JP7451040B2 - Biological state estimation device - Google Patents

Description

The present invention relates to a biological state estimating device, and specifically relates to a technique for estimating the moving state, resting state, and absent state of a living body using a radar sensor.

Pyroelectric sensors, infrared sensors, and cameras are mainly used as conventional human sensors to detect movements of living organisms and monitor them. Specifically, for example, as a technology for monitoring users in nursing care facilities, etc., we use sensors for users installed in each room, video cameras for users installed in each room, and display of user information. A monitoring system is known that includes a terminal for a caregiver (see Patent Document 1).

Japanese Patent Application Publication No. 2018-132795

By the way, in pyroelectric sensors and infrared sensors, presence or absence is determined by utilizing the fact that the reception level changes significantly due to the opening/closing of a door or the walking of a living body. At this time, the threshold regarding the reception level (amplitude) is set in advance and the determination is made by comparing this threshold with the actually observed reception level (amplitude). The setting is difficult, and as a result, there are many cases where a resting state with little movement is mistakenly recognized as an absent state. Furthermore, it is necessary to manually set the reception level (amplitude) threshold for each environment in which the sensor is used, which is time-consuming and can lead to erroneous recognition unless the threshold setting is changed each time the usage environment changes. There is a problem. Furthermore, when a camera is used as in the technique described in Patent Document 1, there is a problem that privacy is violated, and furthermore, it is not possible to distinguish between sleeping and a critical state of the living body using images. There is a problem.

Therefore, the present invention aims to provide a biological system that can accurately determine the moving state, resting state, and absent state of a living body without violating privacy or manually setting thresholds for each usage environment. The purpose of the present invention is to provide a state estimation device.

In order to solve the above problem, the invention according to claim 1 emits a transmitted radio wave, receives a reflected wave of the transmitted radio wave, and mixes the transmitted radio wave and the reflected wave to generate a beat signal. a transmitting/receiving section; a phase calculating section that calculates the phase of the beat signal output from the transmitting/receiving section; data of the beat signal output from the transmitting/receiving section; and data of the phase output from the phase calculating section; a stationary object reflection removing section that removes a component corresponding to a reflected wave from a stationary object from each of the beat signal data output from the transmitting/receiving section and the phase data output from the phase calculating section. a body movement component removing unit that removes a component corresponding to body movement from each; and a speed that calculates a relative velocity between the transmitting and receiving unit and a living body as a target using the phase data after removing the two components. a calculation unit; a moving state estimation unit that determines that the living body is in a moving state when the relative velocity is greater than the breathing rate of the living body; and the relative velocity is less than or equal to the breathing rate of the living body; and the sum of the spectral intensities in a specific frequency range of the frequency spectrum distribution obtained by fast Fourier transforming the data of the beat signal after the removal of the two components is within a predetermined region other than the specific frequency range. a resting state estimation unit that determines that the living body is in a resting state when the sum of spectral intensities in the specific frequency range is greater than the sum of spectral intensities in the specific frequency range; and a spectral intensity in a predetermined region other than the specific frequency range. If the sum is less than or equal to the sum of and an absent state estimating unit that makes a determination , and the specific frequency range is set from 0 Hz to less than 5 Hz .

The invention according to claim 2 is characterized in that, in the biological state estimating device according to claim 1 , the respiration rate of the living body is set to 1 cm to 10 cm/sec.

According to a third aspect of the invention, in the biological state estimation device according to the first or second aspect, the estimation result of the biological state is changed from the resting state to the absent state without passing through the moving state. The present invention is characterized in that it further includes a dangerous state detection unit that determines that the living body is in a dangerous state when the living body changes.

According to the inventions described in claims 1 and 2 , components corresponding to reflected waves from stationary objects such as walls and floors are autonomously removed, so a threshold value can be set for each usage environment. In other words, it is possible to determine with high precision whether the living body is absent, at rest, or moving, without being affected by or depending on the usage environment. According to the invention set forth in claim 1, the states are classified and identified step by step in descending order of movement, such as a moving state, a resting state, and an absent state. Moreover, it becomes possible to estimate the state of the living body with high accuracy.

Moreover, according to the invention described in claims 1 and 2, it becomes possible to appropriately determine whether or not the living body to be observed is in a moving state, especially when the living body to be observed is a human.

Moreover, according to the inventions described in claims 1 and 2 , it becomes possible to appropriately determine whether or not the living body to be observed is in a resting state, especially when the living body to be observed is a human being.

According to the third aspect of the invention, it is possible to detect whether the living body is in a moving state, a resting state, or an absent state, as well as whether the living body is in a dangerous state.

1 is a functional block diagram showing a schematic configuration of a biological state estimation device according to an embodiment of the present invention. FIG. 2 is a flow diagram showing a processing procedure in the biological state estimation device of FIG. 1. FIG. FIG. 3 is a diagram illustrating a process of converting phase displacement into velocity and a process of determining whether a living body is in a moving state. It is a figure explaining the process of determining whether a living body is in a resting state. (A) is a diagram showing a frequency spectrum distribution when a living body is absent. (B) is a diagram showing the frequency spectrum distribution when the living body is in a resting state. It is a figure explaining the process which judges whether a living body is absent.

The present invention will be described below based on the illustrated embodiments. FIG. 1 is a functional block diagram showing a schematic configuration of a biological condition estimation device 1 according to an embodiment of the present invention. Further, FIG. 2 is a flow diagram showing a processing procedure in the biological state estimation device 1 according to the embodiment. The biological state estimating device 1 is a mechanism for estimating the moving state, resting state, and absent state of a living body using a radar sensor, and includes a transmitting/receiving section 2 and an estimating device 3. Note that living organisms include, for example, humans and various animals that are kept or protected.

The biological condition estimation device 1 according to this embodiment includes a transmitting/receiving section 2 that emits a transmitted radio wave, receives a reflected wave of the transmitted radio wave, and mixes the transmitted radio wave and the reflected wave to generate a beat signal. A phase calculation unit (phase calculation task 33a) that calculates the phase of the beat signal output from the transmission and reception unit 2, and data of the beat signal output from the transmission and reception unit 2 and output from the phase calculation unit (phase calculation task 33a) a stationary object reflection removal unit (stationary object reflection removal task 33b) that removes a component corresponding to a reflected wave from a stationary object from the phase data, and a phase calculation unit and the beat signal data output from the transmitting/receiving unit 2. A body motion component removal unit (body motion component removal task 33c) that removes a component corresponding to body motion from each of the phase data output from the phase calculation task 33a, and the phase after the removal of the two components. a speed calculation unit (speed calculation task 33d) that calculates the relative speed between the transmitting/receiving unit 2 and the living body as a target using data; and a movement unit that determines whether or not the living body is in a moving state based on the relative speed. A state estimating unit (moving state estimation task 33e) and a sum of spectrum intensities in a specific frequency range of a frequency spectrum distribution obtained by performing fast Fourier transform on the data of the beat signal after removal of the two components, and the above. a resting state estimation unit (resting state estimation task 33g) that determines whether the living body is in a resting state based on a comparison with the sum of spectral intensities in a predetermined region other than a specific frequency range; an absent state estimation unit (absent state estimation task 33h) that determines whether or not the living body is in the absent state based on whether the frequency distribution shape of each amplitude of the removed beat signal is a normal distribution; , have, like.

The living body state estimation device 1 according to this embodiment further determines that the living body is in a dangerous state when the estimation result of the living body's state changes from a resting state to an absent state without passing through a moving state. It has a dangerous state detection unit (dangerous state detection task 33i) that makes a judgment.

The transmitting/receiving unit 2 is a mechanism for performing a radar scan on a predetermined area and mixing a transmitted wave and a reflected wave to generate a beat signal. It includes a receiving section 22 that detects reflected waves of microwaves. The frequency band of the microwave irradiated from the transmitter 21 is, for example, a 24 GHz band, but is not limited to the 24 GHz band. Further, the predetermined area mentioned above is a space where a living organism to be observed is present, and examples thereof include a living space/activity space of a living organism such as a living space, a rearing space, and a protection area.

The transmitting unit 21 transmits an FMCW (Frequency Modulated Continuous Wave) radar via a transmitting antenna 211 . The FMCW radar is a continuous wave radar that performs frequency modulation over time, and generates and irradiates burst waves containing multiple chirps. Each chirp included in the burst wave is generated with the frequency swept over time (specifically, the frequency changes linearly over time).

The transmitting unit 21 uses a control voltage for generating a transmission signal that changes in the shape of a triangular wave by alternately and continuously providing modulation sections in which the frequency gradually increases and modulation sections in which the frequency gradually decreases on the time axis, for example. A control voltage signal is generated by digital-to-analog conversion of the digital data of the value, and a transmission signal is generated according to the control voltage signal and transmitted to the transmission antenna 211. Further, a part of the transmission signal is also transmitted to the receiving section 22 as a local signal at a predetermined distribution ratio.

The transmitting antenna 211 of the transmitter 21 emits a transmitting radio wave based on the transmitted transmitting signal.

In the FMCW radar, by analyzing the difference between the transmitted wave and the received wave (i.e., the reflected wave), the distance between the transmitting/receiving unit 2 and the target object, the relative speed between the transmitting/receiving unit 2 and the target object, etc. can be determined. Calculated. Note that the modulation width and modulation period of the chirp frequency (that is, the repetition period of the chirp) may be adjusted as appropriate, and it is also possible to make all the chirps have the same waveform or to make some of the chirps have different waveforms. It may be possible to select as appropriate whether to do so or not.

The receiving unit 22 receives reflected waves (specifically, reflected radio waves) of the FMCW radar radiated from the transmitting antenna 211 of the transmitting unit 21 via the receiving antenna 221, and also receives the ToF (ToF) of the reflected waves. It generates and outputs a signal that has been down-converted to an intermediate frequency proportional to Time of Flight (abbreviation for Time of Flight).

For example, the reception antenna 221 of the reception unit 22 receives reflected radio waves that are reflected back from the transmission radio waves radiated from the transmission antenna 211, generates and outputs a reception signal. The received signal is divided at the same level ratio and provided for subsequent processing.

Here, the local signal transmitted from the transmitter 21 to the receiver 22 is divided at the same level ratio, one of the divided signals is supplied as is to subsequent processing, and the other of the divided signals is π The signal is subjected to phase rotation processing of /2 and then supplied to subsequent processing.

Then, the receiving section 22 receives one of the divided signals of the received signal output from the receiving antenna 221 and one of the divided signals of the local signal transmitted from the transmitting section 21 (i.e., π /2 signal that has not been subjected to phase rotation processing) to produce a beat signal that is an intermediate frequency signal (IF signal), specifically, in phase with the transmission signal. Generate component I signals.

The receiving unit 22 also performs π/2 phase rotation on the other divided signal of the received signal output from the receiving antenna 221 and the divided local signal transmitted from the transmitting unit 21. The processed signal is mixed to generate a beat signal that is an intermediate frequency signal (IF signal), specifically, a Q signal that is a quadrature phase component with respect to the transmitted signal. .

The receiving unit 22 then performs unnecessary frequency component removal processing and signal level amplification processing on the I signal as necessary, and then performs analog-to-digital conversion processing to convert the I signal into a digital signal. Generate and output an I signal.

The receiving unit 22 also performs unnecessary frequency component removal processing and signal level amplification processing on the Q signal as necessary, and then performs analog-to-digital conversion processing to convert the Q signal into a digital signal. Generate and output a signal.

Note that ToF is the time it takes for the transmitted radar to be received as a reflected wave, that is, it is a value proportional to the distance between the transmitter/receiver 2 and the object that reflected the radar. By using ToF, the distance between the transmitting/receiving unit 2 and the object (target) that reflected the radar and the speed of the object (target) that reflected the radar are calculated.

Radar scanning by the transmitting/receiving unit 2 is continuously and repeatedly performed at predetermined time intervals. Then, a beat signal converted into a digital signal (specifically, a combination of an I signal and a Q signal) is continuously output from the receiving unit 22 at the predetermined time interval (step S1). .

The estimation device 3 is a mechanism for estimating the state of the living body based on the beat signal output from the transmitting/receiving section 2, and mainly includes an input receiving section 31, a storage section 32, a main task 33, and a central processing unit. A section 34 is provided.

The central processing unit 34 has a function of regulating and controlling each unit constituting the estimation device 3, and is configured by, for example, a processor using a CPU (abbreviation for Central Processing Unit), or is stored in the storage unit 32. It may be configured to realize each function according to a set program.

The input receiving section 31 receives the input of the beat signal (specifically, the combination of the I signal and the Q signal) output from the receiving section 22, and receives the input beat signal (the combination of the I signal and the Q signal). ; "IQ data" in FIG. 2) is stored in the storage unit 32 (step S2).

The storage unit 32 is, for example, a hard disk or a memory, and functions as a storage area/storage device that stores various information, programs, data, and the like. The storage unit 32 may include a RAM (abbreviation for Random Access Memory) that functions as a work area when executing processes related to estimation of the state of the living body.

The main task 33 is a group of tasks and programs for executing processing related to estimating the state of the living body using the beat signal (combination of I signal and Q signal) stored in the storage unit 32. The main tasks 33 mainly include a phase calculation task 33a, a stationary object reflection removal task 33b, a body motion component removal task 33c, a speed calculation task 33d, a moving state estimation task 33e, an FFT processing task 33f, a resting state estimation task 33g, and an absent task. It includes a state estimation task 33h and a dangerous state detection task 33i.

The phase calculation task 33a is a task program for calculating the phase of a beat signal. Specifically, the phase calculation task 33a reads the beat signal (combination of I signal and Q signal) stored in the storage unit 32, and performs orthogonal calculation based on the combination of the mutually corresponding I signal and Q signal. Detection is performed to calculate the phase of each of the combinations (step S3).

The phase calculation task 33a causes the storage unit 32 to store the calculated phase.

The stationary object reflection removal task 33b is a task program for removing components corresponding to reflected waves from stationary objects such as walls and floors, for example, from a signal obtained by performing signal processing on a received signal.

Here, the signal corresponding to the reflected wave (ie, the received signal) output from the reception antenna 221 includes a reflected wave from a stationary object, a reflected wave from a living body, and noise. The stationary object reflection removal task 33b removes the component corresponding to the reflected wave from the stationary object, in order to detect the reflected wave from the living body well and estimate the state of the living body with high accuracy. The stationary object mentioned above is a fixed object that exists in the space where the living organism to be observed resides, and includes, for example, a wall or a floor surrounding the space.

Assuming that the reflected wave from a stationary object is a DC wave (or is almost a DC wave), it is understood as a time average of the phase calculated by the process in step S3. Therefore, the stationary object reflection removal task 33b reads the phase data stored in the storage unit 32, accumulates it by a predetermined amount, calculates the time average of the accumulated phase, and generates a signal corresponding to the calculated time average. , components corresponding to reflected waves from stationary objects are removed by subtracting them from each of the phases calculated by the processing in step S3 (step S4).

The degree of accumulation of phase data to remove components corresponding to reflected waves from stationary objects is not limited to a specific time length or number of data, but is equivalent to reflected waves from stationary objects. The time length and number of data are appropriately set so that an appropriate average can be calculated so that the components can be removed satisfactorily.

The stationary object reflection removal task 33b causes the storage unit 32 to store phase data after removal processing of a component corresponding to a reflected wave from a stationary object (referred to as "first removed phase data").

The body motion component removal task 33c extracts low-period shaking of the entire body (also called "body motion") that occurs regardless of the consciousness of the living organism being observed, from a signal obtained by performing signal processing on the received signal. This is a task program for removing corresponding components.

The method of removing the component corresponding to body movement is not limited to a specific method or procedure, but is any method or procedure that can remove (in other words, suppress) the component corresponding to body movement from phase data. An appropriate one is selected from among them. Specifically, for example, phase data is divided into predetermined time lengths, and components corresponding to body movements are expressed by estimating an N-order function approximation curve (where N is a natural number) of the data in each divided section. A method may be used in which a regression curve is determined and a signal corresponding to the determined regression curve is subtracted from the phase data.

The body motion component removal task 33c reads the first removed phase data stored in the storage unit 32, and removes a component corresponding to body movement from the read first removed phase data (step S5).

The body motion component removal task 33c causes the storage unit 32 to store phase data after the removal process of the component corresponding to the body movement (referred to as "second post-removal process phase data").

The speed calculation task 33d is a task program for calculating the speed related to a living body.

Here, the reflected waves from the living body change in phase and amplitude depending on the movement of the living body, such as the movements of the arms and legs of the living body, breathing, and heartbeat, and the reflected waves corresponding to the movements of the arms and legs of the living body change. , a reflected wave corresponding to the respiratory movement of the living body, and a reflected wave corresponding to the pulsation of the living body. Then, the phase and amplitude of the received signal change with the change in the phase and amplitude of the reflected wave from the living body.

The speed calculation task 33d reads the phase data after the second removal process stored in the storage unit 32, converts the phase displacement into speed (see FIG. 3; the two lines in the left diagram indicate the I signal and the Q signal), and calculate the relative velocity between the transmitter/receiver 2 and the target object (here, particularly a living body) (step S6).

Note that the method for calculating the relative velocity between the transmitting/receiving unit 2 and a living body as a target is not limited to a specific method or procedure, and is based on the movements of the arms and legs of the living body, breathing, and heartbeat. An appropriate method or procedure is selected as appropriate from among the methods and procedures that can calculate the speed associated with the movement of.

The speed calculation task 33d transfers the calculated relative speed to the movement state estimation task 33e.

The movement state estimation task 33e is a task program for determining whether or not a living body serving as a target is moving, based on the relative speed calculation result by the speed calculation task 33d. The movement state estimation task 33e determines whether or not the living body serving as the target is moving based on the relative speed transferred from the speed calculation task 33d (step S7).

Specifically, the movement state estimation task 33e checks the value of the relative velocity transferred from the speed calculation task 33d, and if a speed exceeding the respiration rate of the living body is detected, the movement state estimation task 33e determines that the living body as a target is moving. (See Figure 3). The respiration rate for estimating/discriminating whether it is respiration of a living body or movement is not limited to a specific value, but is appropriately set to an appropriate value based on known knowledge. Specifically, the breathing rate can be set to about 1 cm to 10 cm/sec, for example, because when a person's breathing is difficult, the abdomen expands and contracts approximately once every second, with a maximum width of about 5 cm. It is preferable to set the speed to about 5 cm/sec.

Note that the breathing rate is also related to the process in step S13 later, and is also a rate for distinguishing between a moving state and a resting state. The rate of respiration is therefore set in such a way that the rate of respiration in a resting state can be distinguished from, for example, the rate of turning over. Specifically, for example, when the breathing rate is set to about 5 cm/sec, movements of the living body such as turning over are judged as movement, while movements of the living body accompanying breathing in a resting state are judged as movement. You can prevent it from happening.

If the relative velocity transferred from the velocity calculation task 33d is greater than the breathing velocity (step S7: Yes), the movement state estimation task 33e determines that the living body is moving (in a moving state). Step S8). Then, the main task 33 advances the processing procedure to step S19.

On the other hand, if the relative velocity transferred from the velocity calculation task 33d is less than or equal to the breathing velocity (step S7: No), the movement state estimation task 33e determines that the living body is not moving and moves the processing procedure to step S7. Proceed to S9.

In the processing after step S9, the beat signal (combination of I signal and Q signal: IQ data) stored in the storage unit 32 in the processing in step S2 is used.

The stationary object reflection removal task 33b reads the beat signal (combination of I signal and Q signal: IQ data) from the storage unit 32 (step S9), and extracts the beat signal (combination of I signal and Q signal: IQ data) from each of the I signal and Q signal. Removes components corresponding to reflected waves from objects.

Assuming that the reflected wave from a stationary object is a DC wave (or is almost a DC wave), it is understood as the time average of each of the I signal and Q signal. Therefore, the stationary object reflection removal task 33b reads the data of the I signal and Q signal stored in the storage unit 32, accumulates a predetermined amount, and calculates the time average of each of the accumulated I signal and Q signal. Then, by subtracting a signal corresponding to the calculated time average from each of the I signal and Q signal, components corresponding to reflected waves from stationary objects are removed (step S10).

The degree of accumulation of I signal and Q signal data in order to remove components corresponding to reflected waves from stationary objects is not limited to a specific time length or number of data; An appropriate length of time and number of data are appropriately set so that an appropriate average can be calculated so that components corresponding to waves can be satisfactorily removed.

The stationary object reflection removal task 33b stores data of a combination of an I signal and a Q signal (referred to as "first post-removal IQ data") after processing to remove a component corresponding to a reflected wave from a stationary object in a storage unit. 32 to be memorized.

The body motion component removal task 33c reads the IQ data after the first removal process from the storage unit 32, and extracts from each of the I signal and Q signal the low-period shaking of the entire body that occurs regardless of the consciousness of the living organism being observed. (also called "body movement").

The method for removing components corresponding to body movements is not limited to a specific method or procedure as described above, but specifically, for example, removing I signal data and Q signal data for a predetermined length of time, A regression curve representing a component corresponding to body movement is obtained by estimating an N-order function approximation curve (where N is a natural number) of data for each section and section, and a signal corresponding to the obtained regression curve is A method of subtracting from each of the data of the I signal and the data of the Q signal may be used.

The body movement component removal task 33c reads the IQ data after the first removal process stored in the storage unit 32, and removes the component corresponding to the body movement from the read IQ data after the first removal process (step S11 ).

The body motion component removal task 33c causes the storage unit 32 to store data of the combination of the I signal and the Q signal after the component corresponding to the body motion has been removed (referred to as "second post-removal IQ data"). .

The FFT processing task 33f is a task program for generating a frequency spectrum of a beat signal using fast Fourier transform (FFT) processing.

The FFT processing task 33f generates a frequency spectrum of the beat signal using fast Fourier transform processing for the IQ data after the second removal process as a complex number beat signal with the I signal as the real part and the Q signal as the imaginary part. (Step S12).

Then, the FFT processing task 33f causes the storage unit 32 to store the generated frequency spectrum.

The resting state estimation task 33g is a task program for determining whether the living body is in a resting state. The resting state estimation task 33g determines whether the living body is in a resting state by paying attention to the signal distribution in the frequency domain.

In the frequency spectrum distribution when no living body is present, the spectral intensity is distributed at approximately the same intensity (in other words, without large fluctuations) in all frequency regions (i.e., 0 Hz to ∞ Hz) (Figure 4). In the frequency spectrum distribution when a living body is present and in a resting state, the spectral intensity fluctuates significantly in a specific frequency range compared to other frequencies (see (B) in the same figure).

Therefore, the resting state estimation task 33g calculates the sum of the spectral intensities in a specific frequency range (referred to as "specific range intensity total value") and the sum of the spectral intensities in a predetermined area other than the specific frequency range (referred to as "other area intensity total value"). It is determined whether the living body is in a resting state by comparing the total strength value (called "total strength value").

The frequency range in which the spectral intensity varies when the living body is at rest (referred to as the "specific range") is approximately 0.25 Hz to 3.25 Hz. Specifically, if breathing is approximately once every 3 seconds, the spectral intensity corresponding to the movement of breathing will be distributed around 0.3Hz, and if the heartbeat is approximately once every 1 second, the spectral intensity corresponding to the pulsation will be distributed around 0.3Hz. Distributed around 1Hz. Furthermore, in the case of humans, it is said that the upper limit of the frequency of biological activity that can occur is approximately 200 heartbeats/minute (which can cause convulsions).

Based on the above, for example, the specific range may be set to about 0Hz to less than 5Hz, and a predetermined area other than the specific range (referred to as "other area") may be set to about 5Hz to 20Hz, or the specific range may be set to about 0.25Hz. 3.25 Hz to less than 3.25 Hz, and other regions are set to about 3.25 Hz to 10 Hz.

The resting state estimation task 33g reads the frequency spectrum stored in the storage unit 32, and calculates the specific range intensity total value and the other area intensity total value, respectively. The resting state estimation task 33g then determines whether the specific range total strength value is larger than the other area strength total value (step S13).

If the specific range intensity total value is larger than the other area intensity total value (step S13: Yes), the resting state estimation task 33g determines that the living body is in a resting state (step S14). Then, the main task 33 advances the processing procedure to step S19.

On the other hand, if the specific range intensity total value is less than or equal to the other area intensity total value (step S13: No), the resting state estimation task 33g advances the processing procedure to step S15.

Note that the resting state estimation task 33g compares the average value of the spectral intensity in the specific range with the average value of the spectral intensity in other areas, and determines whether the average value of the spectral intensity in the specific range is higher than the average value of the spectral intensity in the other areas. The living body may be determined to be in a resting state when the amount is also large.

In the processes after step S15, the second removed IQ data ("subtraction signal" in FIG. 2) stored in the storage unit 32 in the process in step S11 is used.

The absent state estimation task 33h is a task program for determining whether or not a living body is absent. The absent state estimation task 33h determines whether or not the living body is absent, paying attention to the fact that the frequency distribution according to the amplitude of the white noise follows a normal distribution (step S16).

The absent state estimation task 33h tests the normality of the frequency distribution according to the amplitude of the beat signal (specifically, the I signal and the Q signal) as the IQ data after the second removal process. Specifically, the absent state estimation task 33h reads the I signal and Q signal as the second post-removal IQ data stored in the storage unit 32, accumulates a predetermined amount, and stores the accumulated I signal. The frequency of each amplitude is calculated for each of the signal and the Q signal (see FIG. 5). Regarding Figure 5, one line in the upper figure represents the temporal change in the amplitude of the I signal, and the lower figure represents the frequency distribution of the I signal by amplitude size. The figure shows an example where the frequency distribution type is a normal distribution, and the lower right figure is an example where the frequency distribution type is not a normal distribution.

The degree of accumulation of I signal and Q signal data for calculating the frequency for each amplitude size is not limited to a specific time length or number of data, but is based on the frequency distribution for each amplitude size. The appropriate time length and number of data are appropriately set so that an appropriate frequency for each amplitude size can be calculated so that it can be determined satisfactorily whether or not the amplitude follows a normal distribution. Alternatively, ranks may be set for the magnitude of the amplitude divided by a predetermined pitch, and the frequency may be calculated for each rank of the magnitude of the amplitude.

If the frequency distribution shape of each amplitude of the I signal or Q signal is a normal distribution, it is determined that a living body is absent. Note that the test of whether the frequency distribution shape of each amplitude of the I signal or Q signal is a normal distribution is to check whether the degree of conformity of the frequency distribution shape to the normal distribution is greater than or equal to a predetermined degree of conformity. It may also be done by Further, the test as to whether or not the frequency distribution shape of each amplitude is a normal distribution may be performed on only one of the I signal and the Q signal, or may be performed on only one of the I signal and the Q signal. The process may be performed for both the Q signal and the Q signal.

If it is determined that the frequency distribution shape according to the magnitude of the amplitude of the beat signal is a normal distribution (step S16: Yes), the absent state estimation task 33h determines that the living body is absent (in an absent state). A judgment is made (step S17). Then, the main task 33 advances the processing procedure to step S19.

As described above, the living body state estimation device 1 first determines whether the living body is making a large movement corresponding to the moving state (step S7), and then it is determined that the living body is not making a large movement. In this case, it is determined whether the living body is making small (or minute) movements corresponding to a resting state (step S13), and further, if it is determined that the living body is not making small movements, it is determined whether the living body is in an absent state or not. It is determined whether or not (step S16). In this way, the state of the living body can be estimated efficiently and accurately without waste by classifying and specifying the states step by step in descending order of movement, such as the moving state, the resting state, and the absent state.

On the other hand, if it is determined that the frequency distribution shape according to the magnitude of the amplitude of the beat signal is not a normal distribution (step S16: No), the absent state estimation task 33h determines that there is interference (step S18). ). Then, the main task 33 suspends the estimation result of the biological state and advances the processing procedure to step S19.

The dangerous state detection task 33i is a task program for detecting that a living body is in a dangerous state. The dangerous state detection task 33i detects whether the living body is in danger based on the combination of the immediately previous state of the living body in the time series and the newly estimated state of the living body, in other words, based on the pattern of changes in the state of the living body in the time series. It is determined whether it is in the state (step S19).

For example, in the time-series transition of the estimated results of the state of the living body, if the state of the living body changes from a resting state to an absent state without passing through a moving state, the dangerous state detection task 33i changes the vital state from a state with vitals to an absent state without passing through a moving state. In other words, there was a direct transition from a breathing state to a non-breathing state, so it is determined that the living body is in a dangerous state. In this case, the dangerous state detection task 33i may notify that the living body is in a dangerous state by emitting an alarm sound or sending an email to a pre-registered address, for example.

In addition, in the time-series transition of the estimation results of the state of the living organism, if the state changes from the absent state to the resting state without passing through the moving state, for example, remote control or autonomous control may be performed in the predetermined area of the observation target. It may be determined that a disturbance such as the activation of electrical equipment or the occurrence of an earthquake has occurred, and it may be determined that another abnormal event that is not related to the state of the living body has occurred.

The main task 33 then returns the processing procedure to step S2. As mentioned above, the radar scan by the transmitting/receiving section 2 is continuously and repeatedly performed at predetermined time intervals, and the combination of the digitalized I signal and Q signal is sent from the receiving section 22 at the predetermined time interval. It is output continuously (step S1).

According to the biological state estimation device 1 described above, components corresponding to reflected waves from stationary objects such as walls and floors are autonomously removed, so a threshold value can be set for each usage environment. In other words, it is possible to determine with high precision whether the living body is absent, at rest, or moving, without being affected by or depending on the usage environment. According to the biological state estimating device 1 described above, the state is classified and specified in stages in descending order of movement, such as a moving state, a resting state, and an absent state, so that efficiency can be achieved without waste. It becomes possible to estimate the condition of the living body in a highly accurate and accurate manner.

Although the embodiments of the present invention have been described above, the specific configuration is not limited to the above embodiments, and even if there are changes in the design within the scope of the gist of the present invention. Included in invention. For example, in the above embodiment, the setting of the threshold value for determining the breathing rate and specific range was mainly explained in the case where the observation target is a human, but the living organism to be observed in this invention is not limited to humans. However, the present invention may be applied to various animals as observation objects.

Further, in the above embodiment, the main task 33 includes the dangerous state detection task 33i, but the dangerous state detection task 33i is not an essential component in the present invention. That is, the basic configuration of the present invention is a configuration for estimating which of the moving state, the resting state, and the absent state.

In addition, multiple transmitter/receivers are installed in a predetermined area (i.e., the space where the living organisms to be observed are present), and the received data obtained is separated for each target object using well-known technology. The states may be estimated simultaneously.

This invention can estimate the condition of a living body based on the vital information of people and various animals, so it can be used, for example, to remotely monitor elderly people at home, observe the condition of patients undergoing medical treatment, and use domestic animals and other animals. It can be applied in fields such as monitoring the condition of animals under protection. It can also be applied to the field of intrusion monitoring into homes and facilities.

1 biological state estimation device 2 transmitting/receiving section 21 transmitting section 211 transmitting antenna 22 receiving section 221 receiving antenna 3 estimating device 31 input receiving section 32 storage section 33 main task 33a phase calculation task 33b stationary object reflection removal task 33c body motion component removal Tasks 33d Speed calculation task 33e Moving state estimation task 33f FFT processing task 33g Resting state estimation task 33h Absence state estimation task 33i Dangerous state detection task 34 Central processing unit

Claims (3)

a transmitting/receiving unit that emits a transmitted radio wave, receives a reflected wave of the transmitted radio wave, and mixes the transmitted radio wave and the reflected wave to generate a beat signal;
a phase calculation unit that calculates the phase of the beat signal output from the transmission and reception unit;
a stationary object reflection removing section that removes a component corresponding to a reflected wave from a stationary object from each of the beat signal data output from the transmitting and receiving section and the phase data output from the phase calculation section;
a body movement component removing unit that removes a component corresponding to body movement from each of the beat signal data output from the transmitting and receiving unit and the phase data output from the phase calculation unit;
a speed calculating unit that calculates a relative speed between the transmitting/receiving unit and a living body as a target using the phase data after removing the two components;
a moving state estimator that determines that the living body is in a moving state when the relative velocity is greater than the respiration rate of the living body ;
The relative velocity is equal to or lower than the respiration rate of the living body, and the frequency range is within a specific frequency range of the frequency spectrum distribution obtained by fast Fourier transforming the beat signal data after the two components have been removed. a resting state estimation unit that determines that the living body is in a resting state when the sum of spectral intensities is greater than the sum of spectral intensities in a predetermined region other than the specific frequency range;
If the sum of the spectral intensities in the specific frequency range is less than or equal to the sum of the spectral intensities in a predetermined area other than the specific frequency range, the frequency of each amplitude of the beat signal after removal of the two components. an absent state estimation unit that determines whether the living body is in an absent state based on whether the distribution shape is a normal distribution;
the specific frequency range is set from 0Hz to less than 5Hz;
A biological state estimation device characterized by: The respiration rate of the living body is set to 1 cm to 10 cm/sec.
The biological condition estimation device according to claim 1, characterized in that: further comprising a dangerous state detection unit that determines that the living body is in a dangerous state when the estimation result of the state of the living body transitions from the resting state to the absent state without passing through the moving state;
The biological condition estimation device according to claim 1 or 2, characterized in that:

Images