Multi-Rate Oversampling With Fixed Anti-Alias Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Magnetic storage devices face challenges with high costs and complexity due to the need for expensive anti-alias filters with variable cut-off frequencies in multi-rate oversampling processes, which are required to prevent unwanted frequencies from appearing in sampled signals.
Innovation Solution
Implementing a simpler anti-alias filter with a fixed cut-off frequency and using multiple over-sampling rates, coupled with decimation filters that maintain a common power spectral density, allowing for variable sampling rates without the need for costly adjustments in circuit implementation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If variable cut-off frequency anti-alias filters are used to prevent unwanted frequencies in multi-rate oversampling, then signal processing reliability is improved, but component cost and circuit complexity increase
Solution Approach 1:
The signal processing system is segmented into distinct functional blocks: a fixed cut-off frequency anti-alias filter stage, a variable-rate ADC stage, and a digital decimation stage. This segmentation allows the analog filter to be simple and fixed, while the complexity is moved to the digital domain where it can be flexibly adjusted through software control of the decimation factor, resolving the contradiction between reliability and complexity.
Solution Approach 2:
The patent introduces an intermediary digital decimation stage between the fixed anti-alias filter and the final output. This intermediary process allows the system to achieve variable effective sampling rates without requiring variable cut-off frequency analog filters, thereby maintaining filter simplicity while achieving the desired signal processing reliability through digital signal processing.
2Reliability
If variable cut-off frequency anti-alias filters are used to prevent unwanted frequencies in multi-rate oversampling, then signal processing reliability is improved, but component cost increases
Solution Approach 1:
The system segments the filtering function into a simple fixed analog anti-alias filter and a digital decimation filter. This segmentation allows the use of a low-cost fixed cut-off frequency analog filter instead of expensive variable cut-off frequency filters, while maintaining signal processing reliability through the addition of the digital decimation stage.
Solution Approach 2:
The patent replaces the mechanical/analog approach of variable cut-off frequency filters with a digital signal processing approach. The variable sampling rate control is achieved through digital decimation rather than analog filter tuning, substituting complex analog components with simpler, lower-cost digital processing that achieves the same functional result.
3Device complexity
If multiple over-sampling rates are used with fixed cut-off frequency anti-alias filter, then component cost and circuit complexity are reduced, but maintaining common power spectral density becomes more difficult
Solution Approach 1:
The system employs feedback control where the decimation factor is adjusted based on the selected oversampling rate to maintain a consistent effective sampling rate and common power spectral density. The controller monitors the oversampling rate and dynamically adjusts the decimation parameter to ensure that the output signal maintains consistent spectral characteristics regardless of the input oversampling rate, resolving the contradiction between simplicity and spectral consistency.
Solution Approach 2:
The patent changes the decimation parameter dynamically based on the selected oversampling rate. By adjusting this parameter, the system maintains a constant effective sampling rate and common power spectral density across different oversampling rates, achieving spectral consistency without requiring complex fixed-frequency filters for each rate.
Data Source
AI summary
Multi-rate oversampling of analog signals in storage devices is described. A method of processing an analog signal derived from a storage medium in a storage device includes: filtering the analog signal with an anti-alias filter having a fixed cut-off frequency related to a target sampling rate; sampling the analog signal using an over-sampling rate of a plurality of over-sampling rates provided by a variable-rate analog-to-digital converter (ADC) to produce an over-sampled digital signal; and filtering the over-sampled digital signal using a decimation filter of a plurality of decimation filters provided by a digital signal processor (DSP) to produce a digital signal having the target sampling rate.