MRI Signal Processing for Large-Tap FIR Filter Computation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
It has been challenging to put a finite impulse response (FIR) digital filter with a large number of taps into practical use due to high computational complexity and significant digit cancellation issues.
Innovation Solution
The signal processing apparatus computes integral values and output signals using representative coefficients determined for ranges of amplification factors, employing a method that adds non-overlapping values from subsequent input sequences to previous integral values and subtracts overlapping values, thereby reducing computational complexity and preventing significant digit cancellation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If an FIR digital filter with a large number of taps is used, then filtering precision is improved, but computational complexity increases
Solution Approach 1:
The patent divides the filter coefficient sequence into multiple segments and processes each segment separately. By segmenting the large-scale filter into smaller manageable parts, the computational complexity is reduced while maintaining the overall filtering precision of the large-tap FIR filter.
Solution Approach 2:
The patent performs preliminary computation of integral values and stores them for reuse. By pre-computing and storing integral values that will be needed for subsequent filtering operations, the actual filtering process requires fewer computations, thus reducing computational complexity while preserving filtering accuracy.
2Measurement precision
If an FIR digital filter with a large number of taps is used, then filtering precision is improved, but significant digit cancellation occurs
Solution Approach 1:
The patent pre-computes and stores integral values before the actual filtering operation. This preliminary action ensures that large cumulative sums are calculated once and stored, avoiding repeated computation that would exacerbate significant digit cancellation and maintain calculation reliability.
Solution Approach 2:
The patent introduces integral values as intermediary quantities that mediate between the input signal and the final filtered output. These intermediary integral values are computed with sufficient precision and stored, serving as a buffer that prevents significant digit cancellation from propagating through the filtering process.
3Productivity
If computational complexity is reduced, then processing speed is improved, but filtering precision may deteriorate
Solution Approach 1:
By segmenting the filter coefficients and processing different segments with different levels of computational effort, the patent achieves a balance between processing speed and filtering precision. Important segments can be processed with higher precision while less critical segments use simplified computation.
Solution Approach 2:
The patent changes computational parameters dynamically, using full precision for critical computations and reduced precision for less critical ones. The integral value computation uses appropriate precision levels based on the specific filtering requirements, optimizing the trade-off between processing speed and filtering precision.
Data Source
AI summary
A signal processing apparatus according to the present embodiment computes a first integral value corresponding to an element in a coefficient sequence of a first input sequence and a second integral value corresponding to the element in a coefficient sequence of a second input sequence next to the first input sequence, and includes a processing circuitry. The processing circuitry adds a value not overlapping the first input sequence in the second input sequence to the first integral value and subtracts a value not overlapping the second input sequence in the first input sequence from the first integral value for the element, thereby computing the second integral value.


