JP2019152441A - Vital sensor - Google Patents

Abstract

To detect biological information with high sensitivity and efficiency in a short processing time without requiring a high-speed CPU or large memory.SOLUTION: There is provided a vital sensor that transmits a modulated wave based on frequency sweep and receives the reflected wave, mixes the transmitted wave and the received wave, and outputs a beat signal having a frequency difference between the two waves. The vital sensor extracts the difference signal by subtraction (difference circuit 27) between the beat signals in the sweep before and after the sweep repeated continuously, calculates an integrated power value obtained by integrating the power of this difference signal for each sweep (power integrating circuit 28), and determines information on minute fluctuations, respiratory rate, heart rate, etc. in the living body by detecting changes in the integrated power value for each sweep (measurement circuit 29).SELECTED DRAWING: Figure 1

Description

  The present invention transmits a vital wave sensor, in particular, a modulated wave in the microwave band / millimeter wave band, and compares this transmitted wave with a received wave reflected from an object to detect minute fluctuations such as respiration and heartbeat of a living body. It relates to vital sensors.

Conventionally, radio wave type non-contact vital sensors have mainly used a Doppler method and a method of measuring a minute phase angle by FMCW (frequency modulation continuous wave). The Doppler method uses a CW signal as a target (for example, In this method, the reflected wave reflected by the human body is received and the phase detection of the received wave and the transmitted wave is performed to obtain information on movement of breathing and heartbeat from the Doppler signal generated by movement.
Breathing refers to the contraction of the lungs of the human body and the movement of the abdomen that fluctuates at the same time, and the heartbeat refers to the heartbeat and blood flow (pulse). These differ in the speed of movement and the distance (amount) of movement, but in either case are much smaller than the movement of a person.

  The Doppler method is mainly used for a single target at a very short distance, and the Doppler signal reacts to all moving objects, and it is difficult to separate unnecessary signals, but there is an advantage that the S / N ratio is easily improved systematically. . That is, if an attempt is made to increase the distance to the target, there may be a case where an erroneous report or the like is generated due to the influence of undesired noise, and it is difficult to extract an accurate behavior of the target.

  On the other hand, in the FMCW system, a radio wave formed by sweeping with a predetermined modulation width and sweep period is radiated from an antenna, a signal reflected by a human body or the like as a target is phase-detected, and the distance is determined from the frequency of the beat signal generated by the distance. This is a calculation method (also called a minute phase angle method). In this beat signal, when the target moves by a very small amount, a phase fluctuation corresponding to the movement occurs, and by extracting the time (cycle) of this phase fluctuation, movements such as respiration and heart rate can be extracted. .

In the case of this FMCW method, distance measurement is possible and it can cope with a relatively long distance, but the distance separation resolution by the sweep bandwidth is c / Δf (c: speed of light, Δf: sweep bandwidth or occupied bandwidth). In Japan, it is 1.5m, and in other countries it is 3m.
Therefore, for example, when detecting a human body in a room, there are various objects such as a wall and furniture in the room, and it is necessary that the reflector and the human body be separated by more than the distance separation resolution. In the case of a reflector within the distance separation resolution, the behavior of minute fluctuations appears due to the fusion of the spectrum after FFT, etc., but the distance is not fixed or it is difficult to extract minute signals because they are buried in a large reflected signal. Or is easily affected by the environment (surroundings) such as large reflectors, and appears as an unstable measurement result.

Japanese Patent No. 5848469 U.S. Pat. No. 9423496

  In the conventional FMCW sensor described above, in the case of a single target at a relatively short distance where radiation from the antenna does not spread, good detection can be performed and practical use under specific conditions. In most cases used as a sensor for measuring the number, heart rate, etc., there is a problem that the sensitivity is not sufficient and false or misreporting occurs.

Moreover, in the living body state detection apparatus of Patent Document 1, distance spectrum intensity data (peak value) is obtained for each distance based on the detection result (FFT processing) of the FMCW radar, and a reference value in a state where no living body exists, The difference from the value of the state at the time of presence is calculated as difference data (frequency spectrum difference), and the posture, respiration, etc. are detected from the time fluctuation of the peak value in the resulting spectrum.
Even in the wireless detection device of Patent Document 2 described above, when a stationary object such as a wall or window and a living body such as a person are at different distances, the peak spectrum with respect to the distance is temporally compared to measure the temporal change of the peak spectrum of the living body. Therefore, living body detection is performed.

  However, in Patent Documents 1 and 2 described above, the spectrum after the FFT processing is compared, and living body detection is performed based on the difference between the spectra. In this method, the reference (background) information is subtracted from the measurement result. There is an inconvenience that the sensitivity of the distance to be detected is remarkably deteriorated.

  Conventionally, since living body detection is performed based on the difference in spectrum obtained by FFT processing, a large memory is required, and calculation and processing until detection are complicated, and there is a problem that processing time is increased.

  The present invention has been made in view of the above problems, and its purpose is to detect living body information efficiently and with high sensitivity in a short processing time without requiring a high-speed CPU or a large memory. It is to provide a vital sensor that can be used.

  In order to achieve the above object, the invention according to claim 1 transmits a modulated wave whose oscillation frequency continuously increases or decreases based on the frequency sweep, and the transmitted wave is reflected by an object existing ahead. In a vital sensor having a transmission / reception circuit that receives a reflected wave and outputs a beat signal having a frequency of a difference between both waves by mixing the transmitted wave and the received wave, the beat in the sweep before and after the continuous repeated sweep The difference signal is extracted by subtraction between the signals, the integrated power value obtained by integrating the power of the difference signal is calculated for each sweep, and information on minute changes in the living body is detected by the change in the integrated power value for each sweep. Features.

The invention of claim 2 is characterized in that information on minute fluctuations of each of a plurality of living bodies is detected by band-limiting the frequency of the beat signal obtained after mixing the transmission wave and the reception wave.
The invention of claim 3 is characterized in that the information on the minute fluctuations of the living body is the presence or absence of a heartbeat or respiration, a heart rate or a respiration rate.
The invention according to claim 4 is characterized in that a distance to the living body is obtained from the difference signal, and information on presence / absence of the living body is also detected.

  According to the above configuration, the modulated wave that rises or falls due to the frequency sweep is sequentially transmitted, the reflected wave from the transmitted wave is received, and the received wave is mixed with the transmitted wave to have a frequency that is the difference between the two waves. A beat signal (IQ signal) is obtained. In this beat signal, a difference signal is extracted by performing subtraction processing with the beat signal of the previous sweep for each successive repeated sweep, and an integrated power value of the difference signal is calculated. By grasping the change of the integrated power value for each sweep by FFT (Fast Fourier Transform) processing, etc., breathing, heartbeat (pulse), presence / absence of minute movement of the living body, heart rate or respiration rate, etc. Is detected.

In addition, the frequency of the beat signal after mixing the transmission wave and the reception wave is band-limited with a filter or the like, and unnecessary signals are excluded, so that information on minute fluctuations in each of a plurality of living bodies can be detected well. The
Furthermore, if the difference signal is subjected to, for example, FFT processing, the distance to the living body can be obtained, and thereby the presence / absence (presence / absence) of the living body itself can be grasped.

According to the present invention, since the difference regarding the frequency spectrum after the FFT processing is not calculated, the number of FFT calculations may be small, and a high-speed CPU and a large memory are not required, and the living body can be efficiently processed in a short processing time. There is an advantage that the information can be detected.
In addition, since the difference between the beat signals is obtained in real time for each sweep, the sensitivity does not change depending on the measurement conditions, and high sensitivity can be obtained.

It is a block diagram which shows the circuit structure of the vital sensor of the Example which concerns on this invention. It is a wave form diagram which shows the signal processing until it obtains the change of an integrated power value from a frequency sweep in an Example. It is a figure which shows the waveform [figure (a)] of the beat signal obtained for every sweep in an Example, and the signal [figure (b)] of an integrated power value change. It is a wave form diagram which shows the frequency of respiration [a figure (a)] and the frequency of a heartbeat [a figure (b)] obtained in an Example. It is explanatory drawing which shows a state when obtaining the information of a biological body with the vital sensor of an Example.

FIG. 1 shows a circuit configuration of a vital sensor according to an embodiment, and this vital sensor is an FMCW sensor using a microwave.
As shown in FIG. 1, the transmission unit 10 includes an oscillator 11, a distributor 12, a transmission amplifier 13, a transmission wave filter 14, an LO amplifier 15, and the like, and performs frequency sweep (sweep of a predetermined modulation width and period). A modulation wave whose oscillation frequency continuously increases or decreases is transmitted, and the reception unit 18 includes an LNA (low noise amplifier) 19 and the like, and receives a reflected wave from an object. The receiving unit 18 is controlled by a control unit (not shown).

  In addition, a mixer 20 is provided for inputting signals from the transmission unit 10 and the reception unit 18, mixing the transmission wave and the reception wave, and outputting a beat signal (I, Q signal). A filter 21, an IF amplifier 22, and an ADC (analog / digital converter) 23 are disposed, and an arithmetic unit 25 including an MCU (micro control unit) is connected to the ADC 23.

  The calculation unit 25 includes a difference (subtraction) circuit 27 that calculates a difference (difference signal) between beat signals obtained in the current sweep and beat signals obtained in the previous sweep for each sweep. A power integration circuit 28 that integrates the power (amplitude value) of the difference signal output from the difference circuit 27, and minute fluctuations of the living body such as breathing, heartbeat, and minute values from the integrated power value output from the power integration circuit 28. A measurement circuit 29 for measuring movement and the like is included. The measurement circuit 29 performs FFT (Fast Fourier Transform) processing on the integrated power signal output from the power integration circuit 28, and determines the presence / absence of respiration, respiration rate, presence / absence of heartbeat, heart rate, etc. from the obtained frequency spectrum. Can be detected. The biological information such as respiration and heartbeat can also be detected by performing pulse counting regardless of the FFT processing.

  The embodiment has the above configuration, and its operation will be described with reference to FIGS. First, in the transmission unit 10, as shown in FIG. 2A, a microwave modulated based on a frequency sweep (up sweep or down sweep) with a repetition time T (several milliseconds) [FIG. 2B]. RF] is continuously transmitted. Based on the transmission of the microwave, the reception unit 18 receives the reflected wave from the object, and the mixer 20 mixes both the waves to obtain a beat signal that is the difference frequency. [FIG. 2 (c)] is sequentially output. That is, since the received wave is delayed compared to the transmitted wave, a beat signal (I, Q signal) having a frequency that is the difference between the two waves is obtained. Since the frequency of the beat signal varies depending on the sweep time, the sweep frequency band (occupied bandwidth), and the distance to the object, the distance of the object can be obtained by analyzing this frequency.

  Next, in the difference circuit 27 of the calculation unit 25, as shown in FIG. 2D, a difference signal (difference beat signal) between beat signals is calculated for each sweep. That is, as shown in FIG. 2 (c), beat signals SW1, SW2, SW3,... SW (n) are obtained for each successive sweep, and the difference between the beat signals, SW1-SW2, SW2-SW3, SW3. -SW4... Are sequentially obtained, and thereafter, in the power integration circuit 28, as shown in FIG. 2 (e), S12 = Σ (SW1-SW2), S23 as the integration processing of the power (amplitude value) of the differential signal. = Σ (SW2-SW3), S34 = Σ (SW3-SW4)... Is calculated, and integrated power values (voltage values) S12, S23, S34.

  Next, the measurement circuit 29 performs, for example, FFT processing on the integrated power values S12, S23, S34..., As shown in FIG. For example, respiration, heartbeat, and minute movement are detected. Then, the respiratory rate and the heart rate can be obtained by performing frequency analysis (for example, FFT processing) on this waveform.

  That is, as shown in FIG. 3A, in beat signals SW1, SW2, SW3,... SW (n) obtained for each sweep, differences Δ1, Δ2, Δ3,. And the integrated power values PΔ1, PΔ2, PΔ3,... PΔ (n) are obtained. As shown in FIG. 3B, the integrated power value PΔ (n) is a value corresponding to the magnitude of the change in the distance information obtained from the difference signal. If this integrated power value is plotted, breathing, A signal including heartbeat and minute movement information is obtained, and by analyzing the frequency of this signal, a predetermined frequency appears, and the respiratory rate, heart rate, and the like are determined based on this frequency. When the distance width of movement (variation) in a living body is large, the integrated power value (voltage value) is large. Conversely, when the distance width is small, the integrated power value is small. Etc. can be detected.

  FIG. 4 shows the breathing and heartbeat frequencies obtained by the measurement circuit 29. The presence of breathing and the number of breaths at the peak frequency in FIG. 4 (a), and the peak frequency in FIG. 4 (b). The presence of the heartbeat and the heart rate will be determined. In general, the respiration rate is 12 to 20 times / min, the heart rate is 50 to 75 times / min, and the frequency is 0.2 Hz to 0.33 Hz and the heart rate is 0.83 Hz to 1.25 Hz. Thus, the respiration rate, the heart rate and other movements are discriminated by decomposing each frequency with a filter or the like.

In the embodiment, the presence / absence (presence / absence) of the moving object can be detected by performing an FFT process or the like on the difference signal obtained by the difference circuit 27.
For example, as shown in FIG. 1, if the difference signal of the difference circuit 27 is output to the measurement circuit 29 without going through the power integrating unit 28 and subjected to FFT processing, the frequency spectrum after the FFT processing is performed in the measurement circuit 29. , It is determined that there is no moving object when the signal intensity at each frequency is substantially 0, and there is a moving object when there is a signal intensity peak at a specific frequency.

  FIG. 5 shows a case where a human sensor is detected by a vital sensor disposed above. When the person is simply moving under the vital sensor as shown on the left side, as described above, the difference signal Thus, the passage of a person is detected, and a person sleeping on a bed or the like, as shown on the right side, is detected based on the integrated power value of the difference signal, such as breathing, heartbeat, and minute movement.

The integrated power value (voltage value) described above is data in which the amplitude value changes in response to minutely changing distance information, and data representing the cycle of the living body in which the cycle (frequency) of this change changes slightly. Become. Such a behavior is exhibited because the repeated measurement time is much shorter than the change time of the living body.
For example, when a person 5 m ahead walks or moves, the frequency (difference signal) obtained by comparison indicates the distance to the person, and the movement of the person in this case is large, and thus indicates the movement distance itself.
On the other hand, the distance does not change when the person is stationary and stays in the place, so it cannot be said that the person is moving, in which case the change in the distance information detected by the difference signal is seen. No, but people usually do breathing and subtle movements (even when sleeping), and at least the heart is constantly moving.
In the present invention, these movements are compared for each sweep, and the resultant integrated power is connected to detect respiration, heartbeat, minute movement, and the like.

Further, in the embodiment, before calculating the difference signal, a band restriction is performed such that a frequency component corresponding to the distance to the target object (living body) is extracted from the beat signal obtained by one sweep, thereby performing a plurality of band limitations. Even when there are living bodies, it is possible to simultaneously detect the heartbeat and respiration of these living bodies.
As a specific method, for example, FFT processing is performed on a beat signal, a distance from a peaked frequency to a target person is measured, and a beat is applied to a bandpass filter that removes other frequency components. This can be realized by passing a signal. For example, even when an object considered to be environmental noise such as a fan or an air conditioner is in the measurement range, the influence can be eliminated by removing the frequency component not caused by the object.

Further, in the embodiment, the distance information of the moving object of the frequency spectrum obtained from the difference signal based on the current sweep and the previous sweep is compared with the distance information of the object of the frequency spectrum of the beat signal obtained by only one sweep. In addition, it is possible to detect the distance of only the stationary body, and thereby, it is possible to clearly distinguish and detect the moving living body and the stationary body.
That is, the vital sensor of the present invention is a radio wave sensor that can detect a minute object when a living object is stationary, a moving object, and a moving object without being affected by the distance separation resolution due to the sweep width, which is a weak point of the FMCW method. is there.

  Further, in the power integrating circuit 28, the value of the difference signal is compared with an absolute value, so that data having a value twice the original frequency is obtained. This is because, in the difference processing, a portion with a small change and a large portion are repeated by comparison at equal intervals, and a small portion exists twice in one cycle (period). It is twice as fast as the heart rate. Therefore, it is easy to perform frequency decomposition of each of breathing and heartbeat, and there is an advantage that the breathing information and the heartbeat information can be reliably separated and extracted.

According to the vital sensor having the above configuration, it is not necessary to calculate and compare the beat signal spectrum for each sweep, and can be detected by simple processing based on the difference signal of the beat signal. Less memory is required for the computing unit 25, and computation processing is also faster.
In addition, since the comparison is performed in real time as compared with the conventional case, there is an advantage that the sensitivity does not change under the measurement condition, detection with high sensitivity is possible, and false or misreporting can be reduced.

10 ... Transmitter, 11 ... Oscillator,
15 ... LO amplifier, 18 ... receiver,
19 ... LNA (low noise amplifier), 20 ... mixer,
21 ... IF filter, 22 ... IF amplifier,
23. Analog to digital converter,
25 ... arithmetic unit, 27 ... difference circuit,
28 ... Power integrating circuit, 29 ... Measuring circuit.

Then, the measuring circuit 29, from the integrated power values S12, S23, S34 ..., as shown in FIG. 2 (f), slight change (change) of a living body that is the measuring object, for example breathing, heartbeat, minute Motion or the like is detected. Then, the respiratory rate and the heart rate can be obtained by performing frequency analysis (for example, FFT processing) on this waveform.

Claims (4)

Based on the frequency sweep, transmit a modulated wave whose oscillation frequency continuously rises and falls, receives a reflected wave reflected by an object present ahead, and mixes the transmitted wave and the received wave In a vital sensor equipped with a transmission / reception circuit that outputs a beat signal having a frequency difference between both waves,
Extract the difference signal by subtraction between the beat signals in the sweep before and after the sweep repeated continuously,
A vital sensor characterized in that an integrated power value obtained by integrating the power of the difference signal is calculated for each sweep, and information on minute fluctuations of a living body is detected by a change in the integrated power value for each sweep.   2. The vital sensor according to claim 1, wherein information on minute fluctuations of each of a plurality of living bodies is detected by band-limiting a frequency of a beat signal obtained after mixing a transmission wave and a reception wave.   The vital sensor according to claim 1, wherein the information on minute fluctuations of the living body is presence or absence of a heartbeat or respiration, a heart rate or a respiration rate.   The vital sensor according to any one of claims 1 to 3, wherein a distance to the living body is obtained from the difference signal, and information on the presence or absence of the living body is also detected.

Images