Radar-Based Non-Contact Heart Rate Estimation With Multi-Band Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing heart rate estimation methods lack robustness and accuracy, particularly in medical and emergency settings, necessitating improved non-contact heart rate estimation techniques.
Innovation Solution
A system and method utilizing radar range history filtering with multiple bandpass filters to form multiple heart rate estimates, including spectrogram analysis and peak counting, followed by a fusion and cross-validation process to enhance accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If existing heart rate estimation methods are used, then the process is simple, but the robustness and accuracy are insufficient
Solution Approach 1:
The patent segments the heart rate estimation process into multiple independent filtering paths: a first path processes high-frequency components (20-200 Hz) to capture cardiac signals, while a second path processes low-frequency components (0.2-5 Hz) to capture respiratory signals. Each path applies specialized filters and analysis methods, allowing the system to optimize for different physiological signals independently, thereby improving overall measurement precision without creating a monolithic complex system
Solution Approach 2:
The patent transforms the traditional single-dimension heart rate estimation into a multi-dimensional approach by simultaneously analyzing both high-frequency cardiac components and low-frequency respiratory components. This dimensional expansion allows the system to extract heart rate information from multiple signal characteristics, improving robustness and accuracy by leveraging additional physiological dimensions rather than relying on a single estimation method
2Reliability
If multiple filtering paths are used, then the accuracy is improved, but the computational complexity increases
Solution Approach 1:
The patent divides the signal processing into segmented frequency bands with dedicated filters: high-pass filter (0.5-2 Hz), band-pass filter (2-15 Hz), and another band-pass filter (15-30 Hz). Each segment targets specific physiological signal components, allowing parallel processing of different frequency ranges. This segmentation enables the system to maintain high reliability through multiple analysis paths while managing computational complexity by processing each segment independently with optimized algorithms
Solution Approach 2:
The patent applies partial action by selectively applying different filtering and analysis methods to different frequency components rather than processing the entire signal uniformly. The high-frequency path (20-200 Hz) focuses specifically on cardiac signals with appropriate filters, while the low-frequency path (0.2-5 Hz) handles respiratory signals separately. This partial processing approach improves reliability for each specific physiological signal type without requiring excessive computational resources to analyze all frequency components with all methods
3Measurement precision
If bandpass filtering with 20 Hz lower edge is applied, then cardiac signals are enhanced, but low-frequency components are lost
Solution Approach 1:
The patent segments the frequency spectrum into distinct processing paths: one path (first filter with 20 Hz lower edge) is dedicated to capturing high-frequency cardiac signals with enhanced precision, while another path (second filter with 0.2-5 Hz passband) is dedicated to capturing low-frequency respiratory signals. By segmenting the processing, the system enhances cardiac signal detection accuracy in the first path without losing low-frequency information, as the second path preserves and analyzes these components separately
Solution Approach 2:
The patent compensates for low-frequency loss in the cardiac signal path by adding another dimension of analysis through the respiratory signal path. The system processes low-frequency components (0.2-5 Hz) separately to capture respiratory information, thereby recovering the information that would be lost if only high-frequency cardiac filtering were applied. This multi-dimensional approach ensures both cardiac and respiratory signals are captured with appropriate frequency characteristics
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
Enhances heart rate estimation accuracy and robustness by combining multiple estimates, providing reliable heart rate data even in challenging environments.
Implementation Method 1
A radar system may be used to illuminate a chest of the subject with radar radiation
Data Source
AI summary
A system and method for non-contact heart rate estimation. In some embodiments, the method includes filtering a radar range history with a first filter, to form a first filtered radar range history; and forming a first estimate of the heart rate of a subject based on the first filtered radar range history, the first filter having a passband including a frequency greater than 20 Hs.