Granted Patent B2

US 12,324,647 · App. 17/960,080 · Granted Jun 10, 2025

Inventors: Beibei Wang (Clarksville, MD); Sakila Sandeepani Jayaweera Samaranayake Arachchige Dona (College Park, MD); Xiaolu Zeng (Beijing, CN); Wei-Hsiang Wang (College Park, MD); K. J. Ray Liu (Potomac, MD); Oscar Chi-Lim Au (Greenbelt, MD)

Assignee: ORIGIN RESEARCH WIRELESS, INC.

A61B5/0002H04B17/309

Abstract

Claims (102)

1. A system for wireless vital sign monitoring, comprising:

a transmitter configured to transmit a wireless signal through a wireless channel of a venue;

a receiver configured to receive the wireless signal through the wireless channel, wherein the received wireless signal differs from the transmitted wireless signal due to the wireless channel that is impacted by a periodic motion of a vital sign of an object in the venue; and

a processor configured for:

obtaining a time series of channel information (CI) of the wireless channel based on the received wireless signal, wherein each CI is a vector associated with a time stamp with N components,

computing a time series of denoised CI (DCI) by denoising the time series of CI (TSCI), wherein each DCI is a vector associated with a time stamp with N components,

computing a time series of concatenated CI (TSCCI) based on a concatenation of the TSCI and the time series of DCI (TSDCI), wherein:

each concatenated CI (CCI) associated with a respective time stamp is a vector comprising 2N components and is computed by concatenating a respective CI associated with the same respective time stamp and a respective DCI associated with the same respective time stamp corresponding to the respective CI,

a first set of N components among the 2N components are associated with the respective CI, each of the N components in the first set being associated with a respective component of the respective CI,

a second set of N components among the 2N components are associated with the respective DCI, each of the N components in the second set being associated with a respective component of the respective DCI,

computing a one dimensional (1D) characteristic function of CCI of the TSCCI in a sliding time window,

computing a two dimensional (2D) decomposition of the TSCCI based on the 1D characteristic function in the sliding time window,

enhancing the 2D decomposition based on a vertical highpass horizontal lowpass filtering, and

monitoring the periodic motion of the vital sign based on the enhanced 2D decomposition.

2. The system of claim 1 , wherein the processor is further configured for:

computing each CCI as a concatenation of a first basic feature of the respective CI and a second basic feature of the corresponding DCI,

wherein the first basic feature of any CI comprises at least one of: the CI, a magnitude of the CI, a phase of the CI, a magnitude of a component of the CI, a phase of the component of the CI, or a function of any magnitude or phase of the CI,

wherein the second basic feature of any DCI comprises at least one of: the DCI, a magnitude of the DCI, a phase of the DCI, a magnitude of a component of the DCI, a phase of the component of the DCI, or a function of any magnitude or phase of the DCI,

wherein the TSDCI is computed by processing the TSCI with a first denoising filter, the first denoising filter comprising at least one of: a low-pass filter, a bandpass filter, a highpass filter, a median filter, or a percentile filter.

3. The system of claim 2 , wherein the processor is further configured for:

determining a series of sliding time windows, each sliding time window being associated with a time stamp;

computing a respective one dimensional (1D) characteristic function of the CCI of the TSCCI in each sliding time window; and

constructing the 2D decomposition based on the respective 1D characteristic function in each sliding time window.

4. The system of claim 3 , wherein:

the respective 1D characteristic function is computed based on a transformation of the CCI of the TSCCI in each sliding time window; and

the transformation comprises at least one of:

a frequency transformation, a transformation to frequency domain, a short-time Fourier transform (STFT), a wavelet transform, a filter-bank representation,

a time transformation, a transformation to a time-shift domain, an auto-correlation function (ACF), an auto-covariance function, or a time-frequency decomposition.

5. The system of claim 4 , wherein the processor is further configured for:

constructing a 2D matrix by lining up and aligning the 1D characteristic functions of the sliding time windows in time order according to the associated time stamps, wherein

i-th column of the 2D matrix represents the 1D characteristic function associated with i-th sliding time window,

j-th row of the 2D matrix is associated with a time-shift of j time units,

a matrix element at the i-th column and the j-th row of the 2D matrix is a value of the 1D characteristic function in the i-th sliding time window with a time-shift of j time units, and

the 2D matrix is a result of the 2D decomposition.

6. The system of claim 5 , wherein the processor is further configured for:

enhancing the 2D decomposition by applying, to each matrix element of the 2D matrix, the vertical highpass horizontal lowpass filtering.

7. The system of claim 6 , wherein the processor is further configured for:

filtering each matrix element of the 2D matrix with a respective 2D vertical highpass horizontal lowpass filter that is either separable or non-separable, either linear or nonlinear.

8. The system of claim 7 , wherein the processor is further configured for:

enhancing the 2D decomposition by filtering each matrix element of the 2D matrix with at least one of: a respective vertical highpass filter or a respective horizontal lowpass filter,

wherein each of the respective vertical highpass filter and the respective horizontal lowpass filter is either linear or nonlinear.

9. The system of claim 8 , wherein the processor is further configured for enhancing the 2D decomposition by at least one of:

filtering part of each column of the 2D matrix with a respective first common vertical highpass filter;

filtering part of each row of the 2D matrix with a respective second common vertical highpass filter;

filtering part of each column of the 2D matrix with a respective first common horizontal lowpass filter; or

filtering part of each row of the 2D matrix with a respective second common horizontal lowpass filter.

10. The system of claim 9 , wherein:

each vertical highpass filter is a directional 1D highpass filter applied in a vertical direction or in a near-vertical direction.

11. The system of claim 10 , wherein:

each horizontal lowpass filter is a directional 1D lowpass filter applied in a horizontal direction or in a near-horizontal direction.

12. The system of claim 11 , wherein the processor is further configured for:

filtering each matrix element of the 2D matrix with a succession of at least two 1D filters, each being either the vertical highpass filter or the horizontal lowpass filter.

13. The system of claim 12 , wherein the processor is further configured for:

enhancing the 2D decomposition by applying histogram equalization to a number of consecutive columns of the 2D matrix to generate an enhanced 2D matrix.

14. The system of claim 13 , wherein the processor is further configured for:

computing a distance score based on dynamic time warping (DTW) between a reference column and a particular column of the enhanced 2D matrix associated with a particular time, wherein:

the reference column is computed in a training session by clustering a number of training columns matched based on DTW,

the training columns are computed based on training TSCI obtained from training wireless signals that are received through the wireless channel when training objects with vital signs are present and impacting the wireless channel, and

when the distance score is larger than a first threshold, the periodic motion of the vital sign of the object is tentatively not detected at the particular time.

15. The system of claim 14 , wherein the processor is further configured for:

subtracting a quantity from each matrix element in the particular column of the enhanced 2D matrix; and

computing a zero-crossing count of the particular column,

wherein when the zero-crossing count is larger than a second threshold, the periodic motion of the vital sign of the object is tentatively not detected at the particular time.

16. The system of claim 15 , wherein the processor is further configured for:

computing multiple peaks of the particular column of the enhanced 2D matrix; and

computing an inter-peak distance based on the multiple peaks.

17. The system of claim 16 , wherein:

when the inter-peak distance is in an acceptable range, the periodic motion of the vital sign of the object is tentatively detected at the particular time, and the inter-peak distance is determined to be equal to a frequency or a period of the vital sign; and

when the inter-peak distance is not in an acceptable range, the periodic motion of the vital sign of the object is tentatively not detected at the particular time.

18. The system of claim 17 , wherein the processor is further configured for:

detecting the vital sign of the object when P is larger than a third threshold, wherein P is a percentage of the periodic motion of the vital sign being tentatively detected in a period of time.

19. A wireless device of a system for wireless vital sign monitoring, comprising:

a processor;

a memory communicatively coupled to the processor; and

a receiver communicatively coupled to the processor, wherein:

an additional wireless device in the system is configured to transmit a wireless signal through a wireless channel of a venue,

the receiver is configured to receive the wireless signal through the wireless channel,

the received wireless signal differs from the transmitted wireless signal due to the wireless channel that is impacted by a periodic motion of a vital sign of an object in the venue, and

the processor is configured for:

obtaining a time series of channel information (CI) of the wireless channel based on the received wireless signal, wherein each CI is a vector associated with a time stamp with N components,

computing a time series of denoised CI (DCI) by denoising the time series of CI (TSCI), wherein each DCI is a vector associated with a time stamp with N components,

computing a time series of concatenated CI (TSCCI) based on a concatenation of the TSCI and the time series of DCI (TSDCI), wherein:

each concatenated CI (CCI) associated with a respective time stamp is a vector comprising 2N components and is computed by concatenating a respective CI associated with the same respective time stamp and a respective DCI associated with the same respective time stamp corresponding to the respective CI,

a first set of N components among the 2N components are associated with the respective CI, each of the N components in the first set being associated with a respective component of the respective CI,

a second set of N components among the 2N components are associated with the respective DCI, each of the N components in the second set being associated with a respective component of the respective DCI,

computing a one dimensional (1D) characteristic function of CCI of the TSCCI in a sliding time window,

computing a two dimensional (2D) decomposition of the TSCCI based on the 1D characteristic function in the sliding time window,

enhancing the 2D decomposition by applying a vertical highpass filter followed by a histogram equalization, and

monitoring the periodic motion of the vital sign based on the enhanced 2D decomposition.

20. A method for wireless vital sign monitoring, comprising:

transmitting a wireless signal through a wireless channel of a venue;

receiving the wireless signal through the wireless channel, wherein the received wireless signal differs from the transmitted wireless signal due to the wireless channel that is impacted by a periodic motion of a vital sign of an object in the venue;

obtaining a time series of channel information (CI) of the wireless channel based on the received wireless signal, wherein each CI is a vector associated with a time stamp with N components;

computing a time series of denoised CI (DCI) by denoising the time series of CI (TSCI), wherein each DCI is a vector associated with a time stamp with N components;

computing a time series of concatenated CI (TSCCI) based on a concatenation of the TSCI and the time series of DCI (TSDCI), wherein:

each concatenated CI (CCI) associated with a respective time stamp is a vector comprising 2N components and is computed by concatenating a respective CI associated with the same respective time stamp and a respective DCI associated with the same respective time stamp corresponding to the respective CI,

a first set of N components among the 2N components are associated with the respective CI, each of the N components in the first set being associated with a respective component of the respective CI,

a second set of N components among the 2N components are associated with the respective DCI, each of the N components in the second set being associated with a respective component of the respective DCI;

computing a one dimensional (1D) characteristic function of CCI of the TSCCI in a sliding time window;

computing a two dimensional (2D) decomposition of the TSCCI based on the 1D characteristic function in the sliding time window;

enhancing the 2D decomposition based on a vertical highpass filter and a histogram equalization; and

monitoring the periodic motion of the vital sign based on the enhanced 2D decomposition.

Assignments (1)

Continuity (40)

References Cited (5)