Biological information processing device and motion information processing device

Abstract

To further enhance separation accuracy of information acquired from a measurement object.SOLUTION: A processor performs: transmission / reception processing for repeatedly transmitting a first signal used for the measurement of biological information on each of a plurality of persons to the plurality of persons and receiving a reflection signal for the repeatedly transmitted first signal; storage processing for calculating the biological information and the position where the biological information is detected from each of the reflection signals for the repeatedly transmitted first signal, and causing the storage part to store the time that the reflection signal is received, and the correspondence between the biological information and the position; extraction processing for, referring to the correspondence, calculating a first time-series change in the biological information for each position, and dividing the positions associated with the biological information into clusters on the basis of the calculated first time-series change; and generation processing for generating a second time-series change in the biological information for each of the clusters.SELECTED DRAWING: Figure 3

Description

[Technical Field]

[0001] The present invention relates to a biometric information processing device and a motion information processing device. [Background technology]

[0002] Techniques for acquiring biometric information of a person to be measured using various sensors and non-contact sensors such as radar are in practical use. For example, Patent Document 1 discloses a technique for detecting respiration and heart rate using radar, in which biometric signals of multiple people are separated using information on respiration rate and position obtained from short-term fluctuations in intensity information, amplitude, and phase in radar images spatially separated by radar signal processing. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Chinese Patent Application Publication No. 114469025 Summary of the Invention [Problem to be solved by the invention]

[0004] The respiratory rates of multiple subjects do not necessarily differ. In addition, multiple subjects may be located close to each other in a small space. Therefore, even if respiratory rate information and location information are used, it may not be possible to separate the biological information of multiple subjects.

[0005] An object of one aspect of the disclosed technology is to provide a biological information processing device and a motion information processing device that can further improve the accuracy of separating information acquired from a measurement object. [Means for solving the problem]

[0006] One aspect of the disclosed technology is exemplified by the following biometric information processing device. The biometric information processing device includes a storage unit and a processor connected to the storage unit. The processor executes the following operations: a transmission / reception process of repeatedly transmitting, to a plurality of persons, first signals used to measure biometric information of the plurality of persons and receiving reflected signals corresponding to the repeatedly transmitted first signals; a storage process of calculating, from each of the reflected signals corresponding to the repeatedly transmitted first signals, the biometric information and a position at which the biometric information was detected, and storing in the storage unit a correspondence relationship between the time of reception of the reflected signal, the biometric information, and the position; an extraction process of calculating, with reference to the correspondence relationship, a first time-series change in the biometric information for each position, and classifying the positions associated with the biometric information into clusters based on the calculated first time-series change; and a generation process of generating, for each of the clusters, a second time-series change in the biometric information.

[0007] When classifying the locations into clusters based on the biometric information, there is a risk that locations of multiple different people whose biometric information coincide with each other may be classified into the same cluster. In this biometric information processing device, the first time-series change in the biometric information is used when classifying the locations into clusters. It is considered that the first time-series change in the biometric information differs for each person. Therefore, even if there is a person whose biometric information temporarily matches, the biometric information processing device can more appropriately classify the locations into clusters. Consequently, the accuracy of separating the biometric information of the person being measured can be further improved.

[0008] The biometric information processing apparatus may further include the following feature: the extraction process extracts, from the correspondence relationship, a detection position where the biometric information has been detected continuously for a predetermined period or more, and extracts the detection position, the biometric information associated with the detection position, and the time associated with the detection position. The method further includes generating a second correspondence relationship between the positions of the biometric information, calculating the first time-series change by referring to the second correspondence relationship, and classifying the positions associated with the biometric information into the clusters based on the calculated first time-series change. The second correspondence relationship has a smaller data size than the first correspondence relationship. Therefore, the calculation load of the classifying process into the clusters is reduced in the biometric information processing device. Furthermore, the extraction process excludes, for example, detection positions where information resembling biometric information is falsely detected due to the influence of suddenly occurring noise or the like, or detection positions with a poor S / N ratio, thereby improving the accuracy of the generated biometric information.

[0009] The biometric information processing device may further include the following feature: The storage process may further include a process of suspending storage of the correspondence relationship when a body movement of any one of the plurality of people is detected based on the reflected signal. When the body movement of the person occurs, the position of the person may change significantly, or the person may move out of the range of the measurement signal. Therefore, if a reflected signal in which a body movement is detected is stored in the storage unit, there is a risk that the accuracy of the generated second time-series change of the biometric information may decrease. Since the biometric information processing device suspends storage of the correspondence relationship when a body movement is detected, the generation of such a low-accuracy second time-series change is prevented.

[0010] Here, the biometric information processing device may execute the extraction process even after the storage of the correspondence relationships has been interrupted, if the correspondence relationships stored in the storage unit are equal to or longer than a predetermined period. Here, the predetermined period is determined in advance as a period during which clustering is possible based on the first time-series variation of the biometric information. By having such a feature, the biometric information processing device can generate the second time-series variation of the biometric information for each of the clusters while reducing the influence of body movement.

[0011] According to another aspect of the disclosed technology, a motion information processing device includes a storage unit and a processor connected to the storage unit, and the processor executes a transmission / reception process of repeatedly transmitting, to a plurality of devices, first signals used to measure motion information of the plurality of devices and receiving reflected signals corresponding to the repeatedly transmitted first signals, a storage process of calculating, from each of the reflected signals corresponding to the repeatedly transmitted first signals, the motion information and a position at which the motion information was detected and storing, in the storage unit, a correspondence relationship between the time of reception of the reflected signal, the motion information, and the position, an extraction process of calculating, with reference to the correspondence relationship, a first time-series change in the motion information for each position and, based on the calculated first time-series change, classifying the positions associated with the motion information into clusters, and a generation process of generating, for each of the clusters, a second time-series change in the motion information.

[0012] When classifying the positions into clusters based on the motion information, there is a risk that the positions of multiple different devices whose motion information coincides may be classified into the same cluster. This motion information processing device uses the first time-series variation of the motion information when classifying the positions into clusters. It is considered that the first time-series variation of the motion information differs for each individual device. Therefore, this motion information processing device can more appropriately classify the positions into clusters even when there are devices whose motion information temporarily coincides. Consequently, it is possible to further improve the accuracy of separating the motion information of the devices to be measured. [Effects of the Invention]

[0013] The present biometric information processing apparatus can further improve the accuracy of separating biometric information relating to a person who is a measurement target. [Brief explanation of the drawings]

[0014] [Figure 1] FIG. 1 is a diagram illustrating an example of a biological information processing apparatus according to an embodiment. [Figure 2] FIG. 2 is a diagram illustrating an example of a hardware configuration of the biological information processing apparatus according to the embodiment. [Figure 3]FIG. 3 is a diagram illustrating an example of a processing block of the biological information processing apparatus according to the embodiment. [Figure 4] FIG. 4 is a first diagram illustrating an example of a time trend map according to the embodiment. [Figure 5] FIG. 5 is a second diagram illustrating an example of a time trend map according to the embodiment. [Figure 6] FIG. 6 is a diagram illustrating an example of a result of clustering performed by the extraction unit in the embodiment. [Figure 7] FIG. 7 is a diagram illustrating a time series change in the respiratory rate generated by the generating unit in the embodiment. [Figure 8] FIG. 8 is a diagram illustrating an example of a processing flow of the biological information processing apparatus according to the embodiment. [Figure 9] FIG. 9 is a diagram illustrating an example of a processing block of a biometric information processing device according to a first modified example. [Figure 10] FIG. 10 is a diagram illustrating an example of a processing flow of the biological information processing apparatus according to the first modified example. [Figure 11] FIG. 11 is a diagram illustrating an example of a monitoring device according to a second modification. DETAILED DESCRIPTION OF THE INVENTION

[0015] <Application example> An application example of the present invention will be described. A biometric information processing device 1 according to this application example is placed in a room 50, as shown in Fig. 1. In Fig. 1, three people 31, 32, and 33 are lying on a bed 52 placed in the room 50.

[0016] The biometric information processing device 1 acquires biometric information from people 31, 32, and 33 as measurement targets. The biometric information processing device 1 repeatedly transmits a measurement signal used to acquire the biometric information, for example, at a predetermined interval. The biometric information processing device 1 then receives reflected signals obtained by reflecting the measurement signal off the people 31, 32, and 33. The reflected signals received by the biometric information processing device 1 are signals onto which information related to the biometric information of the people 31, 32, and 33 is superimposed. Furthermore, the reflected signals received by the biometric information processing device 1 may also be superimposed with signals reflected by the wall 53 and furniture installed in the room 50.

[0017] The biometric information processing device 1 detects biometric information based on the measurement signal and the reflected signal, and estimates the distance and direction from the biometric information processing device 1 to the location where the biometric information was detected. Hereinafter, in this specification, the distance and direction are collectively referred to as location information.

[0018] The biometric information processing device 1 stores the correspondence between the acquired biometric information, position information, and the time at which the reflected signal was received in the storage unit 103 (see FIG. 2). As a method for detecting the biometric information and position information from the reflected signal, various known methods can be adopted.

[0019] When the correspondence relationships for a predetermined analysis period are stored in the storage unit 103, the biological information processing device 1 generates a first time-series change of the biological information for each position based on the correspondence relationships stored in the storage unit 103.

[0020] The biometric information processing device 1 performs clustering using the first time-series changes of the generated biometric information as input data, thereby separating the biometric information of each of the persons 31, 32, and 33. In the clustering, for example, a method in which the number of clusters is input by a user (k-means) or a method in which the number of clusters is automatically estimated (x-means, VBGMM, elbow method, etc.) can be used. In this clustering, for example, positions where the waveforms showing the first time-series changes are similar are classified into the same cluster.

[0021] The biometric information processing device 1 determines biometric information for each cluster at each time. For example, the biometric information processing device 1 determines the statistical value of the biometric information at each position classified into the same cluster at each time as the biometric information of the cluster at that time. For example, the biometric information processing device 1 determines the average value or median value of the statistical value of the biometric information as the biometric information of the cluster. The biometric information processing device 1 generates time-series changes in the biometric information for each cluster based on the determined biometric information.

[0022] <Embodiment> The embodiment will be further described below with reference to the drawings. Fig. 1 is a diagram showing an example of a biometric information processing device 1 according to the embodiment. As described in the application example, the biometric information processing device 1 is installed in a room 50.

[0023] The biometric information processing device 1 is a device that acquires biometric information from people 31, 32, and 33 as measurement targets. The biometric information processing device 1 transmits measurement signals used to acquire the biometric information. The biometric information processing device 1 then receives reflected signals reflected by the people 31, 32, and 33 to acquire the biometric information of the people 31, 32, and 33. As described in the application example, signals reflected by the wall 53 and furniture installed in the room 50 may also be superimposed on the reflected signals. The biometric information processing device 1 is placed on a platform 51 so that its height is approximately the same as that of the people 31, 32, and 33 lying on the bed 52. Examples of the biometric information include information indicating respiration, information indicating heart rate, and information indicating body movement. In this embodiment, the respiratory rate will be described as an example of biometric information.

[0024] Next, a description will be given of the hardware configuration of the biometric information processing device 1. Fig. 2 is a diagram showing an example of the hardware configuration of the biometric information processing device 1 according to the embodiment. The biometric information processing device 1 includes a transmission / reception unit 101, a control unit 102, a storage unit 103, and an output unit 104. The transmission / reception unit 101, the control unit 102, the storage unit 103, and the output unit 104 are connected by a connection bus B1.

[0025] The transmitting / receiving unit 101 includes a transmitting unit 111 and a receiving unit 112. The transmitting unit 111 transmits a measurement signal used to measure the biometric information of the people 31, 32, and 33. The receiving unit 112 receives a reflected signal reflected from the people 31, 32, and 33.

[0026] The transmitter 111 transmits a measurement signal using, for example, radio waves, light waves, sound waves, or ultrasound waves. The transmitter 111 may transmit a measurement signal using a frequency-modulated continuous wave radar (FMCW) method, which sweeps a 4 GHz radio wave having a frequency of 60 to 64 GHz every 100 microseconds. When transmitting radio waves, the transmitter 111 may include a signal generator that generates an unmodulated or frequency-, phase-, or amplitude-modulated signal, and a transmitting antenna that radiates the signal generated by the signal generator into space. The signal generator may be a single oscillator such as a voltage-controlled oscillator (VCO), a combination of a VCO and a phase-locked loop (PLL) circuit, or a modulator that receives a baseband signal from the controller 102. The signal generator may also include a D / A converter that converts a digital signal into an analog signal, or a direct digital synthesizer (DDS). In this embodiment, since the distance and direction from the biometric information processing device 1 are also acquired, radar is suitable as the measurement signal.

[0027] The receiving unit 112 also includes, for example, a receiving antenna that receives a reflected signal from the person 31, a mixer as a multiplication circuit that demodulates the reflected signal and outputs an analog baseband signal or an intermediate frequency (IF) signal, and an A / D conversion unit that converts the analog baseband signal or the IF signal into a digital signal.

[0028] The control unit 102 is, for example, a Central Processing Unit (CPU). The control unit 102 executes programs stored in the storage unit 103 to control each unit in the biometric information processing device 1 and to perform various information processes. For example, the control unit 102 generates the respiratory rates of the persons 31, 32, and 33 using signals received from the persons 31, 32, and 33. The biometric information generation process executed by the control unit 102 will be described in detail later.

[0029] The storage unit 103 stores programs executed by the control unit 102 and various data used in the processes executed by the control unit 102. For example, the storage unit 103 is, for example, a random access memory (RAM), a read only memory (ROM), a hard disk drive, a solid state drive, or the like.

[0030] The output unit 104 notifies the user of the results of the processing executed by the control unit 102, and outputs data related to the results to an external device. The output unit 104 may be configured to output data related to the biometric information of the person being measured to an external device by various communication methods such as various wireless communications and wired communications. Examples of the output unit 104 include a display, a speaker, and a printer.

[0031] <Processing blocks of the biometric information processing device 1> 3 is a diagram showing an example of a processing block of the biometric information processing device 1 according to the embodiment. The biometric information processing device 1 includes a transmission control unit 11, a reflected signal storage unit 12, a map creation unit 13, an extraction unit 14, a generation unit 15, and a notification unit 16. The control unit 102 executes a computer program executable in the storage unit 103, thereby causing the biometric information processing device 1 to perform processing as each unit of the transmission control unit 11, the reflected signal storage unit 12, the map creation unit 13, the extraction unit 14, the generation unit 15, the notification unit 16, etc.

[0032] The transmission control unit 11 controls the transmission and reception of measurement signals by the transmission unit 111. For example, the transmission control unit 11 controls the transmission unit 111 to repeatedly transmit measurement signals used to measure the respiratory rates of the people 31, 32, and 33 a specified number of times.

[0033] The reflected signal storage unit 12 detects a waveform related to the respiratory rate based on the measurement signal transmitted by the transmission control unit 11 and the reflected signal received by the transmission control unit 11, and calculates position information indicating the location where the waveform was detected. The reflected signal storage unit 12 calculates the respiratory rate based on the detected waveform related to the respiratory rate. The reflected signal storage unit 12 stores the correspondence between the calculated respiratory rate, the position information, and the time when the reflected signal was received in the storage unit 103. The processing by the reflected signal storage unit 12 is repeatedly executed each time a measurement signal is transmitted by the transmission control unit 11 and a reflected signal is received.

[0034] When the transmission control unit 11 transmits the measurement signal a specified number of times, the map creation unit 13 generates a time trend map based on the above correspondence stored in the storage unit 103. The time trend map is a two-dimensional map with location information on one axis and time on the other axis. In the time trend map, the respiratory rate is stored in each area defined by the location information and time. In the time trend map, "0" is stored in areas where the respiratory rate is not detected.

[0035] FIG. 4 is a diagram showing an example of the time trend map 131 in the embodiment. The vertical axis of the time trend map 131 is a "time slot" indicating time, and the horizontal axis is a "bin index" indicating location information. One "time slot" corresponds to an observation interval. The "bin index" is a number assigned to each bin that represents the distance and direction included in the location information in one dimension. In other words, in this embodiment, the "bin" can be considered an example of location information. Note that the time trend map 131 illustrated in FIG. 4 illustrates a case where a predetermined analysis period corresponds to 30 "time slots."

[0036] The respiration rate is stored at coordinates partitioned by "time slot" and "bin index." In the time trend map 131, the respiration rate may be indicated, for example, by the shade of color of each coordinate. The time trend map 131 is a correspondence that associates the respiration rate with time and position.

[0037] Returning to FIG. 3, the extraction unit 14 refers to the time trend map 131 and extracts bins in which the respiration rate has been detected consecutively for a predetermined number of slots or more in the "time slot." An example of the predetermined number of slots is 20 slots. The extraction unit 14 generates a time trend map that includes the extracted bins.

[0038] 5 is a diagram showing an example of a time trend map 132 in an embodiment. The time trend map 132 is a time trend map obtained by extracting bins in which the respiratory rate has been detected for a predetermined number of consecutive slots or more from the time trend map 131. Since the time trend map 132 is a map obtained by extracting a portion of the time trend map 131, it can be seen that the amount of data in the time trend map 132 is smaller than that in the time trend map 131.

[0039] The extraction unit 14 generates a time series change in the respiratory rate for each bin based on the time trend map 132. It is believed that the waveform of the time series conversion of the respiratory rate differs for each individual. Therefore, the extraction unit 14 inputs the time series change in the respiratory rate and performs clustering for each coordinate of the time trend map 132. In this clustering, the clustering is performed based on the input waveform indicating the time series change in the respiratory rate. In other words, bins having similar waveforms indicating the time series change in the respiratory rate are classified into the same cluster. Note that various known methods can be used to determine whether the waveforms are similar.

[0040] 6 is a diagram illustrating an example of the result of clustering by the extraction unit 14 in the embodiment. In Fig. 6, the time trend map 132 is clustered into three clusters: "Cluster 1," "Cluster 2," and "Cluster 3." For example, "Cluster 1" corresponds to person 31, "Cluster 2" corresponds to person 32, and "Cluster 3" corresponds to person 33.

[0041] The generating unit 15 generates a time series change in the respiratory rate for each cluster clustered by the extracting unit 14. For example, the generating unit 15 may calculate a statistical value for each time slot for each bin classified into each cluster, and use the calculated statistical value as the respiratory rate for that time slot of that cluster. Examples of statistical values ​​include the mean and median.

[0042] Fig. 7 is a diagram illustrating an example of time series changes in the respiration rate generated by the generation unit 15 in the embodiment. For reference, the example in Fig. 7 also illustrates time series changes in the respiration rate of persons 31, 32, and 33 measured using a respiratory belt that measures respiration rate by wearing a belt around the chest. Referring to Fig. 7, it can be seen that the respiration rate generated by the generation unit 15 matches the respiration rate measured using the respiratory belt with high accuracy.

[0043] The notification unit 16 separates the biometric information generated by the generation unit 15 into clusters and outputs them. Examples of methods for separating the information into clusters include changing the display mode for each cluster or changing the display position for each cluster. The notification unit 16 outputs, for example, the respiratory waveforms of the persons 31, 32, and 33. The notification unit 16 may, for example, display the biometric information on the output unit 104 as a display.

[0044] <Processing flow> 8 is a diagram showing an example of a processing flow of the biometric information processing apparatus 1 according to the embodiment. Hereinafter, an example of a processing flow of the biometric information processing apparatus 1 will be described with reference to FIG.

[0045] The processes from step S1 to step S2 are repeated a specified number of times. In step S1, a measurement signal is transmitted by transmission control unit 11. When a reflected signal of the measurement signal is received by receiving unit 112, reflected signal storage unit 12 detects a waveform indicating the respiratory rate based on the measurement signal and the reflected signal, and calculates the position information of the location where the waveform was detected.

[0046] In step S2, the reflected signal storage unit 12 calculates the respiratory rate based on the waveform indicating the respiratory rate acquired in step S1. The reflected signal storage unit 12 stores in the storage unit 103 the correspondence between the calculated respiratory rate, the position information, and the time when the reflected signal was received.

[0047] In step S3, the map creation unit 13 generates a time trend map 131 based on the correspondence stored in the storage unit 103 in step S2. In step S4, the extraction unit 14 refers to the time trend map 131 generated in step S3 and extracts bins in which the respiratory rate has been detected consecutively in a "time slot" for a predetermined number of slots or more. The extraction unit 14 generates a time trend map 132 that includes the extracted bins.

[0048] In step S5, the extraction unit 14 generates a time series change in the biological information for each bin based on the time trend map 132 generated in step S4, and performs clustering based on the generated time series change in the biological information.

[0049] In step S6, the generating unit 15 generates biometric information for each cluster obtained by clustering in step S5. In step S7, the notifying unit 16 outputs the biometric information generated in step S6 separately for each cluster.

[0050] <Effects of the embodiment> If time-series changes in the biometric information are not used for clustering, and the people 31, 32, and 33 to be measured are located in a small area, and if the biometric information of the people 31, 32, and 33 happens to match, there is a risk that the bins corresponding to the people 31, 32, and 33 will be classified into the same cluster. Therefore, if time-series changes in the biometric information are not used for clustering, there is a risk that the biometric information of the people 31, 32, and 33 will not be separated. There is also a risk that the number of people to be measured will be determined to be less than three.

[0051] In this embodiment, by using time-series changes in the biometric information for clustering, even if the breathing rates of the persons 31, 32, and 33 temporarily become the same, the bins corresponding to the persons 31, 32, and 33 can be classified into different clusters. Therefore, according to this embodiment, the biometric information processing device 1 can further improve the accuracy of separating the biometric information of multiple persons.

[0052] In this embodiment, a time trend map 132 is generated by extracting bins in which the respiration rate has been detected consecutively for a predetermined number of slots or more from the time trend map 131. The time trend map 132 has a smaller amount of data than the time trend map 131. Therefore, in this embodiment, According to the present invention, the biometric information processing device 1 can reduce the computational load of the extraction unit 14 generating time-series changes in biometric information and the generation unit 15 generating time-series changes in respiratory rate. Furthermore, because bins in which respiratory rates are detected consecutively for a predetermined number of slots or more are extracted, bins that are erroneously detected due to factors such as sudden noise, bins with poor S / N ratios, and bins affected by signal superposition are excluded from the generated biometric information. It is assumed that the time ratio of such excluded bins to be extracted as biometric information decreases due to a decrease in signal quality. Furthermore, even if biometric information is extracted, the waveform may not reflect actual body surface displacement. Therefore, in this embodiment, such bins are excluded, thereby improving the accuracy of the generated biometric information.

[0053] <First Modification> In the embodiment described above, the process of storing the correspondence between the respiratory rate, position information, and the time when the reflected signal was received in the memory unit 103 was repeatedly executed until the measurement signal was transmitted a specified number of times. In the first modification, the process of storing the correspondence in the memory unit 103 is continued or interrupted depending on whether or not there is body movement of the persons 31, 32, and 33. Components common to the embodiment are given the same reference numerals, and their description will be omitted. The first modification will be described below with reference to the drawings.

[0054] 9 is a diagram showing an example of a processing block of a biometric information processing device 1A according to Modification 1. The biometric information processing device 1A differs from the biometric information processing device 1 according to the embodiment in that it includes a reflected signal storage unit 12A instead of the reflected signal storage unit 12 and a map creation unit 13A instead of the map creation unit 13.

[0055] When storing the correspondence between the respiratory rate, position information, and the time at which the reflected signal was received in the memory unit 103, the reflected signal memory unit 12A determines whether or not a body movement has occurred in at least one of the persons 31, 32, and 33, based on the received reflected signal. If the reflected signal memory unit 12A determines that a body movement has occurred, it suspends storing the correspondence in the memory unit 103.

[0056] The map creation unit 13A determines whether the correspondence relationships stored in the storage unit 103 are sufficient for generating the time trend map 131, and if they are insufficient, suspends generation of the time trend map 131. For example, the map creation unit 13A sets a threshold value that allows generation of the time trend map 131 even if the number of times is less than a specified number. Then, the map creation unit 13A may generate the time trend map 131 if the correspondence relationships for reflected signals of measurement signals that have been transmitted the threshold number of times or more are stored in the storage unit 103.

[0057] 10 is a diagram showing an example of a processing flow of the biometric information processing apparatus 1A according to Modification 1. Hereinafter, an example of a processing flow of the biometric information processing apparatus 1A will be described with reference to FIG.

[0058] In step S21, the reflected signal storage unit 12A determines whether or not a body movement has occurred in any of the people 31, 32, and 33 based on the reflected signal received in step S1. If it is determined that a body movement has occurred (YES in step S21), the process proceeds to step S22. If it is determined that a body movement has not occurred (NO in step S21), the process proceeds to step S2.

[0059] In step S22, the map creating unit 13A determines whether or not the above-mentioned correspondence relationship for the reflected signals of measurement signals transmitted the threshold number of times or more is stored in the storage unit 103. If the above-mentioned correspondence relationship for the reflected signals of measurement signals transmitted the threshold number of times or more is stored (YES in step S22), the processing from step S3 onwards in Fig. 8 is executed. If the above-mentioned correspondence relationship for the reflected signals of measurement signals transmitted the threshold number of times or more is not stored (YES in step S22), the processing from step S3 onwards in Fig. 8 is executed. 22 NO), the process is terminated.

[0060] When the persons 31, 32, and 33 move, the positions of the persons 31, 32, and 33 may change significantly. Furthermore, due to the body movement, the persons 31, 32, and 33 may move out of the range of the measurement signal. Therefore, when a reflected signal in which body movement is detected is stored in the memory unit 103, there is a risk that the reflected signals before and after the body movement will not be consistent. Consequently, the accuracy of the generated time-series change in the respiratory rate may decrease. According to the first modification, when body movement is detected, the storage in the memory unit 103 is interrupted. Therefore, the consistency of the reflected signal due to the influence of body movement is maintained as much as possible, and the accuracy of the generated time-series change in the respiratory rate is improved.

[0061] <Second Modification> In the embodiment described above, an example in which biological information is acquired by the biological information processing device 1 has been described, but the disclosed technology is not limited to the acquisition of biological information. In the second modification, an example in which a motor device is monitored will be described.

[0062] 11 is a diagram showing an example of a monitoring device 1B according to a second modified example. Monitoring device 1B is placed in a factory 50A and monitors a plurality of motor devices 40A, 40B, and 40C placed within factory 50A. When there is no need to distinguish between the plurality of motor devices 40A, 40B, and 40C, they are also referred to as motor devices 40.

[0063] Monitoring device 1B repeatedly transmits measurement signals used to measure characteristic quantities related to vibrations of motor devices 40A, 40B, and 40C. Monitoring device 1B then receives reflected signals of the repeatedly transmitted measurement signals from motor devices 40A, 40B, and 40C.

[0064] The monitoring device 1B acquires vibration information indicating vibrations of the motor devices 40A, 40B, and 40C from the reflected signal. The monitoring device 1B also estimates the distance and direction from the monitoring device 1B to the location where the vibration information was detected. The vibration information can be acquired using the same method as for acquiring biological information in the embodiment. The distance and direction can also be estimated using the same method as for estimating distance and direction in the embodiment. The monitoring device 1B then stores the correspondence between the time when the reflected signal was received, the vibration information, and the position (distance and direction) in the memory unit 103. The vibration information is an example of "operation information."

[0065] When the correspondence relationships for a predetermined analysis period are stored in the storage unit 103, the monitoring device 1B generates a first time-series change in the vibration information for each position based on the correspondence relationships stored in the storage unit 103.

[0066] Monitoring device 1B performs clustering using the generated first time-series changes in the vibration information as input data, thereby separating the vibration information of each of motor devices 40A, 40B, and 40C.

[0067] The monitoring device 1B determines vibration information for each cluster at each time. For example, at each time, the monitoring device 1B determines the statistical value of vibration information for each position classified into the same cluster as the vibration information of that cluster at that time. The monitoring device 1B determines the average or median of the statistical values ​​of vibration information as the biological information of that cluster. Based on the determined vibration information, the monitoring device 1B generates time-series changes in the vibration information for each cluster.

[0068] The motor device 40 is a device that drives other equipment by rotating its output shaft. Regular or irregular vibrations can occur in the motor device 40 as the output shaft rotates. It is believed that vibration-related features (e.g., frequency, amplitude, period, etc.) vary from device to device due to differences in operating state, load, and individual differences between motor devices 40. Therefore, by generating time-series changes in vibration-related features and clustering them in the same manner as in the embodiment, it is possible to separate vibration-related information for each motor device 40 with high accuracy.

[0069] According to the second modification, by using time-series changes in the vibration information for clustering, even if the vibration frequencies per unit time of motor devices 40A, 40B, and 40C temporarily become the same, the bins corresponding to motor devices 40A, 40B, and 40C can be classified into different clusters. Therefore, according to the second modification, monitoring device 1B can further improve the accuracy of separating the vibration information of multiple motors 40.

[0070] As described above, monitoring device 1B according to the second modified example can monitor motor devices 40A, 40B, and 40C using the same method as in the embodiment.

[0071] In the second modified example, the motor device 40 is cited as an example of an object to be monitored by the monitoring device 1B, but the object to be monitored by the monitoring device 1B is not limited to the motor device 40. Examples of objects to be monitored by the monitoring device 1B include equipment that involves vibrations in the operation of a pump device, production equipment, etc.

[0072] <Other variations> In the embodiment described above, the time trend map 132 is generated from the time trend map 131, but generation of the time trend map 132 may be omitted. The time trend map 132 is obtained by extracting bins from the time trend map 131 in which the respiratory rate has been detected consecutively for a predetermined number of slots or more. Even if a waveform that is accidentally recognized as a respiratory rate occurs in a location other than the persons 31, 32, and 33, such as a wall 53, it is unlikely that such a waveform will continue. Therefore, by generating the time trend map 132, it is possible to efficiently exclude accidental waveforms caused by the wall 53, etc.

[0073] The embodiments and modifications disclosed above can be combined with each other.

[0074] <Appendix 1> A memory unit (103); a processor (102) connected to the storage unit (103), The processor (102) a transmission / reception process (11) for repeatedly transmitting a first signal to a plurality of persons (31, 32, 33) to be used for measuring biometric information of each of the plurality of persons (31, 32, 33) and receiving a reflected signal corresponding to the repeatedly transmitted first signal; a storage process (12) for calculating the biometric information and the position where the biometric information was detected from each of the reflected signals corresponding to the repeatedly transmitted first signal, and storing the correspondence between the time when the reflected signal was received, the biometric information, and the position in the storage unit (103); an extraction process (14) of calculating a first time series change of the biometric information for each of the positions by referring to the correspondence relationship, and classifying the positions associated with the biometric information into clusters based on the calculated first time series change; and executing a generation process (15) for generating a second time-series change of the biological information for each of the clusters. Biometric information processing device (1, 1A). <Appendix 2> The extraction process (14) extracting from the correspondence relationship (131) a detection position where the biometric information has been detected continuously for a predetermined period or more, and generating a second correspondence relationship (132) between the detection position, the biometric information associated with the detection position, and the time associated with the detection position; The method further includes a process of calculating the first time-series change by referring to the second correspondence relationship (132), and dividing the positions associated with the biological information into the clusters based on the calculated first time-series change. A biometric information processing device (1, 1A) according to appendix 1. <Appendix 3> The storage process (12A) further includes a process of suspending storage of the correspondence relationship when a body movement of any one of the plurality of persons is detected based on the reflected signal. A biometric information processing device (1A) according to any one of Supplementary Note 1 and Supplementary Note 2. <Appendix 4> The processor: Even after the storage of the correspondence relationship has been interrupted, if the correspondence relationship stored in the storage unit (103) is for a specified period or more, the extraction process is executed. A biometric information processing device (1A) according to appendix 3. <Appendix 5> A memory unit (103); a processor (102) connected to the storage unit (103), The processor (102) a transmission / reception process (11) for repeatedly transmitting a first signal to a plurality of devices (40A, 40B, 40C) to be used for measuring operational information of each of the plurality of devices (40A, 40B, 40C) and receiving a reflected signal corresponding to the repeatedly transmitted first signal; a storage process (12) for calculating the motion information and the position where the motion information was detected from each of the reflected signals corresponding to the repeatedly transmitted first signal, and storing the correspondence between the time when the reflected signal was received, the motion information, and the position in the storage unit (103); an extraction process (14) of calculating a first time-series change of the motion information for each of the positions by referring to the correspondence relationship, and classifying the positions associated with the motion information into clusters based on the calculated first time-series change; and executing a generation process (15) for generating a second time-series change of the motion information for each of the clusters. Motion information processing device (1B). [Explanation of symbols]

[0075] 1. Biometric information processing device 1A Biometric information processing device 1B...Monitoring device 11 Transmission control section 12...Reflected signal storage section 12A·Reflected signal storage section 13. Map Creation Department 13A Map Creation Section 14...Extraction part 15...Generation part 16·Notification Department 31·person 32·person 33··person 40 Motor device 40A motor unit 40B Motor unit 40C··Motor unit 50 rooms 50A··Factory 51 units 52 Bed 53 Wall 101 Transmitter / Receiver 102 Control unit 103...Storage section 104 Output section 111··Transmitter 112··Receiver 131··Time Trend Map 132··Time Trend Map B1 Connecting bus R1...area

Claims

1. A memory unit; a processor connected to the storage unit, The processor: a transmitting and receiving process of repeatedly transmitting, to a plurality of persons, first signals used to measure biometric information of each of the plurality of persons, and receiving reflected signals corresponding to the repeatedly transmitted first signals; a storage process of calculating the biometric information and the position where the biometric information was detected from each of the reflected signals corresponding to the repeatedly transmitted first signal, and storing in the storage unit a correspondence relationship between the time when the reflected signal was received, the biometric information, and the position; an extraction process of calculating a first time-series change in the biometric information for each of the locations by referring to the correspondence relationship, and classifying the locations associated with the biometric information into clusters based on the calculated first time-series change; and executing a generation process for generating a second time-series change of the biological information for each of the clusters. Biometric information processing device.

2. The extraction process includes: extracting from the correspondence relationship a detection position where the biometric information has been detected continuously for a predetermined period or more, and generating a second correspondence relationship between the detection position, the biometric information associated with the detection position, and the time associated with the detection position; calculating the first time-series change by referring to the second correspondence relationship, and classifying the positions associated with the biological information into the clusters based on the calculated first time-series change; The biometric information processing device according to claim 1 .

3. the storing process further includes a process of suspending storing of the correspondence relationship when a body movement of any one of the plurality of persons is detected based on the reflected signal. The biometric information processing device according to claim 1 .

4. The processor: Even after the storage of the correspondence relationship has been interrupted, if the correspondence relationship stored in the storage unit is for a specified period or more, the extraction process is executed. The biometric information processing device according to claim 3 .

5. A memory unit; a processor connected to the storage unit, The processor: a transmission / reception process of repeatedly transmitting a first signal to a plurality of devices, the first signal being used to measure operational information of each of the plurality of devices, and receiving a reflected signal corresponding to the repeatedly transmitted first signal; a storage process of calculating the motion information and the position where the motion information was detected from each of the reflected signals corresponding to the repeatedly transmitted first signal, and storing a correspondence relationship between the time when the reflected signal was received, the motion information, and the position in the storage unit; an extraction process of calculating a first time-series change of the motion information for each of the positions by referring to the correspondence relationship, and classifying the positions associated with the motion information into clusters based on the calculated first time-series change; and executing a generation process for generating a second time-series change of the motion information for each of the clusters. Motion information processing device.

Images