A low-power wireless detection system and method for nursing

Abstract

The present invention discloses a low-power wireless detection system and method for nursing, including: a power module, an STM core board, a second millimeter wave radar module, a WIFI module, a cloud server, an offline voice module and a client; the power module includes a first millimeter wave radar module, a relay and a battery assembly; the second millimeter wave radar module is connected to the WIFI module and the STM core board; the cloud server is used to receive each data parameter sent by the WIFI module, and judges using a calculation model, and sends the judgment result to the STM core board; the offline voice module is connected to the STM core board; the client is used to configure the wireless detection system, and sends the device number and emergency contact information to the cloud server; wherein, when the first millimeter wave radar module outputs a high level, the relay switch is turned on, and the battery assembly supplies power to all modules of the wireless detection system. The present invention achieves the purpose of reducing nursing power consumption and improving nursing recognition rate.

Description

Technical Field

[0001] The present invention relates to the field of wireless detection technology, and in particular to a low-power wireless detection system and method for nursing. Background Art

[0002] The aging problem in my country is becoming more and more serious, and the number of left-behind elderly people is gradually increasing. Most of the elderly suffer from chronic diseases such as hypertension, hyperlipidemia, diabetes, heart disease, etc. These diseases generally have abnormal characteristics in behavior and body posture. Some elderly people living alone even miss the opportunity for rescue because of falls, breathing difficulties and other emergencies that are not discovered in time.

[0003] Current vital signs detection equipment mainly includes contact and non-contact types. In terms of contact detection, there are mainly ECG monitors, medical sleep monitors, electronic blood pressure monitors, wearable devices, etc. In terms of non-contact detection, most of them are currently through technologies such as depth cameras, infrared thermal imaging systems, ultrasound imaging or WIFI. In the case of daily home monitoring, long-term use of contact equipment can easily cause skin damage and affect daily actions and sleep quality. In terms of non-contact detection, the camera is not satisfactory in conditions of poor light, clothing obstruction, large temperature changes, etc., and it involves issues of privacy protection for the elderly.

[0004] With the development of millimeter wave radar hardware technology, detection equipment has achieved low cost, miniaturization and lightness. Research on detection equipment in vital sign detection, personnel positioning and tracking, and fall detection has become a hot topic.

[0005] At present, most studies use TI's AWR1642-ODS radar, which is based on FMCW technology. It can measure target distance and speed, as well as target azimuth and pitch angle. The board is equipped with the AWR1642 RF chip, which has the control circuit required for the radar RF front end, as well as AD acquisition circuit, ARM processor and DSP processor. The DSP processor and ARM processor can directly process the intermediate frequency signal generated by mixing, and complete radar signal processing algorithms such as distance measurement, speed measurement, and angle measurement. After that, the system uses the three-dimensional detection capability of the radar to obtain the 3D point cloud information of the human body, and then extracts the human body posture features from the 3D point cloud of the human body, and then identifies and determines the changes in human body posture through the calculation model to obtain abnormal behavior information of the elderly.

[0006] In order to generate radar point clouds, millimeter-wave radar monitoring equipment uses MIMO virtual antenna technology and artificial intelligence clustering algorithms to accurately distinguish targets. At the same time, it also uses ARM processors and DSP processors, which increases the power consumption of the equipment. Therefore, millimeter-wave radar monitoring equipment on the market currently basically needs to be powered by an external power supply, which reduces its applicability. In addition, this type of equipment needs to take into account both real-time and reliability, and has high requirements for algorithms and software architecture, so the recognition rate needs to be improved. Summary of the invention

[0007] In order to overcome the shortcomings of the prior art, the present invention provides a low-power wireless detection system and method for nursing, which is used to solve the technical problems of high power consumption and low recognition rate of existing millimeter-wave radar nursing equipment, thereby achieving the purpose of reducing nursing power consumption and improving nursing recognition rate.

[0008] To solve the above problems, the technical solution adopted by the present invention is as follows:

[0009] A low-power wireless detection system for nursing, comprising:

[0010] A power module, comprising a first millimeter-wave radar module, a relay and a battery assembly; the first millimeter-wave radar module is used to output a high level or a low level according to whether there is a moving object within the detection target range;

[0011] An STM core board, a second millimeter-wave radar module and a WIFI module, wherein the second millimeter-wave radar module is connected to the WIFI module and the STM core board, and is used to detect the real-time activities of the cared person, and send various data parameters to the cloud server through the WIFI module;

[0012] The cloud server is used to receive the data parameters sent by the WIFI module, make judgments using the calculation model in the database, and send the judgment results to the STM core board through the WIFI module;

[0013] An offline voice module, connected to the STM core board, for sending an inquiry voice when the STM core board receives a result of abnormal activity judgment of the cared person, and sending a message to the emergency contact through the cloud server when the answer voice is abnormal;

[0014] A client, used to configure the wireless detection system and send the device number and emergency contact information of the wireless detection system to the cloud server;

[0015] Among them, when the first millimeter-wave radar module outputs a high level, the relay switch is opened, and the battery assembly supplies power to all modules of the wireless detection system; when the first millimeter-wave radar module outputs a low level, the relay switch is closed, and the battery assembly only supplies power to the first millimeter-wave radar module.

[0016] As a preferred embodiment of the present invention, the first millimeter wave radar module is a 24G millimeter wave radar sensor; the 24G millimeter wave radar sensor includes: a VCC interface, a GND interface and a VO interface;

[0017] Wherein, the VO interface is a level output interface.

[0018] As a preferred embodiment of the present invention, the battery assembly includes a battery box and a power supply battery, the battery box is used to store the power supply battery, and when the relay switch is turned on, the power supply battery supplies power to all modules of the wireless detection system.

[0019] As a preferred embodiment of the present invention, the second millimeter-wave radar module is a 60GHz millimeter-wave radar AIP chip with a 4T4R antenna, and the AIP chip has embedded computing capabilities.

[0020] As a preferred embodiment of the present invention, the WIFI module is a serial port-wireless module, and the serial port-wireless module has a built-in TCP / IP protocol stack, which can realize the conversion between serial port and wireless, and directly use the configuration instruction to configure the module for use; the serial port-wireless module is used to complete the function of uploading the serial port data of the second millimeter-wave radar module and receiving the cloud server message to send to the STM core board.

[0021] As a preferred embodiment of the present invention, the STM core board is an embedded core board based on an STM32 processor, which is used for the hardware configuration of the second millimeter-wave radar module and the offline voice module, and for conditional branch processing after running the received messages from the cloud server and the offline voice module.

[0022] As a preferred embodiment of the present invention, the offline voice module is connected to the STM core board through a serial port, including a microphone and a speaker, the output end of the microphone is connected to the input end of the offline voice module, and the output end of the offline voice module is connected to the input end of the speaker.

[0023] A low-power wireless detection method for nursing, comprising the following steps:

[0024] After binding the system's device number and emergency contact information through the client, send it to the cloud server;

[0025] After the first millimeter-wave radar module is powered on by the battery assembly, the first millimeter-wave radar module is used to identify whether there is a moving object in the area;

[0026] If not, keep the relay switch closed and continue to supply power to the first millimeter-wave radar module through the battery assembly;

[0027] If it exists, turn on the relay switch, and after all modules are powered by the battery assembly, execute the real-time activity judgment process;

[0028] The real-time activity judgment process includes:

[0029] The speed data, distance data and angle data of the moving object are acquired by using the second millimeter wave radar module, and uploaded to the cloud server via the WIFI module;

[0030] The speed data, distance data and angle data of the moving object are received by the cloud server, and the calculation model in the database is used to make a judgment, and the judgment result is sent to the STM core board through the WIFI module;

[0031] When the STM core board receives the result of abnormal activity judgment of the cared person, a voice question is sent through the offline voice module, and when the answer voice is abnormal, a message is sent to the emergency contact through the cloud server.

[0032] As a preferred embodiment of the present invention, when using the calculation model in the database to make a judgment, it includes:

[0033] Obtain a non-fall data set through a training step;

[0034] According to the non-fall data set, a verification step is performed to construct the calculation model;

[0035] Wherein, the training step comprises:

[0036] The second millimeter-wave radar module obtains speed data, distance data, and angle data of moving objects in the test area multiple times to complete a training, as shown in Formula 1:

[0037]

[0038] In the formula, i is the i-th training, v is the speed data, r is the distance data, and a is the angle data;

[0039] Repeat the training to obtain the training data set, as shown in Formula 2:

[0040] T={U1,U2,…,U n} (2);

[0041] Wherein, n is the number of training times, T is the non-fall data set, and the non-fall data set includes speed data, distance data, and angle data when sitting, squatting, bending over, and walking.

[0042] As a preferred embodiment of the present invention, the verification step includes:

[0043] Sending the data frame acquired by the second millimeter-wave radar module to the cloud server as verification data;

[0044] Assume that the set of verification data is W = {B1, B2, ... B m},in (k=1,2,…,m), if there is a U i , so that U i There is a point If formula 3 is satisfied, then The formula 3 is as follows:

[0045]

[0046] In the formula, ε is the adjustment parameter;

[0047] When W=U i When , it is determined that W∈T, that is, the person has not fallen at this time;

[0048] when When the person falls, it is determined that the person has fallen, and the cloud server uses the WIFI module to send the abnormal activity judgment result of the cared person to the STM core board.

[0049] Compared with the prior art, the present invention has the following beneficial effects:

[0050] (1) The present invention uses a low-frequency millimeter-wave radar to first determine whether there is a moving object in the area. When the determination result is that there is a moving object, the system then supplies power to other corresponding modules to reduce overall power consumption;

[0051] (2) The present invention uses a high-frequency multi-antenna millimeter-wave radar test device to measure several radar data values ​​of a moving object target per unit time, such as speed, distance, azimuth, etc., to generate a corresponding normal behavior data set and store it in a cloud server. The detection data uploaded in real time by the wireless monitoring module device is compared with the data in the data set in the cloud server to obtain a behavior type discrimination result, thereby accelerating the recognition speed;

[0052] (3) The present invention adds an offline voice module to the system. When the cloud server determines that the radar data result is abnormal, the system first issues a question voice to inquire about the health status, and then makes a final judgment result of the abnormal behavior of the user based on the user's answer, and sends a message to the emergency contact through the cloud server to avoid false alarms of fall information;

[0053] (4) The present invention uses a physical energy-saving method. When there is no moving object, all irrelevant modules are powered off, which reduces power consumption to the greatest extent. Therefore, there is no need to use an external power supply. The power supply requirement can be met by using a power supply battery, thereby achieving wiring-free installation;

[0054] (5) The present invention adopts a millimeter-wave radar AIP chip to realize radar signal processing within the chip, directly obtain the v (speed), r (distance), and a (angle) of the moving object, and upload them to the server through the serial port and WIFI, thereby avoiding the high power consumption and high algorithm complexity of signal processing equipment such as DSP.

[0055] The present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 - is a structural diagram of a low-power wireless detection system for care according to an embodiment of the present invention;

[0057] Figure 2 - is an identification flow chart of a low-power wireless detection system for care according to an embodiment of the present invention;

[0058] Figure 3 - is a structural diagram of a power module according to an embodiment of the present invention;

[0059] Figure 4 - is a schematic diagram of a client displaying radar data and corresponding target video according to an embodiment of the present invention;

[0060] Figure 5 - is a step diagram of a low-power wireless detection method for care according to an embodiment of the present invention.

[0061] Explanation of the accompanying figures: 1. Client; 2. Wireless monitoring module equipment; 3. Cloud server; 4. WIFI module; 5. Power module; 6. STM core board; 7. Second millimeter-wave radar module; 8. Offline voice module; 9. Battery assembly; 10. First millimeter-wave radar module; 11. Relay. DETAILED DESCRIPTION

[0062] The low-power wireless detection system for nursing provided by the present invention is as follows: Figure 1 As shown, it includes: a power module 5, an STM core board 6, a second millimeter wave radar module 7, a WIFI module 4, a cloud server 3, an offline voice module 8 and a client 1.

[0063] The power module 5 includes a first millimeter-wave radar module 10, a relay 11 and a battery assembly 9; the first millimeter-wave radar module 10 is used to output a high level or a low level according to whether there is a moving object within the detection target range.

[0064] STM core board 6, second millimeter-wave radar module 7 and WIFI module 4, the second millimeter-wave radar module 7 is connected to the WIFI module 4 and the STM core board 6, and is used to detect the real-time activities of the cared person and send various data parameters to the cloud server 3 through the WIFI module 4.

[0065] The cloud server 3 is used to receive the data parameters sent by the WIFI module 4, and make judgments using the calculation model in the database, and send the judgment results to the STM core board 6 through the WIFI module 4.

[0066] The offline voice module 8 is connected to the STM core board 6 and is used to send an inquiry voice when the STM core board 6 receives the result of abnormal activity judgment of the cared person, and send a message to the emergency contact through the cloud server 3 when the answer voice is abnormal.

[0067] The client 1 is used to configure the wireless detection system and send the device number and emergency contact information of the wireless detection system to the cloud server 3.

[0068] Among them, when the first millimeter-wave radar module 10 outputs a high level, the relay 11 switch is opened, and the battery assembly 9 supplies power to all modules of the wireless detection system; when the first millimeter-wave radar module 10 outputs a low level, the relay 11 switch is closed, and the battery assembly 9 only supplies power to the first millimeter-wave radar module 10.

[0069] Specifically, the power module 5, the STM core board 6, the second millimeter wave radar module 7, the WIFI module 4, and the offline voice module 8 together form a wireless monitoring module device 2, which interacts with the client 1 and the cloud server 3 respectively through the WIFI module 4. Each wireless monitoring module device 2 has a device number, which is used to bind with the emergency contact information and is sent to the cloud server 3 after binding.

[0070] Specifically, the low-power wireless detection system provided by the present invention includes a power module 5, an STM core board 6, a second millimeter-wave radar module 7, a WIFI module 4, an offline voice module 8, a cloud server 3 and a client 1.

[0071] The power supply module 5 includes a first millimeter-wave radar module 10, a relay 11 and a battery assembly 9. The first millimeter-wave radar module 10 is a low-power millimeter-wave radar, which is used to detect whether there is a moving object in the target range. When there is a moving object, the first millimeter-wave radar module 10 outputs a high level. When there is no moving object, the first millimeter-wave radar module 10 outputs a low level. When the first millimeter-wave radar module 10 outputs a high level, the relay 11 switch is turned on, and the battery assembly 9 supplies power to other corresponding modules.

[0072] The second millimeter-wave radar module 7 is connected to the WIFI module 4 and the STM core board 6, and is used to detect the real-time activity status of the person being cared for, send various data parameters to the cloud server 3 through the WIFI module 4, compare them with the calculation model in the database of the cloud server 3, and send the comparison results to the STM core board 6 through the WIFI module 4.

[0073] The calculation model is to obtain several radar data values ​​of the moving target per unit time through the second millimeter wave radar module 7 during the model generation stage, generate corresponding feature action data sets and store them in the cloud server 3 database for test data comparison.

[0074] The offline voice module 8 is connected to the STM core board 6 and is equipped with a speaker and a microphone. When the STM core board 6 receives the result of the abnormal activity judgment of the person being cared for, the offline voice module 8 first sends a voice to inquire about the situation through the speaker, and then the microphone receives the answering voice of the person being cared for. If the answering voice is abnormal, the system sends a message to the mobile phone of the emergency contact through the cloud server 3;

[0075] The client 1 is used to configure the wireless detection system and send the device number and emergency contact information of the wireless monitoring module device 2 in the wireless detection system to the cloud server 3.

[0076] Specifically, the identification process of the wireless detection system provided by the present invention is as follows: Figure 2 shown.

[0077] Further, the first millimeter wave radar module 10 is a 24G millimeter wave radar sensor; the 24G millimeter wave radar sensor includes: a VCC interface, a GND interface and a VO interface;

[0078] Among them, the VO interface is a level output interface.

[0079] Preferably, the first millimeter-wave radar module 10 is a miniaturized, low-cost 24G millimeter-wave radar sensor with three interfaces: a VCC interface, a GND interface and a VO interface, wherein the VO interface is a level output interface, and sensitivity debugging can be achieved by changing the amplification factor of the comparator. After detecting a moving object, the first millimeter-wave radar module 10 outputs a high level, and the maintenance time is adjusted by changing the resistance value.

[0080] Specifically, the 24G millimeter wave radar sensor can adopt the 24G millimeter wave radar sensor of Shenzhen Hailingke, model HLK-LD116-24G.

[0081] Furthermore, the battery assembly 9 includes a battery box and a power supply battery. The battery box is used to store the power supply battery. When the relay 11 switch is turned on, the power supply battery supplies power to all modules of the wireless detection system.

[0082] Specifically, when the level trigger relay 11 receives a high level signal from the first millimeter wave radar module 10, the power supply battery supplies power to all modules of the entire wireless detection system, such as Figure 3 shown.

[0083] Furthermore, the second millimeter-wave radar module 7 is a 60GHz millimeter-wave radar AIP chip with a 4T4R antenna, and the AIP chip has embedded computing capabilities.

[0084] Specifically, the 60GHz millimeter wave radar AIP chip can use the millimeter wave radar module development kit based on the Calterah 4T4R antenna 60GHz millimeter wave radar AIP (on-chip package) chip design, which includes a radar board and a debug board. The debug board is connected to the radar board through a double-row pin. The debug board converts the UART serial port of the radar board to a USB2.0 interface, which can be connected to a computer for debugging.

[0085] Specifically, the above embedded computing capability is: first, the DC component of the intermediate frequency signal of the ADC data obtained by the radar is eliminated, and then 2D-FFT and constant false alarm detection are used to obtain the target distance and speed information, and then 3D-FFT is used to obtain the target azimuth information. Radar signal processing can be completed in the chip and directly output through the serial port without the need for an external processing chip.

[0086] Furthermore, the WIFI module 4 is a serial port-wireless (UART-WIFI) module. The serial port-wireless module has a built-in TCP / IP protocol stack, which can realize the conversion between serial port and wireless, and directly configure the module using configuration instructions; the serial port-wireless module is used to complete the function of uploading serial port data of the second millimeter-wave radar module 7 and receiving messages from the cloud server 3 and sending them to the STM core board 6.

[0087] Specifically, the configuration instruction is an AT instruction, and the UART-WIFI module 4 can adopt a UART-WiFi (serial port-wireless) module launched by ALIENTEK, model ATK-ESP8266.

[0088] Furthermore, the STM core board 6 is an embedded core board based on the STM32 processor, which is used for the hardware configuration of the second millimeter-wave radar module 7 and the offline voice module 8, and for conditional branch processing after receiving messages from the cloud server 3 and the offline voice module 8.

[0089] Furthermore, the offline voice module 8 is connected to the STM core board 6 via a serial port, and includes a microphone and a speaker, wherein the output end of the microphone is connected to the input end of the offline voice module 8, and the output end of the offline voice module 8 is connected to the input end of the speaker.

[0090] Specifically, the offline voice module 8 can realize the voice and text conversion function through the above connection method. The offline voice module 8 can adopt the offline voice module 8 of Shenzhen Hailingke, model HLK-V25.

[0091] Furthermore, the client 1 uses the front-end and back-end separation technology, the back-end service is located in the cloud server 3, and the front-end is located in the app or the web page, which is used to enter user information and wireless detection system hardware information, and reserve the user radar data analysis display content window, such as Figure 4 shown.

[0092] The low-power wireless detection method for nursing provided by the present invention is as follows: Figure 5 As shown, the following steps are included:

[0093] Step S1: Bind the system device number and emergency contact information through client 1 and send them to cloud server 3;

[0094] Step S2: After the first millimeter-wave radar module 10 is powered on by the battery assembly 9, the first millimeter-wave radar module 10 identifies whether there is a moving object in the area;

[0095] Step S3: If not, keep the relay 11 switch closed, and continue to supply power to the first millimeter-wave radar module 10 through the battery assembly 9;

[0096] Step S4: If yes, turn on the relay 11 switch, power all modules through the battery assembly 9, and then execute the real-time activity judgment process;

[0097] The real-time activity judgment process includes:

[0098] The speed data, distance data and angle data of the moving object are acquired by the second millimeter wave radar module 7, and uploaded to the cloud server 3 via the WIFI module 4;

[0099] The speed data, distance data and angle data of the moving object are received through the cloud server 3, and the calculation model in the database is used to make a judgment, and the judgment result is sent to the STM core board 6 through the WIFI module 4;

[0100] When the STM core board 6 receives the result of the abnormal activity judgment of the cared person, it sends a voice question through the offline voice module 8, and sends a message to the emergency contact through the cloud server 3 when the answer voice is abnormal.

[0101] In the above step S4, when the calculation model in the database is used for judgment, it includes:

[0102] Obtain a non-fall data set through a training step;

[0103] Based on the non-fall data set, a validation step is performed to build a computational model;

[0104] The training steps include:

[0105] The second millimeter wave radar module 7 acquires the speed data, distance data and angle data of the moving object in the test area for multiple times to complete a training, as shown in Formula 1:

[0106]

[0107] In the formula, i is the i-th training, v is the speed data, r is the distance data, and a is the angle data;

[0108] Repeat the training to obtain the training data set, as shown in Formula 2:

[0109] T={U1,U2,…,U n} (2);

[0110] Where n is the number of training times, T is the non-fall data set, and the non-fall data set includes speed data, distance data, and angle data when sitting, squatting, bending, and walking.

[0111] Furthermore, the verification step includes:

[0112] The data frame acquired by the second millimeter wave radar module 7 is sent to the cloud server 3 as verification data;

[0113] Assume that the set of verification data is W = {B1, B2, ... B m},in (k=1,2,…,m), if there is a U in the non-fall data set T i , so that U i There is a point If formula 3 is satisfied, then Formula 3 is as follows:

[0114]

[0115] In the formula, ε is the adjustment parameter;

[0116] When W=U i When , it is determined that W∈T, that is, the person has not fallen at this time;

[0117] when When the person falls, it is determined that the person has fallen, and the abnormal activity judgment result of the person being cared for is sent to the STM core board 6 through the cloud server 3 using the WIFI module 4.

[0118] Specifically, the computational model is divided into two steps: training and verification.

[0119] Training phase: The three data of v (speed), r (distance), and a (angle) of a moving object in a test area generated by the second millimeter-wave radar module 7 are processed as follows: Let T be the non-fall data set, T includes data of sitting, squatting, bending and walking, etc., which are stored in the cloud server 3. Let the i-th training data set during the scanning period be in Then T can be expressed as T={U1,U2,…,U n}(n is the number of training times).

[0120] Verification stage: The wireless monitoring module device 2 sends the data frame generated by the second millimeter wave radar module 7 to the cloud server 3 as verification data. At this time, the set of verification data is W = {B1, B2, ... B m},in If there is a U in the set T i , so that U i There is a point Can satisfy the following formula:

[0121]

[0122] It is believed that Where ε is the adjustment parameter. i When , it is determined that W∈T, that is, the person has not fallen. When the person falls, the cloud server 3 sends an activity abnormality message to the STM core board 6.

[0123] The above-mentioned embodiments are only preferred embodiments of the present invention and cannot be used to limit the scope of protection of the present invention. Any non-substantial changes and substitutions made by technicians in this field on the basis of the present invention shall fall within the scope of protection required by the present invention.

Claims

1. A low-power wireless detection system for nursing, characterized in that: include: A power module, comprising a first millimeter-wave radar module, a relay and a battery assembly; the first millimeter-wave radar module is used to output a high level or a low level according to whether there is a moving object within the detection target range; An STM core board, a second millimeter-wave radar module and a WIFI module, wherein the second millimeter-wave radar module is connected to the WIFI module and the STM core board, and is used to detect the real-time activities of the cared person, and send various data parameters to the cloud server through the WIFI module; The cloud server is used to receive the data parameters sent by the WIFI module, make judgments using the calculation model in the database, and send the judgment results to the STM core board through the WIFI module; An offline voice module, connected to the STM core board, for sending an inquiry voice when the STM core board receives a result of abnormal activity judgment of the cared person, and sending a message to the emergency contact through the cloud server when the answer voice is abnormal; A client, used to configure the wireless detection system and send the device number and emergency contact information of the wireless detection system to the cloud server; Among them, when the first millimeter-wave radar module outputs a high level, the relay switch is opened, and the battery assembly supplies power to all modules of the wireless detection system; when the first millimeter-wave radar module outputs a low level, the relay switch is closed, and the battery assembly only supplies power to the first millimeter-wave radar module.

2. The low-power wireless detection system for nursing according to claim 1, characterized in that: The first millimeter wave radar module is a 24G millimeter wave radar sensor; The 24G millimeter wave radar sensor includes: a VCC interface, a GND interface and a VO interface; Wherein, the VO interface is a level output interface.

3. The low-power wireless detection system for nursing according to claim 1, characterized in that: The battery assembly includes a battery box and a power supply battery. The battery box is used to store the power supply battery. When the relay switch is turned on, the power supply battery supplies power to all modules of the wireless detection system.

4. The low-power wireless detection system for nursing according to claim 1, characterized in that: The second millimeter-wave radar module is a 60GHz millimeter-wave radar AIP chip with a 4T4R antenna, and the AIP chip has embedded computing capabilities.

5. The low-power wireless detection system for nursing according to claim 1, characterized in that: The WIFI module is a serial port-wireless module, which has a built-in TCP / IP protocol stack, can realize the conversion between serial port and wireless, and directly configure the module using configuration instructions; the serial port-wireless module is used to complete the function of uploading serial port data of the second millimeter-wave radar module and receiving the cloud server message and sending it to the STM core board.

6. The low-power wireless detection system for nursing according to claim 1, characterized in that: The STM core board is an embedded core board based on the STM32 processor, which is used for the hardware configuration of the second millimeter-wave radar module and the offline voice module, and for conditional branch processing after running the received messages from the cloud server and the offline voice module.

7. The low-power wireless detection system for nursing according to claim 1, characterized in that: The offline voice module is connected to the STM core board via a serial port, and comprises a microphone and a speaker, wherein the output end of the microphone is connected to the input end of the offline voice module, and the output end of the offline voice module is connected to the input end of the speaker.

8. A low-power wireless detection method for nursing, characterized in that: The following steps are involved: After binding the system's device number and emergency contact information through the client, send it to the cloud server; After the first millimeter-wave radar module is powered on by the battery assembly, the first millimeter-wave radar module is used to identify whether there is a moving object in the area; If not, keep the relay switch closed and continue to supply power to the first millimeter-wave radar module through the battery assembly; If it exists, turn on the relay switch, and after all modules are powered by the battery assembly, execute the real-time activity judgment process; The real-time activity judgment process includes: The speed data, distance data and angle data of the moving object are acquired by using the second millimeter wave radar module, and uploaded to the cloud server via the WIFI module; The speed data, distance data and angle data of the moving object are received by the cloud server, and the calculation model in the database is used to make a judgment, and the judgment result is sent to the STM core board through the WIFI module; When the STM core board receives the result of abnormal activity judgment of the cared person, it sends out an inquiry voice through the offline voice module, and sends a message to the emergency contact through the cloud server when the answer voice is abnormal.

9. The low-power wireless detection method for nursing according to claim 8, characterized in that: When using the calculation model in the database to make judgments, it includes: Obtain a non-fall data set through a training step; According to the non-fall data set, a verification step is performed to construct the calculation model; Wherein, the training step comprises: The second millimeter-wave radar module obtains speed data, distance data, and angle data of moving objects in the test area multiple times to complete a training, as shown in Formula 1: （1）； In the formula, For the training, , k=1,2,…,m, is the speed data, is the distance data, is the angle data; Repeat the training to obtain the training data set, as shown in Formula 2: （2）； In the formula, is the number of training times, The non-fall data set includes speed data, distance data and angle data when sitting, squatting, bending over and walking.

10. The low-power wireless detection method for nursing according to claim 9, characterized in that: The verification step includes: Sending the data frame acquired by the second millimeter-wave radar module to the cloud server as verification data; Suppose the set of verification data is ,in , k=1,2,…,m, if the non-fall data set There is a , so that There is a point If formula 3 is satisfied, then , the formula 3 is as follows: （3）； In the formula, To adjust the parameters; when When , that is, the person does not fall at this time; when When the person falls, it is determined that the person has fallen, and the cloud server uses the WIFI module to send the abnormal activity judgment result of the cared person to the STM core board.

Images