MEMS Ballistocardiogram Sensor Heartbeat Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for monitoring the physical and mental state of vehicle occupants using ballistocardiogram (BKG) signals are inefficient due to high noise levels and require complex setups with multiple sensors, leading to delayed and unreliable results.
Innovation Solution
Employing a MEMS-based BKG sensor with an optimal filter that performs cross-correlation with varied heartbeat patterns to identify peak amplitudes, and using an adaptive window function to refine peak localization, reducing computational power and eliminating noise.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a BKG sensor with strain gauges is used to detect heartbeat signals, then the sensor can detect ballistocardiogram signals, but the noise figure becomes very high due to low vibration amplitudes
Solution Approach 1:
The patent combines multiple BKG sensors into a sensor array that collectively detects heartbeat signals. By merging the output signals from multiple sensors, the system achieves better signal-to-noise ratio through signal integration, where the coherent heartbeat signals add up constructively while random noise components tend to cancel out.
Solution Approach 2:
The patent introduces an intermediary processing stage that includes adaptive filtering and signal enhancement algorithms. These intermediary processing steps act as mediators between the raw sensor signals and the final heartbeat detection, selectively amplifying the heartbeat frequency components while attenuating noise components.
2Reliability
If multiple BKG sensors are used to improve signal detection, then measurement reliability improves, but device complexity increases considerably
Solution Approach 1:
The patent makes the BKG sensors multi-functional by designing them to serve both as vibration detectors and as part of a distributed measurement network. The same sensor elements used for detecting seat vibrations are also utilized for heartbeat detection, eliminating the need for separate dedicated sensors and reducing overall system complexity.
Solution Approach 2:
The patent implements self-service through adaptive signal processing where the system automatically adjusts its own parameters based on the detected signal characteristics. The adaptive filters and peak detection algorithms automatically tune themselves to the prevailing conditions, eliminating the need for manual calibration and complex external control systems.
3Measurement precision
If a large database of heartbeat patterns is searched to identify heartbeat signals, then detection accuracy improves, but computation time increases significantly
Solution Approach 1:
The patent performs preliminary action by pre-processing the sensor signals to extract and isolate the heartbeat frequency components before pattern matching. Through bandpass filtering and signal conditioning performed in advance, the system prepares the data in a form that requires minimal computational effort for subsequent pattern recognition, avoiding the need to search through entire raw signal databases.
Solution Approach 2:
The patent applies parameter changes by transforming the heartbeat detection problem from a time-domain pattern matching task to a frequency-domain analysis task. By changing the analysis parameters from temporal patterns to spectral characteristics, the system achieves accurate heartbeat detection through simpler frequency analysis rather than complex temporal pattern database searches.
4Measurement precision
If contact-based EKG or EEG sensors are used to monitor vehicle occupants, then physiological state detection is possible, but the measurement becomes complex and error-prone due to skin resistance
Solution Approach 1:
The patent replaces the electrical contact-based measurement system (EKG/EEG sensors requiring skin contact) with a mechanical vibration-based measurement system (BKG sensors). Instead of measuring electrical potentials through skin contact, the system mechanically detects the vibrations produced by heartbeat through the seat structure, eliminating all complexities related to skin resistance, electrode placement, and contact quality.
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
Enables fast, reliable, and non-intrusive monitoring of vehicle occupants' physical and mental states, allowing for timely warnings and countermeasures, such as alerting drivers to fatigue or stress, without the need for extensive database searches or multiple sensors.
Implementation Method 1
a spring-mass system that records accelerations either by means of changes in capacitance or piezoresistive
Implementation Method 2
a spring-mass system that records accelerations either by means of changes in capacitance or piezoresistive
Implementation Method 3
performs a cross-correlation (convolution) of the BKG signal with heartbeat parameters
Implementation Method 4
The BKG signal can be subjected to an adaptive window function, which is dependent on the length of the heartbeat pattern
Data Source
Figure 1~6
Figure 7
AI summary
The invention relates to a method and an apparatus for recognising the physical and/or psychic condition of a vehicle occupant on the basis of a BCG signal that is obtained by means of a BCG sensor. According to the invention, the BCG sensor is an MEM sensor; an optimum filter (2) is used to perform cross-correlation for the BCG signal with heartbeat parameters that are varied within prescribed limits in order to find a maximum for the cross-correlation function; and probable peaks are located (9) in a cross-correlation function found in this manner, and the heart rate is computed (11) therefrom.