Signal Processor Waveform Similarity Dispersion Correction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing heartbeat measuring devices inaccurately determine waveform similarity due to determining similar figures as similar, which reduces the accuracy of similarity determination.

Innovation Solution

A signal processor calculates the correlation coefficient between waveforms in a standard window and a comparison window, with a lower correlation coefficient value when the difference in amplitude dispersion between the two windows is larger, improving the accuracy of waveform similarity determination.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If the correlation coefficient is calculated using only waveform shape similarity, then the calculation is simple, but the accuracy of waveform similarity determination is reduced due to determining similar figures as similar

Engineering Contradiction:
Improveaccuracy of waveform similarity determinationVSAvoidcomplexity of correlation coefficient calculation
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent modifies the correlation coefficient calculation by introducing amplitude dispersion as an additional parameter. The corrected correlation coefficient is calculated as: corrected correlation coefficient = correlation coefficient × (amplitude dispersion difference factor), where the amplitude dispersion difference factor is derived from the ratio of amplitude dispersions in the first and second windows. This parameter change ensures that waveforms with similar shapes but different amplitude characteristics are properly distinguished, resolving the issue of determining similar figures as similar while maintaining computational feasibility.

Inventive Principle:
Principle #35Parameter changes

2Measurement precision

If the correlation coefficient calculation considers amplitude dispersion differences, then the accuracy of similarity determination is improved, but the calculation complexity increases

Engineering Contradiction:
Improveaccuracy of similarity determinationVSAvoiddifficulty of correlation coefficient calculation
Core Design Contradiction:
Measurement precisionVSDifficulty of detecting and measuring

Solution Approach 1:

The patent segments the correlation coefficient calculation into two distinct parts: (1) the basic correlation coefficient calculation based on waveform shape similarity, and (2) the amplitude dispersion correction factor calculation. By dividing the overall calculation process into these separate components, the system can independently compute each part and then combine them through multiplication. This segmentation makes the enhanced calculation more manageable and easier to implement while achieving improved accuracy in waveform similarity determination.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentUS9962126B2Signal processor, signal processing method, and recording medium
Publication Date: 2018.05.08 FUJITSU LTD
  • US9962126B2 patent drawing
  • US9962126B2 patent drawing
  • US9962126B2 patent drawing

AI summary

A pulse wave detection apparatus detects a signal, and calculates a correlation coefficient between waveforms of the signal included in a first window and a second window having a predetermined duration, the correlation coefficient being smaller when a difference between dispersion of the amplitude value in the first window and dispersion of the amplitude value in the second window becomes larger.