Personnel presence detection device, detection system and detection method

Abstract

The invention discloses a personnel presence detection device, system and method, and the device comprises a signal transmitting module which transmits Wi-Fi signals, a signal receiving module which receives reflected Wi-Fi signals, a processor which can obtain the data of channel state information (CSI) from a plurality of reflected Wi-Fi signals, and a communication module which transmits the Wi-Fi signals to the processor. And an operation module capable of identifying a radio frequency main path of a CSI data set with the largest number of packets from the CSI data. The operation module processes the CSI data set to determine whether the highest frequency value falls within the respiratory frequency range and generate a first detection result, to determine whether the signal variation number is greater than a preset value and generate a second detection result, to determine whether the packet number is less than the preset value and generate a third detection result, and when one of the three detection results is true, the presence of a person is judged.

Description

Personnel presence detection device, detection system and detection method Technical Field

[0001] The present invention relates to a detection device, a detection system and a detection method, and in particular to a detection device, a detection system and a detection method for detecting whether a person is present. Background Art

[0002] Under the scorching sun, a car cabin without air conditioning and with doors and windows closed is like an oven. If a child is forgotten in the car, it may cause heat damage or even heat exhaustion, which may be fatal.

[0003] Currently, the European New Car Assessment Program (Euro NCAP) has included child presence detection as one of its safety testing standards for new cars, and the US National Highway Traffic Safety Administration (NHTSA) has also included rear seat child reminder system functions in its New Car Assessment Program (NCAP) proposal.

[0004] Currently, many in-vehicle occupant detection systems on the market rely solely on breathing patterns to detect a person's presence, but fail to detect other characteristics beyond breathing. Furthermore, breathing patterns may be less noticeable when a person is inside a vehicle due to factors such as movement, crying, or adjusting their sitting position. Consequently, traditional in-vehicle occupant detection systems are prone to misjudgment, rendering them ineffective.

[0005] Summary of the Invention

[0006] The main purpose of the present invention is to provide a person presence detection device, detection system and detection method, which can detect the presence of people in a closed space, thereby preventing children from being left alone in the car and preventing burglars from breaking into the car.

[0007] In one embodiment, the person presence detection device of the present invention includes:

[0008] A signal sending module continuously sends multiple Wi-Fi electromagnetic wave signals to the outside world;

[0009] A signal receiving module continuously receives multiple Wi-Fi reflected electromagnetic wave signals;

[0010] a processor that controls the signal transmitting module and the signal receiving module and obtains channel state information (CSI) data based on the multiple Wi-Fi reflected electromagnetic wave signals; and

[0011] a computing module, connected to the processor, for identifying a primary radio frequency (RF) path from the CSI data, wherein the primary RF path includes a CSI data set having the largest number of packets in the CSI data;

[0012] The computing module is configured to determine, through fast Fourier transform, whether a highest frequency value of the CSI data set falls within a respiratory frequency range and generate a first detection result; determine whether a signal variance of the CSI data set is greater than a preset variation range and generate a second detection result; determine whether the number of packets in the CSI data set is less than a preset packet number threshold and generate a third detection result; and determine the presence of a person when one of the first detection result, the second detection result, and the third detection result is true.

[0013] As described above, the signal receiving module is configured to receive the multiple Wi-Fi reflected electromagnetic wave signals through multiple RF paths, and the computing module is configured to group the multiple Wi-Fi reflected electromagnetic wave signals based on the intensity distribution of each Wi-Fi reflected electromagnetic wave signal in multiple subcarriers, and to use the RF path with the largest number of packets as the primary RF path.

[0014] As described above, the calculation module is configured to group the multiple Wi-Fi reflected electromagnetic wave signals based on the DBSCAN algorithm, and use the RF path with the largest number of packets as the main RF path.

[0015] As mentioned above, the respiratory rate range is 8 to 22 bpm.

[0016] As described above, the computing module is located in the cloud and is connected to the processor via a wired network or a wireless network.

[0017] As mentioned above, it also includes a transmission module, which is connected to the operation module and is configured to send the detection result of whether there is a person present to the outside.

[0018] In one embodiment, the person presence detection system of the present invention includes:

[0019] a first Wi-Fi device that continuously transmits multiple Wi-Fi electromagnetic wave signals;

[0020] a second Wi-Fi device that continuously receives multiple Wi-Fi reflected electromagnetic wave signals and obtains CSI data based on the multiple Wi-Fi reflected electromagnetic wave signals; and

[0021] a computing module connected to the second Wi-Fi device, receiving the CSI data and identifying a primary RF path from the CSI data, wherein the primary RF path includes a CSI data set with the largest number of packets in the CSI data;

[0022] The computing module is configured to determine, through fast Fourier transform, whether a highest frequency value of the CSI data set falls within a respiratory frequency range and generate a first detection result; determine whether a signal variance of the CSI data set is greater than a preset variation range and generate a second detection result; determine whether the number of packets in the CSI data set is less than a preset packet number threshold and generate a third detection result; and determine the presence of a person when one of the first detection result, the second detection result, and the third detection result is true.

[0023] As described above, the second Wi-Fi device is configured to receive the multiple Wi-Fi reflected electromagnetic wave signals through multiple RF paths, and the computing module is configured to group the multiple Wi-Fi reflected electromagnetic wave signals based on the intensity distribution of each Wi-Fi reflected electromagnetic wave signal in multiple small frequency bands, and use the RF path with the largest number of packets as the main RF path.

[0024] As described above, the calculation module is configured to group the multiple Wi-Fi reflected electromagnetic wave signals based on the DBSCAN algorithm, and use the RF path with the largest number of packets as the main RF path.

[0025] As described above, the computing module is set in the cloud and connected to the second Wi-Fi device via a wired network or a wireless network, and the person presence detection system also includes a remote device, which is configured to connect to the computing module via a wired network or a wireless network to query the detection result of the person presence.

[0026] In one embodiment, the method for detecting presence of a person of the present invention includes:

[0027] a) Continuously send multiple Wi-Fi electromagnetic wave signals to the outside world;

[0028] b) a Wi-Fi device continuously receives multiple Wi-Fi reflected electromagnetic wave signals;

[0029] c) obtaining, by the Wi-Fi device, CSI data based on the multiple Wi-Fi reflected electromagnetic wave signals;

[0030] d) identifying, by a computing module, a primary RF path from the CSI data, wherein the primary RF path includes a CSI data set having the largest number of packets in the CSI data;

[0031] e) determining, by fast Fourier transform, whether a highest frequency value of the CSI data set falls within a respiratory frequency range and generating a first detection result;

[0032] f) determining whether a signal variance of the CSI data set is greater than a predetermined variance range and generating a second detection result;

[0033] g) determining whether the number of packets in the CSI data set is less than a preset packet number threshold and generating a third detection result; and

[0034] h) determining that a person is present when one of the first detection result, the second detection result, and the third detection result is true.

[0035] As described above, in step a), the Wi-Fi device continuously sends the multiple Wi-Fi electromagnetic wave signals, and in step b), the same Wi-Fi device continuously receives the multiple Wi-Fi reflected electromagnetic wave signals.

[0036] As described above, step b) includes receiving the multiple Wi-Fi reflected electromagnetic wave signals respectively through multiple RF paths, and step d) includes grouping the multiple Wi-Fi reflected electromagnetic wave signals based on the intensity distribution of each Wi-Fi reflected electromagnetic wave signal in multiple small frequency bands, and using the RF path with the largest number of packets as the main RF path.

[0037] As described above, step d) includes grouping the multiple Wi-Fi reflected electromagnetic wave signals based on the DBSCAN algorithm, and using the RF path with the largest number of packets as the main RF path.

[0038] As mentioned above, this also includes:

[0039] i) The computing module receives a connection and query from a remote device to provide a detection result of whether a person is present.

[0040] Compared with related technologies, the detection device, detection system and detection method of the present invention simultaneously detect a person's breathing characteristics, small movement characteristics and large movement characteristics based on Wi-Fi signals, thereby more accurately determining whether there are people in the enclosed space. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] FIG1 is a specific embodiment of a detection schematic diagram of the present invention;

[0042] FIG2 is a specific embodiment of a detection schematic diagram of the present invention;

[0043] FIG3 is a block diagram of a person presence detection device according to the present invention;

[0044] FIG4 is a flow chart of a method for detecting presence of a person according to the present invention;

[0045] FIG5 is a schematic diagram of a multipath Wi-Fi signal according to the present invention;

[0046] FIG6 is a schematic diagram of CSI data of the present invention;

[0047] 7A-7D are schematic diagrams of call feature conversion according to the present invention;

[0048] FIG8 is a schematic diagram of a small action feature of the present invention;

[0049] FIG9 is a schematic diagram of a large motion characteristic of the present invention;

[0050] FIG10 is a schematic diagram of a person presence detection system according to the present invention.

[0051] In the figures, numerals indicate: 1…person presence detection device; 11…signal sending module; 12…signal receiving module; 13…processor; 14…computing module; 15…transmission module; 20…enclosed space; 2…person; 3…car; 4…house; 5…first Wi-Fi device; 6…second Wi-Fi device; 7…computing module; 8…remote device; P1-P6…packets; RF1…first RF path; RF2…second RF path; RF3…third RF path; T1…first time interval; T2…second time interval; S40-S50…detection steps. DETAILED DESCRIPTION

[0052] A preferred embodiment of the present invention is described in detail below with reference to the accompanying drawings.

[0053] To accurately detect whether a person has been left behind in a vehicle, the present invention discloses a person presence detection device (hereinafter referred to as the detection device). The detection device of the present invention processes and detects signal packets in a confined space, and determines the presence of a person when features associated with a person are detected in these signal packets.

[0054] First, please refer to Figure 1, which is a schematic diagram of a detection embodiment of the present invention. In the embodiment of Figure 1, the enclosed space is, for example, the cabin of a car 3, and the detection device 1 is installed in the cabin to detect a person 2 in the cabin.

[0055] In one embodiment, detection device 1 can be manually activated by the user or connected to the vehicle's onboard system. After vehicle 3 is turned off, the driver exits, and the doors are locked, detection device 1 is activated manually by the user or automatically by the onboard system. This allows detection device 1 to detect the presence of children in the vehicle, preventing careless drivers from leaving children unattended. Because detection device 1 detects the presence of person 2 while vehicle 3 is off, vibrating vehicle components prevent misidentification.

[0056] It's worth noting that, according to FBI statistics, over 3,000 burglaries occur daily in the United States, and similar situations are expected in other countries. The detection device 1 of the present invention is primarily used to detect the presence of a person 2 inside a car 3. Conversely, since a house is also an enclosed space, the detection device 1 of the present invention can also be installed inside a house to detect burglars, thereby improving home security.

[0057] 2 is another embodiment of the detection diagram of the present invention. In the embodiment of FIG2 , the enclosed space is, for example, a house 4 , and the detection device 1 is disposed in one or more specific rooms in the house 4 to detect whether a thief is present in the house 4 .

[0058] In one embodiment, the user can activate detection device 1 while away from home and determined no one is home. In another embodiment, detection device 1 can be connected to a home security system. In this embodiment, detection device 1, after being manually activated by the user or automatically activated by the security system, continuously detects signal packets within house 4 and issues an alert to the user if a signal packet displays characteristics associated with a person, thereby detecting burglars.

[0059] Continuing with Figure 3, a block diagram of the present invention's presence detection device is shown. As shown in Figure 3, the present invention's detection device 1 includes at least a signal transmission module 11, a signal reception module 12, a processor 13, and a computing module 14. Processor 13 is electrically connected to signal transmission module 11 and signal reception module 12 to control signal transmission module 11 to transmit signal packets and to control signal reception module 12 to receive reflected signal packets.

[0060] In one embodiment, the computing module 14 is disposed within the detection device 1 and is electrically connected to the processor 13. In another embodiment, the computing module 14 is disposed outside the detection device 1 (e.g., on an external hard drive, an external computer, or a cloud) and is connected to the processor 13 of the detection device 1 via a wired or wireless connection.

[0061] However, the above are only some specific implementation examples of the present invention, and are not limited thereto.

[0062] In one embodiment, the detection device 1 may be a Wi-Fi device, and the signal sending module 11 and the signal receiving module 12 are the transmitting antenna and the receiving antenna of the Wi-Fi device, respectively. The detection device 1 continuously sends multiple Wi-Fi electromagnetic wave signals to the outside through the signal sending module 11. After the multiple Wi-Fi electromagnetic wave signals are sent to the outside of the detection device 1, they will diverge in the closed space 20 and reflect after touching the walls, objects or people 2 in the closed space 20. The detection device 1 continuously receives multiple Wi-Fi reflected electromagnetic wave signals from the closed space 20 through the signal receiving module 12. In the present invention, the detection device 1 processes the received multiple Wi-Fi reflected electromagnetic wave signals to determine whether there is a person 2 present in the closed space 20.

[0063] Please refer to Figures 3 and 4 simultaneously, wherein Figure 4 is a flow chart of the method for detecting presence of a person of the present invention. Figure 4 discloses the specific steps of the method for detecting presence of a person of the present invention (hereinafter referred to as the detection method), which can be applied to the detection device 1 shown in Figure 3.

[0064] Specifically, the processor 13 of the detection device 1 can be a central processing unit (CPU), a micro control unit (MCU), a programmable logic controller (PLC), a system on chip (SoC) or a field programmable gate array (FPGA), etc., but is not limited to this. Computer executable program code is recorded in the processor 13. When the detection device 1 is powered on and started, and the processor 13 executes the computer executable program code, the signal sending module 11, the signal receiving module 12 and the computing module 14 can be controlled to jointly perform the steps shown in Figure 4 to implement the detection method of the present invention.

[0065] As shown in FIG. 4 , after the detection device 1 is activated (i.e., the user determines that no one is in the enclosed space 20 ), the detection device 1 controls the signal sending module 11 to continuously transmit multiple Wi-Fi electromagnetic wave signals to the outside (step S40 ), and controls the signal receiving module 12 to continuously receive multiple Wi-Fi reflected electromagnetic wave signals from the outside (step S41 ).

[0066] In another embodiment, steps S40 and S41 are performed by the signal transmitting module 11 and the signal receiving module 12 of the same detection device (e.g., the detection device 1 shown in FIG3 ) transmitting multiple Wi-Fi electromagnetic wave signals and receiving multiple Wi-Fi reflected electromagnetic wave signals, respectively. Thus, the detection method of the present invention can be implemented in the enclosed space 20 using a single detection device 1. By using a single, self-contained detection device 1, the cost of implementing the detection method can be reduced, as well as the space required to house the detection device 1.

[0067] In another embodiment, step S40 involves the signal transmission module of the first detection device transmitting multiple Wi-Fi electromagnetic wave signals, while step S41 involves the signal reception module of the second detection device receiving multiple Wi-Fi reflected electromagnetic wave signals. Thus, the detection method of the present invention can be implemented in enclosed space 20 using multiple detection devices (e.g., a Wi-Fi Mesh composed of multiple Wi-Fi devices). By using multiple detection devices to transmit and receive signals, the detection range can be expanded and the detection accuracy can be improved.

[0068] In the present invention, the signal transmission module 11 continuously transmits Wi-Fi electromagnetic wave signals at a specified frequency, and the signal reception module 12 continuously receives Wi-Fi reflected electromagnetic wave signals at a specified frequency. Therefore, the multiple Wi-Fi electromagnetic wave signals and the multiple Wi-Fi reflected electromagnetic wave signals are time-sequential, rather than being transmitted or received simultaneously. Because these signal packets are time-sequential, after step S41, the processor 13 of the detection device 1 can obtain Channel State Information (CSI) data based on the multiple Wi-Fi reflected electromagnetic wave signals (step S42).

[0069] In the field of wireless communication technology, CSI data (as shown in FIG6 ) indicates the known channel characteristics of a communication link and describes how a signal packet is transmitted from the transmitter through a channel (or a path) to the receiver. Since the attenuation of the energy of the signal packet with distance can be obtained from the CSI data, in the present invention, the detection device 1 first converts the multiple Wi-Fi reflected electromagnetic wave signals into CSI data for observation. After obtaining the CSI data, the processor 13 provides the CSI data to the operation module 14, and the operation module 14 identifies the main radio frequency (RF) path from the CSI data (step S43). In the present invention, the main RF path refers to the path used to transmit a CSI data set with the largest number of packets in the CSI data.

[0070] Please also refer to Figure 5, which is a schematic diagram of a multipath Wi-Fi signal according to the present invention. In the present invention, the signal transmission module 11 transmits multiple Wi-Fi electromagnetic wave signals using a scattering transmission method. These signal packets are transmitted and reflected along the same or different paths within the enclosed space 20. Therefore, the signal receiving module 12 can receive multiple reflected Wi-Fi electromagnetic wave signals via multiple RF paths, with the RF path through which the most reflected Wi-Fi electromagnetic wave signals pass being the primary RF path.

[0071] Specifically, after being scattered, these signal packets bounce back after colliding with walls, objects, or person 2 in enclosed space 20. Since signal transmission module 11 continuously transmits signal packets, signal reception module 12 can continuously receive reflected packets via the same or different RF paths. The present invention uses a clustering method to identify the primary RF path containing the most signal packets among multiple RF paths. Based on the multiple signal packets (CSI data) contained within this primary RF path, detection is performed to determine whether they possess characteristics unique to person 2, such as breathing, small movement, or large movement.

[0072] 5 , the first signal packet P1 is transmitted along the first RF path (RF1) in the enclosed space 20 and is received by the signal receiving module 12, the fourth signal packet P4 is transmitted along the second RF path (RF2) in the enclosed space 20 and is received by the signal receiving module 12, and the second, third, fifth, and sixth signal packets P2, P3, P5, and P6 are transmitted along the third RF path (RF3) in the enclosed space 20 and are received by the signal receiving module 12. In other words, the second, third, fifth, and sixth signal packets P2, P3, P5, and P6 have the same transmission path in the enclosed space 20, but are different from the transmission paths of the first signal packet P1 and the fourth signal packet P4.

[0073] By utilizing multipath scattering transmission, the detection device 1 of the present invention ensures that every corner of the enclosed space 20 is detected, thereby reducing the probability of false positives. Furthermore, because multiple signal packets are not transmitted via a single path, even when used within the enclosed space 20, there is no risk of packet loss or interference affecting detection results.

[0074] Please also refer to Figure 6, which illustrates the CSI data of the present invention. As shown in Figure 6, in some cases (e.g., during the first time interval T1, immediately after Person 2 takes his seat), significant multipath transmission leads to large fluctuations, making Person 2's features difficult to distinguish. Directly using raw CSI data for detection (e.g., detecting Person 2's breathing characteristics) can result in misjudgment.

[0075] To overcome these issues, the detection method of the present invention first groups the multiple received Wi-Fi electromagnetic wave signals, identifying the primary RF path containing the most signal packets. It then processes the multiple signal packets within the primary RF path. This method filters out unnecessary noise and makes it easier to detect the characteristics of Person 2.

[0076] It's worth noting that in some cases (e.g., during the third time interval T3 when person 2 is sitting quietly), the multipath effect is minimal, the fluctuations are small and stable, and the characteristics of person 2 are clear. In this case, the computing module 14 can directly use the raw CSI data for detection (e.g., detecting the breathing characteristics of person 2) without performing grouping.

[0077] As previously mentioned, the RF paths traveled by multiple Wi-Fi reflected electromagnetic wave signals within the enclosed space 20 may be the same or different. To effectively extract useful features from the Wi-Fi reflected electromagnetic wave signals, the detection method of the present invention first identifies the primary RF path encompassing the most signal packets and then performs detection based on the CSI data set within the primary RF path (i.e., the set of all Wi-Fi reflected electromagnetic wave signals transmitted along the primary RF path).

[0078] In the present invention, computing module 14 groups multiple Wi-Fi reflected electromagnetic wave signals based on their frequency variations. Multiple signal packets with the same or similar frequency variations are grouped together. Since these signal packets have the same or similar frequency variations, they share the same transmission path.

[0079] In one embodiment, the computing module 14 converts each Wi-Fi reflected electromagnetic wave signal into multiple subcarriers based on the CSI data to observe the intensity distribution of these Wi-Fi reflected electromagnetic wave signals. The computing module 14 then groups the multiple Wi-Fi reflected electromagnetic wave signals based on the intensity distribution. Specifically, the computing module 14 groups multiple signal packets with the same or similar intensity distribution across the multiple subcarriers into the same group (i.e., belonging to the same RF path), and deems the RF path with the largest number of packets as the primary RF path.

[0080] In another embodiment, the calculation module 14 directly groups the multiple Wi-Fi reflected electromagnetic wave signals based on the DBSCAN algorithm, and uses the RF path with the largest number of packets as the main RF path.

[0081] However, the above are only some specific implementation examples of the present invention, and are not limited thereto.

[0082] Returning to Figures 3 and 4 , after step S43 , the computing module 14 obtains the CSI data set for the primary RF path (i.e., CSI data for all signal packets transmitted via the primary RF path) using the clustering method. It then processes the CSI data set using a Fast Fourier Transform (FFT) to determine whether the highest frequency value in the CSI data set falls within the respiratory frequency range, and generates a first detection result (step S44 ).

[0083] Please refer to FIG. 7A to FIG. 7D , which are schematic diagrams of call feature conversion according to the present invention.

[0084] As shown in Figure 7A , in some cases, CSI data can be difficult to detect due to significant multipath transmission and large fluctuations, making it difficult to clearly identify the breathing characteristics of person 2. Therefore, as shown in Figure 7B , computing module 14 converts multiple Wi-Fi reflected electromagnetic wave signals into multiple small frequency bands based on the CSI data to observe the signal intensity distribution. These Wi-Fi reflected electromagnetic wave signals are then grouped based on the intensity distribution. Specifically, multiple Wi-Fi reflected electromagnetic wave signals with the same or similar intensity distribution are grouped together (i.e., belonging to the same RF path).

[0085] Next, as shown in FIG7C , the computation module 14 retains only the multiple Wi-Fi reflected electromagnetic wave signals (i.e., the CSI data set) included in the primary RF path. Finally, as shown in FIG7D , the computation module 14 converts the CSI data set into frequency domain data using a fast Fourier transform and determines whether the highest frequency value of the CSI data set falls within a preset respiratory frequency range.

[0086] In one embodiment, the respiratory rate range is a user-defined range corresponding to the respiratory rate of person 2, for example, 8 bpm to 22 bpm, but not limited thereto. In the embodiment of FIG7D , the highest frequency value of the CSI data set is 12 bpm, which falls within the preset range of 8 bpm to 22 bpm. Therefore, computing module 14 can determine that person 2 is present in enclosed space 20.

[0087] In one embodiment, the computing module 14 sets the first detection result to True when determining that the highest frequency value of the CSI data set falls within the respiratory frequency range (i.e., determining that person 2 is present in the enclosed space 20), and sets the first detection result to False when determining that the highest frequency value of the CSI data set does not fall within the respiratory frequency range (i.e., determining that person 2 is not present in the enclosed space 20).

[0088] Returning to Figure 4, after step S44, the operation module 14 further determines whether the signal variance of the CSI data set is greater than a predetermined variation range, and generates a corresponding second detection result (step S45).

[0089] As shown in Figure 6 , in some situations (e.g., second time interval T2 in Figure 6 ), if person 2 makes small movements within enclosed space 20, such as waving or crying, detection device 1 may be unable to accurately detect breathing characteristics based on CSI data. To avoid misjudgments caused by person 2's small movements, detection device 1 of the present invention further detects small movement characteristics of person 2 based on CSI data.

[0090] Please also refer to Figure 8, which is a schematic diagram of the small motion feature of the present invention. In Figure 8, the dark circles represent the amplitude of the signal packets transmitted via the main RF path, and the light circles represent the amplitude of the signal packets transmitted via other RF paths.

[0091] The small motion signature refers to the signal variance of the CSI data set. In the present invention, the user can set a desired variance range based on variables such as the size and shape of the enclosed space 20, and the number, type, and size of inanimate objects placed within. The calculation module 14 can then determine whether a person 2 is present in the enclosed space 20 based on whether the signal variance of the CSI data set exceeds this variance range.

[0092] It's worth noting that the user can reconfigure the sensitivity of detection device 1 based on environmental variables. If the user believes that the detection results produced by detection device 1 at its current sensitivity are not as expected (e.g., mistaking an inanimate object for a person), the user can adjust the sensitivity of detection device 1 (e.g., modify the amplitude of the change) based on the environmental conditions of enclosed space 20 (e.g., whether new equipment or machinery has been added or removed). This allows the detection device 1 to be calibrated to improve detection accuracy.

[0093] Specifically, the variation amplitude refers to the degree of change in signal amplitude. The computation module 14 calculates the difference between the highest and lowest amplitude peaks of multiple signal packets in the CSI data set and compares this difference with a predetermined variation amplitude. If this difference is greater than the predetermined variation amplitude, it indicates that person 2 is present in the enclosed space 20 and is performing a minor action.

[0094] In one embodiment, the calculation module 14 may calculate the signal variance according to the following formula:

[0095] In the above formula, Var is the signal variance. Q is the CSI data set, where Wherein, m is the number of the small frequency bands.

[0096] In one embodiment, the operation module 14 outputs the second detection result as true when the signal variation number of the CSI data set is greater than a preset variation range (i.e., it is determined that the person 2 is present in the enclosed space 20), and outputs the second detection result as false when the signal variation number of the CSI data set is less than or equal to the preset variation range (i.e., it is determined that the person 2 is not present in the enclosed space 20).

[0097] Returning to Figure 4, after step S45, the calculation module 14 further determines whether the number of signal packets included in the CSI data set is less than a preset number threshold, and generates a corresponding third detection result (step S46).

[0098] As shown in Figure 6 , in some situations (e.g., during the fourth time interval T4 in Figure 6 ), person 2 may make large movements within enclosed space 20, such as getting up or out of a vehicle. In such cases, detection device 1 may be unable to detect breathing characteristics based on CSI data. To avoid misjudgments due to large movements by person 2, detection device 1 of the present invention further detects large movement characteristics of person 2 based on CSI data.

[0099] Please also refer to Figure 9, which illustrates the large motion characteristics of the present invention. In Figure 9, dark circles represent the amplitude of signal packets transmitted via the primary RF path, while light circles represent the amplitude of signal packets transmitted via other RF paths. As can be seen from Figure 9, when person 2 performs a large motion, the number of signal packets transmitted via the primary RF path is significantly less compared to the embodiment of Figure 8.

[0100] Because the number of signal packets transmitted along the same path decreases significantly when person 2 performs large movements, even the primary RF path will still contain fewer packets than expected. In the present invention, the user can preset a desired packet count threshold based on variables such as the size and shape of enclosed space 20, as well as the number, type, and size of inanimate objects placed within. Thus, calculation module 14 can determine whether person 2 is present within enclosed space 20 based on whether the number of signal packets included in the CSI data set is less than the preset packet count threshold.

[0101] As mentioned above, when the user believes that the detection result of the detection device 1 is different from the expectation, the user can adjust the sensitivity of the detection device 1 according to the environmental conditions of the enclosed space 20 (for example, modify the number threshold value), thereby improving the detection accuracy of the detection device 1 through correction.

[0102] Specifically, if the computing module 14 identifies a main RF path with a relatively large number of packets in step S43 of FIG. 4 , but the number of signal packets covered in this main RF path is less than the preset threshold, it means that a person 2 is present in the enclosed space 20 and this person 2 is performing a large movement.

[0103] In one embodiment, the operation module 14 outputs the third detection result as true when the number of signal packets included in the CSI data set is less than the number threshold value (i.e., it is determined that person 2 is present in the closed space 20), and outputs the third detection result as false when the number of signal packets included in the CSI data set is greater than or equal to the number threshold value (i.e., it is determined that person 2 is not present in the closed space 20).

[0104] It's worth noting that the present invention simultaneously detects breathing, small motion, and large motion features based on CSI data in the primary RF path, thereby avoiding misjudgments. However, the detection of these three features is not in a specific order. That is, steps S44, S45, and S46 in Figure 4 are not executed in a fixed order and are not limited to the one shown in Figure 4.

[0105] Return to Figure 4. After step S46, the operation module 14 determines whether one of the first detection result (corresponding to the breathing feature), the second detection result (corresponding to the small movement feature) and the third detection result (corresponding to the large movement feature) is true (step S47). In the present invention, the operation module 14 simultaneously detects the breathing feature, the small movement feature and the large movement feature based on the same CSI data, and determines that there is a person 2 present in the closed space 20 when any detection result is true (step S48). On the contrary, if the first detection result, the second detection result and the third detection result are all false, it means that the operation module 14 cannot detect the breathing feature, the small movement feature and the large movement feature from the CSI data. At this time, the operation module 14 determines that there is no person 2 present in the closed space 20 (step S49).

[0106] After generating the detection result of whether person 2 is present or absent, the computing module 14 of the present invention can receive a connection and query from a remote device via a wired or wireless method to provide a detection result of whether person 2 is present (step S50 ).

[0107] Specifically, as shown in FIG3 , the detection device 1 may further include a transmission module 15 connected to the operation module 14. The transmission module 15 may be, for example, a Wi-Fi module, a Bluetooth module, an infrared module, or a connector module, etc., for connecting to an external remote device via a wired or wireless method. In the present invention, the user can use a remote device to connect to the detection device 1 (or a cloud provided with the operation module 14), thereby obtaining a warning notification of the presence of the person 2 at the remote end. In this way, whether a child is left alone in the car 2 or a thief invades an unoccupied house 3, the user can obtain a warning notification in the first place to avoid unfortunate events.

[0108] In one embodiment, the detection device 1 of the present invention may also be provided with a display module (not shown). When the computing module 14 generates a detection result of whether the person 2 is present or absent, the display module (e.g., a display screen or a speaker) may be used to directly display the result.

[0109] In the aforementioned embodiment, detection device 1 is a single Wi-Fi device. The detection method of the present invention is implemented by independently transmitting and receiving signal packets within the single detection device 1, using the signal transmitting module 11 and the signal receiving module 12. Consequently, the detection method of the present invention can be implemented without requiring a shared oscillator between the signal transmitting module 11 and the signal receiving module 12 within the detection device 1.

[0110] In the aforementioned embodiment, the signal transmitting module 11 and the signal receiving module 12 can use the 5.8 GHz frequency band to connect to the user's mobile device (not shown) to provide networking functionality, and simultaneously use the 5.8 GHz frequency band to transmit and receive Wi-Fi electromagnetic wave signals, thereby detecting whether a person is present in the enclosed space 20.

[0111] In addition to the embodiment of realizing personnel detection by using a single detection device 1 to transmit and receive signal packets, the present invention further discloses a personnel presence detection system (hereinafter referred to as the detection system), which uses at least two detection devices to transmit and receive signal packets to realize the detection method of the present invention.

[0112] Please refer to Figure 10, which is a schematic diagram of a person presence detection system according to the present invention. In the embodiment of Figure 10, the detection system includes a first Wi-Fi device 5, a second Wi-Fi device 6, and a computing module 7, wherein the first Wi-Fi device 5 and the second Wi-Fi device 6 are disposed in the same enclosed space 20. In this embodiment, the first Wi-Fi device 5, the second Wi-Fi device 6, and the computing module 7 each execute corresponding computer-executable code to collectively implement the steps of the detection method shown in Figure 4.

[0113] In one embodiment, the computing module 7 is connected to both the first Wi-Fi device 5 and the second Wi-Fi device 6 via a wired or wireless connection. In another embodiment, the computing module 7 is connected to a Wi-Fi device (hereinafter, the second Wi-Fi device 6 is used as an example) for receiving Wi-Fi reflected electromagnetic wave signals via a wired or wireless connection.

[0114] In this embodiment, the first Wi-Fi device 5 and the second Wi-Fi device 6 connect to each other via the 2.4 GHz frequency band, thereby establishing a Wi-Fi Mesh network within the enclosed space 20. Furthermore, the first Wi-Fi device 5 and the second Wi-Fi device 6 each connect to a user's mobile device (not shown) via the 5.8 GHz frequency band to provide networking functionality. Similar to the previous embodiment, the first Wi-Fi device 5 and the second Wi-Fi device 6 also use the 5.8 GHz frequency band to transmit and receive Wi-Fi electromagnetic wave signals, thereby detecting the presence of person 2 within the enclosed space 20.

[0115] In the embodiment of Figure 10 , after the user leaves enclosed space 20, the detection system can be activated. At this point, the detection system controls the first Wi-Fi device 5 to continuously transmit multiple Wi-Fi electromagnetic wave signals and controls the second Wi-Fi device 6 to continuously receive multiple Wi-Fi reflected electromagnetic wave signals. Furthermore, the detection system can use the second Wi-Fi device 6 to obtain CSI data based on the multiple Wi-Fi reflected electromagnetic wave signals. The Wi-Fi electromagnetic wave signals, Wi-Fi reflected electromagnetic wave signals, and CSI data are the same as those described in the previous embodiment and are not further described here.

[0116] In the first embodiment, the computing module 7 may be disposed in the second Wi-Fi device 6 responsible for receiving the Wi-Fi reflected electromagnetic wave signal, and implemented by the second Wi-Fi device 6. In the second embodiment, the computing module 7 may be independently disposed in the enclosed space 20 (e.g., implemented by a smart mobile device, tablet computer, or personal computer), and connected to the first Wi-Fi device 5 and / or the second Wi-Fi device 6 via wired or wireless means.

[0117] In the third embodiment, the computing module 7 can be located in the cloud (e.g., implemented by a cloud server) and connected to the first Wi-Fi device 5 and / or the second Wi-Fi device 6 via a wired or wireless network. Thus, the computing module 7 can receive the CSI data from the first Wi-Fi device 5 or the second Wi-Fi device 6 and identify the primary RF path from the CSI data. The primary RF path is the same as in the previous embodiment and will not be further described here.

[0118] As in the previous embodiment, after computing module 7 obtains the CSI data and identifies the primary RF path, it processes the CSI data set and performs the following operations: (1) detecting breathing characteristics and generating a first detection result; (2) detecting small motion characteristics and generating a second detection result; and (3) detecting large motion characteristics and generating a third detection result. The breathing characteristics, small motion characteristics, large motion characteristics, and their detection methods are the same as those described in the previous embodiment and are not further described here.

[0119] When the computing module 7 determines that at least one of the first detection result, the second detection result, and the third detection result is true, the computing module 7 generates a detection result indicating that a person is present. Conversely, when the first detection result, the second detection result, and the third detection result are all false, the computing module 7 generates a detection result indicating that no person is present.

[0120] In the present invention, the user can remotely operate the remote device 8 (eg, a smart mobile device) and connect to the computing module 7 via a wired network or a wireless network to query the computing module 7 for the detection result.

[0121] In one embodiment, a user can operate a remote device 8 to log in to the computing module 7. When the computing module 7 determines that a person is present in the enclosed space 20 based on the first, second, and third detection results, it can immediately and automatically send an alert message to the remote device 8 to notify the user, thereby preventing accidents.

[0122] The above description is only a preferred embodiment of the present invention, and does not limit the claims of the present invention. Therefore, all equivalent changes made by applying the content of the present invention are similarly included in the scope of the present invention and are hereby stated.

Claims

1. A person presence detection device, characterized in that: include: The signal sending module continuously sends multiple Wi-Fi electromagnetic wave signals to the outside world; The signal receiving module continuously receives multiple Wi-Fi reflected electromagnetic wave signals; a processor, controlling the signal sending module and the signal receiving module, and obtaining communication status information data based on the multiple Wi-Fi reflected electromagnetic wave signals; and A computing module connected to the processor, identifying a main radio frequency path from the communication status information data, wherein the main radio frequency path includes a communication status information data set having the largest number of packets in the communication status information data; Among them, the operation module is configured to determine whether the highest frequency value of the communication status information data set falls within the breathing frequency range through fast Fourier transform and generate a first detection result, determine whether the signal variation number of the communication status information data set is greater than a preset variation range and generate a second detection result, determine whether the number of packets of the communication status information data set is less than a preset number threshold value and generate a third detection result, and determine that there is a person present when one of the first detection result, the second detection result and the third detection result is true.

2. The person presence detection device according to claim 1, characterized in that: The signal receiving module is configured to receive the multiple Wi-Fi reflected electromagnetic wave signals through multiple radio frequency paths, and the calculation module is configured to group the multiple Wi-Fi reflected electromagnetic wave signals based on the intensity distribution of each of the Wi-Fi reflected electromagnetic wave signals in multiple small frequency bands, and use the radio frequency path with the largest number of packets as the main radio frequency path.

3. The person presence detection device according to claim 1, characterized in that: The operation module is configured to group the multiple Wi-Fi reflected electromagnetic wave signals based on a DBSCAN algorithm, and use the radio frequency path with the largest number of packets as the main radio frequency path.

4. The person presence detection device according to claim 1, characterized in that: The respiratory rate range is 8 to 22 bpm.

5. The person presence detection device according to claim 1, characterized in that: The computing module is disposed in the cloud and is connected to the processor via a wired network or a wireless network.

6. The person presence detection device according to claim 1, characterized in that: It also includes a transmission module, which is connected to the operation module and is configured to send the detection result of the presence of personnel to the outside.

7. A person presence detection system, characterized in that: include: A first Wi-Fi device continuously sends multiple Wi-Fi electromagnetic wave signals; a second Wi-Fi device, continuously receiving a plurality of Wi-Fi reflected electromagnetic wave signals, and obtaining communication status information data based on the plurality of Wi-Fi reflected electromagnetic wave signals; and a computing module connected to the second Wi-Fi device, receiving the communication status information data and identifying a main radio frequency path from the communication status information data, wherein the main radio frequency path includes a communication status information data set with the largest number of packets in the communication status information data; Among them, the operation module is configured to determine whether the highest frequency value of the communication status information data set falls within the breathing frequency range through fast Fourier transform and generate a first detection result, determine whether the signal variation number of the communication status information data set is greater than a preset variation range and generate a second detection result, determine whether the number of packets of the communication status information data set is less than a preset number threshold value and generate a third detection result, and determine that there is a person present when one of the first detection result, the second detection result and the third detection result is true.

8. The personnel presence detection system according to claim 7, characterized in that: The second Wi-Fi device is configured to receive the multiple Wi-Fi reflected electromagnetic wave signals through multiple radio frequency paths, and the operation module is configured to group the multiple Wi-Fi reflected electromagnetic wave signals based on the intensity distribution of each of the Wi-Fi reflected electromagnetic wave signals in multiple small frequency bands, and use the radio frequency path with the largest number of packets as the main radio frequency path.

9. The personnel presence detection system according to claim 7, characterized in that: The operation module is configured to group the multiple Wi-Fi reflected electromagnetic wave signals based on a DBSCAN algorithm, and use the radio frequency path with the largest number of packets as the main radio frequency path.

10. The personnel presence detection system according to claim 7, characterized in that: The computing module is arranged in the cloud and connected to the second Wi-Fi device via a wired network or a wireless network, and the presence detection system further comprises a remote device, which is configured to connect to the computing module via a wired network or a wireless network to query the detection result of presence of a person.

11. A method for detecting presence of a person, characterized in that: include: a) Continuously send multiple Wi-Fi electromagnetic wave signals to the outside world; b) The Wi-Fi device continuously receives multiple Wi-Fi reflected electromagnetic wave signals; c) obtaining communication status information data by the Wi-Fi device based on the multiple Wi-Fi reflected electromagnetic wave signals; d) The operation module identifies the main radio frequency path from the communication status information data, wherein The main radio frequency path includes a communication status information data set having the largest number of packets in the communication status information data; e) determining by fast Fourier transform whether the highest frequency value of the communication status information data set falls within the respiratory frequency range and generating a first detection result; f) determining whether the signal variation number of the communication status information data set is greater than a preset variation range and generating a second detection result; g) determining whether the number of packets in the communication status information data set is less than a preset number threshold and generating a third detection result; and h) determining that a person is present when one of the first detection result, the second detection result and the third detection result is true.

12. The method for detecting presence of a person according to claim 11, characterized in that: The step a) is that the Wi-Fi device continuously sends the multiple Wi-Fi electromagnetic wave signals to the outside, and the step b) is that the same Wi-Fi device continuously receives the multiple Wi-Fi reflected electromagnetic wave signals.

13. The method for detecting presence of a person according to claim 11, characterized in that: The step b) includes receiving the multiple Wi-Fi reflected electromagnetic wave signals respectively through multiple radio frequency paths, and the step d) includes grouping the multiple Wi-Fi reflected electromagnetic wave signals based on the intensity distribution of each of the Wi-Fi reflected electromagnetic wave signals in multiple small frequency bands, and using the radio frequency path with the largest number of packets as the main radio frequency path.

14. The method for detecting presence of a person according to claim 11, characterized in that: The step d) includes grouping the multiple Wi-Fi reflected electromagnetic wave signals based on the DBSCAN algorithm, and taking the radio frequency path with the largest number of packets as the main radio frequency path.

15. The method for detecting presence of a person according to claim 11, characterized in that: Also includes: i) The computing module accepts the connection and query of the remote device to provide the detection result of the presence of personnel.

Images