Biological State Estimation via Frequency Slope Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current techniques for monitoring a driver's biological state during driving, such as those using time-series waveforms of pulse waves and air pressure fluctuations, lack precision in distinguishing between different states like fatigue and sleep, and do not effectively combine multiple methods for accurate human state estimation.
Innovation Solution
A biological body state estimation device that analyzes a time-series waveform of a biological signal from the upper body, specifically identifying frequency components at 0.0017 Hz, 0.0035 Hz, and 0.0053 Hz to determine the state of a human, using frequency slope analysis and power spectrum changes to differentiate between fatigue, activity, and functional adjustment signals, thereby improving the accuracy of state estimation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional pulse wave analysis methods are used, then the monitoring process is simple, but the precision of state distinction between fatigue and sleep is insufficient
Solution Approach 1:
The patent segments the biological signal analysis into multiple frequency bands (0.0017 Hz, 0.0035 Hz, and 0.0053 Hz) and analyzes each band separately to distinguish different states. By dividing the frequency spectrum into specific segments, the system achieves more precise state distinction without requiring complete signal reprocessing, thus improving measurement precision while managing complexity through structured analysis.
Solution Approach 2:
The patent introduces frequency domain analysis as an additional dimension to traditional time-domain pulse wave monitoring. By transforming the signal from time-domain to frequency-domain and analyzing power spectrum changes across multiple frequency components, the system gains new insights for state distinction. This dimensional transformation enables precise differentiation between fatigue and sleep states while maintaining computational efficiency through targeted frequency band analysis.
2Measurement precision
If single-method analysis is used, then the processing is efficient, but the accuracy of human state estimation is insufficient
Solution Approach 1:
The patent merges multiple analysis methods by combining frequency spectrum analysis with power spectrum changes and phase relationship detection. Instead of using a single method, the system integrates multiple analytical approaches to examine different aspects of the biological signal simultaneously. This combination achieves higher state estimation accuracy while maintaining processing efficiency through parallel analysis of multiple frequency components and their relationships.
Solution Approach 2:
The patent creates a multi-functional analysis system that can distinguish between multiple states (fatigue, sleep, and other conditions) using a unified framework. The same frequency band analysis and power spectrum monitoring mechanism serves multiple purposes: detecting fatigue levels, identifying sleep onset, and differentiating between various physiological states. This universality improves accuracy without requiring separate specialized systems for each state detection.
3Measurement precision
If frequency components are analyzed in detail, then the state differentiation is improved, but the calculation time increases
Solution Approach 1:
The patent extracts and focuses analysis on specific frequency bands (0.0017 Hz, 0.0035 Hz, and 0.0053 Hz) that are most indicative of sleep and fatigue states. Instead of analyzing the entire frequency spectrum, the system selectively processes only these critical frequency components, reducing computational load while maintaining high differentiation precision. This extraction approach eliminates unnecessary calculations and speeds up processing without sacrificing state detection accuracy.
Solution Approach 2:
The patent applies partial action by analyzing only the necessary frequency components and power spectrum changes required for state differentiation, rather than performing complete signal processing. The system processes sufficient frequency data to achieve accurate state distinction while avoiding excessive computation on irrelevant frequency bands. This partial analysis strategy balances precision requirements with computational efficiency, reducing calculation time while maintaining adequate state differentiation capability.
Data Source
AI summary
A device to detect a state of a human being is provided. The device determines a time-series waveform of a frequency from a time-series waveform of a biological signal acquired from an upper body of a human being, and determines time-series waveforms of a frequency slope and frequency variation to analyze these frequencies. Upon analyzing the frequencies, it determines power spectrums of frequencies corresponding to a preset functional adjustment signal, fatigue reception signal, and activity adjustment signal, and then determines a state of the human being from a time-series change in each power spectrum. Dominant degrees of the functional adjustment signal and activity adjustment signal are compared as distribution rates thereof, in addition to a degree of progress of fatigue, to determine a state of the human being, for example a relaxed state, fatigued state, prominent state of sympathetic nervous, prominent state of parasympathetic nervous.