Heart Rate Measurement via mmWave Radar and Camera Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for measuring heart rate using millimeter-wave radar struggle to accurately distinguish between heartbeat and respiration signals due to interference from higher-order harmonics, often relying on periodicity assumptions that can introduce inaccuracies, especially when respiratory patterns are irregular.
Innovation Solution
A contactless heterogeneous-sensing system employing both high-resolution mmWave radar and lower-resolution cameras to track chest displacement, utilizing Short Time Fourier Transforms and harmonic cancellation techniques to separate respiration and heartbeat signals based on resolution differences between sensors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If millimeter-wave radar is used to measure heart rate, then contactless measurement is achieved, but measurement precision deteriorates due to interference from respiration harmonics
Solution Approach 1:
The patent combines millimeter-wave radar with camera sensors to create a heterogeneous sensing system. The radar provides contactless heart rate measurement while the camera captures chest displacement to characterize respiration patterns. By merging these complementary sensors, the system achieves both contactless operation and improved measurement precision through multi-source data fusion.
Solution Approach 2:
The patent introduces respiration signal characterization as an intermediary step between raw radar signals and final heart rate measurement. By first characterizing respiration patterns using camera data and then using this characterization to filter radar signals, the system eliminates respiration interference and achieves accurate heart rate measurement.
2Device complexity
If periodicity assumptions are used to distinguish heartbeat and respiration, then signal separation is simplified, but measurement precision deteriorates when respiratory patterns are irregular
Solution Approach 1:
The patent implements a feedback mechanism where camera-based respiration characterization continuously informs the radar signal processing. The system monitors actual respiration patterns and adjusts the filtering parameters accordingly, allowing it to adapt to irregular respiratory patterns without relying on fixed periodicity assumptions.
Solution Approach 2:
The patent dynamically changes processing parameters based on detected respiration characteristics. Instead of using fixed periodicity assumptions, the system adjusts filtering parameters in real-time according to the actual respiration pattern detected by the camera, enabling accurate heart rate measurement regardless of respiratory regularity.
3Quantity of substance
If higher-order harmonics of respiration are present in radar signals, then respiration interference increases, but no effective filtering method is available
Solution Approach 1:
The patent extracts and removes respiration-related signal components from the radar data using camera-based characterization. By identifying and isolating the respiration signal patterns, the system can selectively remove these components and their harmonics, leaving only the heart rate signal for accurate measurement.
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 provides more accurate heart rate measurements by effectively removing signal artifacts associated with respiration, improving the reliability of heart rate detection even with irregular breathing patterns.
Implementation Method 1
A millimeter-wave (mmWave) radar sensor transmits pulses of electromagnetic waves to a target and detects signals reflected by the target via an antenna
Implementation Method 2
utilizing Short Time Fourier Transforms and harmonic cancellation techniques to separate respiration and heartbeat signals
Data Source
AI summary
Systems, apparatus, and methods for measuring heart rate are disclosed. An example system includes a transmitter to emit electromagnetic waves; a first sensor to output signals representative of the electromagnetic waves reflected by a subject; a second sensor to generate image data, the image data including data corresponding to a chest of the subject; machine readable instructions; and processor circuitry to at least one of instantiate or execute the machine readable instructions to generate heartbeat data by cancelling harmonics associated with respiration by the subject from data corresponding to the output signals of the first sensor based on the image data, and determine a heart rate for the subject based on the heartbeat data.


