Optical Sensor Array Biometric Noise Filtering via Singular Value Decomposition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing detection devices struggle to accurately acquire biometric information due to noise interference from body motion, unintended biosignals, and commercial power frequency noise.
Innovation Solution
A detection device comprising a plurality of optical sensors, a light source, an analog front-end circuit, and a signal processing circuit that converts time-domain data into a matrix for singular value decomposition, allowing for the removal of noise components and the acquisition of biometric information as image data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional optical detection is used to acquire biometric information, then the detection process is simple, but noise from body motion, unintended biosignals, and commercial power frequency interference degrades measurement precision
Solution Approach 1:
The patent transforms time-domain detection data into frequency-domain representation through spectral analysis. By converting the detection time series data into frequency components, the system can identify and separate noise signals (such as 50Hz/60Hz commercial power frequency) from the desired biometric signals based on their frequency characteristics, thereby improving measurement precision despite noise interference
Solution Approach 2:
The patent extracts specific frequency components from the composite signal using spectral analysis. By identifying the frequency range where biometric information (pulse waves, blood flow) is concentrated and separating it from other frequency components containing noise, the system isolates the useful signal from harmful interference, enhancing measurement accuracy
2Measurement precision
If multiple optical sensors are used to improve detection accuracy, then measurement precision improves, but device complexity increases
Solution Approach 1:
The patent divides the detection area into multiple regions with optical sensors arranged in a matrix configuration. Each sensor detects light intensity independently, and the system processes signals from multiple sensors to acquire comprehensive biometric information. This segmentation allows parallel detection across different spatial locations, improving measurement precision while maintaining manageable device complexity through modular sensor arrangement
Solution Approach 2:
The patent employs multiple optical sensors that can detect various types of biometric information (pulse waves, blood flow, oxygen saturation) simultaneously using the same hardware platform. By processing detection values from multiple sensors through spectral analysis, the system achieves multi-functional capability without proportionally increasing device complexity, as the same sensor array serves multiple detection purposes
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
The solution effectively filters out noise and enhances the accuracy of biometric information acquisition, such as pulse waves and blood oxygen saturation levels, by transforming and processing time-domain data using singular value decomposition.
Implementation Method 1
a plurality of photoelectric conversion elements such as photodiodes are arranged on a semiconductor substrate
Data Source
AI summary
According to an aspect, a detection device includes: a plurality of optical sensors arranged in a detection area; a light source configured to emit light to the optical sensors; an analog front-end (AFE) circuit configured to acquire a detection value of each of the optical sensors; and a signal processing circuit configured to acquire predefined biometric information based on first time-domain data obtained by acquiring the detection values in chronological order. The signal processing circuit is configured to: convert the first time-domain data into a time-domain matrix and perform singular value decomposition on the time-domain matrix, and inversely calculate second time-domain data based on a predetermined singular value among a plurality of singular values obtained as a result of the singular value decomposition; and acquire the biometric information that changes in chronological order as image information using the second time-domain data.