Signal Feature Extraction Apparatus for Accurate Peak Positioning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing signal feature extraction techniques, such as those using filter banks, often produce erroneous peak positions due to noise interference, leading to inaccurate analysis of waveform patterns in signals like electrocardiograms.
Innovation Solution
A signal feature extraction apparatus that employs a local maximum value detection part and a feature peak value selection part to identify and select feature peak values based on time interval and amplitude/phase differences, distinguishing desired peaks from noise-induced peaks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a feature extraction filter is used to extract waveform patterns from input signals, then the desired waveform pattern can be extracted, but noise may cause erroneous peak positions leading to inaccurate feature analysis
Solution Approach 1:
The patent applies feedback by using the average time interval information from previously detected feature peak values to evaluate and select the current peak candidate. The selection process feeds back the time interval difference and amplitude/phase differences to determine whether to accept the current peak, creating a closed-loop verification system that improves peak detection accuracy and eliminates noise-induced erroneous peaks.
2Reliability
If multiple peak detection criteria are applied to eliminate noise peaks, then feature analysis reliability improves, but detection complexity increases
Solution Approach 1:
The patent changes parameters by evaluating peak candidates based on multiple parameters simultaneously: time interval difference from average time interval, amplitude difference from average amplitude, and phase difference from average phase. This multi-parameter evaluation approach systematically distinguishes valid feature peaks from noise peaks while maintaining clear selection criteria.
Data Source
AI summary
The purpose of the present invention is to accurately identify an appearance position of a desired waveform pattern. A signal feature extraction apparatus extracts a feature peak value of a waveform pattern from an output signal of a feature extraction filter that extracts a waveform pattern of an input signal. A local maximum value detection part 1 stores a time index and a value of the output signal when an absolute value of the output signal is equal to or more than a predetermined threshold value and takes a local maximum value. A feature peak value selection part 2 selects a feature peak value on the basis of a difference between a time interval between the stored time index and a time index of a feature peak value selected immediately before and an average time interval at which the feature peak value can occur, and a difference between a logarithmic amplitude and phase of the output signal corresponding to the stored time index and an average logarithmic amplitude and average phase that the feature peak value can take.