Biological information detection device and biological information detection method

Abstract

To provide a biological information detection device capable of accurately detecting biological information by using an electric wave.SOLUTION: A biological information detection device includes: an array antenna including a plurality of transmission antenna elements 21 and a plurality of reception antenna elements 22; a synthetic output acquisition unit (S12) that acquires an array synthetic output represented by a complex signal form on the basis of a reception complex signal corresponding to a reception radio signal from each reception antenna element 22 in a status where a radio signal of a predetermined frequency is transmitted from each transmission antenna element 21 to a living body being a detection target of biological information; an output adjustment part (S13 to S14) that adjusts the array synthetic output such that a phase and an amplitude represented by a signal point on a complex plane of the array synthetic output are in a predetermined state based on a predetermined amplitude value and a predetermined phase fluctuation on the complex plane; and a biological information extraction unit (S15 to S17) that extracts target biological information from the array synthetic output that has been adjusted.SELECTED DRAWING: Figure 13

Description

[Technical field]

[0001] The present invention relates to a biological information detection device and a biological information detection method for detecting biological information based on periodic body movements such as heartbeat and respiratory movement of a living body (human or animal). [Background technology]

[0002] For the purpose of monitoring human health, methods of detecting vital signs using wearable devices and cameras have been studied (see Non-Patent Documents 1 and 2). The methods described in these non-patent documents impose a large physical and mental burden due to forgetting to wear the device or wearing it for a long time. The method described in Non-Patent Document 2 has problems of privacy issues and weakness in blind spots and dark places. Therefore, in order to solve these problems, vital sign detection using radio waves has been studied (see Non-Patent Documents 3, 4, and 5). This is a method of irradiating a living body with radio waves and extracting vital signs from the Doppler shift contained in the reflected waves. The amplitude and phase of the reflected waves derived from the vital signs change due to fluctuations on the surface of the living body such as breathing and heartbeat, and the Doppler shift contains time fluctuations and frequency fluctuations. However, in complex radio wave propagation environments such as indoors, it becomes difficult to accurately detect vital signs due to reflected waves and noise other than those of the living body. Therefore, an arctangent demodulation method has been proposed to improve the accuracy of vital sign detection (see Non-Patent Documents 5 and 6). This arctangent demodulation is a method for detecting vital signs by observing the phase angle of arc-shaped signal points caused by body surface displacement, and has the advantage of being able to reduce harmonics.

[0003] In addition, research is being conducted on increasing the number of degrees of freedom of the propagation path using MIMO (Multiple-Input Multiple-Output) to dramatically improve the vital sign detection sensitivity (see Non-Patent Documents 7, 8, and 9). By using beamforming to control the radiation directivity of the array antenna, it is possible to emphasize the vital sign and reduce clutter. However, even during quiet breathing, the direction and distance of the object to be observed change due to unconscious body movements, so beam tracking that continuously emphasizes the vital sign is difficult. In addition, if the arc-shaped response waveform caused by the body surface displacement cannot be observed, the arctangent demodulation, which requires estimation of the center coordinate of the arc, does not function, making it difficult to remove the DC offset, and the vital sign cannot be detected accurately. Such a problem may also occur in the biosignal processing device described in Patent Document 1. Therefore, in order to solve this problem, pseudo arctangent demodulation (see Non-Patent Document 10) and CMA-like (Constant Modulus Algorithm) adaptive array algorithm (see Non-Patent Document 11), which do not require estimation of the center coordinate of the arc, have been proposed. The CMA-like method is a method that sequentially determines weights to suppress clutter based on the fact that the vital sign trajectory becomes linear. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent No. 7217011 [Non-patent literature]

[0005] [Non-Patent Document 1] Y.-L. Zheng, X.-R. Standing, CCY Poon, BPL Lo, H. Zhang, X.-L. Zhou, G.-Z. Yang, N. Zhao, and Y.-T. Zhang, “Unobtrusive sensing and wearable devices for health informatics,” IEEE Transactions on Biomedical Engineering, vol. 61, no. 5, pp. 1538?1554, May 2014.

Outdoor Tool2

Outdoor Tools3

Outdoor Tools 4

[0006] The above-mentioned CMA-like (Constant Modulus Algorithm) adaptive array algorithm is a method for determining weights sequentially to suppress clutter based on the linearity of vital sign trajectories, but because it does not take into account the characteristics of vital signs, it has the problem of not necessarily obtaining the desired response.

[0007] The present invention has been made in view of the above circumstances, and provides a biological information detection device that can detect biological information (vital signs) of humans or animals with higher accuracy by using radio waves. [Means for solving the problem]

[0008] A biological information detection device according to the present invention is a biological information detection device that detects biological information based on periodic body movements of a living organism, and includes an array antenna including any one of a set of a transmitting antenna element and a plurality of receiving antennas, a set of a plurality of transmitting antenna elements and receiving antenna elements, or a set of a plurality of transmitting antenna elements and a plurality of receiving antenna elements, a combined output acquisition unit that acquires an array combined output expressed in a complex signal format based on a received complex signal corresponding to a received radio wave signal from each receiving antenna element in the array antenna under a condition in which a radio wave signal of a predetermined frequency is transmitted from each transmitting antenna element in the array antenna toward a living organism that is a detection target for biological information, an output adjustment unit that adjusts the array combined output so that the phase and amplitude represented by a signal point on a complex plane of the array combined output acquired by the combined output acquisition unit fall within a predetermined state based on a predetermined amplitude value and a predetermined phase fluctuation width on the complex plane, and a biological information extraction unit that extracts target biological information from the adjusted array combined output obtained by the output adjustment unit.

[0009] With this configuration, under the condition that radio wave signals of a predetermined frequency are transmitted from each transmitting antenna element in the array antenna toward a living body as a detection target of biological information, an array combined output expressed in a complex signal format (data of an in-phase component (I) with a carrier wave and data of a quadrature component (Q)) is acquired based on a received signal from each receiving antenna element in the array antenna. Then, the array combined output is adjusted so that the phase and amplitude represented by the signal point on the complex plane of the array combined output (IQ data) fall within a predetermined state based on a predetermined amplitude value and a predetermined phase fluctuation width on the complex plane. From the array combined output that falls within the predetermined state obtained by the adjustment, the target biological information (for example, information related to heartbeat or respiratory movement) is extracted.

[0010] In the biological information detection device according to the present invention, the composite output acquisition unit can include a transfer function estimation unit that estimates a complex transfer function between a transmitting antenna element and a receiving antenna element for each of a plurality of combinations of each transmitting antenna element and each receiving antenna element in the array antenna, and a composite output generation unit that treats each of the plurality of complex transfer functions obtained by the transfer function estimation unit as a complex signal and generates an array composite output from the plurality of complex signals.

[0011] With this configuration, under the condition that each transmitting antenna element in the array antenna transmits a radio wave signal of a predetermined frequency toward a living body as a detection target for biological information, a complex transfer function between each transmitting antenna element and each receiving antenna element is estimated for each combination of the transmitting antenna element and the receiving antenna element based on a received signal from each receiving antenna element in the array antenna. Then, each of the multiple complex transfer functions obtained by the estimation is treated as a complex signal, and an array synthesis output is generated from the multiple complex signals.

[0012] In the biological information detection device according to the present invention, the output adjustment unit can be configured to adjust a signal point on a complex plane of the array combined output so that the signal point falls within an arc-shaped locus with the predetermined amplitude value and a predetermined phase fluctuation width centered on the origin of the complex plane.

[0013] A complex signal (array combined output) reflected from the surface of a living body has the property of drawing an arc on a complex plane due to fluctuations on the body surface caused by vital signs such as respiration, heart rate, etc. By utilizing this property, the array combined output is adjusted by the above configuration so that the signal point on the complex plane of the obtained array combined output falls within an arc-shaped locus with the predetermined amplitude value and the predetermined phase fluctuation width centered on the origin of the complex plane.

[0014] In the biological information detection device according to the present invention, the output adjustment unit can be configured to include a preprocessing unit that determines a complex weight vector and a complex DC component used to define the array combined output in a complex signal format, and an improved CMA adjustment unit that sequentially adjusts the complex weight vector and the complex DC component so that a signal point represented by the array combined output falls within an arc-shaped locus centered on the origin on a complex plane.

[0015] With this configuration, the complex weight vector and the complex DC component used to define the array synthesis output in the form of a complex signal are obtained. The complex signal reflected from the surface of the living body described above has the property of drawing an arc on a complex plane due to the fluctuation of the body surface caused by vital signs such as breathing and heart rate, and the complex weight vector and the complex DC component are sequentially adjusted so that the signal point represented by the array synthesis output falls within a circular locus centered on the origin on the complex plane. The array synthesis output defined using the complex weight vector and the complex DC component obtained by this adjustment can be obtained as an adjusted array synthesis output.

[0016] In the biological information detection device according to the present invention, the improved CMA adjustment unit can be configured to adjust the complex weight vector and the complex DC component so that a central angle of an arc-shaped locus of the signal points of the array combined output on the complex plane falls within a predetermined range.

[0017] With this configuration, the complex weight vector and the complex DC component are sequentially adjusted so that the central angle of the arcuate locus within which the signal points represented by the array combined output should fall on the complex plane falls within a predetermined range.

[0018] Further, a biological information detection method according to the present invention is a biological information detection method for detecting biological information based on periodic body movement of a living body, and includes a composite output acquisition step of acquiring an array composite output expressed in a complex signal format based on a received complex signal corresponding to the received radio wave signal from each receiving antenna element in the array antenna under a condition in which a radio wave signal of a predetermined frequency is transmitted from each transmitting antenna element in the array antenna toward a living body as a detection target for biological information, an output adjustment step of adjusting the array composite output so that the phase and amplitude represented by a signal point on a complex plane of the array composite output acquired in the composite output acquisition step fall within a predetermined state based on a predetermined amplitude value and a predetermined phase fluctuation width on the complex plane, and a biological information extraction step of extracting target biological information from the adjusted array composite output obtained by the output adjustment step. Effect of the Invention

[0019] According to the biological information detection device and biological information detection method of the present invention, it is possible to detect biological information (vital signs) of a human or animal with higher accuracy by using radio waves. [Brief description of the drawings]

[0020] [Figure 1] FIG. 1 is a conceptual diagram of a vital signs detection system. [Diagram 2] FIG. 2 shows the response of a complex signal containing a DC offset. [Diagram 3] FIG. 3 shows the convergence of the signal response by the proposed method. [Figure 4] FIG. 4 is a diagram showing an experimental scene. [Diagram 5] FIG. 5 is a block diagram showing the experimental system. [Figure 6] FIG. 6 is a conceptual diagram of the heart rate estimation accuracy evaluation method. [Figure 7A] FIG. 7A is a diagram showing an IQ response waveform according to the maximum ratio combining method. [Figure 7B] FIG. 7B is a diagram showing an IQ response waveform according to the CMA-like method. [Figure 7C] FIG. 7C is a diagram showing an IQ response waveform according to the proposed method. [Figure 8A] FIG. 8A is a diagram showing body surface displacement estimated by the maximum ratio combining method. [Figure 8B] FIG. 8B is a diagram showing body surface displacement estimated by the CMA-like method. [Figure 8C] FIG. 8C is a diagram showing the body surface displacement estimated by the proposed method. [Figure 9A] FIG. 9A is a diagram showing a heart rate response estimated by the maximum ratio combining method. [Figure 9B] FIG. 9B is a diagram showing a cardiac response estimated by the CMA-like method. [Figure 9C] FIG. 9C is a diagram showing the heart rate response estimated by the improved CMA method (proposed method). [Figure 10A] FIG. 10A is a diagram showing the time response of the heart rate by the maximum ratio combining method. [Figure 10B] FIG. 10B is a diagram showing the time response of heart rate using the CMA-like method. [Figure 10C] FIG. 10C shows the time response of the heart rate using the improved CMA method (proposed method). [Figure 11] FIG. 11 is a diagram showing the heart rate estimation error CDF by each method (maximum ratio combining, CMA-like method, and proposed method). [Figure 12] FIG. 12 is a block diagram showing a configuration of a biological information detecting device according to an embodiment of the present invention. [Figure 13] FIG. 13 is a flowchart showing a procedure of processing in the processing unit in the biological information detecting device shown in FIG. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0021] First, the contents of the paper published by the inventor of the present application (including some revised parts) will be described below.

[0022] 1. Introduction In recent years, the spread of COVID-19 has progressed, and it is known that the older the person, the higher the age-specific mortality rate of infected individuals [1]. In addition, the mortality rate by underlying disease shows that patients with underlying heart disease have a mortality rate that is about 10 times higher than those without underlying diseases [2]. Furthermore, the number of elderly people living alone is also increasing, and is predicted to reach about 9 million by 2040 [3]. Therefore, it is considered necessary to measure vital signs that can monitor the health status of elderly people on a daily basis. Currently, methods using wearable devices and cameras [4] [5] are being considered as vital sign detection technologies, but the former are physically and mentally burdensome due to forgetting to wear them or wearing them for long periods of time, and the latter have problems such as privacy issues and weakness in blind spots and dark places. Therefore, to solve these problems, vital sign detection using radio waves is being considered [6]~[8]. This is a method in which radio waves are irradiated to a living body and vital signs are extracted from the Doppler shift contained in the reflected waves. The amplitude and phase of the reflected waves from vital signs change due to fluctuations on the surface of the body, such as breathing and heartbeat, and the Doppler shift includes time and frequency fluctuations. However, in complex radio wave propagation environments such as indoors, it becomes difficult to accurately detect vital signs due to reflected waves and noise other than those of the body. Therefore, the arctangent demodulation method has been proposed to improve the accuracy of vital sign detection [8] [9]. Arctangent demodulation is a method of detecting vital signs by observing the phase angle of arc-shaped signal points caused by body surface displacement, and has the advantage of reducing harmonics. In addition, research has been conducted to dramatically improve the sensitivity of vital sign detection by increasing the number of degrees of freedom of the propagation path using MIMO (Multiple-Input Multiple-Output)

[10] ~

[12] . It is possible to emphasize vital signs and reduce clutter by using beamforming to control the radiation directivity of the array antenna. However, even during quiet breathing, the direction and distance of the observed target change due to unconscious body movements, making it difficult to perform beam tracking that continuously emphasizes vital signs.In addition, if the arc-shaped response waveform caused by the body surface displacement cannot be observed, the arctangent demodulation, which requires the estimation of the center coordinate of the arc, does not work, making it difficult to remove the DC offset and making it impossible to accurately detect vital signs. To solve this problem, pseudo arctangent demodulation

[13] , which does not require the estimation of the center coordinate of the arc, and the CMA-like (Constant Modulus Algorithm) adaptive array algorithm

[14] have been proposed. The CMA-like method is a method that sequentially determines weights to suppress clutter based on the fact that the vital sign trajectory becomes linear. However, since it does not take into account the characteristics of the vital signs, it has the problem that the desired response is not necessarily obtained.

[0023] Therefore, we propose an improved beamforming method based on CMA to track vital signs by blind processing in MIMO (Multiple-Input Multiple-Output) radar. Specifically, we remove DC offset and update weights considering vital sign characteristics so that the phase fluctuation width due to breathing becomes the desired value. We also compare the heart rate estimation accuracy with the conventional beamforming method based on maximum ratio combining

[12] and the CMA-like method

[14] , and explain the effectiveness of our method.

[0024] 2. Vital sign detection method using the proposed method Figure 1 shows a conceptual diagram of the vital sign detection system discussed in this paper. In this case, we assume that the reflected waves from the human body while breathing quietly in front of the antenna are used. T The transmitting antenna of element and M R The propagation channel matrix X(t) in a MIMO radar consisting of element receiving antennas is given by

number

[0025] Next, the propagation channel matrix X(t) is converted into a column vector by a virtual array, and the channel elements are treated as signals. At that time, the signal x(t) is expressed as follows:

number

number

number

[0026] Also, h vital (t) is

number

[0027] Therefore, in order to track environmental and positional changes, the proposed improved CMA method sequentially updates the array synthesis weight and offset, optimizing it to draw a circular arc with the origin as the reference point so as to keep the amplitude and variance of the phase fluctuation width constant. First, we explain the method of emphasizing the vital sign component from the observed propagation channel response, and then we describe the method of extracting the heart rate component from the emphasized vital sign component by using arctangent demodulation.

[0028] 2.1 Vital sign emphasis method Here, we will discuss signal processing using CW-MIMO radar in vital sign detection. The reflected signal generated by the body surface displacement of the human body has the property of drawing an arc on the complex plane due to the body surface fluctuation caused by vital signs such as breathing and heart rate. However, as shown in Figure 2, the presence of DC offset γ causes the center coordinate of the signal trajectory to shift from the origin. Therefore, in order to accurately detect the vital signs, it is necessary to remove the DC offset. In addition, in order to take into account the characteristics of the vital signs, it is necessary to control the signal with a phase fluctuation width corresponding to the body surface displacement d(t) using the amplitude value equivalent to the amplitude Av representing the strength of the Doppler component, the instantaneous phase θ of the signal trajectory, and the average phase ?θ. Therefore, in the proposed improved CMA, in addition to the amplitude smoothing that is the original function of CMA, as shown in Figure 3, the phase fluctuation width corresponding to the body surface displacement caused by breathing is also taken into account to calculate the array weight that tracks the vital signs and remove the DC offset at the same time. Here, the array weight and DC offset value to be determined are expressed as a function,

number

number

number

number

number

number

number

number

[0029] 2.2 Heart rate component extraction method Here, we will describe a method for extracting the heartbeat component using arctangent demodulation, a technique for observing the phase angle from the array combined output obtained in the previous section. Complex signals reflected from the surface of a living body have the property of drawing an arc on the complex plane due to fluctuations on the body surface caused by vital signs such as breathing and heartbeat. Arctangent demodulation is a method for detecting vital signs by observing the phase angle of signal points on the arc, and has the advantage of being able to reduce harmonics. Here, the amount of phase rotation obtained from the array combined output y(t) by arctangent demodulation is given by:

number

[0030] 3. Experiment 3.1 Experimental conditions and environment Table 1 shows the experimental conditions, and Figure 4 shows the experimental scene. The measurements were taken indoors, with 16 subjects breathing quietly (Iwate University ethical review: 202019). A patch array antenna was used, and the antenna configuration was 16 × 16 MIMO with 16 receiving antenna elements and 16 transmitting antenna elements. The array element spacing was 0.5 wavelengths, the antenna-subject distance was 0.5 m, and the transmitting and receiving antennas were installed so that the height of the center of each antenna matched the height of the subject's chest. In this paper, a high-pass filter of 0.7 Hz, a low-pass filter of 20 Hz, and a weighting coefficient α of the improved CMA method of 8.0 × 10 ?2 It was decided. [Table 1]

[0031] Figure 5 shows the experimental setup. In this experiment, a SP64T (Single-Pole 64Throw) switch is used to rapidly switch the transmitting antenna elements while simultaneously measuring the receiving side. A continuous wave of 5.018 GHz is used for transmission, with a switching speed of 3.2 kHz. In order to accurately synchronize transmission and reception, the transmitting signal is input from the SG (Signal Generator) to the downconverter by a distributor on the transmitting side, and the signal from the receiving antenna is input to the LNA (Low Noise Amplifier) ​​on the receiving side. The downconverted baseband signal is divided into real and imaginary parts and collected by the DAQ (Data-Acquisition unit) at a sampling frequency of 32 kHz. On the receiving side, the observation data is divided for each switched transmitting antenna element, and downsampled by performing time averaging processing, thereby acquiring the MIMO channel at a frequency of 200 Hz while improving the signal-to-noise ratio. This is fast enough to observe the low-frequency components that indicate biological activity.

[0032] Figure 6 shows a conceptual diagram of the heart rate estimation accuracy evaluation method. For the heart rate response waveform acquired by the MIMO radar and the waveform measured by the reference ECG (Electrocardiogram), bpm (beats per minute) is calculated from the peak interval. In addition, the response times of the heart rates obtained from the MIMO radar and the ECG are different. Therefore, spline interpolation is applied to each time-varying heart rate, and the time-varying heart rate estimated from the MIMO radar is compared with the time-varying heart rate obtained from the ECG at the corresponding time to calculate the error.

[0033] 3.2 Experimental results Figures 7A to 7C (Figure 7) show the IQ response waveforms by each method. Figure 7A (Figure 7(a)) shows the response waveform after applying maximum ratio combining and performing origin correction by the least squares method to remove the DC offset. However, since the response waveform fluctuates in an arc while shifting in the amplitude direction, the origin correction by the least squares method does not work. In the response waveform by the CMA-like method in Figure 7B (Figure 7(b)), it can be confirmed that the array output converges to fluctuate in the imaginary part direction on the complex plane. In addition, in the response waveform by the proposed method in Figure 7C (Figure 7(c)), the array output converges to the set amplitude value and control phase angle (σy = 1.0, σθ = 30 deg.). This confirms that the DC offset is removed and the phase fluctuation width due to breathing is controlled to the desired value.

[0034] Figures 8A to 8C (Figure 8) show the body surface displacement estimated by each method. In this paper, the distance displacement obtained by the laser range finder, which is the reference, and the phase information obtained by the radar were converted to distance and compared. In Figure 8A (Figure 8 (a)), the arctangent demodulation does not work as in Figure 7A (Figure 7 (a)), and the origin is corrected to the center of the entire IQ response waveform, and the phase response waveform cannot be accurately extracted. Figures 8B and 8C (Figures 8 (b) and (c)) show the imaginary part fluctuation extraction waveform and the phase extraction waveform by arctangent demodulation, respectively. The trends of the phase response waveform and the laser waveform by both methods are similar, and it can be confirmed that the body surface displacement can be estimated with high accuracy using microwaves. In addition, by comparing Figures 8A to 8C (Figures 8 (a)-(c)), it can be confirmed that the proposed method can estimate the trend of the body surface displacement with the highest accuracy.

[0035] Figures 9A to 9C (Figure 9) show the heart rate response estimated by each method. In Figure 9A (Figure 9(a)), it can be seen that the peaks of the heart rate response waveform do not follow the ECG waveform. This is thought to be because, as in Figure 8A (Figure 8(a)), it is not possible to extract the phase response caused by the body surface displacement, and the heart rate components buried in the clutter components (unnecessary signal components) cannot be accurately extracted. In Figures 9B and 9C (Figures 9(b) and (c)), it can be seen that the heart rate response waveform has peaks that follow the ECG waveform. Comparing the two methods, it can be seen that the peak positions of the heart rate response waveform by the proposed method follow the ECG peaks.

[0036] Figures 10A to 10C (Figure 10) show the time response of heart rate by each method. The average heart rate error by each method was 30.1 bpm for maximum ratio combining, 6.46 bpm for the CMA-like method, and 1.48 bpm for the proposed improved CMA method. This confirms that the proposed method is able to estimate the time-varying heart rate with the highest accuracy.

[0037] Figure 11 shows the CDF of the heart rate estimation error for all subjects using each method. The 50% heart rate error value for each method was 11.1 bpm for maximum ratio synthesis, 10.5 bpm for the CMA-like method, and 3.47 bpm for the proposed improved CMA method. The 90% heart rate error value was 28.9 bpm for maximum ratio synthesis, 31.3 bpm for the CMA-like method, and 25.2 bpm for the proposed improved CMA method. This confirms that the proposed method is able to estimate heart rate with the highest accuracy.

[0038] literature [1] Ministry of Health, Labor and Welfare, Domestic trends of COVID-19 infections, 2020. [2] Ministry of Health, Labor and Welfare Quarantine Station FORTH, Underlying diseases and fatality rate of novel coronavirus patients in China, 2020. [3] Ministry of Health, Labor and Welfare, National Institute of Population and Social Security Research, Future Estimates of the Number of Households in Japan, 2018. [4] Y.-L. Zheng, X.-R. Ding, C. C. Y. Poon, B. P. L. Lo, H. Zhang, X.-L. Zhou, G.-Z. Yang, N. Zhao, and Y.-T. Zhang, “Unobtrusive sensing and wearable devices for health informatics,” IEEE Transactions on Biomedical Engineering, vol. 61, no. 5, pp. 1538?1554, May 2014. [5] N. Bernacchia, L. Scalise, L. Casacanditella, I. Ercoli, P. Marchionni, and E. P. Tomasini, “Non contact measurement of heart and respiration rates based on KINECT?,” IEEE International Symposium on Medical Measurements and Applications (MeMeA), pp. 1?5, Jun. 2014. [6] A. Droitcour, V. Lubecke, J. Lin, and O. Boric-Lubecke, “A microwave radio for Doppler radar sensing of vital signs,” IEEE MTT-S International Microwave Sympsoium Digest, vol. 1, pp. 175?178, May 2001. [7] B. Lohman, O. Boric-Lubecke, V. Lubecke, P. Ong, and M. Sondhi, “A digital signal processor for Doppler radar sensing of vital signs,” IEEE Engineering in Medicine and Biology Magazine, vol. 21, no. 5, pp. 161?164, Dec. 2002. [8] B.-K. Park, O. Boric-Lubecke, and V. M. Lubecke, “Arctangent demodulation with DC offset compensation in quadrature Doppler radar receiver systems,” IEEE Transactions on Microwave Theory and Techniques, vol. 55, no. 5, pp. 1073?1079, May 2007. [9] C. Li and J. Lin, “Random body movement cancellation in Doppler radar vital sign detection,” IEEE Transactions on Microwave Theory and Techniques, vol. 56, no. 12, pp. 3143?3152, Dec. 2008.

[10] M. Nosrati, S. Shahsavari, S. Lee, H.Wang, and N. Tavassolian, “A concurrent dual-beam phased-array Doppler radar using MIMO beamforming techniques for short-range vitalsigns monitoring,” IEEE Transactions on Antennas and Propagation, vol. 67, no. 4, pp. 2390?2404, Apr. 2019.

[11] C. Feng, X. Jiang, M.-G. Jeong, H. Hong, C.-H. Fu, X. Yang, E. Wang, X. Zhu, and X. Liu, “Multitarget vital signs measurement with chest motion imaging based on MIMO radar,” IEEE Transactions on Microwave Theory and Techniques, vol. 69, no. 11, pp. 4735?4747, Nov. 2021.

[12] Yuta Ogawa, Kota Sasaki, Naoki Homma, Kentaro Murata, Morio Iwai, Koichiro Kobayashi, and Atsushi Sato, “Evaluation of Heart Rate Estimation Using Rayleigh Quotient Beamforming Method,” IEICE Society Conference, pp. B?1?104, Sep. 2021.

[13] Jun Sato, Naoki Homma, Morio Iwai, Koichiro Kobayashi, and Atsushi Sato, “Evaluation of simultaneous detection method of multiple human heartbeats using amplitude equivalent component of pseudo arctangent demodulation,” IEICE Technical Report, AP2019-135, pp. 157-162, Nov. 2019.

[14] Naoki Homma, Kentaro Murata, Morio Iwai, Koichiro Kobayashi, and Atsushi Sato, “An adaptive array algorithm for vital sign detection suitable for low microwave bands,” IEICE Technical Report, SRW2021-11, pp. 43-48, Jun. 2021.

[0039] Hereinafter, an embodiment of the present invention based on the contents of the above-mentioned paper will be described with reference to the drawings.

[0040] A biological information detecting device according to an embodiment of the present invention is configured as shown in FIG.

[0041] This biological information detection device has a transmitter 11 (SG), a distributor 12, an attenuator 13, a transmitter 23, a receiver 24 (including a low noise amplifier (LNA)), a down converter 14 (DC), a terminal board 15, and a processing unit 17 (PC, NI DAQ (Data Acquisition Unit)), similar to the experimental system shown in FIG. 5, and detects biological information (heartbeat signal, heart rate, etc.) based on the periodic body movement of a subject (person), for example, heartbeat. A transmitting patch array antenna 21 is connected to the transmitter 23. The transmitting patch array antenna 21 includes, for example, 16 transmitting antenna elements. A receiving patch array antenna 22 is connected to the receiver 24. The receiving patch array antenna 22 includes, for example, 16 receiving antenna elements. The transmitting array antenna 21 and the receiving patch array antenna 22 form a 16×16 MIMO array antenna (MIMO radar).

[0042] The transmitter 23 transmits radio signals of a predetermined frequency (e.g., continuous waves of 5.018 GHz) from each of the multiple transmitting antenna elements to the subject while sequentially switching (e.g., switching speed of 3.2 kHz) the transmitting antenna elements of the transmitting patch array antenna 21 based on the oscillation signal from the oscillator 11 provided via the distributor 12 and the attenuator 13. In order to synchronize the transmitting side and the receiving side, the signal from the oscillator 11 distributed by the distributor 12 is provided to the down converter 14 as a synchronization signal.

[0043] Each of the multiple receiving antenna elements of the receiving patch array antenna 22 outputs a receiving signal corresponding to the received radio wave. The receiver 24 includes a low noise amplifier (LNA), takes in the receiving signals from the multiple receiving antenna elements (receiving patch array antenna 22) and processes these receiving signals. A plurality of signals from the receiver 24 corresponding to the receiving signals from the multiple receiving antenna elements are down-converted by the down-converter 14 in synchronization with the signal on the transmitting side. A plurality of baseband signals corresponding to the multiple antenna elements from the down-converter 14 are serially supplied to a processing unit 17 via a terminal board 15 which is switched by a relay box 16 so as to correspond to the switching operation of radio wave transmission in the transmitter 23. On the receiving side, the switched transmitting antenna By dividing the observation data for each element, performing time averaging and downsampling, the signal-to-noise ratio is improved and the MIMO channel is acquired at a frequency of, for example, 200 Hz, which is fast enough to observe the low-frequency components that indicate biological activity.

[0044] The processing unit 17 includes a computer system configured by software and hardware (including a CPU, memory, etc.), and collects baseband signals sequentially provided via the terminal board 15 at a predetermined sampling frequency (e.g., 32 kHz) by dividing the signals into real and imaginary parts. The processing unit 17 then processes the baseband signals collected according to a procedure to be described later, and detects periodic body movements of the subject (living body), for example, biological information (heartbeat signal, heart rate) based on heartbeat.

[0045] The processing unit 17 executes the process according to the procedure shown in FIG.

[0046] In FIG. 13, the processing unit 17 samples signals based on received signals from each receiving antenna element (receiving patch array antenna 22) as complex signals (real part, imaginary part) (S11). Based on the sampled complex signals, the processing unit 17 estimates complex transfer functions between each transmitting antenna element of the transmitting patch array antenna 21 and each receiving antenna element of the receiving patch array antenna 22 for each of a plurality of combinations of the transmitting antenna elements and the receiving antenna elements (S12: transfer function estimator: combined output acquirer). The plurality of complex transfer functions X(t) are expressed in the matrix format of equation (1). The processing unit 17 further converts the complex transfer function X(t) expressed in the matrix format into an array signal x(t) in the row vector format of equation (2) (S12). Here, the array combined output y(t) can be expressed by equation (3) using the array signal x(t) as a complex signal and the array weight w as a complex signal (array combined output generator).

[0047] Next, the processing unit 17 executes an optimization process for the array combining output (S13, S14). This optimization process for the array combining output (output adjustment unit) is performed as follows.

[0048] In this optimization process of the array combining output, as shown in Fig. 3, the phase and amplitude represented by the signal point on the complex plane of the array combining output y(t) expressed in a complex signal format are optimized so as to be in a predetermined state based on a predetermined amplitude value (σy) and a predetermined phase fluctuation width (σθ) on the complex plane, specifically, to be within an arc locus of a predetermined phase fluctuation width (σθ: predetermined central angle) with the predetermined amplitude value (σy) as the radius, based on the property that a complex signal reflected from the surface of a living body draws an arc on the complex plane due to the fluctuation of the body surface caused by vital signs such as breathing and heart rate.

[0049] In the above optimization process, the processing unit 17 obtains the complex weight vector w and the complex DC component γ used in the definition of the array combining output y(t) (see equations (3) and (4)) (preprocessing unit). Then, the processing unit 17 sequentially updates the complex weight vector w and the complex DC component γ using the steepest descent method in accordance with the arithmetic expressions shown in the above-mentioned equations (6) to (13) so that the evaluation function Q (see equation (6)) is minimized, thereby sequentially adjusting the array combining output y(t) (see equations (3) to (5)) (improved CMA adjustment unit).

[0050] When the adjusted array synthesis output y(t) (see equations (3) to (5)) is obtained by the optimization process described above, the processing unit 17 calculates the phase rotation amount β(t) (phase fluctuation width corresponding to body movement) according to equation (14) (S15). The processing unit 17 extracts a signal corresponding to heartbeat vibration by performing a filter process corresponding to the body movement to be detected, for example, heartbeat vibration, on the movement rotation amount β(t) (S16). Then, the processing unit 17 generates heartbeat information (for example, heart rate: biological information) from the extracted signal corresponding to the heartbeat vibration (S17: S15 and S16 constitute a biological information extraction unit). Thereafter, the processing unit 17 repeatedly executes the above-mentioned process until receiving an instruction to end the measurement (YES in S18). In the process, the processing unit 17 generates biological information (heartbeat information) that changes from moment to moment.

[0051] According to the above-described biological information detection device, it is possible to detect biological information (heartbeat information, etc.) of a subject with high accuracy by using radio waves without contacting the subject.

[0052] By changing the characteristics of the filter processing during the processing by the processing unit 17 described above, it is possible to extract biological information (respiratory movement information) based on the respiratory movement of the subject instead of cardiac movement information.

[0053] In the above-mentioned biological information detection device, a MIMO antenna including a transmitting patch array antenna 21 having a plurality of transmitting antenna elements and a receiving patch array antenna 22 having a plurality of receiving antenna elements is used, but the present invention is not limited to this. A SIMO antenna including a plurality of transmitting antenna elements and a single receiving antenna element, or a MISO antenna including a single transmitting antenna element and a plurality of receiving antenna elements may also be used.

[0054] The living body from which biological information is detected may be an animal other than a human.

[0055] Although the embodiment of the present invention has been described above, the embodiment and the modified examples of each part are presented as examples and are not intended to limit the scope of the invention. These novel embodiments described above can be embodied in various other forms, and various omissions, substitutions, and modifications can be made without departing from the gist of the invention. [Industrial Applicability]

[0056] The bioinformation detection device of the present invention has the effect of being able to detect bioinformation (vital signs) of humans or animals more accurately using radio waves, and is useful as a bioinformation detection device that detects bioinformation based on periodic bodily movements such as heartbeat and breathing of a living body (human or animal). [Explanation of symbols]

[0057] 11 Transmitter 12 Distributor 13 Attenuator 14 Down Converter 15 Terminal Board 16 Relay box 17 Processing Unit 21 Transmitting patch array antenna 22 Receiving patch array antenna 23 Transmitter 24 Receiver

Claims

1. A biological information detection device that detects biological information based on periodic body movements of a living body, an array antenna including any one of a set of a transmitting antenna element and a plurality of receiving antennas, a set of a plurality of transmitting antenna elements and receiving antenna elements, and a set of a plurality of transmitting antenna elements and a plurality of receiving antenna elements; a composite output acquisition unit that acquires an array composite output expressed in a complex signal format based on a received complex signal corresponding to a received radio wave signal from each receiving antenna element in the array antenna under a condition in which a radio wave signal of a predetermined frequency is transmitted from each transmitting antenna element in the array antenna toward a living body that is a detection target for biological information; an output adjustment unit that adjusts the array combined output so that the phase and amplitude represented by a signal point on a complex plane of the array combined output acquired by the combined output acquisition unit fall within a predetermined state based on a predetermined amplitude value and a predetermined phase fluctuation width on the complex plane; a biometric information extraction unit that extracts target biometric information from the adjusted array combined output obtained by the output adjustment unit.

2. The composite output acquisition unit a transfer function estimating unit that estimates a complex transfer function between each of a plurality of combinations of each transmitting antenna element and each receiving antenna element in the array antenna; 2. The biological information detection device according to claim 1, further comprising a combined output generating unit which treats each of a plurality of complex transfer functions obtained by said transfer function estimating unit as a complex signal and generates an array combined output from said plurality of complex signals.

3. 2. The biological information detection device according to claim 1, wherein the output adjustment unit adjusts a signal point on a complex plane of the array synthesis output so that the signal point falls within an arc-shaped locus having the predetermined amplitude value and a predetermined phase fluctuation width centered on an origin of the complex plane.

4. The output adjustment unit is a pre-processing unit for determining a complex weight vector and a complex DC component used to define the array synthesis output in the form of a complex signal; and an improved CMA adjustment unit that sequentially adjusts the complex weight vector and the complex DC component so that the signal points represented by the array synthesis output fall within an arc-shaped locus centered on the origin on a complex plane.

5. The improved CMA adjustment unit includes:

5. The biological information detection device according to claim 4, wherein the complex weight vector and the complex DC component are adjusted so that a central angle of a circular arc locus of the signal points of the array combined output on the complex plane falls within a predetermined range.

6. A biological information detection method for detecting biological information based on periodic body movements of a living body, comprising: An array antenna including any one of a set of a transmitting antenna element and a plurality of receiving antennas, a set of a plurality of transmitting antenna elements and a receiving antenna element, and a set of a plurality of transmitting antenna elements and a plurality of receiving antenna elements is used, a composite output acquisition step of acquiring an array composite output expressed in a complex signal format based on a received complex signal corresponding to a received radio wave signal from each receiving antenna element in the array antenna under a condition in which a radio wave signal of a predetermined frequency is transmitted from each transmitting antenna element in the array antenna toward a living body that is a detection target for biological information; an output adjustment step of adjusting the array combined output so that the phase and amplitude represented by the signal point on the complex plane of the array combined output acquired in the combined output acquisition step fall within a predetermined state based on a predetermined amplitude value and a predetermined phase fluctuation on the complex plane; a biometric information extraction step of extracting target biometric information from the adjusted array combined output obtained in the output adjustment step.

Images