signal, which is converted into ADC samples. The ADC samples are analyzed, in-line, to determine the type of wavelet, level of decomposition, and threshold level for a wavelet transform of the particular readback signal. The wavelet transform provides details and/or approximations (wavelet coefficients) that are analyzed to determine the type, location, and duration of any identified defects. Any noise in the details and/or approximations (wavelet coefficients) is removed by a wavelet based denoising operation. Flags indicating the type, location, and duration of any defects are generated so that the defects may be mapped.