Detection device, detection system, and model generation device

Abstract

To provide a detection device capable of reducing a communication load.SOLUTION: A detection device includes: a sensor that detects first information regarding a user's biological activity; and a processing unit that acquires second information D10b, D10c from another detection device that detects the second information D10b, D10c regarding the user's biological activity, generates third information D12 based on the first information and the second information D10b, D10c, and transmits the third information D12 to a server installed separately.SELECTED DRAWING: Figure 7

Description

[Technical field]

[0001] The present invention relates to a detection device, a detection system and a model generating device. [Background technology]

[0002] 2. Description of the Related Art There is known a device that uses a microwave Doppler sensor to check the safety of a user. There is also known a system that monitors or checks the safety of a user in a toilet or bathroom. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2012-75861 A [Patent Document 2] Patent Publication No. 2021-149220 [Patent Document 3] JP 2016-218773 A [Patent Document 4] Special Publication No. 2022-547258 Summary of the Invention [Problem to be solved by the invention]

[0004] When monitoring the status of users in each of a plurality of partitions, if a plurality of detection devices in the same partition each communicate with a server, the communication load becomes large.

[0005] The present invention has been made in consideration of the above problems, and has an object to reduce the communication load. [Means for solving the problem]

[0006] The present invention is a detection device that includes a sensor that detects first information regarding a user's biological activity, and a processing unit that acquires second information regarding the user's biological activity from another detection device that detects the second information, generates third information based on the first information and the second information, and transmits the third information to a separately installed server.

[0007] The present invention is a detection system comprising a plurality of first detection devices, one installed in each of a plurality of sections and acquiring first information regarding the vital activity of a user in the corresponding section, and a plurality of second detection devices, one or more installed in each of the plurality of sections and acquiring second information regarding the vital activity of a user in the corresponding section, wherein the plurality of first detection devices acquire the second information from a second detection device among the plurality of second detection devices that is installed in the same section, generate third information based on the first information and the second information, and transmit the third information to a server installed outside the plurality of sections.

[0008] The present invention is a model generation device comprising a memory and a processing unit that acquires multiple pieces of information output by multiple detection devices that are installed in multiple compartments and detect information regarding the biological activity of users in the corresponding compartments, generates a model for determining the status of users in the multiple compartments based on the multiple pieces of information, and transmits the model to the multiple detection devices.

[0009] The present invention is a detection device comprising a sensor that transmits a first electromagnetic wave, receives a second electromagnetic wave reflected by an object from the first electromagnetic wave, and generates an analog signal related to a movement of the object based on the first electromagnetic wave and the second electromagnetic wave, a processing unit that converts the analog signal into a digital signal and generates information related to the movement of the object based on the digital signal, and an adjuster that adjusts the amplitude of the analog signal based on the digital signal and outputs the adjusted analog signal to the processing unit.

[0010] The present invention is a detection device that includes a sensor that transmits a first electromagnetic wave, receives a second electromagnetic wave reflected from the first electromagnetic wave by an object, and generates information about the object based on the first electromagnetic wave and the second electromagnetic wave, a window through which the first electromagnetic wave and the second electromagnetic wave pass, and a heater that heats the window to suppress condensation on the window. Effect of the Invention

[0011] According to the present invention, the communication load can be reduced. [Brief description of the drawings]

[0012] [Figure 1] FIG. 1 is a block diagram of a detection device according to a first embodiment. [Diagram 2] FIG. 1 is a block diagram showing a detection system according to a first embodiment. [Diagram 3] FIG. 2 is a plan view showing an example of a section in the first embodiment. [Figure 4] FIG. 2 is a plan view showing an example of a section in the first embodiment. [Diagram 5] FIG. 2 is a block diagram of a server according to the first embodiment. [Figure 6] FIG. 2 is a sequence diagram according to the first embodiment. [Figure 7] 4 is a flowchart of a process executed by a processing unit of the detection device in the first embodiment. [Figure 8] 13A is a diagram showing the voltage versus time of the signal 26e in the first embodiment, and FIG. 13B is a table showing the signal 26f. [Figure 9] 1 is a table showing information D10a to D10c in the first embodiment. [Figure 10] 13 is a table showing information D12 in the first embodiment. [Figure 11] 11 is a flowchart of a process executed by a processor of a server according to the first embodiment. [Figure 12] FIG. 2 is a sequence diagram according to the first embodiment. [Figure 13]4 is a flowchart of a process executed by a processing unit of the detection device in the first embodiment. [Figure 14] 11 is a table showing information D16, D20, and D22 in the first embodiment. [Figure 15] 11 is a flowchart of a process executed by a processor of a server according to the first embodiment. [Figure 16] 4 is a flowchart showing a process executed by a processing unit of the detection device in the first embodiment. [Figure 17] FIG. 11 is a schematic diagram of a vehicle in a second embodiment. [Figure 18] FIG. 11 is a sequence diagram according to the second embodiment. [Figure 19] FIG. 11 is a cross-sectional view of a detection device according to a third embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0013] Facilities such as hotels and other accommodation facilities, hospitals, schools, police stations, or public institutions are divided into multiple sections. One or multiple users (users) are staying in each section. For example, in a hotel, multiple sections correspond to multiple rooms. In a hotel, each room has a toilet and a bathroom. For this reason, if only one detection device for detecting a user's biometric information is installed in one room, the biometric information of the user may not be detected when the user is in the toilet or bathroom. Therefore, it is possible to install multiple detection devices in one room. When multiple detection devices are installed in one room, the communication load increases if information from the multiple detection devices in the multiple rooms is sent individually to a server.

[0014] In the following embodiment, among multiple detection devices installed in one room, only one detection device communicates with the server, which reduces the communication load between the detection device and the server.

[0015] Hereinafter, embodiments of the present invention will be described with reference to the drawings. EXAMPLES

[0016] FIG. 1 is a block diagram of a detection device according to a first embodiment. The detection devices 10a to 10c each include a sensor 12, a PGA (Programmable Gain Amplifier) ​​18, and a memory 22. The sensor 12 includes a transmitting antenna 13a, a receiving antenna 13b, a high-frequency circuit 11, an amplifier 16, and an LPF (Low Pass Filter) 17. The high-frequency circuit 11 includes an oscillator 14 and a mixer 15. The high-frequency circuit 11 also includes an amplifier and the like, but the description thereof will be omitted. The antennas 13a and 13b are, for example, patch antennas provided on a substrate. The antenna 13a transmits a signal 26a generated by the oscillator 14 to a user 25 or the like. The antenna 13b receives a signal 26b reflected by the user 25 when the signal 26a is irradiated to the user 25. The signals 26a and 26b are electromagnetic waves, for example, microwaves or millimeter waves. The frequency of the signal 26a is, for example, 10 GHz or more and 120 GHz or less, for example, around 24 GHz. The mixer 15 mixes the signals 26a and 26b and outputs the converted signal 26c. The frequency of the signal 26c output by the mixer 15 corresponds to the difference between the frequency of the signal 26a and the frequency of the signal 26b. As a result, the signal 26a becomes an analog signal corresponding to the movement of the area irradiated with the signal 26a.

[0017] The amplifier 16 amplifies the signal 26c. The LPF 20 suppresses signals from the amplified signal 26c that have a higher frequency than the biovibration signal, passes signals with a frequency of the biovibration signal (e.g., 10 Hz or less), and outputs a filtered signal 26d. The PGA 18 amplifies the signal 26d, and outputs the amplified signal as a signal 26e. The signal 26e is an analog signal that mainly contains bioinformation among the analog signals corresponding to the movement of the user 25.

[0018] The processing unit 20 is a processor such as a CPU (Central Processing Unit) or a microcomputer, and executes processing in cooperation with software. The processing unit 20 includes an A / D (Analog-Digital) converter 21 and interfaces (I / F) 23a to 23c. The A / D converter 21 converts an analog signal 26e into a digital signal 26f. The I / F 23a transmits and receives information to the server 30. The I / F 23b transmits and receives information to the other detection devices 10a to 10c. The I / F 23c transmits and receives information to the sensor 28. Of the detection devices 10a to 10c, the detection devices 10b and 10c may not be provided with the I / Fs 23a and 23c. The processing unit 20 generates information such as the heart rate, respiration rate, body movement, and heart rate variability of the user 25 based on the signal 26f. The processing unit 20 also determines the state of the user from the generated information such as the heart rate, respiration rate, body movement, and heart rate variability. The memory 22 is a non-volatile or volatile memory, and stores setting conditions for performing processing, data during the calculation of information, programs, and the like.

[0019] FIG. 2 is a block diagram showing a detection system in the first embodiment. A facility is provided with a plurality of sections 40a to 40f. In the detection system 100, a plurality of detection devices 10a and 10b are installed in the plurality of sections 40a to 40d, respectively, and a plurality of detection devices 10a to 10c are installed in the plurality of sections 40e and 40f, respectively. A sensor 28 is installed in each of the plurality of sections 40a to 40f. The sections 40a to 40f correspond to rooms in a lodging facility, for example. The sections 40a and 40b are symmetrical rooms in the lodging facility, the sections 40c and 40d are symmetrical rooms, and the sections 40e and 40f are symmetrical rooms.

[0020] The detection device 10a corresponds to a parent device, and the detection devices 10b and 10c correspond to child devices. The sensor 28 detects indicators such as temperature, humidity, illuminance, sound volume, sound frequency, door opening / closing, and / or light switch on / off in each of the sections 40a to 40f. A plurality of sensors for detecting different indicators may be installed in one of the sections 40a to 40f. The detection device 10a, the detection devices 10b and 10c, and the sensor 28 in one of the sections 40a to 40f are connected wirelessly or wired, such as Wi-Fi (registered trademark) or Bluetooth (registered trademark). The detection device 10a in each of the sections 40a to 40f is connected to the server 30 and the management terminal 32 via a network 33. The network 33 is a wireless or wired network, such as a LAN (Local Area Network) or a wireless LAN.

[0021] 3(a) to 4(b) are plan views showing examples of sections in the first embodiment. Sections 40a and 40b in FIG. 3(a) and FIG. 3(b) are examples of single rooms. In sections 40a and 40b, one bed 80, a pillow 81, a desk 82, and a chair 83 are set. The bed 80 is, for example, a single bed. In a bathroom 89, a toilet seat 84, a washstand 85, and a bathtub 86 are provided. The bathroom 89 is a sub-section that serves both as a toilet and a bathroom. A door 87 is provided in the bathroom 89, and a door 88 is provided in sections 40a and 40b. The detection device 10a is installed at the corner of sections 40a and 40b, and the detection device 10b is installed in the bathroom 89. When only one detection device 10a is provided in section 40a, if the detection device 10a is installed at the corner of section 40a, the signal 26a will spread throughout section 40a. However, when the user is in the bathroom 89, if the door 87 is closed or there are water droplets on the curtain or the like, the detection device 10a may not be able to obtain the user's biometric information. Therefore, the detection device 10b is installed in the bathroom 89. As a result, the signal 26b transmitted by the detection device 10b spreads throughout the bathroom 89 in the section 40a. The sections 40a and 40b are symmetrical rooms.

[0022] Section 40c in FIG. 4(a) is an example of a double room. In section 40c, bed 80 is larger than sections 40a and 40b, and bed 80 is, for example, a double bed. Also, chair 83 is installed next to desk 82. Therefore, the area of ​​section 40c is larger than the areas of sections 40a and 40b. Detection device 10a is installed at the corner of section 40c, and detection device 10b is installed in bathroom 89. Section 40d in FIG. 2 is a section symmetrical to section 40c. Many rooms in lodging facilities such as hotels or residential facilities such as apartments have symmetrical structures and arrangements with respect to a wall.

[0023] Section 40e in FIG. 4(b) is an example of a triple room. Three beds 80 are provided in section 40e. Therefore, the area of ​​section 40e is larger than that of section 40c. Because the area of ​​section 40e is larger, three detection devices 10a to 10c are provided in section 40e. Section 40f in FIG. 2 is a section symmetrical to section 40e.

[0024] A compartment may be a patient room in a hospital, a classroom in a school, or a room or seat in police and public institutions, for example.

[0025] FIG. 5 is a block diagram of the server in the first embodiment. The server 30 includes a processor 34, a memory 35, an input / output device 36, and an internal bus 37. The processor 34 is, for example, a CPU, and executes processes such as model generation. The memory 35 is, for example, a volatile memory or a non-volatile memory, and stores data and the like used when the processor 34 executes processes. The memory 35 may store a program executed by the processor 34. The input / output device 36 inputs data and information acquired by the processor 34 from an external device, and outputs data output by the processor 34 to an external device. The internal bus 37 connects the processor 34, the memory 35, and the input / output device 36, and transmits data and the like. The server 30 cooperates with software to grasp the status of each of the partitions 40a to 40f, and also functions as a model generation device.

[0026] First, the process of each device when the server 30 grasps the status of the sections 40a to 40f and generates a model will be described. FIG. 6 is a sequence diagram in the first embodiment. Three detection devices 10a to 10c are installed in one section. The detection device 10a generates information D10a (S12). Next, the detection device 10a makes an information request R10b to the detection device 10b. The detection device 10b transmits information D10b on the user's biological activity acquired by the detection device 10b to the detection device 10a. Next, the detection device 10a makes an information request R10c to the detection device 10c. Next, the detection device 10c transmits information D10c on the user's biological activity acquired by the detection device 10c to the detection device 10a.

[0027] Next, the detection device 10a generates information D12 based on the information D10a to D10c (S18). Next, the detection device 10a transmits the information D12 to the server 30. The server 30 receives the information D12 from the detection device 10a of the multiple sections 40a to 40f and stores the information D12 in the memory 35. Next, the server 30 generates a model D14 for the detection device 10a to determine the state of the user based on the stored information D12 while referring to sections with similar structures (S36). Next, the server 30 transmits the generated model D14 to the detection device 10a, and the detection device 10a receives the model D14. In the example of FIG. 2, the sections with similar structures are sections 40a and 40b, sections 40c and 40d, sections 40e and 40f.

[0028] Fig. 7 is a flowchart of the process executed by the processing unit of the detection device in the first embodiment. Fig. 8(a) is a diagram showing the voltage with respect to time in the signal 26e in the first embodiment, and Fig. 8(b) is a table showing the signal 26f. Figs. 9(a) to 9(c) are tables showing the information D10a to D10c in the first embodiment. Fig. 10 is a table showing the information D12 in the first embodiment.

[0029] As shown in FIG. 7, the processing unit 20 of the detection device 10a acquires a signal 26f (S10). The acquisition of the signal 26f will be described. The A / D converter 21 acquires a signal 26e from the PGA 18. As shown in FIG. 8(a), the signal 26e is an analog signal of voltage against time, and a waveform corresponding to the user's biological information such as heart rate and respiration is superimposed. The A / D converter 21 converts the analog signal 26e into a digital signal 26f. As shown in FIG. 8(b), the signal 26f is a voltage V(1)...V(i)...V(n). i is an integer from 1 to n, and corresponds to the time in FIG. 8(a). That is, the signal 26f is information in which V(i) is arranged in a time series. For example, when the sampling interval of the A / D converter 21 is t, the time interval between V(i) and V(i+1) is t.

[0030] Next, the processing unit 20 generates information D10a based on the signal 26f (S12). As shown in FIG. 9(a), the information D10a stores "10a" as the device ID, and stores "heart rate," "respiratory rate," "body movement," and "heart rate variability" as data. The device ID is an identification code indicating the detection device that acquired the signal 26e. The "heart rate" is data corresponding to the user's heart rate per minute. The "respiratory rate" is data corresponding to the user's respiratory rate per minute. The "body movement" is the user's body movement other than the heart rate and breathing, such as the number and strength of movements per minute. The "heart rate variability" is equivalent to the heart rate variability. These data are generated by Fourier transforming the signal 26f and then analyzing the Fourier transformed signal by the processing unit 20.

[0031] As shown in Fig. 6 and Fig. 7, the processing unit 20 acquires information D10b from the detection device 10b and acquires information D10c from the detection device 10c (S14). As shown in Fig. 9(b) and Fig. 9(c), the information D10b and D10c each store "10b" as the device ID, and each store "heart rate", "respiratory rate", "body movement", and "heart rate variability" as data. The processing units 20 of the detection devices 10b and 10c generate the information D10b and D10c, respectively, in the same manner as in S10 and S12 of Fig. 7, and transmit the information D10b and D10c, respectively, to the detection device 10a based on the requests R10b and R10c of Fig. 6.

[0032] 7, the processing unit 20 acquires sensor data from the sensor 28 (S16). The sensor data is data indicating, for example, temperature, humidity, illuminance, sound volume, sound frequency, whether a door is open or closed, and / or whether a light switch is on or off. The order of executing S10, S12, S14, and S16 can be set appropriately.

[0033] Next, the processing unit 20 generates information D12 based on the information D10a to D10c and the sensor data (S18). As shown in FIG. 10, the information D12 stores "XX" as a header, "YY" as a packet ID, "10a", "10b" and / or "10c" as a device ID, and "40f" as a section ID, and stores "heart rate", "respiratory rate", "body movement", "heart rate variability", and "sensor data" as data. The header is the header of the packet transmitted by the server 30, and the packet ID is an identification code indicating the packet. The device ID is an identification code of the detection device that acquired the signal 26e. The section ID is an identification code of the detected section. The section ID may use the MAC (Media Access Control) address or IP (Internet Protocol) address of the detection device 10a of each section. When one of the detection devices 10a to 10c mainly detects the biometric information of the user, the processing unit 20 causes the information D12 to include one of the information D10a to D10c corresponding to the one detection device that detected the biometric information of the user. When multiple detection devices of the detection devices 10a to 10c detect the biometric information of one or multiple users, the processing unit 20 causes the information D12 to include multiple pieces of information D10a to D10c corresponding to the multiple detection devices that detected the biometric information of the user.

[0034] As shown in Fig. 7, the processing unit 20 transmits information D12 to the server 30 (S20). The communication between the processing unit 20 and the server 30 may be encrypted. Next, the processing unit 20 judges whether to end the process (S22). For example, if the processing unit 20 is instructed to end the process by the server 30, the processing unit 20 judges Yes, otherwise it judges No. If No, the process returns to step S10. If Yes, the process ends.

[0035] FIG. 11 is a flowchart of a process executed by the processor of the server in the first embodiment. First, the processor 34 acquires information D12 corresponding to each of the sections 40a to 40f from each of the detection devices 10a installed in the sections 40a to 40f in FIG. 2 (S30). Next, the processor 34 stores the information D12 in the memory 35 for each group (S32). For example, in FIG. 2, the sections 40a are one group, and the sections 40b are one group other than the section 40a. Similarly, each of the sections 40a to 40f is a different group. At this time, information on the user's state is stored in the memory 35 in association with each information D12. The user's state is, for example, a state in which the user is relaxed, a state in which the user is asleep, a state in which the user is tense, a state in which the user is abnormal, etc. The information on the user's state may be included in the information D12 or may be acquired from elsewhere.

[0036] Next, the processor 34 determines whether or not to generate a model (S34). For example, when a predetermined period of time has elapsed or when an external device instructs the generation of a model, the processor 34 determines Yes, otherwise it determines No. If No, the process returns to S30, and in S30 and S32, the information D12 is accumulated.

[0037] If the answer is Yes in S34, the processor 34 generates a model D14 for each group based on the information D12 (S36). For example, the processor 34 uses the information D12 for each group and information on the state of the user when the information D12 is acquired as training data, and generates the model D14 by machine learning. The training data may include the sensor data in the information D12, information on the date such as weekdays and holidays or days of the week, information on the time period, and / or the number of users in the section. For example, even if the heart rate of a user is the same, the user is often asleep when the illuminance is low or late at night, and is often relaxed when the illuminance is high or until noon. Next, the processor 34 transmits the model D14 generated for each group to the detection device 10a of the sections 40a to 40f included in the corresponding group (S38). Then, the process ends.

[0038] The process of each device when the detection device 10a uses the model to determine the state of the user will be described. FIG. 12 is a sequence diagram in the first embodiment. Three detection devices 10a to 10c are installed in one section. As in FIG. 6, the detection device 10a generates information D10a (S12). The detection device 10a receives information D10b and D10c from the detection devices 10b and 10c, respectively.

[0039] Next, the detection device 10a uses the model to determine the state of the user based on the information D10a to D10c (S52). When the detection device 10a determines that the user (or the section) is abnormal, it transmits information D16 indicating the abnormality to the server 30. At this time, the information D16 indicating the abnormality may be, for example, information with a small amount of information including information indicating "abnormality" without including data. When the server 30 receives the information D16, it transmits information D17 confirming receipt to the detection device 10a. The detection device 10a may transmit the information D16 multiple times until it receives the information D17. In this case, when the detection device 10a receives the information D17, it stops transmitting the information D16. This makes it possible to prevent the information D16, which is important information, from not reaching the server 30 even if the network 33 is congested. Next, the server 30 determines whether to request detailed information based on the information D16 (S72). When the server 30 determines that detailed information is requested, it transmits a request R18 for detailed information to the detection device 10a. Next, the server 30 transmits the gaze information D24 to the management terminal 32. The manager operating the management terminal 32 recognizes that gaze is required.

[0040] Next, the detection device 10a transmits a request R20 for detailed information to the detection device 10c that detected the biometric information of the user. Next, the detection device 10c transmits the detailed information D20 to the detection device 10a. Next, the detection device 10a transmits information D22 including the information D20 to the server 30. Next, the server 30 determines whether the user (or the section) is abnormal or not based on the information D22 (S80). When the server 30 determines that the user (or the section) is abnormal, the server 30 transmits abnormality information D26 to the management terminal 32. The manager operating the management terminal 32 recognizes that the abnormality has occurred. The management terminal 32 or the server 30 may transmit an alarm to a mobile terminal such as a smartphone of the person who notifies the abnormality, and notify the abnormality.

[0041] Fig. 13 is a flowchart of the process executed by the processing unit of the detection device in Example 1. Fig. 14(a) to Fig. 14(c) are tables showing information D16, D20 and D22 in Example 1, respectively.

[0042] As shown in FIG. 13, the processing unit 20 of the detection device 10a acquires a model D14 from the server 30 (S50). Next, the processing unit 20 executes S10, S12, S14, and S16 in the same manner as in FIG. 7. Next, the processing unit 20 judges the state of the user (or the section) based on the information D10a to D10c and the sensor data (S52). The model D14 is the model generated in step S36 of FIG. 11. For example, when the information D10a to D10c and the sensor data are input to the model, the processing unit 20 can judge the state of the user (or the section). Next, the processing unit 20 judges whether the state of the user (or the section) is abnormal or not (S54). If the result is No, the process returns to S10.

[0043] If the result is Yes, the processing unit 20 transmits information D16 to the server 30 (S56). As shown in FIG. 14(a), the information D16 stores "XX" as a header, "YY" as a packet ID, "10a", "10b" and / or "10c" as a device ID, "40f" as a section ID, and "abnormal" as a state, and stores "heart rate", "respiratory rate", "body movement", "heart rate variability", and "sensor data" as data. The information D16 includes the user's state in addition to the information D12 in FIG. 10. The device ID "10c" indicates that the detection device 10c has detected the user's biometric information. The user's state "abnormal" indicates that the state of the user (or section) determined using the model D14 is "abnormal". The information D16 may not include data, and may only include information indicating that the user (or section) is abnormal. If the information D16 includes urgent information, the server 30 transmits information D17 to the detection device 10a upon receiving the information. The detection device 10a transmits packets including the information D16 multiple times until it receives the information D17. This allows the information D16 to reach the server 30 even if packets of the information D16 are lost. The urgent information D16 is, for example, information indicating that the heart rate or breathing of a user whose heart rate had been measured has stopped. The detection device 10a may continue to transmit the information D16 to the server 30 a finite number of times as short packets using a connectionless protocol such as UDP (User Datagram Protocol).

[0044] Returning to FIG. 13, the processing unit 20 receives a request R18 from the server 30 (S58). Next, the processing unit 20 judges whether detailed information of the detection device 10a is requested (S60). If Yes, proceed to S64. If No, the processing unit 20 transmits a request R20 to the detection device from which detailed information is requested, among the detection devices 10b and 10c, and acquires information D20 from the detection device 10b or 10c (S62). As shown in FIG. 14(b), the information D20 stores "10c" as the device ID and V(1)...V(i)...V(n) as data. V(i) corresponds to the signal 26f (see FIG. 8(b)) acquired by the processing unit 20 of the detection device 10c.

[0045] Returning to FIG. 13, the processing unit 20 generates information D22 (S64). As shown in FIG. 14(c), the information D22 stores "XX" as a header, "YY" as a packet ID, "10a", "10b" and / or "10c" as a device ID, and "40f" as a partition ID, and stores V(1)...V(i)...V(n) as data. When transmitting data V(i) of the detection device 10a, the data V(i) corresponds to the signal 26f acquired in S10. When transmitting data V(i) of the detection device 10b or 10c, the data V(i) corresponds to the information D20 acquired in S62.

[0046] 13, the processing unit 20 transmits the information D22 to the server 30 (S66). Next, the processing unit 20 judges whether or not to end the process (S68). If the answer is No, the process returns to step S10. If the answer is Yes, the process ends.

[0047] FIG. 15 is a flowchart of the process executed by the processor of the server in the first embodiment. First, the processor 34 acquires the information D16 from the detection device 10a (S70). Next, the processor 34 determines whether or not a request for detailed information is necessary based on the information D16 (S72). For example, when the information D16 is received from the detection device 10a for a section where no users are supposed to be present, the processor 34 determines that a request for detailed information is unnecessary. If the answer is No, the process ends. When the information D16 is received without performing S72, the process may proceed to S74.

[0048] In S72, if the answer is Yes, the processor 34 transmits a detailed information request R18 to the detection device 10a (S74). Next, the processor 34 transmits gaze information D24 to the management terminal 32 (S76). The order of S74 and S76 may be reversed, and S76 may not be performed. Next, the processor 34 receives detailed information D22 from the detection device 10a (S78). Next, the processor 34 determines whether the user (or the section) is abnormal or not based on the information D22 (S80). For example, the processor 34 compares the information with that of another section in the same group, and if the detailed information of the other section in the same group is similar, it determines that the user is not abnormal. If the answer is No, the process ends. If the answer is Yes, the processor 34 transmits abnormality information D26 to the management terminal 32 (S82). Then, the process ends.

[0049] When the detection devices 10a to 10c each communicate with the server 30 via the network 33, the traffic on the network 33 increases, and the communication between the detection devices 10a to 10c and the server 30 becomes unstable.

[0050] Therefore, according to the first embodiment, one detection device 10a (first detection device) is installed in each of the multiple sections 40a to 40f, and one or more detection devices 10b and 10c (second detection devices) are installed in each of the multiple sections 40a to 40f. As shown in S10 of FIG. 7, the sensor 12 of the detection device 10a detects a signal 26f. The signal 26f is first information regarding the biological activity of the user in the section. As shown in S14, the processing unit 20 acquires information D10b and D10c from the detection devices 10b and 10c. The information D10b and D10c is second information regarding the biological activity of the user in the section. The processing unit 20 generates information D10a based on the signal 26f as shown in S12, and further generates information D12 (third information) based on the information D10a to D10c as shown in S18. As shown in S20, the processing unit 20 transmits the information D12 to a server 30 separately installed outside the section. In this way, the detecting device 10a generates information D12 to be transmitted to the server 30 based on the information D10b and D10c of the detecting devices 10b and 10c. This makes it possible to suppress traffic on the network 33 and reduce the communication load compared to a case in which each of the detecting devices 10a to 10c transmits the information D10a to D10c to the server 30.

[0051] 9(a) to 10, the processing unit 20 generates only a part of the information D10a to D10c as the information D12. This can further reduce the communication load between the detection device 10a and the server 30.

[0052] The signal 26f is information in which values ​​corresponding to the movement of at least a part of the user's body are arranged in a time series. As shown in FIG. 10, the information D12 includes information corresponding to the user's heart rate and / or respiratory rate, but does not include the signal 26f. In this way, the processing unit 20 does not transmit the signal 26f to the server 30, but transmits the user's heart rate, respiratory rate, etc. This can reduce the communication load between the detection device 10a and the server 30.

[0053] The processing unit 20 generates information D10a (fourth information) corresponding to the heart rate and / or respiratory rate of the user based on the signal 26f, as in S12 of Fig. 7. This can reduce the communication load between the detection device 10a and the server 30.

[0054] The detection devices 10b and 10c may transmit the data V(i) to the detection device 10a. However, in this case, the communication load between the detection device 10a and the detection devices 10b and 10c increases. Therefore, as shown in FIG. 9(b) and FIG. 9(c), the information D10b and D10c are information corresponding to the heart rate and / or respiratory rate of the user generated by the detection devices 10b and 10c. This allows the detection devices 10b and 10c to operate autonomously. Therefore, the communication load between the detection device 10a and the detection devices 10b and 10c can be reduced. As shown in FIG. 10, the information D12 includes at least one of the information D10a to D10c, and does not include the signal 26f. This allows the communication load between the detection device 10a and the server 30 to be reduced.

[0055] As shown in S30 of FIG. 11, the processor 34 of the server 30 acquires a plurality of pieces of information D12 transmitted by the detection devices 10a installed in the plurality of sections 40a to 40f. As shown in S36, a model D14 for determining the state of the user in the plurality of sections 40a to 40f is generated based on the plurality of pieces of information D12. As shown in S38, the model D14 is transmitted to the plurality of detection devices. As shown in S52 of FIG. 13, the processing unit 20 uses the model D14 received from the server 30 to determine the state of the user based on the information D10a to D10c. In this way, the server 30 generates the model D14, and the information D12 of many sections 40a to 40f is used as teacher data to generate the model D14 for determining the state of the user. Therefore, the accuracy of the model D14 can be improved. In addition, if the detection device 10a in each of the sections 40a to 40f creates a model, the load on the memory capacity of the detection device 10a and the load on the processing unit 20 become large, and the cost of the detection device 10a increases. By having the server 30 create the model D14, the load on the detection device 10a can be reduced.

[0056] Also, as a comparative example, when creating a model using information D12 of the detection device 10a in one section as teacher data, it takes time to collect the teacher data, and it takes time to create the model D14. In the first embodiment, the model D14 can be created using not only the information D12 of the detection device 10a in one section, but also the information D12 of many similar sections 40f as teacher data. Therefore, it is possible to collect more teacher data in a short period of time. Therefore, it is possible to create a highly accurate model D14 in a short period of time. In addition, the created model D14 can be transmitted to the detection devices 10a in many similar sections via the network 33. Furthermore, the accuracy of the model D14 can be improved by adding the time, the day of the week, and information of other sensors 28 to the teacher data.

[0057] As shown in FIG. 2 to FIG. 4(b), when the size, shape, and arrangement of the equipment in the rooms are different in the sections 40a to 40f, the reflection state of the signal 26b, for example, is different. For example, furniture may reflect the signal 26b less well than the wall. Therefore, it may be better to use different models for determining the state of the user for each type of the sections 40a to 40f. In the first embodiment, the sections 40a to 40f are divided into a plurality of groups. The processor 34 generates a plurality of models D14 based on the information D12 transmitted by the detection device 10a corresponding to each of the plurality of groups, and transmits the models D14 to the detection device 10a of the corresponding group. This allows a model suitable for the type of the sections 40a to 40f to be created, improving the accuracy of the model.

[0058] Rooms that are symmetrical on the left and right, such as sections 40a and 40b, sections 40c and 40d, and sections 40e and 40f, may be grouped together in the same group, or in three groups. Also, rooms that are symmetrical on the left and right may be grouped together in different groups, making a total of six groups.

[0059] As shown in FIG. 13, the processing unit 20 transmits the signal 26f to the server 30 in S64 and S66 based on an instruction from the server 30. Alternatively, the processing unit 20 acquires the information D20 (fifth information) of FIG. 14(b) detected by the detection device 10b or 10c from the detection device 10b or 10c in S62. As shown in S66, the processing unit 20 transmits the information D20 to the server 30. In this way, when the server 30 wants to determine an abnormality of the user using more detailed information, the detection device 10a can transmit the data V(i) to the server 30. The data amount of the data V(i) is very large. Therefore, as shown in FIG. 6 and FIG. 7, in normal operation, the detection device 10a transmits the information D12 with a small data amount to the server 30. This allows a margin in the load of communication between the server 30 and the multiple partitions 40a to 40f. As shown in Figures 12 and 13, when an abnormality is detected in one of the sections 40a to 40f, even if the detection device 10a that detected the abnormality transmits information D22 containing a large amount of data to the server 30, it is possible to prevent the communication capacity from being exceeded.

[0060] Sensor 12 transmits signal 26a (first electromagnetic wave) within the zone, receives signal 26b (second electromagnetic wave) reflected from signal 26a by an object such as a user, and generates signal 26e based on signals 26a and 26b. In this manner, by using a radar using microwave or millimeter wave electromagnetic waves, information regarding the movement of users within the zone can be detected with high accuracy. Sensor 12 may be a sensor other than a sensor using electromagnetic waves.

[0061] Next, a flow in which the processing unit 20 adjusts the gain of the PGA 18 in FIG. 1 will be described. FIG. 16 is a flowchart showing the process executed by the processing unit of the detection device in the first embodiment. The processing unit 20 acquires the signal 26f (S90). Next, the processing unit 20 judges whether the voltage V of the signal 26f is equal to or greater than the threshold value Th1 (S92). The processing unit 20 may compare the maximum value of V(i) (i is an integer from 1 to n) of the signal 26f with the threshold value Th1. The processing unit 20 may also judge Yes when a predetermined number or more of points in V(i) are equal to or greater than the threshold value Th1. Furthermore, the threshold value Th1 is set to the maximum value (e.g., FFFF) of the data converted by the A / D converter 21, and when V(i) that is the maximum value continues, it is considered that the A / D converter 21 is saturated. Therefore, in such a case, the processing unit 20 may judge Yes. If Yes, the processing unit 20 reduces the gain of the PGA 18 (S94). Then, the process proceeds to S99.

[0062] If the answer is No in S92, the processing unit 20 judges whether or not the voltage V is equal to or less than the threshold value Th2 (S96). The processing unit 20 may compare the minimum value of V(i) (i is an integer from 1 to n) of the signal 26f with the threshold value Th2. The processing unit 20 may also judge Yes if the number of points in V(i) that are equal to or less than the threshold value Th2 is a predetermined number or more. If the answer is No, the processing unit 20 proceeds to S99. If the answer is Yes, the processing unit 20 increases the gain of the PGA 18 (S98). Then, the processing unit 20 proceeds to S99. Then, the processing unit 20 judges whether or not to end (S99). If the answer is No, the processing unit 20 returns to step S90. If the answer is Yes, the processing unit 20 ends.

[0063] The sensor 12 generates an analog signal 26e related to the user's biological activity. The A / D converter 21 of the processing unit 20 converts the analog signal 26e into a digital signal 26f, and the processing unit 20 generates information D10a related to the user's biological activity based on the digital signal 26f. In such detection devices 10a to 10c, for example, when the user approaches the detection devices 10a to 10c, the amplitude of the signal 26e increases. This causes clipping or saturation in the A / D converter 21, and the analog signal 26e is not normally converted into a digital signal. Also, when the user moves away from the detection devices 10a to 10c or when the user is asleep, the amplitude of the signal 26e decreases. This reduces the accuracy of the digital signal 26f. Therefore, as shown in FIG. 16, the PGA 20 (adjuster) adjusts the amplitude of the analog signal 26e based on the digital signal 26f, and outputs the adjusted analog signal 26e to the processing unit 20. This can increase the sensitivity of the sensor 12. The PGA 20 may be used in a detection system in which the sensor 12 generates an analog signal 26e related to the movement of an object, and the processor 20 generates information related to the movement of the object. EXAMPLES

[0064] The second embodiment is an example in which the detection devices 10a to 10c are installed in a vehicle such as a school bus. FIG. 17 is a schematic diagram of a vehicle in the second embodiment. A plurality of seats 92 are provided in a vehicle 90 such as a school bus. A plurality of detection devices 10a to 10c are provided on the ceiling of the vehicle 90. The detection device 10a is installed in the center of the vehicle 90, and the detection devices 10b and 10c are installed in the front and rear of the vehicle, respectively. The detection device 10a is connected to the detection devices 10b and 10c wirelessly or by wire. The detection device 10a is connected to an external mobile terminal, for example, wirelessly. The wireless method is, for example, a mobile communication method or Wi-Fi Direct (registered trademark). The structure of the detection devices 10a to 10c is the same as that of the first embodiment.

[0065] FIG. 18 is a sequence diagram in the second embodiment. The detection devices 10a to 10c, a mobile terminal 94, and a terminal 96 are provided. The mobile terminal 94 is a terminal capable of communicating with the detection device 10a, and is, for example, a mobile phone, a smartphone, or a tablet. The terminal 96 is a mobile phone, a smartphone, a landline phone, or a personal computer, and is a terminal managed by an administrator. In FIG. 18, S12 to S52 are the same flow as FIG. 12 in the first embodiment, and the description is omitted. When the detection device 10a determines in S52 that the user (or the inside of the vehicle 90) is abnormal, the detection device 10a transmits information D16 indicating the abnormality to the mobile terminal 94. The mobile terminal 94 transmits information D26 indicating that there is an abnormality to the terminal 96. The administrator who manages the terminal 96 recognizes that there is an abnormality in the vehicle 90. For example, the administrator can recognize abnormalities such as the user being left behind in the vehicle 90, the user being injured or ill, or the user not moving.

[0066] As in the second embodiment, the detection devices 10a to 10c may be installed on vehicles other than buildings. EXAMPLES

[0067] Example 3 is an example in which a window for passing electromagnetic waves output from an antenna is heated. FIG. 19 is a cross-sectional view of the detection devices 10a to 10c according to Example 3. A window 50 is provided in a housing 52. A substrate 54 is provided in the housing 52. The antennas 13a and 13b, the sensor 12, and the processing unit 20 are provided on the substrate 54. A heat-conducting plate 56 is provided to thermally connect the substrate 54 and the window 50. The housing 52 is a metal plate such as stainless steel. The window 50 is a material that allows the signals 26a and 26b to pass and has good thermal conductivity, such as a thermally conductive resin plate. The heat-conducting plate 56 is a metal plate with good thermal conductivity, such as a copper plate. The thermal conductivity of the heat-conducting plate 56 is higher than the thermal conductivity of the housing 52, the window 50, and the substrate 54, for example. The other configurations of the detection devices 10a to 10c are the same as those of Example 1.

[0068] In the third embodiment, when the detection device 10b is installed in the bathroom 89, water droplets and the like adhere to the window 50. This makes it difficult for the signal 26a transmitted by the antenna 13a and the signal 26b received by the antenna 13b to pass through the window 50. Therefore, the heat conductive plate 56 (heater) heats the window 50 to suppress condensation on the window 50. This makes it possible to suppress the adhesion of water droplets and the like to the window 50. This makes it possible to suppress the signals 26a and 26b from being difficult to pass through the window 50.

[0069] As the heater, a heater may be provided on the window 50. As in the third embodiment, the heat generated by the sensor 12 is used to suppress condensation on the window 50, thereby reducing power consumption associated with heating. The heater may be provided in a detection device in which the sensor 12 generates an analog signal 26e related to the movement of the object and the processing unit 20 generates information related to the movement of the object.

[0070] The present invention is not limited to the above-described embodiment, but can be modified in various ways without departing from the spirit and scope of the present invention. [Explanation of symbols]

[0071] 10a, 10b, 10c Detector, 11 High frequency circuit, 12 Sensor, 13a, 13b Antenna, 16, 18 Amplifier, 20 Processing section, 25 User, 26a, 26b, 26c, 26d, 26e, 26f Signal, 21 A / D converter, 22 Memory, 26a, 26b, 26c, 26d, 26e, 26f Signal, 28 Sensor, 30 Server, 32 Management terminal, 33 Network, 34 Processor, 40a, 40b, 40c, 40d, 40e, 40f Partition, 50 Window, 52 Housing, 54 Board, 56 Heat conductive plate

Claims

1. A sensor that detects the user's movements, A processing unit generates third information based on first information regarding the user's body movements obtained based on the output of the sensor and second information regarding the user's biological activity obtained from a separately installed detection device, and transmits the third information to a separately installed server. A detection device equipped with the following features.

2. The detection device according to claim 1, wherein the processing unit generates only a portion of the first information and the second information as the third information.

3. The first information is information in which values ​​corresponding to the movement of at least a part of the user's body are arranged in a time series. The detection device according to claim 1, wherein the third information includes information corresponding to the user's heart rate and / or respiratory rate, but does not include the first information.

4. The first information is information in which values ​​corresponding to the movement of at least a part of the user's body are arranged in a time series. The processing unit generates fourth information corresponding to the user's heart rate and / or respiratory rate based on the first information, The second information is information corresponding to the user's heart rate and / or respiratory rate generated by the separately installed detection device, The detection device according to claim 1, wherein the third information includes at least one of the second information and the fourth information, and does not include the first information.

5. The processing unit, based on instructions from the server, Transmit the first information to the server, and / or The detection device according to claim 4, wherein the detection device obtains from the separately installed detection device a fifth piece of information in which values ​​corresponding to the movement of at least a part of the user's body detected by the separately installed detection device are arranged in a time series, and transmits the fifth piece of information to the server.

6. The detection device according to any one of claims 1 to 5, wherein the processing unit uses a model received from the server to determine the user's status based on the first information and the second information.

7. The detection device according to any one of claims 1 to 5, wherein the sensor transmits a first electromagnetic wave, receives a second electromagnetic wave reflected from the object, and generates the first information based on the first electromagnetic wave and the second electromagnetic wave.

8. Multiple first detection devices are installed, one in each of the multiple sections, to acquire first biological activity information regarding the biological activity of users in the corresponding section. Multiple second detection devices are installed in each of the aforementioned multiple sections, and acquire second biological activity information relating to the biological activity of users in the corresponding section. Equipped with, A detection system in which the plurality of first detection devices are master units, acquires second biological activity information from second detection devices installed in the same section among the plurality of second detection devices which are slave units, generates third biological activity information based on the first biological activity information and the second biological activity information, and transmits the third biological activity information to a server installed outside the plurality of sections.

9. Having the server, the server is Memory and A processing unit that acquires a plurality of third biological activity information outputs from a plurality of first detection devices, each of which is installed in a plurality of compartments and detects information on the biological activity of users in the corresponding compartment, generates a model for determining the state of users in the plurality of compartments based on the plurality of third biological activity information, and transmits the model to the plurality of first detection devices, The detection system according to claim 8, comprising:

10. The aforementioned multiple sections are divided into multiple groups, The detection system according to claim 9, wherein the processing unit generates a plurality of models corresponding to each of the plurality of groups based on the third biological activity information output by the first detection device corresponding to each of the plurality of groups, and transmits the plurality of models to the first detection device of each corresponding group.

11. The sensor comprises a transmitting antenna that transmits a first electromagnetic wave and a receiving antenna that receives a second electromagnetic wave reflected by the user from the first electromagnetic wave, and generates an analog signal relating to the user's movement based on the first electromagnetic wave and the second electromagnetic wave. The processing unit includes an A / D converter that converts the analog signal into a digital signal, and generates the first information relating to the user's body movements based on the digital signal. The detection device according to claim 1, further comprising a regulator that adjusts the amplitude of the analog signal based on the digital signal and outputs the adjusted analog signal to the processing unit.

12. The sensor comprises a transmitting antenna that transmits a first electromagnetic wave and a receiving antenna that receives a second electromagnetic wave reflected by the user of the first electromagnetic wave, The detection device further, A window through which the first electromagnetic wave and the second electromagnetic wave pass, A heater that suppresses condensation on the window by heating the window, The detection device according to claim 1, comprising:

13. The detection device according to claim 12, wherein the heater uses the heat generated by the sensor to suppress condensation on the window.

Images