Biological State Estimation via Frequency Band Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for monitoring a driver's biological state during driving, such as those using pulse waves and air pressure fluctuations, lack accuracy in distinguishing between different states like fatigue, sleep prediction, and autonomic nerve system activity, necessitating a more precise technique for state estimation.
Innovation Solution
A biological body state estimation device that analyzes time-series waveforms of biological signals from the upper body, employing frequency calculations, slope analysis, and power spectrum analysis to identify specific frequency bands indicative of fatigue, activity, and functional adjustment signals, allowing for more accurate determination of a person's state by analyzing zero-crossing points, peak detection, and frequency fluctuations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If pulse wave analysis or air pressure fluctuation monitoring is used to monitor driver state, then the monitoring function is provided, but the accuracy in distinguishing between different states (fatigue, sleep prediction, autonomic nerve system activity) is insufficient
Solution Approach 1:
The patent segments the frequency spectrum into distinct bands (ultra-low frequency 0.002-0.005 Hz, low frequency 0.005-0.03 Hz, high frequency 0.03-0.5 Hz) and analyzes each band separately to identify specific physiological states. This segmentation allows precise differentiation between fatigue, sleep prediction, and autonomic nerve system activity by examining characteristic patterns in each frequency range independently.
Solution Approach 2:
The patent transforms the analysis from time-domain signal processing to frequency-domain analysis by applying Fourier transform and examining power spectral density across multiple frequency dimensions. This dimensional transformation enables simultaneous monitoring of multiple physiological parameters (heart rate variability, respiratory rate, autonomic tone) that are otherwise difficult to distinguish in the time domain.
2Measurement precision
If comprehensive frequency analysis including ultra-low frequency bands is performed, then state estimation accuracy is improved, but computational load and processing time increase
Solution Approach 1:
The patent applies preliminary filtering and pre-processing steps to the biological signal before performing comprehensive frequency analysis. By pre-processing the signal to remove artifacts and normalize data, the system reduces computational requirements for the subsequent ultra-low frequency analysis, enabling accurate state estimation without excessive processing time.
Solution Approach 2:
The patent dynamically adjusts analysis parameters such as window size, frequency resolution, and threshold values based on the current driving conditions and signal quality. This adaptive parameter adjustment optimizes the balance between analysis accuracy and processing speed, allowing the system to maintain high precision in state estimation while minimizing computational overhead during real-time monitoring.
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
A technology to grasp a state of a human being more accurately is provided. The technology is provided with means for acquiring a time-series waveform of a frequency from a time-series waveform of a biological signal sampled from the upper body of a human being and for further acquiring a time-series waveform of frequency slope and a time-series waveform of frequency fluctuation and for applying frequency analysis to them. In the frequency analysis, a power spectrum of each frequency corresponding to a functional adjustment signal, a fatigue reception signal, and an activity adjustment signal, respectively, determined in advance is acquired. Then, a state of a human being is determined from a time-series change of each power spectrum. The fatigue reception signal indicates a degree of progress of fatigue in a usual active state and thus, by comparing it with degrees of predominance of the functional adjustment signal and the activity adjustment signal as their distribution rates, a state of a human being (relaxed state, fatigued state, state in which sympathetic nerve is predominant, a state in which parasympathetic nerve is predominant and the like) can be determined more accurately.