Fractional Resampling with Two-Stage Filter Tap Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing digital signal resampling methods are computationally intensive and require significant memory and processing resources, making them inefficient in terms of precision and resource utilization.
Innovation Solution
A two-stage resampling method is employed, where the input signal is first divided into groups and resampled using a fixed number of filter taps, followed by a second resampling stage that phase-shifts the intermediate signal to align with the output signal, using a smaller number of filter taps that can fit within the L1 cache.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If a large number of filter taps are used to achieve high resampling precision, then manufacturing precision is improved, but device complexity and memory requirements increase
Solution Approach 1:
The patent divides the input signal into multiple groups and processes each group separately with a reduced set of filter taps. By segmenting the signal processing into stages (initial resampling followed by phase shifting), the system achieves high overall precision while using fewer filter taps at each stage, thus reducing device complexity and memory requirements.
2Manufacturing precision
If more memory resources are allocated for filter storage, then resampling precision is improved, but loss of substance increases
Solution Approach 1:
The patent segments the filter processing into multiple passes, where each pass uses a reduced set of filter taps that can be stored in L1 cache. The first pass performs initial resampling with a subset of taps, and the second pass performs phase shifting with another subset. This segmentation allows the system to maintain high precision while using far less memory than a single large filter would require.
3Manufacturing precision
If computational resources are increased to improve resampling accuracy, then manufacturing precision is improved, but productivity decreases
Solution Approach 1:
The patent segments the computationally intensive filtering operation into two lighter passes: an initial resampling pass and a phase shifting pass. Each pass uses fewer filter taps and can be optimized for cache efficiency. This segmentation reduces the computational burden per pass while maintaining overall accuracy, thereby improving processing efficiency and productivity.
4Manufacturing precision
If a single-stage resampling with many filter taps is used, then resampling precision is improved, but device complexity and memory usage increase
Solution Approach 1:
The patent replaces a single-stage resampling approach with multiple filter taps with a two-stage approach: initial resampling followed by phase shifting. Each stage uses a reduced number of filter taps that fit within L1 cache, simplifying the filter implementation. The phase shifting stage compensates for the segmentation, maintaining overall precision while reducing device complexity.
Data Source
AI summary
Methods and systems for resampling an input signal that includes a first plurality of values. A first resampling is performed to obtain an intermediate signal, which includes dividing the first plurality of values into a plurality of groups of values, resampling values in each group according to a first filter tap set to obtain an intermediate group of values. The first value of each group is aligned in time with the first value of each intermediate group. A second resampling is then performed on the intermediate signal to obtain an output signal, which performs a phase shift of each of the intermediate groups of values to align in time with a respective group of values of the output signal. The output signal is output by wired or wireless means.