Sensor, estimation device, estimation method, and program recording medium

By calculating the complex transfer function matrix using an N×M antenna element sensor, extracting the components of biological activity, and using the Doppler RCS threshold for determination, the problem of low accuracy and long time required for wireless signal estimation of organism posture and movement in existing technologies is solved, thus achieving efficient detection of organism status.

CN110673135BActive Publication Date: 2024-11-08PANASONIC INTELLECTUAL PROPERTY MANAGEMENT CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN201910573748.3
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Priority Date
2019-03-12
Filing Date
2019-06-28
Publication Date
2024-11-08
Estimated Expiration
2039-06-28

AI Technical Summary

Technical Problem

Existing technologies for estimating the posture and movement of organisms using wireless signals suffer from problems such as low detection accuracy, long detection time, and inability to accurately determine the movement state of the organism.

Method used

Using a sensor with N transmitting antenna elements and M receiving antenna elements, the components affecting life activities are extracted by calculating the complex transfer function matrix. The position and posture of the organism are calculated by combining the Doppler RCS, and the movement state of the organism is determined by using the Doppler RCS threshold.

Benefits of technology

It enables the rapid and highly accurate estimation of an organism's location, posture, and movements, allowing for efficient detection of organism conditions while protecting privacy.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN110673135B_ABST
    Figure CN110673135B_ABST
Patent Text Reader

Abstract

The present invention provides a sensor, an estimation device, an estimation method and a storage medium. By using wireless signals, the posture of a living body can be estimated in a short time and with high accuracy, and is characterized by comprising: a transmission signal generator having N (N is a natural number greater than 3) transmission antenna elements, each of which transmits a signal to a specified range where a living body may exist; a receiver having M (M is a natural number greater than 3) receiving antenna elements, each of which receives N reception signals, and the N reception signals include a reflection signal obtained by a part of the N transmission signals transmitted by the N transmission antenna elements being reflected by the living body.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present disclosure relates to a sensor for estimating conditions such as posture and behavior of a living body by using wireless signals. Background Art

[0002] As a method of knowing the position of a person, etc., methods using wireless signals have been studied (for example, refer to patent documents 1 to 3). Patent document 1 discloses a method of detecting a living body using a Doppler sensor, and patent document 2 discloses a method of detecting a person's movements and living body information using a Doppler sensor and a filter. Patent document 3 discloses a technology for knowing the position and state of a person who is the detection object by analyzing a component containing a Doppler frequency shift using Fourier transform. In addition, patent document 4 discloses a device for estimating the state of a person using a Doppler sensor, patent document 5 discloses a device for estimating the posture of a person using a state estimation device, and patent document 6 discloses a device for estimating the position of a person or an object using a camera or an RF tag.

[0003] Prior art literature

[0004] Patent Literature

[0005] Patent Document 1: Japanese Patent Application No. 2014-512526

[0006] Patent Document 2: International Publication No. 2014 / 141519

[0007] Patent Document 3: Japanese Patent Application Publication No. 2015-117972

[0008] Patent Document 4: Japanese Patent Application Publication No. 2011-215031

[0009] Patent Document 5: Japanese Patent Application Publication No. 2018-8021

[0010] Patent Document 6: Japanese Patent Application Publication No. 2008-275324 Summary of the invention

[0011] Problems to be solved by the present invention

[0012] However, further improvements are needed to estimate the condition of a living body, such as the posture and behavior of the living body, by utilizing wireless signals.

[0013] Means of solving the problem

[0014] In order to achieve the above-mentioned object, a sensor of one embodiment of the present invention comprises: a transmission signal generator having N transmission antenna elements, wherein the N transmission antenna elements respectively transmit transmission signals to a specified range where a biological body may exist, and N is a natural number greater than 3; a receiver having M receiving antenna elements, wherein the M receiving antenna elements respectively receive N reception signals including reflection signals, wherein the reflection signals are obtained when a part of the N transmission signals respectively transmitted by the N transmission antenna elements is reflected or scattered by the biological body, and M is a natural number greater than 3; a circuit; and a memory, wherein the circuit comprises: a first matrix calculation unit, which calculates the matrix of the transmission signals received by each of the M receiving antenna elements during a specified period of time; For each of the N received signals, for each N×M combination formed by combining the N transmitting antenna elements and the M receiving antenna elements one-to-one, a complex transfer function representing the propagation characteristics between the transmitting antenna element and the receiving antenna element in the combination is calculated, and a first N×M matrix having the calculated N×M complex transfer functions as components is calculated one by one; a second matrix extraction unit is configured to successively extract the second matrix corresponding to the component affected by life activities by successively extracting the second matrix corresponding to the specified frequency range in the first matrix successively calculated by the first matrix calculation unit, wherein the life activities include at least one of the breathing, heartbeat and body movement of the organism. a presence determination unit for determining whether a living body is present within the specified range by a specified method; a position estimation unit for successively estimating the position of the living body relative to the sensor using the second matrix successively extracted by the second matrix extraction unit after the presence determination unit determines that the living body is present within the specified range; a Doppler RCS calculation unit for (i) successively calculating a first distance and a second distance based on the position of the living body successively estimated by the position estimation unit, the position of the transmitting signal generator, and the position of the receiver, wherein the first distance represents the distance between the successively estimated position of the living body and the transmitting signal generator, and the second distance represents the distance between the successively estimated position of the living body and the transmitting signal generator, and the second distance represents the distance between the successively estimated position of the living body and the transmitting signal generator. The distance represents the distance between the organism and the receiver, and (ii) using the calculated first distance and the second distance, the Doppler radar cross-section (RCS) value for the organism is calculated successively; a movement determination unit, which stores the successively estimated positions of the organism in the memory a predetermined number of times in the estimated order, and uses the positions of the organism stored in the memory the predetermined number of times to determine that the organism is moving when the displacement of the position of the organism is greater than a predetermined value, and determines that the organism is not moving when the displacement is less than the predetermined value; a Doppler RCS threshold setting unit, which sets the Doppler RCS threshold by a predetermined method;A Doppler RCS threshold determination unit compares the calculated Doppler RCS value with the set Doppler RCS threshold to determine whether the Doppler RCS value is less than the Doppler RCS threshold; and a situation estimation unit selectively estimates at least one of the posture of the living body and the action of the living body based on at least one of the determination result of the movement determination unit and the determination result of the Doppler RCS threshold determination unit. ;

[0015] In addition, the present disclosure can be implemented not only as a sensor, but also as an integrated circuit having a processing unit of such a sensor, or as a method of setting the processing unit constituting the device as steps, or as a program that causes a computer to execute these steps, or as information, data or signals representing the program. Moreover, these programs, information, data and signals can also be distributed via a recording medium such as a CD-ROM or a communication medium such as the Internet.

[0016] Effects of the Invention

[0017] According to the present disclosure, by using wireless signals, the condition of a living body can be estimated in a short time and with high accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 This is a block diagram showing an example of the configuration of a sensor in the first embodiment.

[0019] Figure 2 This is a block diagram showing the functional configuration of the circuit and memory in the first embodiment.

[0020] Figure 3 This is a diagram showing a Doppler RCS threshold setting value table in the sensor according to the first embodiment.

[0021] Figure 4 This is a diagram showing an example of posture estimation by the posture estimating unit in the first embodiment.

[0022] Figure 5 This is a block diagram showing an example of the configuration of the behavior estimating unit in the first embodiment.

[0023] Figure 6 This is a diagram for explaining an example of extracting an action period from time series data of an altitude position (Height) or a Doppler RCS value.

[0024] Figure 7 This is a diagram for explaining the process of converting into a direction vector.

[0025] Figure 8 This is a diagram showing an example of a direction code table.

[0026] Fig. 9 This is a diagram showing an example of time series data of calculated direction codes and distances.

[0027] Fig.10 This is a diagram showing test data obtained by measurement and model data as a model code.

[0028] Fig.11 This is a flowchart showing an example of the operation of the sensor in the first embodiment.

[0029] Fig.12 This is a flowchart showing a first example of the behavior estimation process in the first embodiment.

[0030] Fig.13 This is a flowchart showing a second example of the behavior estimation process in the first embodiment.

[0031] Fig.14 This is a diagram showing an example of a previous posture, a transitionable posture, and a behavior in the behavior estimating unit according to the first embodiment.

[0032] Fig.15 This is a flowchart showing a third example of the behavior estimation process in the first embodiment.

[0033] Fig.16 This is a flowchart showing another example of the operation of the sensor in the first embodiment.

[0034] Fig.17 This is a flowchart showing another example of the operation of the sensor in the first embodiment.

[0035] Fig.18 This is a flowchart showing another example of the operation of the sensor in the first embodiment.

[0036] Fig.19 This is a block diagram showing the functional configuration of a circuit and a memory in the second embodiment.

[0037] Fig. 20 This is a flowchart showing an example of the operation of the sensor in the second embodiment.

[0038] Fig.21 This is a flowchart showing an example of the operation of the posture probability estimation unit in the second embodiment.

[0039] Fig. 22 This is a diagram showing an example of teacher data processing used for posture probability estimation in the second embodiment.

[0040] Fig.23 This is a diagram showing an example of normalization of teacher data obtained by posture probability estimation and posture estimation in the second embodiment.

[0041] Fig.24 This is a diagram showing an example of action probability estimation in the action probability estimation unit in the second implementation mode.

[0042] [Description of Reference Numerals]

[0043] 10 Sensors

[0044] 20 Send signal generator

[0045] 21 Transmitting antenna element

[0046] 30 Receiver

[0047] 31 Receiving antenna element

[0048] 40, 40A circuit

[0049] 41 Memory

[0050] 42 Information indicating the corresponding status

[0051] 50 Organisms

[0052] 410 First matrix calculation unit

[0053] 420 Second matrix extraction unit

[0054] 430 Number Estimation Department

[0055] 440 Position Estimation Unit

[0056] 450 Doppler RCS calculation unit

[0057] 460 Movement Determination Unit

[0058] 470 Doppler RCS threshold setting unit

[0059] 480 Doppler RCS threshold determination unit

[0060] 490 Posture Estimation Unit

[0061] 491 Posture Probability Estimation Unit

[0062] 500 Action Presumption Department

[0063] 501 Action Probability Estimation Department

[0064] 510 Action period extraction unit

[0065] 511 Direction code calculation unit

[0066] 512 Action Comparison Unit

[0067] 520 Situation Estimates Department DETAILED DESCRIPTION

[0068] (Foundation that forms the basis of the present invention)

[0069] The inventors have studied in detail the prior art related to estimating the posture, behavior, etc. of a living body using wireless signals. As a result, it is known that the methods of Patent Documents 1 and 2 can detect the presence or absence of a person, but have a problem of being unable to detect the direction, position, size, posture, etc. of the person.

[0070] In addition, the method of Patent Document 3 has the problem that it is difficult to detect the direction or position of a living body such as a person in a short time and with high accuracy. This is because the frequency change caused by the Doppler effect derived from the activity of the living body is extremely small, and in order to observe the frequency change by Fourier transform, it is necessary to observe for a long time (e.g., tens of seconds) in a static posture of the living body. In addition, it is usually rare for a living body to maintain the same posture or position for tens of seconds.

[0071] In addition, Patent Document 4 discloses a device that uses a Doppler sensor to estimate the presence, rest, and activity of a person, but there is a problem that the user's posture is unknown. In addition, in Patent Document 5, although a sensor can be used to estimate the posture of a person, there is a problem that it is unknown whether the person is moving or in action. Patent Document 6 discloses a device that uses an electronic tag, camera image, etc. to estimate the movement of a person or goods, but in this technology, there is a privacy problem because images are captured.

[0072] The inventors have repeatedly studied the above problems and found that by using the propagation characteristics and scattering cross-sectional area (Doppler RCS (Rader cross-section) value) of the reflected signal sent from a transmission signal generator including antenna elements placed at different positions and reflected by the organism, the direction, position and size of the organism can be estimated. The inventors have thus completed the present invention by estimating the condition of the organism as shown by the movement of the organism, the posture of the organism at the position where the organism exists, and the action of the organism in a short time and with high accuracy.

[0073] That is, a sensor of one embodiment of the present disclosure comprises: a transmission signal generator having N transmission antenna elements, wherein the N transmission antenna elements respectively transmit transmission signals to a specified range where a biological body may exist, wherein N is a natural number greater than 3; a receiver having M reception antenna elements, wherein the M reception antenna elements respectively receive N reception signals including reflection signals, wherein the reflection signals are obtained by reflecting or scattering a part of the N transmission signals respectively transmitted by the N transmission antenna elements by the biological body, wherein M is a natural number greater than 3; a circuit; and a memory, wherein the circuit comprises: a first matrix calculation unit, which calculates a matrix based on the matrix received by each of the M reception antenna elements during a specified period. For each of the N received signals, for each N×M combination formed by combining the N transmitting antenna elements and the M receiving antenna elements one-to-one, a complex transfer function representing the propagation characteristics between the transmitting antenna element and the receiving antenna element in the combination is calculated, and a first N×M matrix having the calculated N×M complex transfer functions as components is calculated one by one; a second matrix extraction unit is configured to successively extract the second matrix corresponding to the component affected by life activities by successively extracting the second matrix corresponding to the specified frequency range in the first matrix successively calculated by the first matrix calculation unit, wherein the life activities include at least one of the breathing, heartbeat and body movement of the organism; without In the determination unit, a presence determination is performed by a predetermined method to determine whether a living body exists within the predetermined range; a position estimation unit, after the presence determination unit determines that the living body exists within the predetermined range, successively estimates the position of the living body relative to the sensor using the second matrix successively extracted by the second matrix extraction unit; a Doppler RCS calculation unit, (i) based on the position of the living body successively estimated by the position estimation unit, the position of the transmitting signal generator and the position of the receiver, successively calculates a first distance and a second distance, wherein the first distance represents the distance between the successively estimated position of the living body and the transmitting signal generator, and the second distance represents the distance between the successively estimated position of the living body and the transmitting signal generator, and the second distance represents the distance between the successively estimated position of the living body and the transmitting signal generator. indicating the distance between the organism and the receiver, and (ii) using the calculated first distance and the second distance, successively calculating the Doppler radar cross-section (RCS) value for the organism; a movement determination unit, storing the successively estimated positions of the organism in the memory a predetermined number of times in the estimated order, and using the positions of the organism stored in the memory the predetermined number of times, determining that the organism is moving when the displacement of the position of the organism is greater than a predetermined value, and determining that the organism is not moving when the displacement is less than the predetermined value; a Doppler RCS threshold setting unit, setting the Doppler RCS threshold by a predetermined method;A Doppler RCS threshold determination unit compares the calculated Doppler RCS value with the set Doppler RCS threshold to determine whether the Doppler RCS value is less than the Doppler RCS threshold; and a situation estimation unit selectively estimates at least one of the posture of the living body and the action of the living body based on at least one of the determination result of the movement determination unit and the determination result of the Doppler RCS threshold determination unit. ;

[0074] Therefore, the condition of the living body, which is any one of the movement of the living body, the posture of the living body, and the behavior of the living body, can be estimated in a short time and with high accuracy.

[0075] In addition, it may be that when the movement determination unit determines that the organism is not moving and the Doppler RCS value is determined by the Doppler RCS threshold determination unit to be below the Doppler RCS threshold, the condition estimation unit estimates the posture of the organism; when the Doppler RCS threshold determination unit determines that the Doppler RCS value is greater than the Doppler RCS threshold, the condition estimation unit estimates the action of the organism.

[0076] Therefore, when the Doppler RCS value is below the Doppler RCS threshold, it is determined that the organism is stationary and the posture of the organism is estimated, so the posture of the organism can be estimated with good accuracy. In addition, when the Doppler RCS value is greater than the Doppler RCS threshold, it is determined that the organism is moving and the action of the organism is estimated, so the action of the organism can be estimated with good accuracy.

[0077] Furthermore, when the movement determination unit determines that the living body is moving, the situation estimation unit may determine that the behavior of the living body is moving.

[0078] Therefore, it is possible to accurately estimate that the living body is moving.

[0079] In addition, it may also be that at least three of the N transmitting antenna elements are respectively arranged at different positions in the vertical direction and the horizontal direction, and at least three of the M receiving antenna elements are respectively arranged at different positions in the vertical direction and the horizontal direction, and the memory stores a Doppler RCS value and information indicating the correspondence between the position of the organism in the vertical direction relative to the sensor, i.e., the vertical position, and the posture of the organism, and the position estimation unit estimates a three-dimensional position as the position of the organism, the three-dimensional position including the position of the organism in the vertical direction relative to the sensor, i.e., the vertical position, and the condition estimation unit estimates the posture of the organism using the calculated Doppler RCS value, the vertical position, and the information indicating the correspondence stored in the memory.

[0080] Therefore, the position where the living body exists and the posture of the living body at that position can be estimated in a short time and with high accuracy.

[0081] In addition, it may also be that at least three of the N transmitting antenna elements are respectively arranged at different positions in the vertical direction and the horizontal direction, and at least three of the M receiving antenna elements are respectively arranged at different positions in the vertical direction and the horizontal direction, the memory stores a Doppler RCS value and information indicating the correspondence between the position of the organism in the vertical direction relative to the sensor, i.e., the vertical position, and the posture of the organism, the position estimation unit estimates a three-dimensional position as the position of the organism, the three-dimensional position including the position of the organism in the vertical direction relative to the sensor, i.e., the vertical position, and the condition estimation unit uses the calculated Doppler RCS value, the vertical position, and the information indicating the correspondence to estimate the posture probability as the posture of the organism, the posture probability being the respective probabilities of one or more predetermined postures that the organism can take.

[0082] Therefore, it is possible to estimate the position where the living body exists and the probabilities of each of one or more postures that the living body at the position can take, that is, the posture probabilities, in a short time and with high accuracy.

[0083] In addition, it may be that the situation estimation unit successively estimates the posture of the organism using the successively calculated Doppler RCS values ​​and the successively estimated position of the organism, and then stores the successively estimated posture of the organism in the memory, and the Doppler RCS threshold setting unit sets the Doppler RCS threshold based on the posture of the organism estimated at the latest timing among the postures of the organism stored in the memory.

[0084] Therefore, the Doppler RCS threshold can be set to an appropriate value according to the posture of the living body.

[0085] In addition, it may also be that the situation estimating unit, after successively estimating the posture of the organism using the successively calculated Doppler RCS values ​​and the successively estimated position of the organism, stores the successively estimated posture of the organism in the memory, then, before estimating the posture of the organism, waits until the successively calculated Doppler RCS values ​​become below the Doppler RCS threshold, estimates the posture of the organism estimated based on the N received signals when the Doppler RCS value becomes below the Doppler RCS threshold as a subsequent posture, and uses the last estimated posture of the organism, i.e., the previous posture, and the estimated subsequent posture stored in the memory to estimate the movement of the organism.

[0086] Therefore, the behavior of a living body can be estimated in a short time and with high accuracy.

[0087] In addition, it may also be that the situation estimating unit, after successively estimating the posture of the organism using the successively calculated Doppler RCS values ​​and the successively estimated position of the organism, stores the successively estimated posture of the organism in the memory, waits until the successively calculated Doppler RCS values ​​become below the Doppler RCS threshold, and when the Doppler RCS value becomes below the Doppler RCS threshold, estimates one or more next actions of the organism using the last estimated posture of the organism stored in the memory, i.e., the previous posture, and estimates the action of the organism by determining one action from the one or more estimated next actions using the successively calculated Doppler RCS values ​​and the successively estimated position of the organism.

[0088] Therefore, the behavior of a living body can be estimated in a short time and with high accuracy.

[0089] In addition, it may also be that the situation estimating unit, after successively estimating the posture of the organism using the successively calculated Doppler RCS values ​​and the successively estimated position of the organism, stores the successively estimated posture of the organism in the memory, waits until the successively calculated Doppler RCS values ​​become below the Doppler RCS threshold, and when the Doppler RCS value becomes below the Doppler RCS threshold, estimates one or more next actions of the organism using the last estimated posture of the organism, i.e., the previous posture, stored in the memory, and estimates the respective probabilities of the estimated one or more next actions, i.e., the action probabilities, as the actions of the organism by using the successively calculated Doppler RCS values ​​and the successively estimated position of the organism.

[0090] Therefore, the probability of each of one or more next actions, that is, the action probability can be estimated in a short time and with high accuracy.

[0091] Therefore, the movement of a living body, the posture or behavior of a living body at a position where the living body exists can be estimated in a short time and with high accuracy.

[0092] In addition, the presence / absence determination unit may presume the number of the organisms present in the specified range, and when the number of the organisms is 0, determine that the organisms are not present in the specified range, and when the number of the organisms is 1 or more, determine that the organisms are present in the specified range.

[0093] Therefore, it is possible to determine whether or not a living body exists in a predetermined range by using the estimation of the number of living bodies present in the predetermined range. Thus, the number of living bodies can be estimated simultaneously with the estimation of the living body conditions.

[0094] In addition, another embodiment of the present invention is an estimation device having a circuit and a memory, wherein the circuit comprises: an acquisition unit, which acquires N reception signals received in a specified period in each of the M reception antenna elements from a receiver having M reception antenna elements, wherein the M reception antenna elements respectively receive the N reception signals, wherein the N reception signals include a reflection signal, wherein the reflection signal is obtained by a part of the N transmission signals transmitted by a transmission signal generator having N transmission antenna elements being reflected or scattered by a biological body, wherein the N transmission antenna elements respectively transmit transmission signals to a specified range where the biological body may exist, wherein M is a natural number greater than 3, and N is a natural number greater than 3; a matrix calculation unit, which calculates, for each of the N received signals obtained, a complex transfer function representing a propagation characteristic between the transmitting antenna element and the receiving antenna element in each N×M combination formed by combining the N transmitting antenna elements and the M receiving antenna elements one-to-one, and successively calculates an N×M first matrix having the calculated N×M complex transfer functions as components; and a second matrix extraction unit, which successively extracts second matrices corresponding to components affected by life activities by successively extracting second matrices corresponding to a specified frequency range in the first matrix successively calculated by the first matrix calculation unit, wherein the life activities include breathing, heartbeat, and body movement of the organism. at least one of the movements; a presence determination unit, which determines whether a living body is within the specified range by a specified method; a position estimation unit, which, after the presence determination unit determines that the living body is within the specified range, successively estimates the position of the living body relative to the transmitting signal generator and the receiver using the second matrix successively extracted by the second matrix extraction unit; a Doppler RCS calculation unit, which (i) successively calculates a first distance and a second distance based on the position of the living body, the position of the transmitting signal generator, and the position of the receiver successively estimated by the position estimation unit, wherein the first distance represents the distance between the living body and the transmitting signal generator, The second distance represents the distance between the organism and the receiver, and (ii) using the calculated first distance and the second distance, successively calculating the Doppler radar cross-section (RCS) value for the organism; a movement determination unit, storing the successively estimated positions of the organism in the memory a predetermined number of times in the estimated order, and using the positions of the organism stored in the memory the predetermined number of times, when the displacement of the position of the organism is greater than a predetermined value, determining that the organism is moving, and when it is less than the predetermined value, determining that the organism is not moving; a Doppler RCS threshold setting unit, setting the Doppler RCS threshold by a predetermined method;A Doppler RCS threshold determination unit compares the calculated Doppler RCS value with the set Doppler RCS threshold to determine whether the Doppler RCS value is less than the Doppler RCS threshold; a situation estimation unit selectively estimates the posture of the living body and the behavior of the living body based on at least one of the determination result of the movement determination unit and the determination result of the Doppler RCS threshold determination unit.

[0095] Therefore, the condition of the living body, which is any one of the movement of the living body, the posture of the living body, and the behavior of the living body, can be estimated in a short time and with high accuracy.

[0096] In addition, the present disclosure can be implemented not only as a sensor, but also as an integrated circuit having a processing unit of such a sensor, or as a method of setting the processing unit constituting the device as steps, or as a program that causes a computer to execute these steps, or as information, data or signals representing the program. Moreover, these programs, information, data and signals can also be distributed via a recording medium such as a CD-ROM or a communication medium such as the Internet.

[0097] Hereinafter, the embodiments of the present disclosure will be described in detail using the accompanying drawings. In addition, the embodiments described below all represent a preferred specific example of the present disclosure. The numerical values, shapes, materials, structural elements, configuration positions of structural elements, connection methods, steps, order of steps, etc. shown in the following embodiments are examples and are not intended to limit the present disclosure. In addition, regarding the structural elements in the following embodiments that are not recorded in the independent claims representing the highest concept of the present disclosure, they are described as arbitrary structural elements that constitute a more preferred method. Note that in this specification and the drawings, the same figure marks are given to structural elements having substantially the same functional structure, and repeated descriptions of these structural elements are omitted.

[0098] (Implementation Method 1)

[0099] Figure 1 This is a block diagram showing an example of the configuration of a sensor in the first embodiment.

[0100] like Figure 1 As shown, the sensor 10 includes a transmission signal generator 20, a receiver 30, and an estimation device 100. The estimation device 100 includes a circuit 40 and a memory 41. The sensor 10 transmits microwaves from the transmission signal generator 20 to a living body 50 such as a person, and the receiver 30 receives reflected waves reflected or scattered by the living body 50.

[0101] The transmission signal generator 20 has N (N is a natural number greater than or equal to 3) transmission antenna elements 21. The transmission signal generator 20 has an array antenna having N antenna elements 21 arranged in the horizontal direction (x direction). x and arranged in the vertical direction (z direction) N z N (N = N x ×N z ) transmitting antenna elements 21. That is, at least three transmitting antenna elements 21 among the N transmitting antenna elements 21 are arranged at different positions in the vertical direction and the horizontal direction.

[0102] The receiver 30 has M (M is a natural number greater than or equal to 3) receiving antenna elements 31. The receiver 30 has an array antenna having M antenna elements 31 arranged in the horizontal direction (x direction). x and arranged in the vertical direction (z direction) M z The M (M = M x ×M z ) receiving antenna elements 31. That is, at least three receiving antenna elements 31 among the M receiving antenna elements 31 are arranged at different positions in the vertical direction and the horizontal direction.

[0103] Here, the angle formed by the first reference direction and the first biological body direction is φ T , the first reference direction is a direction on the horizontal plane arbitrarily set relative to the signal generator 20, and the first biological direction is a direction from the signal generator 20 to the biological body 50. In addition, the angle formed by the vertical direction and the first biological direction, that is, the elevation angle of the biological body 50 is θ T In addition, the angle formed by the second reference direction and the second biological body direction is φ R , the second reference direction is a direction on the horizontal plane arbitrarily set relative to the receiver 30, and the second biological body direction is a direction from the receiver 30 to the biological body 50. In addition, the angle formed by the vertical direction and the second biological body direction is θ R If the central coordinates of the part where the organism 50 performs life activities are set to (x b ,y b , z b ), then the direction (θ T ,θ R ,φ T ,φ R ) and coordinates (x b ,y b , z b ) can be transformed into each other according to the positional relationship among the transmitting signal generator 20, the receiver 30 and the biological body 50.

[0104] The N transmitting antenna elements 21 transmit transmission signals to the predetermined range where the living body may exist. That is, the transmission signal generator 20 transmits N transmission signals to the predetermined range from N different positions. The predetermined range where the living body may exist refers to the detection range where the sensor 10 detects the presence of the living body.

[0105] Specifically, the N transmitting antenna elements 21 each emit a microwave as a transmission signal to a biological body 50 such as a human being. The N transmitting antenna elements 21 may also transmit a signal that has been subjected to different modulation processing for each transmitting antenna element 21 as a transmission signal. In addition, the N transmitting antenna elements 21 may each switch a modulated signal or an unmodulated signal successively for transmission. The modulation processing may also be performed by the transmission signal generator 20. In this way, by setting the transmission signals transmitted from the N transmitting antenna elements 21 to different transmission signals for each of the N transmitting antenna elements 21, the transmitting antenna element 21 that transmitted the transmission signal received by the receiver 30 can be determined. In this way, the transmission signal generator 20 may also include a circuit for performing modulation processing.

[0106] The M receiving antenna elements 31 receive N receiving signals respectively, and the N receiving signals are signals reflected or scattered by the biological body 50, i.e., reflected signals, among the N transmitting signals respectively transmitted by the N transmitting antenna elements 21. The receiver 30 performs frequency conversion on the receiving signal composed of microwaves and converts it into a low-frequency signal. The receiver 30 outputs the signal obtained by converting it into a low-frequency signal to the circuit 40. That is, the receiver 30 may also include a circuit for processing the receiving signal.

[0107] The circuit 40 performs various processes for operating the sensor 10. The circuit 40 is composed of, for example, a processor that executes a control program and a volatile storage area (main storage device) used as a work area when executing the control program. The volatile storage area is, for example, a RAM (Random Access Memory). In addition, the circuit 40 may also be composed of a dedicated circuit for performing various processes for operating the sensor 10. That is, the circuit 40 may be a circuit for performing software processing or a circuit for performing hardware processing.

[0108] The memory 41 is a nonvolatile storage area (auxiliary storage device), and is, for example, a ROM (Read Only Memory), a flash memory, a HDD (Hard Disk Drive), etc. The memory 41 stores, for example, information useful for various processes to operate the sensor 10 .

[0109] Next, use Figure 2 The functional structure of the circuit 40 will be described.

[0110] Figure 2This is a block diagram showing a functional structure implemented by the circuit and memory in the first embodiment.

[0111] The circuit 40 includes a first matrix calculation unit 410 , a second matrix extraction unit 420 , a number of people estimation unit 430 , a position estimation unit 440 , a Doppler RCS calculation unit 450 , a movement determination unit 460 , a Doppler RCS threshold setting unit 470 , a Doppler RCS threshold determination unit 480 , and a situation estimation unit 520 .

[0112] The first matrix calculation unit 410 calculates a complex transfer function based on the received signal converted into a low-frequency signal. The complex transfer function refers to a function that represents the propagation loss and phase rotation between each transmitting antenna element 21 and each receiving antenna element 31. When the number of transmitting antenna elements is N and the number of receiving antenna elements is M, the complex transfer function is a complex matrix having M×N components. Hereinafter, this complex matrix is ​​referred to as a complex transfer function matrix. The estimated complex transfer function matrix is ​​output to the second matrix extraction unit 420. That is, the first matrix calculation unit 410 sets each complex transfer function representing the propagation characteristics between each of the N transmitting antenna elements 21 and each of the M receiving antenna elements 31 as a component based on each of the multiple received signals received in each of the M receiving antenna elements 31 during a specified period, and calculates the N×M first matrix one by one. In addition, the propagation characteristics between each of the N transmitting antenna elements 21 and each of the M receiving antenna elements 31 refer to the propagation characteristics between the transmitting antenna element 21 and the receiving antenna element 31 in each of the N×M combinations in which the N transmitting antenna elements 21 and the M receiving antenna elements 31 are combined in a one-to-one manner.

[0113] The second matrix extraction unit 420 separates the complex transfer function matrix components obtained from the received signal passing through the biological body 50 and the complex transfer function matrix components obtained from the received signal not passing through the biological body 50. The components passing through the biological body 50 refer to components that change with time due to the activity of the biological body. Therefore, for example, when the part other than the biological body 50 is stationary, the components passing through the biological body 50 can be extracted by extracting components other than direct current from the components obtained by Fourier transforming the components of the complex transfer function matrix in the time direction. In addition, for example, the components passing through the biological body 50 can be extracted by extracting components whose difference from the result observed when the biological body 50 does not exist in a prescribed range exceeds a prescribed threshold. As described above, the second matrix extraction unit 420 extracts the complex transfer function matrix components obtained from the received signal including the reflected signal passing through the biological body 50, and calculates the extracted complex transfer function matrix components as the second matrix. That is, the second matrix extraction unit 420 successively extracts second matrices corresponding to components affected by life activities by successively extracting second matrices corresponding to specified frequency ranges in the first matrices successively calculated in the first matrix calculation unit 410, wherein the life activities include at least one of the breathing, heartbeat and body movement of the organism.

[0114] The prescribed frequency range is, for example, a frequency derived from life activities, and the life activities include at least one of the above-mentioned breathing, heartbeat, and body movement of the organism. The prescribed frequency range is, for example, a frequency in the range of 0.1 Hz to 3 Hz. The second matrix extraction unit 420 can extract the biological component affected by the life activities of the parts of the biological body 50 caused by the movement of the heart, lungs, diaphragm, and internal organs or the life activities caused by the hands, feet, etc. by extracting the second matrix corresponding to the above-mentioned prescribed frequency range. In addition, the parts of the biological body 50 caused by the movement of the heart, lungs, diaphragm, and internal organs are, for example, the human heart.

[0115] Here, the second matrix extracted by the second matrix extraction unit 420 is a matrix having N×M components, and is extracted from a complex transfer function obtained from a received signal observed in the receiver 30 during a predetermined period. Therefore, the second matrix has frequency response or time response information. In addition, the predetermined period is a period of approximately half of a cycle of at least one of breathing, heartbeat, and body movement of a living body.

[0116] The second matrix calculated by the second matrix extraction unit 420 is output to the number estimation unit 430. The number estimation unit 430 may also perform number estimation using, for example, an eigenvalue or an eigenvector obtained using the first matrix calculated by the first matrix calculation unit 410 or the second matrix extracted by the second matrix extraction unit 420. The number estimation unit 430 is an example of a presence determination unit that determines whether a living body exists within a specified range. The number estimation unit 430 may also estimate the number of living bodies existing within a specified range, and when the estimated number of living bodies is 0, determine that there are no living bodies within the specified range, and when the estimated number of living bodies is 1 or more, determine that there are living bodies within the specified range.

[0117] After the person estimation unit 430 determines that a living body exists within the predetermined range, the position estimation unit 440 estimates the position of the living body 50 relative to the sensor 10 using the second matrix successively extracted by the second matrix extraction unit 420. The position estimation unit 440 estimates the departure angle θ from the transmission signal generator 20. T and the angle of arrival θ to the receiver 30 R The angles of the two, and the estimated departure angle θ T and the angle of arrival θ R The position of the living body 50 is estimated by using triangulation. Here, the estimated position of the living body 50 is a three-dimensional position, which includes the position of the living body 50 in the vertical direction relative to the sensor 10, that is, the vertical position. The positions of the living body 50 successively estimated by the position estimating unit 440 may also be stored in the memory 41.

[0118] The Doppler RCS calculation unit 450 uses the successively extracted second matrix and the successively estimated position to successively calculate the Doppler scattering cross-sectional area (Doppler RCS: Radar Cross Section) value. Specifically, in order to calculate the Doppler RCS value, the Doppler RCS calculation unit 450 successively calculates the distance RT and the distance RR based on the successively estimated position by the position estimation unit 440, the position of the transmission signal generator 20, and the position of the receiver 30. The distance RT represents the distance between the successively estimated position of the biological body 50 and the transmission signal generator 20, that is, the first distance, and the distance RR represents the distance between the successively estimated position of the biological body 50 and the receiver 30, that is, the second distance. The Doppler RCS calculation unit 450 calculates the propagation distance based on the successively calculated distance RT and distance RR, and successively calculates the Doppler RCS value using the calculated propagation distance and the strength of the biological component. In addition, the position of the transmission signal generator 20 and the position of the receiver 30 can also be stored in the memory 41 in advance.

[0119] The movement determination unit 460 stores the positions of the living body 50 successively estimated by the position estimation unit 440 in the memory 41 in the estimated order. The movement determination unit 460 uses the positions of the living body 50 stored in the memory 41 for a predetermined number of times (for example, L times) to determine whether the living body 50 is moving. The movement determination unit 460 refers to the L positions of the living body 50 estimated in a predetermined amount of time such as the past 5 seconds among the stored plurality of positions of the living body 50, and sequentially calculates the distance between the latest position and the previous position, the distance between the previous position and the previous two positions, ..., and the distance between the previous L-1 position and the previous L positions, and calculates the sum of all the calculated L-1 distances as the moving distance. That is, the movement determination unit 460 uses all the L positions sampled in the predetermined time to accumulate the distance between two points adjacent in time, thereby calculating the moving distance. In addition, the moving distance can also be said to be the displacement of the position where the living body 50 exists.

[0120] Then, for example, when the calculated moving distance is greater than a specified value such as 1 meter, the movement determination unit 460 determines that the biological body 50 has been moving within the specified period of the past five seconds. When the calculated moving distance is less than the specified value, the movement determination unit 460 determines that the biological body has not moved within the specified time.

[0121] In addition, the movement determination unit 460 uses all the multiple positions sampled within the specified time to calculate the movement distance, but is not limited to this. The distance between the latest position and the position before the specified time can also be calculated as the movement distance, and several points in the L positions corresponding to the specified time can be extracted, and the movement distance can be calculated by accumulating the distance between two points adjacent in time in the positions of the extracted points.

[0122] The Doppler RCS threshold setting unit 470 sets the Doppler RCS threshold using a predetermined method. Specifically, the Doppler RCS threshold setting unit 470 sets the Doppler RCS threshold based on the state of the living body estimated at the latest timing accumulated in the memory 41 .

[0123] The Doppler RCS threshold may be set to a threshold corresponding to each condition of a plurality of different living organisms, or may be set to a threshold corresponding to each condition of a plurality of different living organisms, for example. Figure 3 As shown, the value obtained by multiplying the average of the ten Doppler RCS values ​​in a certain previous posture calculated ten times before by a fixed ratio such as 1.5 times (150%) is set as the threshold. Figure 3 This is a diagram showing a Doppler RCS threshold setting value table in the sensor according to the first embodiment.

[0124] At this time, the ratio to be multiplied by the threshold value has an optimal value that differs depending on the front posture of the application object, and therefore the ratio may be appropriately determined for each front posture of the application object. Figure 3 The ratio of the Doppler RCS threshold value when the bottom front posture is a prone position and the next posture is a rollover is set to 1.1 times (110%), and the ratio of the Doppler RCS threshold value used to calculate other situations is set to 1.2 times (120%). This is because when the bottom front posture is a prone position and the next posture is a rollover, the calculated Doppler RCS value is a value smaller than the Doppler RCS value in other postures.

[0125] The Doppler RCS threshold determination unit 480 compares the Doppler RCS value calculated by the Doppler RCS calculation unit 450 with the Doppler RCS threshold value set by the Doppler RCS threshold value setting unit 470 , and determines whether the Doppler RCS value is equal to or less than the Doppler RCS threshold value.

[0126] The situation estimation unit 520 estimates the situation of the living body, which is either the posture of the living body 50 or the behavior of the living body 50, based on at least one of the determination result of the movement determination unit 460 and the determination result of the Doppler RCS threshold determination unit 480. The situation estimation unit 520 includes a posture estimation unit 490 and a behavior estimation unit 500.

[0127] The posture estimation unit 490 estimates the posture of the living body 50 when the movement determination unit 460 determines that the living body 50 is not moving and the Doppler RCS value is determined to be less than the Doppler RCS threshold value by the Doppler RCS threshold determination unit. In this case, the posture estimation unit 490 can estimate that the position of the living body 50 is almost unchanged and the posture of the living body 50 is still, and thus can estimate the posture of the living body 50 that is still at one position. The posture estimation unit 490 estimates the posture of the living body 50 using the Doppler RCS value calculated by the Doppler RCS calculation unit 450, the vertical position of the living body 50, and the information 42 indicating the correspondence between the Doppler RCS value and the vertical position of the living body 50 and the posture of the living body 50 stored in the memory 41, for example, using the method described in Patent Document 5.

[0128] In Patent Document 5, the posture is estimated based on the vertical position and the area of ​​the Doppler RCS value stored in the memory 41, but it is also possible to use, for example, Figure 4 As shown, when K=1 in the K nearest neighbor method, the posture of the living body associated with the teacher data closest to the vertical position and Doppler RCS value of the living body is estimated as the posture of the living body 50. Figure 4 This is a diagram showing an example of posture estimation by the posture estimating unit in the first embodiment.

[0129] When the Doppler RCS threshold determination unit 480 determines that the Doppler RCS value is greater than the Doppler RCS threshold, the behavior estimation unit 500 estimates the behavior of the living body 50. In this case, since it can be estimated that the posture of the living body 50 is in motion, the behavior estimation unit 500 can estimate the behavior of the living body 50. The behavior estimation unit 500 refers to the calculated Doppler RCS value and the estimated vertical position of the living body 50 stored in the memory 41, calculates the correlation between the Doppler RCS value and the transition vector of the vertical position of the living body 50 and the pre-stored teacher vector, and estimates the behavior of the living body 50.

[0130] Figure 5 This is a block diagram showing an example of the configuration of the behavior estimating unit 500 in the first embodiment.

[0131] The behavior estimation unit 500 estimates the behavior of the living body 50 using the position estimated by the position estimation unit 440, the time series data indicating the time change of the Doppler RCS value calculated by the Doppler RCS calculation unit 450, and the information 42 indicating the corresponding status pre-stored in the memory 41. The behavior estimation unit 500 includes a behavior period extraction unit 510, a direction code calculation unit 511, and a behavior comparison unit 512.

[0132] The action period extraction unit 510 extracts Figure 6 The period during which the position of the biological body 50 estimated by the position estimation unit 440 or the time variation of the Doppler RCS value calculated by the Doppler RCS calculation unit 450 is greater than a predetermined value is regarded as the period during which the user is taking action, i.e., the action period. Figure 6 This is a diagram for explaining an example of extracting an action period from time series data of an altitude position (Height) or a Doppler RCS value.

[0133] When the action period extraction unit 510 extracts the action period using the vertical position, i.e., the height position, or the Doppler RCS value, for example, in order to avoid the influence of instantaneous noise, the height position and the noise component of the Doppler RCS value can be removed from the obtained time series data of the position of the biological body 50 or the Doppler RCS value, for example, using a median filter, an FIR filter, an average value, etc., and the change interval of the height information or the Doppler RCS change interval after the filtering process can be extracted as the action period of the biological body. That is, the action period extraction unit 510 can also extract the action period using the time series data obtained by removing the instantaneous noise component from the multiple vertical positions or multiple Doppler RCS values ​​obtained in the time series using a predetermined filter.

[0134] The action period extraction unit 510 is effective when limiting the period to be estimated for the purpose of reducing the amount of calculation, but is not necessarily required. That is, when the posture estimation is performed for the entire interval, the action period extraction unit 510 can of course be omitted and the estimation can be performed for the entire interval.

[0135] The direction code calculation unit 511 converts the vertical position (height position) obtained from the estimated position of the biological body 50 and the time change of the calculated Doppler RCS value, which are time changes during the motion period extracted by the motion period extraction unit 510, into a direction vector using a predetermined method. Specifically, the direction code calculation unit 511 converts the vertical position (height position) obtained from the estimated position of the biological body 50 and the time change of the Doppler RCS value into a direction vector. Figure 7 The height position and the Doppler RCS value are two-dimensionally plotted as shown, and the distance ΔP of the trajectory and the direction θ of the trajectory are calculated in the trajectory of the time change. The direction code calculation unit 511 calculates the distance ΔP between the first coordinate p1 (H1, R1) and the second coordinate p2 (H2, R2) and the direction θ when observing the second coordinate p2 (H2, R2) from the first coordinate p1 (H1, R1) in the trajectory from the first coordinate p1 (H1, R1) represented by the height position H1 and the Doppler RCS value R1 at the first timing to the second coordinate p2 (H2, R2) represented by the height position H2 and the Doppler RCS value R2 at the second timing next to the first timing, and converts it into a direction vector. Figure 7 This is a diagram for explaining the process of converting into a direction vector.

[0136] Next, the direction code calculation unit 511 calculates the direction code by normalizing the transformed direction vector. Figure 8 For example, the direction code calculation unit 511 determines the direction θ among the direction codes 1 to 8 as the nearest direction code. Figure 8 This is a diagram showing an example of a direction code table.

[0137] The direction code calculation unit 511 calculates the direction code and the distance ΔP as described above to obtain Fig. 9 Time series data of the direction codes shown. Fig. 9 : is a diagram showing an example of time series data of calculated direction codes and distances. In addition, at this time, the direction code calculation unit 511 may standardize the direction codes in order to avoid the influence of individual differences.

[0138] The action comparison unit 512 determines the action corresponding to the time series data in the information 42 representing the corresponding status by comparing the time series data of the direction code calculated by the direction code calculation unit 511 with the information 42 representing the corresponding status stored in the memory, thereby inferring the action of the organism 50.

[0139] In addition, the information 42 representing the corresponding conditions stored in the memory 41 is information representing the corresponding conditions of a plurality of model codes and the actions of the organism 50, wherein the plurality of model codes represent the position of the organism 50 in the vertical direction relative to the sensor 10, i.e., the vertical position, and the time change of the Doppler RCS value. In addition, the actions of the organism 50 corresponding to the information 42 representing the corresponding conditions include falling, sitting on a chair, sitting on the floor, standing up from a chair, standing up from the floor, jumping, and changing directions. That is, the action estimation unit 500 uses the position estimated by the position estimation unit 440 and the time change of the Doppler RCS value calculated by the Doppler RCS calculation unit 450, and the information 42 representing the corresponding conditions pre-stored in the memory 41, to estimate which action the organism 50 has taken, including falling, sitting on a chair, sitting on the floor, standing up from a chair, standing up from the floor, jumping, and changing directions. In addition, the model code is represented as follows: Fig.10 The time series data shown.

[0140] In addition, when the movement determination unit 460 determines that the living body 50 is moving, the behavior estimation unit 500 can estimate that the behavior of the living body 50 is moving. In this case, if the posture of the living body 50 estimated immediately before the determination that the living body 50 is moving is standing, the behavior estimation unit 500 can estimate that the behavior of the living body 50 is walking. In addition, if the posture of the living body 50 estimated immediately before the determination that the living body 50 is moving is sitting, the behavior estimation unit 500 can estimate that the behavior of the living body 50 is moving using a wheelchair. In addition, if the posture of the living body 50 estimated immediately before the determination that the living body 50 is moving is lying, the behavior estimation unit 500 can estimate that the behavior of the living body 50 is turning over or crawling.

[0141] In addition, in the circuit 40, the processing in each unit 410 to 500 described above is repeatedly performed at multiple different timings to obtain time series data. For example, the circuit 40 obtains time series data consisting of multiple positions in the time series and multiple Doppler RCS values ​​in the time series by repeatedly performing processing at a predetermined sampling period.

[0142] Next, the operating principle of the sensor 10 of the first embodiment is described in detail using mathematical formulas. In addition, a method of extracting biological components using Fourier transform is described here. The processing described here is performed by the circuit 40. The complex transfer function matrix between the transmission signal generator 20 and the receiver 30 is defined as

[0143]

Mathematical formula 1

[0144]

[0145] Here, t represents the time. If the components of equation 1 are Fourier transformed, we get

[0146]

Mathematical formula 2

[0147]

[0148] Such a frequency response matrix. Here, f represents frequency, and each component of the frequency response matrix is ​​a complex number. The frequency response matrix includes both propagation components through the biological body 50 and propagation components outside the biological body 50. When the outside of the biological body is considered to be stationary, the DC component of the frequency response matrix, that is, G(0), is considered to mainly include propagation components outside the biological body. This is because, since the life activities including at least one of the breathing, heartbeat and body movement of the biological body produce Doppler frequency shift, it is considered that the component passing through the biological body is included outside f=0. In addition, considering the frequency of the breathing or heartbeat of the biological body and its higher harmonics, it is considered that there are more components originating from the biological body in the range of f<3[Hz]. Therefore, for example, if G(f) in the specified frequency range of 0[Hz]<f<3[Hz] is taken out, the biological component can be effectively extracted.

[0149] Next, a method for estimating the position of a living body using the living body component G(f) is described. The generated living body component matrix G(f) is rearranged as follows:

[0150]

Mathematical formula 3

[0151] g(f)=[g 11 (t), ..., g M1 (t), g 12 (t), ..., g N2 (t), ..., g 1M (t), ..., g NM (t)] T (Formula 3)

[0152] The vector form of . It is defined as the biological component vector. Here {·} T represents transposition. According to the biological component vector g(f),

[0153]

Mathematical formula 4

[0154]

[0155] Calculate the correlation matrix. Here, {·} Hrepresents the complex conjugate transpose. In addition, R is averaged between 0 [Hz] < f < 3 [Hz]. It is known that this averaging improves the accuracy of the position estimation described later. Next, by performing eigenvalue decomposition on the calculated correlation matrix R, the eigenvector U of the correlation matrix R and its complex conjugate transpose vector U are calculated. H .

[0156]

Mathematical formula 5

[0157] R=UDU H (Formula 5)

[0158] In addition, the eigenvector in Formula 5 is expressed by Formula 6 shown below.

[0159]

Mathematical formula 6

[0160] U=[u1,...,u MN ] (Formula 6)

[0161] Here, u i represents the intrinsic vector of the i-th column, and the number of elements is NM. D is a diagonal matrix whose diagonal elements are eigenvalues, expressed as

[0162]

Mathematical formula 7

[0163] D = diag[λ1, ..., λ MN ] (Formula 7)

[0164] Here, diag[·] represents a diagonal matrix whose diagonal terms have elements in [·]. The circuit 40 uses the above information to estimate the position of the biological body 50 to be detected. Here, as an example, a position estimation method based on the MUSIC method is described. The MUSIC method is a method of estimating direction and position by using a direction vector called a steering vector and an eigenvector shown in Formula 6. The biological component vector shown in Formula 3 is obtained by deforming the original M×N matrix. In order to estimate the position of the biological body 50, it is necessary to define a steering vector corresponding to it. The steering vector representing the (θT, φT) direction from the transmitting signal generator 20 toward the biological body 50 and the steering vector representing the (θR, φR) direction from the receiver 30 toward the biological body 50 are respectively expressed as

[0165]

Mathematical formula 8

[0166]

[0167]

Mathematical formula 9

[0168]

[0169] Here, it is expressed as

[0170]

Mathematical formula 10

[0171]

[0172]

Mathematical formula 11

[0173]

[0174] Here, k represents the wave number, d Tx and d Tz denote the element spacing of the transmitting antenna element 21 in the x direction and the z direction, d Rx and d Rz 1 and 2 show the element spacings in the x direction and the z direction of the receiving antenna element 31. In addition, in this embodiment, it is assumed that the linear array antenna is a linear array antenna in which the spacings between elements in the same direction are constant.

[0175] d Tx For example, d represents the interval between two transmitting antenna elements 21 adjacent to each other in the x direction among the plurality of transmitting antenna elements 21. Tz For example, d represents the interval between two transmitting antenna elements 21 adjacent to each other in the z direction among the plurality of transmitting antenna elements 21. Rx For example, d represents the interval between two receiving antenna elements 31 adjacent to each other in the x direction among the plurality of receiving antenna elements 31. Rz For example, it represents the interval between two receiving antenna elements 31 adjacent to each other in the z direction among the plurality of receiving antenna elements 31. If the Kronecker product of these steering vectors is calculated, it becomes

[0176]

Mathematical formula 12

[0177]

[0178] Steering vector a(θ T ,φ T ,θ R ,φ R ) is a vector with MN×1 elements, and is a function of four variables: departure angles θT, φT and arrival angles θR, φR. T ,φ T ,θ R ,φ R ) is defined as the turning vector. If the number of organisms within the detection range is set to L, then the evaluation function

[0179]

Mathematical formula 13

[0180]

[0181] Here, the evaluation function of formula 11 is called MUSIC spectrum, which searches for the combination of directions from the transmission signal generator 20 and the receiver 30 toward the detection object (θ T ,θ R ) can be obtained by using trigonometric methods according to the θ corresponding to the maximum point. T and θ R The three-dimensional position of the living body 50 including the vertical position as viewed from the transmission signal generator 20 and the receiver 30 is determined.

[0182] The Doppler RCS value is then calculated from equation 12, and the posture of the organism 50 is estimated from the position of the organism 50 and the Doppler RCS value. If the frequency range of the above extraction is set to f1 to f2 (f1 < f2), when the power transmission coefficient is calculated based on the channel component reflected and observed from the organism, it can be calculated

[0183]

Mathematical formula 14

[0184]

[0185] Here, ρ ij is a matrix

[0186]

Mathematical formula 15

[0187]

[0188] On the other hand, the power from the jth transmitting antenna element 21 to the ith receiving antenna element 31 via the biological body 50 is expressed as

[0189]

Mathematical formula 16

[0190]

[0191] Here, P t It is assumed that equal power is transmitted from all the transmitting antenna elements 21. G t represents the operating gain of the transmission signal generator 20, G r represents the operating gain of the receiver 30, R1 represents the distance from the transmission signal generator 20 to the biological body 50, and R2 represents the distance from the biological body 50 to the receiver 30. The distance R1 and the distance R2 can be easily calculated based on the position estimated by mathematical formula 13. Therefore, since the power transfer coefficient defined by formula 12 is expressed by ρ ij =P rij / P t The scattering cross-sectional area can be expressed as

[0192]

Mathematical formula 17

[0193]

[0194] In addition, the scattering cross-sectional area processed here is not the scattering cross-sectional area of ​​the entire organism 50, but only the Doppler scattering area corresponding to the variable component caused by the influence of the life activities of the organism 50 such as breathing, heartbeat and body movement. Then, all elements are averaged and obtained by

[0195]

Mathematical formula 18

[0196]

[0197] Calculate the average scattering cross-sectional area.

[0198]

Mathematical formula 19

[0199]

[0200] It is called Doppler RCS.

[0201] In addition, for example, in the case of further separating and measuring the life component and body movement component of a living organism, it is of course sufficient to adjust the frequency range of formula 14 and perform subsequent processing.

[0202] The method of action presumption is described below.

[0203] If the Doppler RCS value σ and the height z of the organism 50 are continuously estimated at multiple different timings by the above method, the trajectory of the σ-z characteristic can be observed. At this time, when the organism is in a stationary state, the Doppler RCS value hardly changes, so the flow of trajectory points when the organism is moving, that is, the Doppler RCS value changes greatly, is extracted. For this trajectory point group, the distance between the i-1th trajectory point and the i-th trajectory point, that is, the trajectory point movement amount, is set to ΔP i , the angle between the i-1th trajectory point and the i-th trajectory point is set as the angle parameter α i , defined as Equation 20 and Equation 21 respectively.

[0204]

Mathematical formula 20

[0205]

[0206]

Mathematical formula 21

[0207]

[0208] Next, a direction code is assigned to the moving direction of the trajectory point using the value of the angle parameter. The direction code is assigned relative to the angle by dividing 360° into 8 parts and assigning a number from 1 to 8 to each direction. At this time, it is also possible to set it in a way that no code boundary is set in the direction so that the code does not change frequently when moving in the vertical or horizontal direction. Since there are individual differences in the speed of movement of the organism 50, in order to avoid erroneous recognition caused by different numbers of trajectory points due to different movement speeds even in the same movement, the direction code can be standardized by considering the difference in movement speed. For example, based on the original direction code string c obtained by direction estimation, j (j=1~j max ), considering the ratio of the trajectory point movement amount to the sum of the trajectory point movement amounts, a standardized code string consisting of K items is generated. max is the number of trajectory points, which varies according to the action time. Standardized code string C k (k=1, 2, ..., K) is the sum of the movement amount of the trajectory point ΔPi and the movement amount of the trajectory point ΔP according to the i-th trajectory movement amount ΔPi sum It was prepared as follows.

[0209] 1) In j=1, the normalized code string that satisfies the k range of Mathematical Formula 22 is set as the code of j=1 in the original direction code string.

[0210]

Mathematical formula 22

[0211]

[0212] 2) When j is 2~j max In the range of , the standardized code string in the k range that satisfies the following mathematical formula 23 is set as the code in each j. sum Satisfies mathematical formula 24.

[0213]

Mathematical formula 23

[0214]

[0215]

Mathematical formula 24

[0216]

[0217] The behavior of the organism 50 is estimated by comparing a standardized sequence (test data) consisting of K items made from trajectory point data with a model data sequence consisting of K items corresponding to each of a plurality of behaviors. The model data sequence is a sequence of behavior measurements performed multiple times in advance during each of a plurality of behaviors, and the direction code with the most number of items in the standardized code string obtained through the multiple behavior measurements is set as the model data in the item. Since the direction code is circular, the maximum difference is 4. In order to calculate the difference between the direction code number and the actual number, it is set as mathematical formula 25. Among them, in δC i When >4, it is set to mathematical formula 26.

[0218]

Mathematical formula 25

[0219] δC i =|C test,1 -C model,i | (Formula 24)

[0220]

Mathematical formula 26

[0221] δC i =8-δC i (Formula 25)

[0222] Among them, C test,i represents the element of the i-th item in the test data column, C model,i represents the element of the i-th item in the model data column. Next, the sum of the squares of the direction code differences between the test data and the model data is calculated as the deviation shown in Mathematical Formula 27. For example, Fig.12 As shown, by comparing the test data with the model data, the deviation E for action determination is calculated by mathematical formula 27. Fig.12 This is a diagram showing test data obtained by measurement and model data as a model code.

[0223]

Mathematical formula 27

[0224]

[0225] Then, the action corresponding to the model data string with the smallest deviation is output as the estimation result.

[0226] Next, the operation of the sensor 10 in the first embodiment will be described using a flowchart.

[0227] Fig.11 This is a flowchart showing an example of the operation of the sensor in the first embodiment.

[0228] In the sensor 10 , the N transmission antenna elements 21 of the transmission signal generator 20 transmit N transmission signals using the N transmission antenna elements 21 to a predetermined range where the living body 50 may be present.

[0229] The M receiving antenna elements 31 of the receiver 30 receive N receiving signals including a plurality of reflected signals obtained when the N transmission signals transmitted by the transmission signal generator 20 are reflected by the biological body 50 .

[0230] The circuit 40 successively calculates an N×M first matrix based on each of the N received signals received by each of the M receiving antenna elements 31 within a specified period, and the N×M first matrix sets each complex transfer function representing the propagation characteristics between each of the N transmitting antenna elements 21 and each of the M receiving antenna elements 31 as a component (S11).

[0231] The circuit 40 successively extracts second matrices corresponding to components affected by life activities including at least one of breathing, heartbeat, and body movement of the biological body 50 by successively extracting second matrices corresponding to a prescribed frequency range in the first matrix (S12).

[0232] The circuit 40 uses the first matrix or the second matrix to perform a presence determination (S13) to determine whether the biological body 50 is present within the specified range as the effective range of the sensor 10. In step S13, a number estimation may be performed to estimate the number of biological bodies present within the specified range. After the circuit 40 determines that the biological body 50 is present within the specified range (Yes in S13), it proceeds to step S14. After the circuit 40 determines that the biological body 50 is not present within the specified range (No in S13), it returns to step S11 (S13).

[0233] In step S14, the circuit 40 successively estimates the position of the living body 50 relative to the sensor 10 using the successively extracted second matrix (S14). Then, the circuit 40 stores the successively estimated positions of the living body 50 in the memory 41 a predetermined number of times in the estimated order.

[0234] The circuit 40 successively calculates the distance RT representing the distance between the organism 50 and the transmitting signal generator 20, and the distance RR representing the distance between the organism 50 and the receiver 30 based on the successively estimated positions of the organism 50, the transmitting signal generator 20, and the receiver 30, and uses the calculated distance RT and distance RR to successively calculate the Doppler RCS value relative to the organism 50 (S15).

[0235] The circuit 40 uses the positions of the organism 50 stored in the memory 41 for a predetermined number of times to determine whether the displacement of the position where the organism 50 exists is less than a predetermined value (S16). If the displacement is more than 1 m in 5 seconds (No in S16), the circuit 40 determines that the organism 50 is moving and returns to step S11. In addition, it is also possible to determine that the action of the organism 50 is moving by using this result. If the displacement is less than 1 m in 5 seconds (Yes in S16), the circuit 40 determines that the organism 50 is not moving and proceeds to the next step S17.

[0236] The circuit 40 sets the Doppler RCS threshold value by a predetermined method with reference to the previous posture and the average value of the Doppler RCS values ​​of the predetermined number of times in the past stored in the memory 41 (S17). At this time, the Doppler RCS threshold value may be set to a predetermined value for each posture in the past, or may be set to a value obtained by multiplying the average value of the Doppler RCS values ​​in the past by a predetermined ratio such as 1.5 times.

[0237] The circuit 40 compares the calculated Doppler RCS value with the set Doppler RCS threshold value, and determines whether the Doppler RCS value is equal to or less than the Doppler RCS threshold value (S18).

[0238] When the calculated Doppler RCS value is equal to or smaller than the Doppler RCS threshold value (Yes in S18 ), the circuit 40 performs a posture estimation process in which the posture estimation unit 490 estimates the posture of the living body 50 ( S19 ).

[0239] When the calculated Doppler RCS value is larger than the Doppler RCS threshold value (No in S18), the circuit 40 performs a behavior estimation process in which the behavior estimation unit 500 estimates the behavior of the living body 50 (S20).

[0240] Since a specific example of the processing in the circuit 40 has been described in the description of the functional structure of the circuit 40, it is omitted.

[0241] As described above, in steps S16 to S20 , at least one of the posture estimation of the living body 50 and the behavior estimation of the living body is selectively performed based on at least one of the determination result in the movement determination and the determination result in the Doppler RCS threshold determination.

[0242] Fig.12 This is a flowchart showing a first example of the detailed behavior estimation processing.

[0243] The circuit 40 extracts a period during which the vertical position in the estimated position or the time change of the calculated Doppler RCS value is larger than a predetermined value as an action period during which the living body 50 moves ( S21 ).

[0244] The circuit 40 converts the extracted time variation during the action period, i.e., the vertical position obtained from the estimated position and the time variation of the calculated Doppler RCS value, into a direction vector using a predetermined method, and calculates a direction code (S22) by normalizing the converted direction vector.

[0245] The circuit 40 compares the calculated time series data of the direction code with the information 42 representing the corresponding status stored in the memory, determines the action corresponding to the time series data in the information 42 representing the corresponding status, and thereby estimates the action of the organism 50 (S23).

[0246] Next, the operation of the preliminary learning of the sensor 10 for acquiring the information 42 indicating the corresponding status will be described.

[0247] The circuit 40 receives input for designating a prescribed action through an input unit (not shown). Thus, the circuit 40 recognizes that the action performed within the prescribed period is the action indicated by the received input. Next, the same processing as steps S11 to S22 and steps S21 and S22 of the action of the sensor 10 described above is sequentially performed.

[0248] Next, the circuit 40 associates the motion received by the input unit with the calculated time series data of the direction code, and stores the information indicating the correspondence status as teacher data in the memory 41. By performing the above prior learning for each motion of the organism, the information indicating the correspondence status is generated and stored in the memory 41.

[0249] According to the sensor 10 of this embodiment, at least one of the estimation of the posture of the living body 50 and the estimation of the action of the living body 50 is selectively performed based on at least one of the determination result in the movement determination and the determination result in the Doppler RCS threshold determination. Therefore, the estimation of the living body condition as any one of the movement of the living body 50, the posture of the living body 50, and the action of the living body 50 can be performed in a short time and with high accuracy.

[0250] According to the sensor 10 of this embodiment, the position where the living body 50 exists and the presence, stillness, movement, posture, behavior, etc. of the living body at that position can be estimated in a short time and with high accuracy.

[0251] The sensor 10 detects the existence of the living body 50 by detecting the moving part. Therefore, for example, by using this detection, it is possible to infer whether the person is alive and in which posture he is standing, sitting, sitting cross-legged, or lying on his back. In this way, the existence of the person can be effectively confirmed. In addition, since the existence of the person can be confirmed without performing image analysis on the image taken by the camera, the existence of the person can be confirmed while protecting the privacy of the person.

[0252] (Variation 1 of Implementation Example 1)

[0253] In the first embodiment, in the behavior estimation process of the circuit 40, the trajectory of the combination of the vertical position of the living body 50 and the Doppler RCS value is stored in the memory 41 as the teacher data in advance, and the behavior is estimated by calculating the correlation between the teacher data and the estimated vertical position of the living body 50 and the calculated Doppler RCS value combination trajectory, but the present invention is not limited to this. The behavior estimation unit 500 can perform, for example, Fig.13 The process shown in the flowchart is used as the process of estimating the behavior of the living body 50.

[0254] Fig.13 This is a flowchart showing a second example of the behavior estimation process in the first embodiment.

[0255] The posture estimating unit 490 of the circuit 40 stores the estimated posture in the memory 41 every time the posture is estimated.

[0256] The behavior estimating unit 500 refers to the latest posture of the living body 50 stored in the memory 41 and acquires the latest posture as a previous posture for estimating the behavior of the living body 50 ( S31 ).

[0257] Then, the behavior estimation unit 500 sequentially acquires the Doppler RCS values ​​calculated one by one (S32). The behavior estimation unit 500 acquires the Doppler RCS values ​​stored in the memory 41 from the memory 41 in the order of calculation.

[0258] The behavior estimating unit 500 determines whether the acquired Doppler RCS value is equal to or less than the Doppler RCS threshold value ( S33 ).

[0259] When it is determined that the Doppler RCS value is below the Doppler RCS threshold value (Yes in S33), the behavior estimation unit 500 causes the posture estimation unit 490 to estimate the posture of the living body 50, and obtains the estimated posture as the post posture in the behavior (S34). That is, the behavior estimation unit 500 waits until the Doppler RCS value becomes below the Doppler RCS threshold value, and thus waits until the living body 50 is stationary, and estimates the posture of the living body 50 estimated from the N received signals when the Doppler RCS value becomes below the Doppler RCS threshold value as the post posture. By obtaining the posture of the living body 50 after the living body 50 is stationary as the post posture, the behavior estimation unit 500 can obtain the post posture with good accuracy.

[0260] Then, the behavior estimation unit 500 estimates the behavior of the living body 50 using the acquired front posture and the estimated rear posture (S35). Fig.14 The front posture, the rear posture, and the action shown in the table are used to establish a corresponding state, and a combination of the front posture and the rear posture is determined to establish a corresponding action in the corresponding state, thereby estimating the action of the biological body 50. Fig.14 This is a diagram showing an example of the correspondence between the combination of the front posture and the rear posture and the behavior in another example of the behavior estimation process in the first embodiment.

[0261] (Variation 2 of Implementation Example 1)

[0262] In addition, for example, Fig.14 As shown, since the previous posture can be used to define the posture after the action that has occurred, that is, the next posture, after determining the previous posture, the action estimation unit 500 can narrow down the options for the next possible action. Fig.14 The back pose in can also be replaced with a transferable pose. By using this information, such as Fig.15 In the flowchart shown, when the previous posture is determined, the action can be estimated by dividing the options of the action that occurs after the determined previous posture.

[0263] Fig.15 This is a flowchart showing a third example of the behavior estimation process in the first embodiment.

[0264] The action estimation unit 500 refers to the latest posture of the organism 50 stored in the memory 41 and obtains the latest posture as the previous posture for estimating the action of the organism 50 (S41). If the previous posture is posture A, the action estimation unit 500 proceeds to step S42. For example, posture A is a standing posture.

[0265] The behavior estimation unit 500 executes steps S42 and S43. Steps S42 and S43 are the same as steps S32 and S33, respectively, and thus description thereof is omitted.

[0266] In step S44, since the acquired previous posture is a standing posture, the action estimation unit 500 refers to Fig.14 The next back posture is reduced to the lying posture or the sitting posture according to the corresponding condition in the figure, and then the action is estimated (S44). Specifically, the posture estimation unit 490 for estimating the back posture then estimates the posture with the closest estimation result in the sitting posture and the lying posture as the back posture. For example, by comparing the teacher data of the sitting posture and the lying posture with the combination of the vertical position and the Doppler RCS value of the biological body 50, it is estimated that any one of the sitting posture and the lying posture is the back posture of the biological body 50.

[0267] Similarly, if the previous posture is posture B, the behavior estimation unit 500 proceeds to step S45. Since only the previous posture is different, the processing of steps S45 to S47 itself is the same as that of steps S42 to S44, and thus the description thereof is omitted.

[0268] (Variation 3 of Implementation Example 1)

[0269] The flowchart showing the operation of the sensor 10 shown in the above-mentioned embodiment 1 is an example. Fig.16 as well as Fig.17 As shown, by storing the first matrix, the second matrix, the position information of the biological body 50, the Doppler RCS, etc. in the memory 41, the subsequent processing can be shared to the server or other computers, and the Fig.18 As shown, the Doppler RCS threshold is first determined.

[0270] Fig.16 This is a flowchart showing another example of the operation of the sensor in the first embodiment.

[0271] like Fig.16 As shown, the circuit 40 of the sensor 10 can also be separated Fig.11 The processes of steps S11 to S15 and steps S16 to S20 in the flowchart described in . That is, the first device that executes the processes of steps S11 to S15 and the second device that executes the processes of steps S16 to S20 may be different.

[0272] First, the first device executes the processing of steps S11 to S15. Then, the processing results obtained by steps S11 to S15 are saved in a memory (S101). In addition, the memory used in step S101 can be a memory of the first device, a memory of the second device, or a memory of a device different from the first device and the second device.

[0273] Next, the second device reads the latest result from the memory (S102). Then, the second device performs the processing of steps S16 to S20.

[0274] In addition, the first means and the second means are each constituted by a circuit and a memory.

[0275] Fig.17 This is a flowchart showing another example of the operation of the sensor in the first embodiment.

[0276] like Fig.17 As shown, the circuit 40 of the sensor 10 can also be separated Fig.11 That is, the third device that executes the processes of steps S11 to S19 and the fourth device that executes the process of step S20 may be different from each other.

[0277] First, the first device performs the processing of steps S11 to S19. Then, the processing results obtained by steps S11 to S19 are saved in the memory (S111). The memory used in step S101 can be the memory of the third device, the memory of the fourth device, or the memory of a device different from the third device and the fourth device.

[0278] Next, the fourth device reads the latest result from the memory (S112).

[0279] Then, the fourth device sets a Doppler RCS threshold value (S113), and determines whether the Doppler RCS value in the most recent result is less than the Doppler RCS threshold value (S114).

[0280] The fourth device performs the process of step S20 when the Doppler RCS value is greater than the Doppler RCS threshold value (No in S114), and returns to step S112 when the Doppler RCS value is less than the Doppler RCS threshold value (Yes in S114).

[0281] Fig.18 This is a flowchart showing another example of the operation of the sensor in the first embodiment.

[0282] like Fig.18 As shown in steps S121 and S122 , the setting of the Doppler RCS threshold and the determination of the Doppler RCS threshold may be performed before the determination of whether the living body 50 is moving.

[0283] In this case, similarly to step S17 , the circuit 40 refers to the previous posture and the average value of the Doppler RCS values ​​of a predetermined number of past times stored in the memory 41 , and sets the Doppler RCS threshold value by a predetermined method ( S121 ).

[0284] Next, similarly to step S18, the circuit 40 compares the calculated Doppler RCS value with the set Doppler RCS threshold value, and determines whether the Doppler RCS value is equal to or less than the Doppler RCS threshold value (S122).

[0285] When the calculated Doppler RCS value is equal to or smaller than the Doppler RCS threshold value (Yes in S122 ), the circuit 40 performs a posture estimation process ( S19 ) in which the posture estimation unit 490 estimates the posture of the living body 50 , and returns to step S11 .

[0286] When the calculated Doppler RCS value is greater than the Doppler RCS threshold value (No in S18), the circuit 40 executes step S16. If the movement amount is less than the threshold value (Yes in S16), the circuit 40 estimates the behavior of the biological body 50 (S20) and returns to step S11.

[0287] If the movement amount is greater than the threshold value (No in S16), the circuit 40 determines that the object is moving and returns to step S11.

[0288] (Implementation Method 2)

[0289] In the circuit 40 of the first embodiment, a posture or an action of the organism 50 at a certain moment is estimated, but the present invention is not limited thereto, and the posture probability or action probability of the organism 50 may be estimated as the posture and action of the organism 50. The posture probability of the organism 50 refers to the probability of occurrence of one or more pre-set postures estimated to be taken by the organism 50. And, the action probability of the organism 50 refers to the probability of occurrence of each of the one or more actions that the organism 50 may take.

[0290] Fig.19 This is a block diagram showing the functional configuration of a circuit and a memory in the second embodiment.

[0291] The circuit 40A in the second embodiment is different from the circuit 40 in the first embodiment in that the posture estimating unit 490 and the action estimating unit 500 are replaced by a posture probability estimating unit 491 and an action probability estimating unit 501 , respectively.

[0292] Fig. 20 This is a flowchart showing an example of the operation of the sensor in the second embodiment.

[0293] In the second embodiment, the difference is that steps S19a and S20a are performed instead of steps S19 and S20. Figure 21 to Figure 23 The details of step S19a will be described.

[0294] Fig.21 This is a flowchart showing an example of posture probability estimation processing in the posture probability estimation unit. Fig. 22 This is a diagram showing an example of teacher data used for posture probability estimation in the second embodiment.

[0295] The posture probability estimation unit 491 first normalizes the teacher data for estimating postures (S51). The posture probability estimation unit 491 may perform step S51 every time the teacher data is updated. In normalizing the teacher data, for example, Fig. 22 As shown, in order to suppress the deviation in the height direction in the original teacher data, for example, the data above 70% and below 30% of the teacher data for each posture is deleted. The range of deleted data set at this time varies depending on the installation status and purpose of the sensor 10A. Next, using equations 27 and 28, the height information and Doppler RCS information are standardized.

[0296]

Mathematical formula 28

[0297]

[0298]

Mathematical formula 29

[0299]

[0300] Among them, Z i std represents the standardized teacher data of the i-th vertical position, Z i represents the i-th teacher data of the plumb position, μ z represents the mean value of the vertical position, σz represents the variance of the vertical position, σ i std represents the ith, normalized teacher data of Doppler RCS, σ i represents the i-th teacher data of Doppler RCS, μ σ represents the average value of Doppler RCS, σ σ represents the variance of Doppler RCS. An example of the calculation result is shown in Fig.23 .also, Fig.23 This is a diagram showing an example of normalization of teacher data obtained by posture probability estimation and posture estimation in the second embodiment.

[0301] exist Fig.23 In the example, when the measured data is located at the asterisk, the posture probability estimation unit 491 extracts data of 100 points in the vicinity thereof, and performs posture probability estimation (S52). For example, when 70 points of the extracted 100 points represent standing postures and 30 points represent sitting postures, the posture probability estimation unit 491 calculates a posture probability estimation value consisting of 70% standing postures and 30% sitting postures as the posture of the biological body 50.

[0302] Next, Fig.24: is a flowchart showing an example of the action probability estimation process in the action probability estimation unit.

[0303] In implementation mode 2, if the above Figures 21 to 23 As described above, the posture probability estimation unit 491 calculates the posture probability estimation value, and the calculated posture probability estimation value is sequentially stored in the memory 41 .

[0304] The action probability estimation unit 501 refers to the latest posture probability estimation value of the living body 50 stored in the memory 41, and obtains a plurality of initial postures respectively indicated by a plurality of posture probabilities included in the latest posture probability estimation value (S61).

[0305] The action probability estimation unit 501 calculates the possible action probability for each of the plurality of initial postures. For example, in step S61, the posture probability of the initial posture A is estimated to be P a1 , the posture probability of the initial posture B is estimated to be P b1 , the posture probability of the initial posture C is estimated to be P c1 , the posture probability of the initial posture D is estimated to be P d1 .

[0306] Hereinafter, the initial position A will be described.

[0307] The action probability estimation unit 501 executes steps S62 and S63. Steps S62 and S63 are different from steps S32 and S33 respectively in that the vertical position is also recorded in step S62, and the others are the same. That is, the action probability estimation unit 501 executes steps S62 and S63, and waits while recording until the Doppler RCS value becomes below the Doppler RCS threshold, and when the Doppler RCS value becomes below the Doppler RCS threshold, it proceeds to the next step S64. In step S64, the action determination deviation E is calculated using a prescribed method or formula 26.

[0308] Next, the action probability estimation unit 501 estimates the posture probability value P of the initial posture based on the posture probability value P referred to from the memory 41. i , use Equation 29 to find the initial posture correction term B as the posture probability i .

[0309]

Mathematical formula 30

[0310]

[0311] In this case, C can be any natural number. In addition, the posture probability after the action is completed is obtained by the posture probability estimation unit 491, and the post-action posture correction term A is obtained by the same method as equation 29. i(S64) In addition, in this embodiment, since the action determination deviation E is a variable indicating the correlation between the teacher data and a certain code, the action when E is minimum becomes the result. i Any formula may be used as long as it is a variable that becomes the smallest when the most accurate posture is achieved.

[0312] Then, the probability-corrected deviation Epi is obtained by equation 30. Thus, P in step S66 of the initial position A is calculated. A .

[0313]

Mathematical formula 31

[0314]

[0315] Here, E i This is the result of obtaining the action determination deviation E for each code of each teacher data.

[0316] Thus, in steps S61 to S66, the posture probability P of the initial posture A is calculated in step S64. a1 , the action probability P of each of the multiple actions determined based on the initial posture A a21 , P a22 , ..., in step S65, the posture probability P of the subsequent posture as a result of performing a plurality of actions after the initial posture A is calculated. a31 , P a32 Then, in step S66, the posture probability P of the initial posture A is calculated. a1 , multiple action probabilities P a21 , P a22 One of the following poses, and the pose probability P of the subsequent pose a31 , P a32 The minimum value of the product of all possible combinations of these three probabilities is taken as the action probability P A .

[0317] Similar to the initial posture A, steps S67 to S71 corresponding to steps S62 to S66 are repeated for the initial posture B, and similarly, the processing is repeated for the other initial postures C and D.

[0318] The action probability estimation unit 501 compares the calculated probability-corrected deviation E in all the initial postures A to D, . . . pi P A , P B , P C , P D , ..., determine the smallest action as the action of the organism (S72).

[0319]

Industrial Applicability

[0320] The present disclosure can be used for sensors and estimation methods for estimating the direction and position of a moving object using wireless signals. In particular, the present invention can be used for a direction estimation method, a position estimation method, and a height estimation device using a scattering cross-sectional area, which are installed in a measuring device for measuring the direction and position of a moving object including a living body and a machine, a household appliance that performs control according to the direction and position of a moving object, a monitoring device for detecting the intrusion of a moving object, and the like.

Claims

1. A sensor comprising: A transmission signal generator having N transmission antenna elements, wherein the N transmission antenna elements respectively transmit transmission signals to a specified range where a biological body may exist, where N is a natural number greater than 3; A receiver having M receiving antenna elements, wherein the M receiving antenna elements respectively receive N receiving signals including reflected signals, wherein the reflected signals are obtained by reflection or scattering of a part of the N transmitting signals respectively transmitted by the N transmitting antenna elements by the biological body, wherein M is a natural number greater than or equal to 3; Circuit; as well as Memory, The circuit has: A first matrix calculation unit calculates, based on each of the N reception signals received by each of the M reception antenna elements in a predetermined period, a complex transfer function representing a propagation characteristic between the transmission antenna element and the reception antenna element in each N×M combination in which the N transmission antenna elements and the M reception antenna elements are combined one-to-one, and sequentially calculates an N×M first matrix having the calculated N×M complex transfer functions as components; a second matrix extracting unit that successively extracts the second matrix corresponding to the component affected by life activities by successively extracting the second matrix corresponding to the prescribed frequency range in the first matrix successively calculated by the first matrix calculating unit, the life activities including at least one of breathing, heartbeat and body movement of the organism; A presence / absence determination unit performs presence / absence determination by a predetermined method to determine whether the living body is present within the predetermined range; a position estimating unit that, after the presence determination unit determines that the living body exists within the predetermined range, sequentially estimates the position of the living body relative to the sensor using the second matrix sequentially extracted by the second matrix extracting unit; A Doppler RCS calculation unit, (i) based on the position of the living body, the position of the transmission signal generator, and the position of the receiver, successively calculates a first distance and a second distance, wherein the first distance indicates the distance between the successively estimated position of the living body and the transmission signal generator, and the second distance indicates the distance between the living body and the receiver, and (ii) successively calculates a Doppler radar cross-sectional area (RCS) value for the living body using the calculated first distance and second distance; a movement determination unit that stores the successively estimated positions of the living body in the memory a predetermined number of times in the estimated order, and uses the positions of the living body stored in the memory the predetermined number of times to determine that the living body is moving when the displacement of the position of the living body is greater than a predetermined value, and determines that the living body is not moving when the displacement is less than the predetermined value; A Doppler RCS threshold setting unit, which sets the Doppler RCS threshold by a prescribed method; a Doppler RCS threshold determination unit, which compares the calculated Doppler RCS value with the set Doppler RCS threshold, and determines whether the Doppler RCS value is less than the Doppler RCS threshold; as well as The situation estimation unit selectively estimates at least one of the posture of the organism and the action of the organism, wherein the estimation of the action of the organism is based on at least one of the determination result of the movement determination unit and the determination result of the Doppler RCS threshold determination unit, and the estimation of the posture of the organism is based on the determination result of the Doppler RCS threshold determination unit.

2. The sensor according to claim 1, wherein: the situation estimating unit, When the movement determination unit determines that the living body is not moving and the Doppler RCS threshold determination unit determines that the Doppler RCS value is less than the Doppler RCS threshold, the posture of the living body is estimated, When the Doppler RCS threshold determination unit determines that the Doppler RCS value is greater than the Doppler RCS threshold, the behavior of the living body is estimated.

3. The sensor according to claim 1, wherein: the situation estimating unit, When the movement determination unit determines that the living body is moving, it is determined that the behavior of the living body is moving.

4. The sensor according to any one of claims 1 to 3, wherein: At least three of the N transmitting antenna elements are respectively arranged at different positions in the vertical direction and the horizontal direction. At least three of the M receiving antenna elements are respectively arranged at different positions in the vertical direction and the horizontal direction. The memory stores a Doppler RCS value and information indicating a correspondence between a position of the living body in the vertical direction relative to the sensor, that is, a vertical position and a posture of the living body. The position estimating unit estimates a three-dimensional position as a position where the living body exists, wherein the three-dimensional position includes a vertical position, that is, a position of the living body in a vertical direction relative to the sensor. The situation estimating unit estimates the posture of the living body using the calculated Doppler RCS value, the vertical position, and the information indicating the correspondence relationship stored in the memory.

5. The sensor according to any one of claims 1 to 3, wherein: At least three of the N transmitting antenna elements are respectively arranged at different positions in the vertical direction and the horizontal direction. At least three of the M receiving antenna elements are respectively arranged at different positions in the vertical direction and the horizontal direction. The memory stores a Doppler RCS value and information indicating a correspondence between a position of the living body in the vertical direction relative to the sensor, that is, a vertical position and a posture of the living body. The position estimating unit estimates a three-dimensional position as a position where the living body exists, wherein the three-dimensional position includes a vertical position of the living body in a vertical direction relative to the sensor. The situation estimating unit estimates a posture probability as a posture of the living body, the posture probability being a probability of each of one or more predetermined postures that the living body can take, using the calculated Doppler RCS value, the vertical position, and the information indicating the correspondence relationship.

6. The sensor according to any one of claims 1 to 3, wherein: The situation estimating unit sequentially estimates the posture of the living body using the sequentially calculated Doppler RCS values ​​and the sequentially estimated position of the living body, and stores the sequentially estimated posture of the living body in the memory. The Doppler RCS threshold value setting unit sets the Doppler RCS threshold value based on the posture of the living body estimated at the latest timing among the postures of the living body stored in the memory.

7. The sensor according to any one of claims 1 to 3, wherein: the situation estimating unit, After successively estimating the posture of the living body using the successively calculated Doppler RCS values ​​and the successively estimated position of the living body, the successively estimated posture of the living body is stored in the memory, Next, before estimating the posture of the living body, the apparatus waits until the Doppler RCS values ​​calculated successively become below the Doppler RCS threshold, and estimates the posture of the living body estimated based on the N received signals when the Doppler RCS value becomes below the Doppler RCS threshold as a subsequent posture. The behavior of the living body is estimated using the previous posture of the living body, that is, the front posture and the estimated rear posture stored in the memory.

8. The sensor according to any one of claims 1 to 3, wherein: the situation estimating unit, After successively estimating the posture of the living body using the successively calculated Doppler RCS values ​​and the successively estimated position of the living body, the successively estimated posture of the living body is stored in the memory, Waiting until the Doppler RCS value calculated successively becomes less than the Doppler RCS threshold, When the Doppler RCS value becomes less than the Doppler RCS threshold, one or more next actions of the living body are estimated using the last estimated posture of the living body stored in the memory, that is, the previous posture, The behavior of the living body is estimated by using the Doppler RCS values ​​calculated successively and the positions of the living body estimated successively to determine one behavior from the one or more estimated next behaviors.

9. The sensor according to any one of claims 1 to 3, wherein: the situation estimating unit, After successively estimating the posture of the living body using the successively calculated Doppler RCS values ​​and the successively estimated position of the living body, the successively estimated posture of the living body is stored in the memory, Waiting until the Doppler RCS value calculated successively becomes less than the Doppler RCS threshold, When the Doppler RCS value becomes less than the Doppler RCS threshold, one or more next actions of the living body are estimated using the last estimated posture of the living body stored in the memory, that is, the previous posture, By using the Doppler RCS values ​​calculated successively and the positions of the living body estimated successively, the estimated probabilities of the one or more next actions, that is, the action probabilities, are estimated as the actions of the living body.

10. The sensor according to any one of claims 1 to 3, wherein: The presence / absence determination unit estimates the number of the living beings present in the specified range, and when the number of the living beings is 0, determines that the living beings do not exist in the specified range, and when the number of the living beings is 1 or more, determines that the living beings exist in the specified range.

11. An estimation device comprising a circuit and a memory, wherein: The circuit has: an acquisition unit, acquiring, from a receiver having M receiving antenna elements, N receiving signals received in a predetermined period in each of the M receiving antenna elements, the M receiving antenna elements respectively receiving the N receiving signals, the N receiving signals including a reflection signal, the reflection signal being obtained by a part of the N transmission signals transmitted by a transmission signal generator having N transmitting antenna elements being reflected or scattered by a living body, the N transmitting antenna elements respectively transmitting transmission signals to a predetermined range in which the living body may exist, M being a natural number greater than or equal to 3, and N being a natural number greater than or equal to 3; A first matrix calculation unit calculates, based on each of the N received signals obtained, a complex transfer function representing a propagation characteristic between the transmitting antenna element and the receiving antenna element in each N×M combination in which the N transmitting antenna elements and the M receiving antenna elements are combined one-to-one, and successively calculates an N×M first matrix having the calculated N×M complex transfer functions as components; a second matrix extracting unit that successively extracts the second matrix corresponding to the component affected by life activities by successively extracting the second matrix corresponding to the prescribed frequency range in the first matrix successively calculated by the first matrix calculating unit, the life activities including at least one of breathing, heartbeat and body movement of the organism; A presence / absence determination unit performs presence / absence determination to determine whether a living body exists within the specified range by a specified method; a position estimating unit that, after the presence determination unit determines that the living body exists within the predetermined range, sequentially estimates the position of the living body relative to the transmission signal generator and the receiver using the second matrix sequentially extracted by the second matrix extraction unit; A Doppler RCS calculation unit, (i) based on the position of the living body, the position of the transmission signal generator, and the position of the receiver, which are successively estimated by the position estimation unit, successively calculates a first distance and a second distance, wherein the first distance indicates the distance between the living body and the transmission signal generator, and the second distance indicates the distance between the living body and the receiver, and (ii) successively calculates a Doppler radar cross-sectional area (RCS) value for the living body using the calculated first distance and second distance; a movement determination unit that stores the successively estimated positions of the living body in the memory a predetermined number of times in the estimated order, and uses the positions of the living body stored in the memory the predetermined number of times to determine that the living body is moving when the displacement of the position of the living body is greater than a predetermined value, and determines that the living body is not moving when the displacement is less than the predetermined value; A Doppler RCS threshold setting unit, which sets the Doppler RCS threshold by a prescribed method; a Doppler RCS threshold determination unit, which compares the calculated Doppler RCS value with the set Doppler RCS threshold, and determines whether the Doppler RCS value is less than the Doppler RCS threshold; The condition estimation unit selectively estimates the posture of the organism and estimates the behavior of the organism, wherein the estimation of the behavior of the organism is based on at least one of the determination result of the movement determination unit and the determination result of the Doppler RCS threshold determination unit, and the estimation of the posture of the organism is based on the determination result of the Doppler RCS threshold determination unit.

12. An estimation method, performed in an estimation device having a circuit and a memory, wherein: Acquire N reception signals received in a predetermined period at each of the M reception antenna elements from a receiver having M reception antenna elements, the M reception antenna elements respectively receiving the N reception signals, the N reception signals including reflection signals, the reflection signals being obtained when a part of the N transmission signals transmitted by a transmission signal generator having N transmission antenna elements is reflected or scattered by a living body, the N transmission antenna elements respectively transmitting transmission signals to a predetermined range where the living body may exist, M is a natural number greater than or equal to 3, and N is a natural number greater than or equal to 3, Based on each of the N received signals obtained, for each N×M combination of the N transmitting antenna elements and the M receiving antenna elements in a one-to-one combination, a complex transfer function representing a propagation characteristic between the transmitting antenna element and the receiving antenna element in the combination is calculated, and a first N×M matrix having the calculated N×M complex transfer functions as components is successively calculated, By successively extracting second matrices corresponding to prescribed frequency ranges in the successively calculated first matrices, successively extracting the second matrices corresponding to components affected by life activities, wherein the life activities include at least one of breathing, heartbeat, and body movement of the organism, Determine the presence or absence of the organism within the specified range by a specified method, After determining that a living body exists within the predetermined range, the position of the living body relative to the transmission signal generator and the receiver is estimated one by one using the second matrix extracted one by one, Based on the successively estimated positions of the organism, the positions of the signal transmitter, and the positions of the receiver, a first distance and a second distance are successively calculated, wherein the first distance represents the distance between the organism and the signal transmitter, and the second distance represents the distance between the organism and the receiver. Using the calculated first distance and the calculated second distance, successively calculating Doppler radar cross-sectional area (RCS) values ​​for the biological body, storing the successively estimated locations of the living body in the memory a predetermined number of times in the estimated order, A movement determination is performed, using the position of the living body stored in the memory at the specified number of times, and when the displacement of the position of the living body is greater than a specified value, it is determined that the living body is moving, and when the displacement is less than the specified value, it is determined that the living body is not moving. The Doppler RCS threshold is set by the prescribed method. Performing a Doppler RCS threshold determination, comparing the calculated Doppler RCS value with the set Doppler RCS threshold, and determining whether the Doppler RCS value is below the Doppler RCS threshold, Selectively perform one of the estimation of the posture of the organism and the estimation of the movement of the organism, wherein the estimation of the movement of the organism is performed based on at least one of the determination result of the movement determination and the determination result of the Doppler RCS threshold determination, and the estimation of the posture of the organism is performed based on the determination result of the Doppler RCS threshold determination.

13. A program recording medium storing a program for causing a computer to execute an estimation method, wherein the estimation method is performed in an estimation device having a circuit and a memory, wherein: Acquire N reception signals received in a predetermined period at each of the M reception antenna elements from a receiver having M reception antenna elements, the M reception antenna elements respectively receiving the N reception signals, the N reception signals including reflection signals, the reflection signals being obtained when a part of the N transmission signals transmitted by a transmission signal generator having N transmission antenna elements is reflected or scattered by a living body, the N transmission antenna elements respectively transmitting transmission signals to a predetermined range where the living body may exist, M is a natural number greater than or equal to 3, and N is a natural number greater than or equal to 3, Based on each of the N received signals obtained, for each N×M combination of the N transmitting antenna elements and the M receiving antenna elements in a one-to-one combination, a complex transfer function representing a propagation characteristic between the transmitting antenna element and the receiving antenna element in the combination is calculated, and a first N×M matrix having the calculated N×M complex transfer functions as components is successively calculated, By successively extracting second matrices corresponding to prescribed frequency ranges in the successively calculated first matrices, successively extracting the second matrices corresponding to components affected by life activities, wherein the life activities include at least one of breathing, heartbeat, and body movement of the organism, Determine the presence or absence of the organism within the specified range by a specified method, After determining that a living body exists within the predetermined range, the position of the living body relative to the transmission signal generator and the receiver is estimated one by one using the second matrix extracted one by one, Based on the successively estimated positions of the organism, the positions of the signal transmitter, and the positions of the receiver, a first distance and a second distance are successively calculated, wherein the first distance represents the distance between the organism and the signal transmitter, and the second distance represents the distance between the organism and the receiver. Using the calculated first distance and the calculated second distance, successively calculating Doppler radar cross-sectional area (RCS) values ​​for the biological body, storing the successively estimated locations of the living body in the memory a predetermined number of times in the estimated order, A movement determination is performed, using the position of the living body stored in the memory at the specified number of times, and when the displacement of the position of the living body is greater than a specified value, it is determined that the living body is moving, and when the displacement is less than the specified value, it is determined that the living body is not moving. The Doppler RCS threshold is set by the prescribed method. Performing a Doppler RCS threshold determination, comparing the calculated Doppler RCS value with the set Doppler RCS threshold, and determining whether the Doppler RCS value is below the Doppler RCS threshold, Selectively perform one of the estimation of the posture of the organism and the estimation of the movement of the organism, wherein the estimation of the movement of the organism is performed based on at least one of the determination result of the movement determination and the determination result of the Doppler RCS threshold determination, and the estimation of the posture of the organism is performed based on the determination result of the Doppler RCS threshold determination.