Granted Patent B2

US 12,433,498 · App. 17/784,038 · Granted Oct 7, 2025

Inventors: Dongeek Shin (San Jose, CA); Ivan Poupyrev (Mountain View, CA); Brandon Barbello (Mountain View, CA); Shwetak Patel (Mountain View, CA)

Assignee: Google LLC

A61B5/02438A61B5/1102A61B5/6898A61B5/7267G01S7/414G01S7/415G01S7/417G01S13/34G01S13/583G01S13/88

Abstract

Claims (45)

1. A smartphone, comprising:

a housing;

a display screen housed by the housing;

a radar integrated circuit housed by the housing such that an antenna array of the radar integrated circuit is configured to emit radio waves into an ambient environment of the smartphone in front of the display screen and the radar integrated circuit is configured to receive reflected radio waves from the ambient environment; and

output a raw radar waterfall based on the received reflected radio waves;

one or more processors that are in communication with the display screen and the radar integrated circuit, wherein the one or more processors are configured to:

receive the raw radar waterfall from the radar integrated circuit;

analyze the raw radar waterfall to create a ballistocardiography waveform, wherein the one or more processors being configured to analyze the raw radar waterfall comprises the one or more processors being configured to:

perform a background clutter removal process that removes radar data from the raw radar waterfall attributed to a static object to create a foreground radar waterfall;

determine relative phase differences of foreground chirp frames present within the foreground radar waterfall; and

perform a phase unwrapping process on the determined relative phase differences to create the ballistocardiography waveform; and

output data based on the ballistocardiography waveform.

2. The smartphone of claim 1 , wherein the radar integrated circuit is located behind a top bezel of the housing of the smartphone.

3. The smartphone of claim 1 , wherein the radar integrated circuit emits frequency-modulated continuous-wave radar (FMCW).

4. The smartphone of claim 1 , wherein the data based on the ballistocardiography waveform is output to an application executed by the smartphone.

5. The smartphone of claim 4 , further comprising a non-transitory processor-readable medium on which the application is installed, wherein the application comprises a machine learning component that is trained to detect a heart condition based on the ballistocardiography waveform.

6. The smartphone of claim 1 , wherein the one or more processors being configured to output data based on the ballistocardiography waveform comprises the one or more processors being configured to cause the ballistocardiography waveform to be presented on the display screen of the smartphone.

7. The smartphone of claim 1 , wherein:

the one or more processors are further configured to analyze the ballistocardiography waveform to determine a heartrate; and

the one or more processors being configured to output data based on the ballistocardiography waveform comprises outputting the heartrate.

8. The smartphone of claim 1 , further comprising a proximity sensor, wherein the one or more processors are further configured to determine a front surface of the smartphone through which the antenna array of the radar integrated circuit is pointed to the ambient environment is placed against a body part of a user.

9. A method for performing ballistocardiography, the method comprising:

emitting, by a radar integrated circuit of a mobile device, frequency-modulated continuous-wave (FMCW) radar;

receiving, by the radar integrated of the mobile device, reflected radio waves based on the FMCW radar being reflected off objects to create a raw radar waterfall;

analyzing, by the mobile device, the raw radar waterfall to create a ballistocardiography waveform, wherein analyzing the raw radar waterfall comprises:

performing a background clutter removal process that removes radar data from the raw radar waterfall attributed to a static object to create a foreground radar waterfall;

determining relative phase differences of foreground chirp frames present within the foreground radar waterfall; and

performing a phase unwrapping process on the determined relative phase differences to create the ballistocardiography waveform;

outputting, by the mobile device, data based on the ballistocardiography waveform to a machine-learning application installed on the mobile device;

classifying, using the machine-learning application, the ballistocardiography waveform based on a trained machine-learning model of the machine-learning application; and

outputting, by the mobile device, an indication of a classification based on the classifying the ballistocardiography waveform.

10. The method for performing ballistocardiography of claim 9 , wherein outputting data based on the ballistocardiography waveform comprises outputting the data based on the ballistocardiography waveform to the machine-learning application that was installed by a user on the mobile device.

11. The method for performing ballistocardiography of claim 10 , wherein the machine-learning application comprises a machine-learning model trained to detect a particular heart condition.

12. The method for performing ballistocardiography of claim 9 , wherein outputting data based on the ballistocardiography waveform comprises causing the ballistocardiography waveform to be presented on a display screen of the mobile device.

13. The method for performing ballistocardiography of claim 9 , further comprising:

analyzing the ballistocardiography waveform to determine a heartrate, wherein outputting data based on the ballistocardiography waveform comprises outputting an indication of the determined heartrate.

14. The method for performing ballistocardiography of claim 9 , further comprising determining a surface of the mobile device through which an antenna array of the radar integrated circuit is pointed to the ambient environment is placed against a body part of a user.

15. A non-transitory processor-readable medium comprising processor-readable instructions configured to cause one or more processors to:

cause frequency-modulated continuous-wave (FMCW) radio waves to be emitted;

create a raw radar waterfall based on received reflected radio waves;

analyze the raw radar waterfall to create a ballistocardiography waveform, wherein the processor-readable instructions configured to analyze the raw radar waterfall comprises processor-readable instructions that cause the one or more processors to:

perform a background clutter removal process that removes radar data from the raw radar waterfall attributed to a static object to create a foreground radar waterfall;

determine relative phase differences of foreground chirp frames present within the foreground radar waterfall; and

perform a phase unwrapping process on the determined relative phase differences to create the ballistocardiography waveform; and

output data based on the ballistocardiography waveform.

Assignments (1)

Continuity (1)

References Cited (183)