Biometric Feature Extraction via Waveform Decomposition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for non-invasive biometric information detection, such as blood pressure monitoring, face challenges in accurately extracting features from bio-signals like photoplethysmography (PPG) signals, especially in mobile healthcare settings where signal quality can be affected by noise and variability.
Innovation Solution
A feature extraction apparatus and method that decomposes bio-signals into component pulses using a waveform decomposer, which models the component pulse waveform function based on a Gaussian function and an asymmetry factor, allowing for the extraction of features like time, amplitude, and standard deviation from these pulses for biometric information detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If bio-signals are directly processed without decomposition, then the processing is simple and fast, but the measurement precision and reliability of biometric information detection deteriorates due to noise and signal variability
Solution Approach 1:
The patent applies segmentation by decomposing the complex bio-signal waveform into multiple component pulses (propagation wave and reflection waves). The waveform decomposer separates the PPG signal into distinct physiological wave components, allowing precise extraction of features from each component. This segmentation enables accurate blood pressure detection by analyzing characteristic points of individual pulses rather than processing the entire complex waveform at once.
2Measurement precision
If noise filtering is applied to improve signal quality, then the measurement precision improves, but the processing time and computational resources increase
Solution Approach 1:
The patent implements preliminary action by performing waveform decomposition into component pulses as the first step in signal processing. By pre-separating the bio-signal into propagation and reflection waves before feature extraction, the system establishes a structured foundation that facilitates efficient noise filtering. The decomposition itself acts as a form of preliminary signal organization that reduces subsequent processing complexity and enables faster extraction of relevant features from cleaned signal components.
3Reliability
If feature extraction is performed on the full bio-signal waveform, then all information is captured, but the reliability of specific biometric parameters deteriorates due to overlapping signal components
Solution Approach 1:
The patent applies the extraction principle by isolating specific component pulses from the full bio-signal waveform. The waveform decomposer extracts individual propagation waves and reflection waves as separate entities. This allows the feature extractor to obtain characteristic points (amplitude, time, shape features) from specific pulse components rather than analyzing the entire overlapping waveform. By taking out and analyzing individual pulse components separately, the system achieves reliable blood pressure, vascular age, and arterial stiffness detection without the interference of overlapping signal elements.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach enables accurate and reliable detection of biometric information, including blood pressure, vascular age, and arterial stiffness, by effectively filtering noise and improving signal processing in mobile healthcare environments.
Implementation Method 1
a measurer configured to emit light onto a user's skin, detect the light reflecting from the user's skin, and measure a bio-signal based on the detected light
Data Source
AI summary
A feature extraction apparatus configured to perform biometric information detection includes a bio-signal obtainer configured to acquire a bio-signal; and a processor configured to decompose a waveform of the acquired bio-signal into component pulses and extract a feature for the biometric information detection based on characteristic points of the component pulses.