[go: up one dir, main page]
More Web Proxy on the site http://driver.im/

WO2024002029A1 - 呼吸检测方法、电子设备、存储介质及程序产品 - Google Patents

呼吸检测方法、电子设备、存储介质及程序产品 Download PDF

Info

Publication number
WO2024002029A1
WO2024002029A1 PCT/CN2023/102515 CN2023102515W WO2024002029A1 WO 2024002029 A1 WO2024002029 A1 WO 2024002029A1 CN 2023102515 W CN2023102515 W CN 2023102515W WO 2024002029 A1 WO2024002029 A1 WO 2024002029A1
Authority
WO
WIPO (PCT)
Prior art keywords
frequency
respiratory
signal set
respiratory frequency
domain candidate
Prior art date
Application number
PCT/CN2023/102515
Other languages
English (en)
French (fr)
Inventor
孙超
王皓
张大庆
郭兴民
孙雪
张博
Original Assignee
华为技术有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 华为技术有限公司 filed Critical 华为技术有限公司
Publication of WO2024002029A1 publication Critical patent/WO2024002029A1/zh

Links

Classifications

    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/08—Measuring devices for evaluating the respiratory organs
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/08—Measuring devices for evaluating the respiratory organs
    • A61B5/0816—Measuring devices for examining respiratory frequency
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H40/00—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices
    • G16H40/60—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices for the operation of medical equipment or devices
    • G16H40/67—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices for the operation of medical equipment or devices for remote operation
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B17/00—Monitoring; Testing
    • H04B17/30—Monitoring; Testing of propagation channels
    • H04B17/309—Measuring or estimating channel quality parameters
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W24/00—Supervisory, monitoring or testing arrangements
    • H04W24/10—Scheduling measurement reports ; Arrangements for measurement reports
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/30—Services specially adapted for particular environments, situations or purposes

Definitions

  • This application relates to the field of communication technology, and in particular to breath detection methods, electronic devices, storage media and program products.
  • Long-term respiratory frequency monitoring has a very high medical guidance value.
  • the use of electronic equipment to monitor respiratory frequency for a long time is very helpful for the early detection of some diseases.
  • the contact methods are somewhat intrusive. It is not easy for users to insist on using monitoring for a long time.
  • Wi-Fi wireless network
  • CSI Channel State Information
  • the channel state information describes the attenuation and phase experienced by the signal from the signal transmitter to the signal receiver along the multipath. Phase rotation, any change in the signal propagation path will affect the channel status information obtained by the signal receiving end. For example, the rise and fall of the chest caused by human breathing will cause the measured value of the channel status information to show periodic changes. This is a good way to use the channel status information to analyze human breathing. It is possible to conduct detection.
  • the inventor found that the channel state information extracted from the Wi-Fi signal contains noise.
  • the extracted channel state information is layered, and the channel state information based on the layer cannot accurately detect breathing. , even breath detection cannot be achieved.
  • respiration detection methods, electronic equipment, storage media and program products that can process hierarchical channel state information, and can perform respiration detection based on the hierarchical channel state information to obtain accurate respiration detection information.
  • embodiments of the present application provide a breathing detection method, which method includes: a first device receives channel state information of a Wi-Fi signal, and the first device determines N signal sets based on the phase difference or amplitude of the channel state information, The phase differences or amplitudes within the N signal sets respectively belong to N different numerical ranges, where N is an integer greater than or equal to 2.
  • the first device determines the environment change information based on the N signal sets, and the environment change information includes breathing detection information.
  • the data in each signal set includes the phase difference or amplitude of the channel state information. If N signal sets are determined based on the phase differences of the channel state information, the phase differences within the N signal sets respectively belong to N different numerical ranges. If N signal sets are determined based on the amplitude of the channel state information, the amplitudes in the N signal sets respectively belong to N different numerical ranges.
  • N signal sets are determined based on the phase difference or amplitude of the channel state information, so that when processing layered channel state information, the layered channel state information is processed according to the phase difference of the channel state information. Or the amplitude is divided into N signal sets, and each signal set is extracted. Respiration detection can be performed based on the N signal sets, so as to implement respiration detection based on the layered channel state information and obtain accurate respiration detection information.
  • determining the N signal sets according to the phase difference or amplitude of the channel state information includes: determining the phase difference of the channel state information of each subcarrier; dividing the phase difference into N signal sets; or determining the phase difference of each subcarrier.
  • the amplitude of the channel state information divide the amplitude into a signal set to obtain N signal sets.
  • determining the environmental change information based on the N signal sets includes: obtaining the target signal set based on the N signal sets; obtaining the frequency domain candidate respiratory frequency and the respiratory energy ratio corresponding to the frequency domain candidate respiratory frequency based on the target signal set; according to the target signal
  • the time domain candidate respiratory frequency and the first autocorrelation coefficient corresponding to the time domain candidate respiratory frequency are collected together to obtain the environmental change information according to the frequency domain candidate respiratory frequency, respiratory energy ratio, time domain candidate respiratory frequency and the first autocorrelation coefficient.
  • the target signal set of each subcarrier is obtained by extracting the sub-signals of the channel state information of each subcarrier layer. And perform fast Fourier transform and autocorrelation function calculation on the target signal set of each subcarrier to simultaneously obtain the frequency domain candidate respiratory frequency, respiratory energy ratio, time domain candidate respiratory frequency and first autocorrelation coefficient of the subcarrier.
  • the possibility of the subcarrier carrying the sensing object's respiratory information is determined to verify the frequency domain candidate Whether the respiratory frequency or time-domain candidate respiratory frequency is the respiratory frequency of the sensing object, double verification ensures the accuracy of the respiratory detection information.
  • outputting the environmental change information according to the frequency domain candidate respiratory frequency, the respiratory energy ratio, the time domain candidate respiratory frequency and the first autocorrelation coefficient includes: when it is detected that the respiratory energy ratio is greater than the corresponding first threshold and the first autocorrelation coefficient is greater than When the corresponding second threshold is reached, the frequency domain candidate respiratory frequency or the time domain candidate respiratory frequency is output as respiratory detection information.
  • outputting the environmental change information based on the frequency domain candidate respiratory frequency, the respiratory energy ratio, the time domain candidate respiratory frequency and the first autocorrelation coefficient includes: when it is detected that the frequency domain candidate respiratory frequency is close to the time domain candidate respiratory frequency, and the respiratory energy When one or more of the ratio and the first autocorrelation coefficient are greater than the corresponding threshold, the frequency domain candidate respiratory frequency or the time domain candidate respiratory frequency is output as respiratory detection information.
  • the frequency domain candidate respiratory frequency or the time domain candidate is Outputting the respiratory frequency as respiratory detection information includes: when it is detected that the candidate respiratory frequency in the frequency domain is close to the candidate respiratory frequency in the time domain, and one or more of the respiratory energy ratio and the first autocorrelation coefficient is greater than the corresponding threshold, recording the respiratory energy ratio and the sum of the first autocorrelation coefficient; add the sum of all recorded respiratory energy ratios and the first autocorrelation coefficient as the denominator; take the maximum value of the sum of the recorded respiratory energy ratio and the first autocorrelation coefficient as Numerator; when it is detected that the ratio of the numerator to the denominator is less than the third threshold, the frequency domain candidate respiratory frequency or the time domain candidate respiratory frequency is output as respiratory detection information.
  • filtering the N signal sets to obtain the target signal set includes: for each subcarrier, deleting signal sets whose phase differences or amplitudes are less than the preset threshold in the N signal sets to obtain the remaining signal sets. ; Obtain the target signal set based on the remaining signal sets.
  • obtaining the target signal set according to the remaining signal sets includes: performing one or more of the following operations on each remaining signal set, and using the final remaining signal set as the target signal set: judging the remaining Whether the phase difference or amplitude within the preset time period is missing in the signal set; if so, delete the remaining signal set; or, determine the first missing amount and the second missing amount in the remaining signal set every preset time interval Whether the difference is higher than the preset missing threshold; wherein, the first missing amount corresponds to the highest value of the missing amount of phase difference or the missing amount of amplitude within the preset time interval, and the second missing amount corresponds to the missing amount of phase difference within the preset time interval or the minimum value of the missing amount of amplitude; if so, delete the remaining signal set.
  • performing analysis and processing on the target signal set in the frequency domain to obtain the frequency domain candidate respiratory frequency and the respiratory energy ratio corresponding to the frequency domain candidate respiratory frequency includes: performing fast Fourier transform on the target signal set to obtain the target signal set The corresponding frequency domain signal; according to the respiratory frequency band and the frequency domain signal, the frequency domain candidate respiratory frequency and the respiratory energy ratio corresponding to the frequency domain candidate respiratory frequency are obtained.
  • obtaining the frequency domain candidate respiratory frequency and the respiratory energy ratio corresponding to the frequency domain candidate respiratory frequency according to the respiratory frequency band and the frequency domain signal includes obtaining a sub-frequency domain signal in the frequency domain signal with a frequency located in the respiratory frequency band; converting the amplitude of the sub-frequency domain signal Sum of values and frequency domain signal The ratio of the sum of the amplitudes is used as the respiratory energy ratio; the frequency corresponding to the maximum amplitude in the sub-frequency domain signal is used as the frequency domain candidate respiratory frequency.
  • performing analysis and processing on the target signal set in the time domain to obtain the time domain candidate respiratory frequency and the first autocorrelation coefficient corresponding to the time domain candidate respiratory frequency includes: performing autocorrelation calculation on the target signal set to obtain the corresponding The autocorrelation coefficient; according to the respiratory frequency band and the autocorrelation coefficient, the time domain candidate respiratory frequency and the first autocorrelation coefficient corresponding to the time domain candidate respiratory frequency are obtained.
  • obtaining the time domain candidate respiratory frequency and the first autocorrelation coefficient corresponding to the time domain candidate respiratory frequency based on the respiratory frequency band and the autocorrelation coefficient includes: using the frequency corresponding to the maximum autocorrelation coefficient in the respiratory frequency band as the time domain candidate respiratory frequency; The autocorrelation coefficient corresponding to the candidate respiratory frequency in the time domain is used as the first autocorrelation coefficient.
  • inventions of the present application provide an electronic device.
  • the electronic device includes at least one processor and a memory, where the memory is used to store instructions, and the processor is used to execute the instructions to implement any of the above methods.
  • the electronic device includes a terminal device or a wireless access node.
  • the electronic device is used to present an interface or output speech based on the breath detection information.
  • embodiments of the present application provide a computer-readable storage medium.
  • the computer-readable storage medium stores a program, and the program causes the electronic device to perform any of the above methods.
  • inventions of the present application provide a computer program product.
  • the computer program product includes computer-readable instructions. When the computer-readable instructions are executed by one or more processors, any one of the above methods is implemented.
  • the electronic device provided by the second aspect, the computer-readable storage medium provided by the third aspect, and the computer program product provided by the fourth aspect correspond to the method provided by the first aspect or the second aspect, and therefore, they are The beneficial effects that can be achieved or the various implementation methods can be referred to above and will not be described again here.
  • Figure 1 is a schematic diagram of the application scenario of the breath detection method provided by this application.
  • Figure 2A is a schematic diagram of channel state information provided by this application.
  • FIG. 2B is a schematic diagram of another channel state information provided by this application.
  • Figure 2C is a schematic diagram of the result of processing the channel state information in Figure 2B;
  • Figure 3A is a schematic diagram of the result of signal extraction of the channel state information in Figure 2B by the breathing detection method provided by this application;
  • Figure 3B is a schematic diagram of the results of fast Fourier transform and autocorrelation calculation on the channel state information in Figure 3A using the breathing detection method provided by this application;
  • Figure 4 is a schematic flow chart of a breath detection method provided by an embodiment of the present application.
  • Figure 5A is a schematic diagram of subcarrier No. 6, subcarrier No. 16 and subcarrier No. 26 provided by this application;
  • Figure 5B is a schematic diagram of extracting the sub-signal of sub-carrier No. 6 in Figure 5A;
  • Figure 6 is a histogram obtained by counting the phase differences of subcarriers provided by this application.
  • Figure 7 is a schematic diagram of the results obtained by performing fast Fourier transform on the target signal set of all subcarriers
  • Figure 8 is a schematic flow chart of a method for obtaining frequency domain candidate respiratory frequency and respiratory energy ratio provided by this application;
  • Figure 9 is a schematic diagram of the results obtained after performing autocorrelation calculation on the target signal set of a subcarrier
  • Figure 10 is a schematic flow chart of the method for obtaining the time domain candidate respiratory frequency and the first autocorrelation coefficient provided by this application;
  • Figure 11 is a method provided by this application based on frequency domain candidate respiratory frequency, respiratory energy ratio, time domain candidate respiratory frequency and the third A schematic flow chart of the method for outputting respiratory detection information by autocorrelation coefficient;
  • Figure 12 is a schematic flow chart of another method provided by this application for outputting respiratory detection information based on frequency domain candidate respiratory frequency, respiratory energy ratio, time domain candidate respiratory frequency and first autocorrelation coefficient;
  • Figure 13 is a schematic diagram of the main interface of the user equipment provided by the embodiment of the present application.
  • Figure 14 is a schematic diagram of the breath detection information interface of user equipment provided by an embodiment of the present application.
  • Figure 15 is a schematic diagram of another breath detection information interface of user equipment provided by an embodiment of the present application.
  • Figure 16 is a schematic flow chart of another breath detection method provided by an embodiment of the present application.
  • Figure 17 is a schematic flow chart of another breath detection method provided by an embodiment of the present application.
  • Figure 18 is a schematic flow chart of another breath detection method provided by an embodiment of the present application.
  • Figure 19 is a schematic diagram of the software architecture of the electronic device provided by the embodiment of the present application.
  • Figure 20 is a schematic structural diagram of the second device provided by the embodiment of the present application.
  • Figure 21 is a schematic structural diagram of the first device provided by the embodiment of the present application.
  • Figure 22 is a schematic structural diagram of user equipment provided by an embodiment of the present application.
  • Figure 1 is a schematic diagram of an application scenario according to an embodiment of the present application.
  • the application scenario includes the second device 101, the first device 102 and the sensing object 103.
  • Figure 1 takes the application scenario including a second device 101, a first device 102 and a sensing object 103 as an example.
  • the actual application may include two or more second devices 101, two or more third devices A device 102.
  • the number and form of devices, the number and form of objects shown in Figure 1 do not constitute a limitation on the embodiment of the present application.
  • the second device 101 may include one or more transmitting antennas.
  • the first device 102 may include two or more receiving antennas, which is not specifically limited in this application.
  • a transmitting antenna of the second device 101 and a receiving antenna of the first device 102 are an antenna pair, and the antenna pair corresponds to a path.
  • a path Starting from the frequency domain, a path includes multiple subcarriers, and each subcarrier is a frequency domain. Channel; if starting from the time domain, a path includes multiple paths, including straight-line propagation paths and other transmission paths (i.e. multipath propagation).
  • the second device 101 sends a Wi-Fi signal to the first device 102, and the Wi-Fi signal reaches the first device 102 after passing through the sensing object 103.
  • the first device 102 obtains channel state information based on the Wi-Fi signal, and determines environmental change information of the environment where the first device 102 is located based on the channel state information, including the sensing result of sensing the sensing object 103 .
  • the sensing result can be used to indicate the breathing status of the sensing object 103, such as whether the sensing object 103 is breathing or the breathing frequency of the sensing object 103, etc.
  • the application scenario may also include user equipment 104.
  • the user equipment 104 is configured to receive the environment change information transmitted by the first device 102 and notify the user of the environment change information. Among them, the notification method can be interface display or voice playback, etc.
  • the user equipment 104 communicates with the first device 102.
  • the connection method between the first device 102 and the user equipment 104 may be a wired connection (network cable, PLC and other wired connection methods) or a wireless connection (Wi-Fi, Bluetooth and other wireless transmission methods).
  • This application The communication method is not limited.
  • the environmental change information may be the breathing detection information of the sensing object 103, as shown in Figures 14 and 15.
  • the user device 104 and the second device 101 in the application scenario are the same device, that is, the first device 102 can send the perceived environment change information to the second device 101 (user device 104).
  • the user device 104 in the application scenario and the first device 102 are the same device, and the first device 102 notifies the user of the calculated environment change information. That is, the user equipment 104 can perform the breathing detection method provided by the embodiment of the present application. When the user carries (such as holds) the user equipment 104 and moves to a new environment where the first device 102 or the second device 101 is arranged, the user equipment 104 It is still possible to communicate with the second device 101 or the first device 102 in the environment.
  • the user equipment 104 can perform breathing detection based on the received Wi-Fi signal sent by the second device 101 and sense the environment change information of the environment in which it is located, or the user equipment 104 can receive the environment change information transmitted by the first device 102 in the new environment. .
  • both the sensing object 103 and the first device 102 are within the coverage of the second device 101 .
  • the application scenario may also include a cloud server 105.
  • the cloud server 105 may communicate with the first device 102 and be used to receive and store the channel state information or environment change information of the Wi-Fi signal transmitted by the first device 102.
  • the cloud server 105 can communicate with the user equipment 104 to transmit channel state information or environmental change information of the Wi-Fi signal to the user equipment 104 .
  • the first device 102, the cloud server 105 or the user equipment 104 may respond to the received The information (channel state information of Wi-Fi signal or environmental change information) is analyzed and processed to obtain the breathing detection information of the sensing object 103.
  • the user equipment 104 obtains the breathing detection information or the user equipment 104 from the cloud server 105 or the first device 102. After calculating the respiration detection information, the user equipment 104 presents the respiration detection information, as shown in Figures 14 and 15.
  • the first device 102, the cloud server 105 or the user device 104 can perform fusion processing on the received information (channel state information of the Wi-Fi signal or environmental change information), such as based on the received information and Other fusion information is fused and calculated to output analysis results such as work and rest detection, sedentary reminder, breathing in the room, respiratory frequency, etc.
  • the user equipment 104 presents the analysis result, as shown in Figures 14 and 15.
  • the breath detection method of the embodiment of the present application can be divided into one or more modules.
  • the one or more modules can be a series of computer program instruction segments capable of completing specific functions.
  • the instruction segments are The execution process of the breath detection method according to the embodiment of the present application.
  • the one or more modules may be stored in the first device 102 and/or the user device 104 and/or the cloud service 105. That is to say, some steps in the breath detection method in the embodiment of this application can be executed by the first device 102, some steps can be executed by the user device 104, and some steps can be executed by the cloud server 105. This application does not specifically limit this. .
  • the second device 101 is an entity for transmitting Wi-Fi signals.
  • the second device 101 may be, for example, Bluetooth glasses, a Bluetooth watch, a Bluetooth headset, a Bluetooth speaker, a Bluetooth smart screen, a tablet, a Bluetooth desk lamp, a Bluetooth door lock, a Bluetooth socket, a Bluetooth electronic scale, a mobile phone, a wearable device, a tablet, a wireless Computers with transceiver functions, routers, wireless access points AP, LTE indoor signal transmission base stations, virtual reality (VR) terminal equipment, augmented reality (AR) terminal equipment, self-driving (self-driving) wireless terminals, etc.
  • VR virtual reality
  • AR augmented reality
  • the first device 102 is an entity for receiving Wi-Fi signals.
  • the first device 102 may be, for example, a wireless router or wireless access point AP, an LTE indoor signal transmitting base station, a mobile phone, a wearable device, a tablet, a computer with wireless transceiver functions, etc.
  • the user device 104 is a terminal device used to notify the user of the sensing results.
  • the user device 104 can be, for example, a mobile phone, a wearable device, a tablet, a computer with a wireless transceiver function, a virtual reality (VR) terminal device, an enhanced Augmented reality (AR) terminal equipment, etc.
  • VR virtual reality
  • AR enhanced Augmented reality
  • the above application scenario can be an environment where the above devices can communicate through Wi-Fi, such as a home environment, an office environment, etc.
  • the following takes the application scenario in Figure 1 as a home environment as an example.
  • the second device 101 sends a Wi-Fi signal
  • the first device 102 receives the Wi-Fi signal
  • the breath detection method provided by the embodiment method to obtain the breathing detection information of the sensing object 103 in the home environment.
  • the first device 102 receives the Wi-Fi signal
  • the first device 102 sends the channel state information of the Wi-Fi signal to the user equipment 104 or the cloud service 105
  • the user equipment 104 or the cloud service 105 executes the method provided by the embodiment of the present application.
  • the breathing detection method obtains the breathing detection information of the sensing object 103 in the home environment based on the channel state information of the Wi-Fi signal. If the first device 102 executes the breathing detection method provided by the embodiment of the present application, the user device 104 obtains the breathing detection information of the sensing object 103 from the first device 102 . If the cloud server 105 executes the breathing detection method provided by the embodiment of the present application, the user equipment 104 obtains the breathing detection information of the sensing object 103 from the cloud server 105 .
  • the user lives alone.
  • the first device 102 can monitor the environmental change information of the sensing object 103 (i.e., the user) in the home environment during sleep, and then obtain the breathing detection information based on the environmental change information.
  • the first device 102, the second device 101, the cloud server 105 or the user device 104 can help the user to know whether there is an apnea disorder during sleep through the obtained breathing detection information. From this, it can be seen that users who suffer from sleep apnea disorder can avoid It is difficult to voluntarily detect the time and severity of apnea, which may lead to the problem that the symptoms cannot be treated and improved in time.
  • the breath detection information obtained by the breath detection method of the embodiment of the present application is highly accurate, thereby improving the accuracy and success rate of users' human health status detection.
  • the sensing object 103 in Figure 1 is a person is only used as an example and does not constitute a limitation on the embodiments of the present application.
  • the sensing object 103 may also be an animal, etc.
  • the embodiment of this application does not limit the specific technology and device form used by the second device 101, the first device 102, the user device 104, and the sensing object 103.
  • the current Wi-Fi breathing detection method can only process noise-free or slightly noisy channel state information.
  • Figure 2A there is no layering of channel state information.
  • the inventor also found that the channel state information extracted from Wi-Fi signals is generally noisy.
  • the reasons for the presence of noise in the channel state information include but are not limited to: in order to maximize communication efficiency, devices (such as routers) enable automatic gain control. Automatic gain control will affect the channel status information, causing noise in the channel status information.
  • the channel state information with noise is layered, and the channel state information includes multiple layers of sub-signals. As shown in Figure 2B, the channel state information includes four layers of sub-signals, which respectively correspond to the curves S1, S2, S3 and S4 of Figure 2B.
  • curve S1 has a signal between 0-3 seconds, a signal between 6-8 seconds, and a signal between 11-15 seconds.
  • Curve S5 in Figure 2C indicates that after using the existing breathing detection method to perform Fast Fourier Transform (FFT) on the channel state information in Figure 2B, the amplitude corresponding to each frequency is 0. .
  • Curve S6 in Figure 2C indicates that after using the existing respiration detection method to calculate the autocorrelation function (ACF) of the channel state information in Figure 2B, the autocorrelation coefficient corresponding to the respiration frequency band is a negative number. It can be seen from this that the current breathing detection method cannot obtain frequency information from the layered channel state information in Figure 2B, resulting in the inability to perform breathing detection based on the multi-layered channel state information, and there is a large gap between it and commercial implementation.
  • FFT Fast Fourier Transform
  • the breathing detection method provided by the embodiment of the present application can be used to process noise-free or slightly noisy channel state information, and can also be used to process layered channel state information, and can be based on the layered channel state information. breath test, get Accurate breath detection information.
  • Figure 3A shows the channel state information obtained by processing the channel state information in Figure 2B using the breathing detection method provided by the embodiment of the present application, which is the extracted sub-signal.
  • the channel state information data obtained after processing is more uniform and the data distribution is clearer.
  • Figure 3B shows the results obtained by performing fast Fourier transform and autocorrelation calculation on the channel state information in Figure 3A using the breathing detection method provided by the embodiment of the present application. After extracting the sub-signal using the respiration detection method provided by the embodiment of the present application, and performing fast Fourier transform on the sub-signal, a higher amplitude characteristic value can be obtained in the respiration frequency band.
  • a higher autocorrelation coefficient is obtained in the respiratory frequency band. That is to say, based on the respiration detection method provided by the embodiment of the present application, frequency information can be obtained from the layered channel state information in Figure 2B, and then respiration detection can be performed based on the multi-layered channel state information.
  • the breathing detection method provided by the embodiment of this application can be applied to the first device 102, the user equipment 104 and the cloud server 105, and this application does not specifically limit this.
  • FIG. 4 is a schematic flow chart of a breath detection method provided by an embodiment of the present application.
  • the breath detection method is applied to the first device shown in Figure 1 as an example.
  • Step S41 Obtain channel state information of each subcarrier.
  • the second device sends out data packets through M subcarriers at a specific rate (such as 200 Hz per second).
  • the two antennas of the first device simultaneously receive data packets on multiple subcarriers and measure the data packets generated by each subcarrier.
  • the first device extracts the preamble part in the message, divides the received preamble part by the known sequence stored locally, and obtains the corresponding channel state information.
  • the leading part in the message is a sequence known to both the second device and the first device.
  • M that is, the number of subcarriers
  • M is related to the Wi-Fi protocol, bandwidth and the number of antennas used.
  • 802.11a/g has 52 subcarriers in 20MHz mode
  • 802.11n has 56 subcarriers in 20MHz mode.
  • the embodiment of this application does not specifically limit the number of subcarriers.
  • the packet can be a data packet carrying a special training symbol, or a null data packet (NDP), or a physical layer protocol data unit (physical layer protocol date units, PPDU) .
  • NDP null data packet
  • PPDU physical layer protocol data unit
  • the first device samples each packet once and performs sampling at a specific sampling frequency (such as 50 Hz). In other embodiments, the first device performs sampling once each time it receives a message, and continuously samples over time, such as collecting for 15 seconds, to obtain a sequence of channel state information in time order.
  • a specific sampling frequency such as 50 Hz.
  • the first device performs sampling once each time it receives a message, and continuously samples over time, such as collecting for 15 seconds, to obtain a sequence of channel state information in time order.
  • the channel state information is used to reflect the current wireless channel conditions.
  • measurements are made for each orthogonal frequency division multiplexing (OFDM) subcarrier group to obtain the CSI matrix corresponding to the OFDM subcarrier group.
  • the number of rows of the CSI matrix is the number of transmitting antennas, and the number of columns of the CSI matrix is the number of receiving antennas.
  • the element value of each CSI matrix represents the channel response on a subcarrier.
  • H is the CSI matrix
  • hi represents the channel state information of the i-th subcarrier
  • M is the number of subcarriers.
  • each packet index to be analyzed will eventually obtain K (number of transmitting antennas) ⁇ J (number of receiving antennas) ⁇ M (number of subcarriers) channel state information.
  • a subcarrier is a complex number containing a real part and an imaginary part. The real part corresponds to the amplitude of the channel state information, and the real part corresponds to the phase of the channel state information.
  • Step S42 For each subcarrier, process the phase difference or amplitude of the channel state information to obtain N signal sets.
  • the phase differences or amplitudes in the N signal sets respectively belong to N different numerical ranges, where N is greater than Or an integer equal to 2.
  • the phase difference or the amplitude of the channel state information includes the amplitude, phase and phase difference of the channel state information.
  • step S41 for each subcarrier, continuous sampling is performed over time to obtain the channel state information sequence of each subcarrier.
  • the channel state information obtained by the first device in step S41 is hierarchical. Therefore, in step S42, the first device extracts the sub-signals in the channel state information to obtain a corresponding signal set. Based on the fact that subcarriers are complex numbers, for each subcarrier, extract the phase or amplitude in the channel state information of the subcarrier according to the time sequence, and obtain the phase sequence or amplitude sequence of the channel state information of the subcarrier, which can be expressed as a phase sequence or amplitude Sequences represent sub-signals. The first device can extract the sub-signals from the channel state information based on the phase sequence or the amplitude sequence.
  • the second device includes one transmitting antenna
  • the first device includes two receiving antennas (antenna A and antenna B)
  • the sampling time is 15 seconds
  • subcarrier No. 6 is used as an example.
  • Q1 is the phase sequence of subcarrier No. 6 received by antenna A
  • ai represents the phase corresponding to subcarrier No. 6 collected at the i-th sampling point within 15 seconds
  • T is the sampling point within 15 seconds. quantity.
  • Q2 is the phase sequence of subcarrier No. 6 received by antenna B
  • bi represents the phase corresponding to subcarrier No. 6 collected at the i-th sampling point within 15 seconds
  • T is the sampling point within 15 seconds. quantity.
  • C is the phase difference sequence of subcarrier No. 6
  • ci represents the phase difference corresponding to subcarrier No. 6 collected at the i-th sampling point within 15 seconds
  • T is the number of sampling points within 15 seconds.
  • the channel state information in which a single subcarrier contains breathing information may include multiple sub-signals, where the sub-signals can be understood as the smallest subset of the noisy (layered) channel state information containing breathing frequency information,
  • the reason for the layering of channel state information may be due to the influence of automatic gain control.
  • the inventor found that the sub-signal also has the ability to reflect the current wireless channel conditions.
  • the sub-signals of each layer can be extracted, so that the layered channel state information can be processed.
  • subcarrier No. 6 shows the sampling time is 15 seconds and the corresponding subcarriers No. 6, 16 and 26 among the 52 subcarriers. Phase difference sequence. It can be seen from Figure 5A that the channel state information of each subcarrier includes two layers, the two layers are spaced apart from each other, and each layer represents a sub-signal. Take the 6th in Figure 5A To illustrate subcarrier No. 6, subcarrier No. 6 includes two curves S7 and S8. The curves S7 and S8 are spaced apart from each other. The curve S7 represents one sub-signal and the curve S8 represents another sub-signal.
  • each subcarrier multiple sub-signals included in its channel state information are extracted respectively to obtain multiple signal sets.
  • N signal sets can be extracted.
  • Figure 5B shows how to extract the sub-signal of sub-carrier No. 6 in Figure 5A. For example, extract curve S7, obtain each phase difference sequence on curve S7, and obtain a signal set.
  • the curve S8 is extracted, each phase difference sequence on the curve S8 is obtained, and another signal set is obtained.
  • the sampling time of both signal sets is 15 seconds.
  • One ups and downs of the sub-signal of subcarrier No. 6 represents one breath of the sensing object.
  • window 51 intercepts one breathing cycle of the sub-signal of subcarrier No. 6. This breathing cycle can be exhalation first and then inhalation, or it can be inhalation first and then exhalation. That is, the curve 51a in the window 51 corresponds to the phase difference sequence during exhalation, and the curve 51b in the window 51 corresponds to the phase difference sequence during inhalation. Alternatively, the curve 51a in the window 51 corresponds to the phase difference sequence during inhalation, and the curve 51b in the window 51 corresponds to the phase difference sequence during exhalation.
  • the phase difference or the amplitude sequence in time order can be obtained, such as the above phase difference sequence.
  • the phase differences or the amplitudes with similar values in the phase difference or the amplitude sequence are divided into the same signal set, thereby dividing the phase differences or the amplitudes with different values through different signal sets.
  • the phase differences on the curve S7 are numerically closer to other phase differences on the curve S7, while the phase differences on the curve S7 are numerically farther from the phase differences on the curve S8.
  • phase difference on curve S7 is divided into the same signal set, and the phase difference on curve S7 belongs to the same numerical range, such as (-4.5 to -4.6] degrees.
  • the phase difference on curve S8 is divided into another signal set, curve The phase difference on the S8 belongs to the same numerical range, such as (-5 to -4.9] degrees.
  • N signal sets have different corresponding numerical ranges. That is to say, for each subcarrier, if it includes N sub-signals, N signal sets can be obtained through step S42, which correspond to N different value ranges.
  • the phase differences corresponding to all sampling points during the sampling time can be counted and counted, the number of sampling points corresponding to each phase difference can be counted, and then values with similar values are merged into a set.
  • the phase differences corresponding to all sampling points within 15 seconds can be counted and counted, and merge the phase differences with similar values into a group to obtain a histogram as shown in Figure 6.
  • four signal sets (groups) can be obtained.
  • the first signal set corresponds to a numerical range of (2.5-2.8] degrees.
  • the second signal set corresponds to a numerical range of (3.4-4.1] degrees. .
  • the numerical range corresponding to the third signal set is (4.5-4.9] degrees.
  • the numerical range corresponding to the fourth signal set is (5.1-5.4] degrees.
  • antenna A can obtain a sequence of 52 subcarrier channel state information amplitudes, where the channel state information of each subcarrier includes multiple sub-signals. For each subcarrier, numerically similar amplitudes are grouped into the same signal set.
  • antenna B can obtain 52 subcarrier channel state information amplitude sequences, perform the same operation on each subcarrier obtained by antenna B, and extract the signal set of each subcarrier, which will not be described again here.
  • multiple sub-signals of the channel state information of each subcarrier can be extracted separately based on the phase difference, or based on the amplitude, or both based on the phase difference and amplitude. This application does not specifically limit this.
  • Step S43 Filter the N signal sets to obtain the target signal set.
  • the first device filters N signal sets in the subcarrier to obtain a target signal set that can represent the subcarrier, that is, it can be obtained based on the target signal set.
  • the subcarrier carries the breathing frequency information of the sensing object, and the data of the target signal set is evenly distributed.
  • step S43 can be implemented as: for each subcarrier, delete the signal sets in the N signal sets whose phase difference or amplitude is less than the preset threshold to obtain the remaining signal sets. That is, for each subcarrier, each signal set of the subcarrier is screened to determine whether the phase difference or the number of amplitudes in each signal set is less than the preset threshold. If so, delete the signal set, If not, keep the signal set. Then, the target signal set is obtained based on the remaining signal set after screening.
  • the preset threshold is set to 10% of the sampling amount (such as 2000), where the sampling amount is the subcarrier obtained by the first device within the sampling time for each subcarrier.
  • the amount of channel state information for the carrier It is determined whether the total number of phase differences or amplitudes included in each signal set is less than 200, and if so, the signal set is deleted.
  • the preset threshold is 100
  • the phase difference or the total amplitude of the third signal set and the fourth signal set are both less than the preset threshold
  • the third signal set and the fourth signal set are deleted.
  • the first signal set and the second signal set remain, that is, the data of curve S3 and curve S4 in Figure 2B are retained, and the data of curve S1 and curve S2 with sparse data distribution are deleted.
  • the preset threshold can be set according to the actual situation, and it can also be set to 20% of all sampling volumes. In other embodiments, the preset threshold can be set to 10% or 20% of the sampling points, etc. This application is This is not specifically limited.
  • the remaining signal sets are obtained, and each remaining signal set is Perform one or more of the following operations (first operation and second operation) to use the final remaining signal set as the target signal set.
  • First operation For each remaining signal set, determine whether the phase difference or the amplitude within the preset time period is missing in the signal set, and if so, delete the signal set. If not, keep the signal set.
  • the preset time period can be 2 seconds, 5 seconds or 6 seconds.
  • the preset time period can be set according to the sampling time, for example, set to 10% or 20% of the sampling time. This application does not specifically limit this.
  • Second operation For each remaining signal set, determine whether the difference between the first missing amount and the second missing amount of the signal set at a preset time interval is higher than the preset missing threshold; where, the first missing amount is The highest value of the phase difference or the amplitude missing amount should be within the preset time interval, and the second missing amount corresponds to the lowest value of the phase difference or the amplitude missing amount within the preset time interval. If so, delete the signal collection, if not, retain the signal collection.
  • the preset time interval can be a fixed value, such as 1 second, 2 seconds or 3 seconds, and the preset missing threshold can be 40%, 50% or 60%, which is not specifically limited in this application.
  • the phase difference or the amplitude missing amount may be determined based on the sampling point or the sampling amount. For example, if the phase difference or the number of amplitudes collected at each sampling point is set to 100, and if the phase difference or the number of amplitudes at a certain sampling point on the signal set is 60, then Determine the phase difference or the amplitude at the sampling point corresponding to the signal set The missing amount is 40%.
  • the second operation for each remaining signal set, obtain the first missing amount per second of the signal set and The second missing amount. For example, within the first second, the highest value of the phase difference or the amplitude loss amount (first loss amount) of the signal set at that second is 60%. When the minimum value (second missing amount) of the phase difference or the amplitude missing amount of the signal set in this second is 30%, then the difference between 60% and 30% is less than 50%, then the signal set is judged to be At the 1st second, the signal set is retained. In the same way, continue to judge the second second. If it is determined in 2 seconds that the difference between the first missing amount and the second missing amount in the signal set is higher than the preset missing threshold, then the signal set is deleted and the next signal set of the subcarrier is continued to be judged.
  • the first device can perform the above determination on the signal sets of all subcarriers at the same time, or can determine each subcarrier in order of subcarriers to obtain the target signal set of each subcarrier.
  • the final remaining signal set is used as the target of the subcarrier.
  • the number of target signal sets may be multiple, such as 2.
  • the first device after the first device performs one or more of the first operation and the second operation on multiple signal sets of each subcarrier, it detects whether the number of remaining signal sets is 2 or 2. greater than 2. If so, the remaining signal sets are further screened and a signal set is selected as the target signal set. If not, the final remaining signal set is directly used as the target signal set.
  • the first device uses the signal set with the smallest value range among the final remaining signal sets as the target signal set. For example, as shown in Figure 6, the final remaining signal set includes a first signal set and a second signal set, wherein the value range of the first signal set is 0.3 (2.8-2.5) smaller than the second signal set.
  • the value range corresponding to the signal set is 0.7 (4-3.4.1), and the first signal set is regarded as the target signal set.
  • the remaining first signal set and the second signal set correspond to curves S3 and S4 respectively, that is, S4 with the smallest curve line width is selected as the target signal set.
  • the first device after the first device selects the target signal set, it performs data processing on the target signal set, such as linear interpolation to complete the data, and then uses the processed target signal set for processing in the following steps.
  • data processing such as linear interpolation to complete the data
  • Step S44 Perform fast Fourier transform on the target signal set to obtain a frequency domain signal corresponding to the target signal set.
  • the first device performs discrete Fourier transform on the target signal set of the subcarrier, extracts the frequency domain information of the target signal set, and obtains the frequency domain signal corresponding to the target signal set. . After converting to frequency domain signals, the frequency components in the target signal set can be easily analyzed for processing in the frequency domain.
  • the first device can obtain the target signal set corresponding to each subcarrier.
  • the first device performs fast Fourier transform on the target signal sets of all subcarriers to obtain the frequency domain signal corresponding to each target signal set.
  • Figure 7 shows the results obtained by fast Fourier transform on the target signal set of all subcarriers. Each curve represents the result of fast Fourier transform on the target signal set corresponding to one subcarrier. It can be seen that in the frequency It has good amplitude characteristics between 10 times/minute and 20 times/minute.
  • Step S45 Obtain the frequency domain candidate respiratory frequency and the respiratory energy ratio corresponding to the frequency domain candidate respiratory frequency according to the respiratory frequency band and the frequency domain signal.
  • the first device after obtaining the frequency domain signal corresponding to the target signal set of each subcarrier, the first device obtains the frequency domain candidate respiratory frequency corresponding to each subcarrier and the frequency domain candidate respiratory frequency corresponding to the frequency domain candidate respiratory frequency according to each frequency domain signal. Respiratory energy ratio.
  • the respiratory frequency band is the respiratory power value of the human body under normal conditions.
  • the respiratory frequency band may be 0.1-0.6Hz.
  • Step S45 can be specifically implemented as the following steps:
  • Step S81 Obtain the sub-frequency domain signal in the frequency domain signal whose frequency is located in the respiratory frequency band.
  • the first device uses the signal in the frequency domain signal with a frequency within 0.1-0.6Hz as the subfrequency domain of the subcarrier. signal to obtain the sub-frequency domain signal of the sub-carrier.
  • Step S82 Use the ratio of the sum of the amplitudes of the sub-frequency domain signals to the sum of the amplitudes of the frequency domain signals as the respiratory energy ratio.
  • the first device uses the ratio of the sum of the amplitudes of the subcarrier's sub-frequency domain signals to the sum of the amplitudes of the frequency domain signals as the breathing energy ratio. Specifically, the first device The sum of the amplitudes of the sub-frequency domain signals of the sub-carrier is added to obtain the sum of the amplitudes of the sub-frequency domain signals. The sum of the amplitudes of the frequency-domain signals of the sub-carrier is summed to obtain the sum of the amplitudes of the frequency-domain signals. Calculate the sum of the sub-frequency domain signals. The ratio of the sum of the amplitudes of the frequency domain signals to the sum of the amplitudes of the frequency domain signals is used as the breathing energy ratio, and the breathing energy ratio corresponds to the subcarrier.
  • calculating the respiratory energy ratio is: summing the amplitudes of all signals in the frequency domain signal with frequencies located in the respiratory frequency band as the numerator, summing all the amplitudes of the frequency domain signals as the denominator, and the ratio of the numerator to the denominator That is the respiratory energy ratio.
  • Step S83 Use the frequency corresponding to the maximum amplitude in the sub-frequency domain signal as the frequency domain candidate respiratory frequency.
  • the first device obtains the signal with the largest amplitude among the sub-frequency domain signals of the subcarrier, and uses the frequency corresponding to the signal as the frequency domain candidate breathing frequency.
  • the first device calculates the breath-to-noise ratio (BNR) of each subcarrier, and uses the frequency corresponding to the maximum breath-to-noise ratio of the subcarrier as the frequency of the subcarrier.
  • BNR breath-to-noise ratio
  • Frequency domain candidate respiratory frequency is calculated as the ratio of the maximum amplitude in the sub-frequency domain signal to the sum of the amplitudes of the frequency domain signal.
  • Step S46 Perform autocorrelation calculation on the target signal set to obtain the autocorrelation coefficient corresponding to the target signal set.
  • the first device performs autocorrelation calculation on the target signal set of each subcarrier, that is, calculates the autocorrelation function (Autocorrelation Function, ACF) on the target signal set, and obtains the autocorrelation function corresponding to the target signal set of each subcarrier.
  • Autocorrelation Coefficient AC
  • Figure 9 is the result obtained after performing autocorrelation calculation on the target signal set of a subcarrier.
  • the curve represents the autocorrelation analysis result of the subcarrier. It can be seen that when the frequency is 12 times/each The autocorrelation coefficient is larger between minutes and 16 times/minute.
  • Step S47 Obtain the time domain candidate respiratory frequency and the first autocorrelation coefficient corresponding to the time domain candidate respiratory frequency according to the respiratory frequency band and the autocorrelation coefficient.
  • the first device after obtaining the autocorrelation coefficient corresponding to the target signal set of each subcarrier, the first device obtains the time domain candidate respiratory frequency corresponding to each subcarrier and the corresponding time domain candidate respiratory frequency according to each autocorrelation coefficient.
  • Step S47 can be specifically implemented as:
  • Step S101 Use the frequency corresponding to the maximum autocorrelation coefficient in the respiratory frequency band as the time domain candidate respiratory frequency.
  • the respiratory frequency band is 0.1-0.6 Hz
  • the signals in the target signal set with frequencies within 0.1-0.6 Hz are divided into the first set to obtain the first set.
  • Each channel state information in the first signal set has its corresponding autocorrelation coefficient, and the frequency corresponding to the signal with the largest autocorrelation coefficient in the first signal set is used as the time domain candidate respiratory frequency.
  • the frequency corresponding to the autocorrelation coefficient with the largest value among the autocorrelation coefficients in the respiratory frequency band is used as the candidate respiratory frequency in the time domain.
  • Step S102 Use the autocorrelation coefficient corresponding to the candidate respiratory frequency in the time domain as the first autocorrelation coefficient.
  • the largest autocorrelation coefficient in the first signal set is regarded as the first autocorrelation coefficient.
  • the autocorrelation coefficient with the largest value among the autocorrelation coefficients in the respiratory frequency band is used as the first autocorrelation coefficient.
  • Step S48 Output respiratory detection information based on the frequency domain candidate respiratory frequency, the respiratory energy ratio, the time domain candidate respiratory frequency and the first autocorrelation coefficient.
  • the first device performs frequency domain and time domain calculations on the target signal set of each subcarrier through the above steps S44-S47, and obtains the frequency domain candidate breathing frequency and frequency domain candidate breathing frequency of each subcarrier.
  • the first device pairs the obtained frequency-domain candidate respiratory frequency, respiratory energy ratio, time-domain candidate respiratory frequency, and the third An autocorrelation coefficient is used for joint judgment in the time domain and frequency domain to obtain reliable respiratory detection information.
  • step S48 can be implemented as: when it is detected that the respiratory energy ratio is greater than the corresponding first threshold and the first autocorrelation coefficient is greater than the corresponding second threshold, changing the frequency domain candidate respiratory frequency or the time domain candidate respiratory frequency Output as respiratory detection information.
  • the first threshold and the second threshold may be the same or different, and the first threshold and the second threshold may be determined according to actual conditions.
  • the first threshold and the second threshold may be set to a value range of 0.1-10.
  • the respiratory energy ratio When it is detected that the respiratory energy ratio is greater than the corresponding first threshold, it indicates that the frequency domain candidate respiratory frequency is highly likely to be the respiratory frequency of the sensing object.
  • the first autocorrelation coefficient When it is detected that the first autocorrelation coefficient is greater than the corresponding second threshold, it indicates that the time domain candidate respiratory frequency is highly likely to be the respiratory frequency of the sensing object.
  • the respiratory energy ratio is greater than the corresponding first threshold and the first autocorrelation coefficient is greater than the corresponding second threshold, it means that the frequency domain candidate respiratory frequency and the time domain candidate respiratory frequency are highly likely to be the respiratory frequency of the sensing object. It also means that The consistency between the time domain and the frequency domain can further ensure that the candidate respiratory frequency in the frequency domain or the candidate respiratory frequency in the time domain is the respiratory frequency of the sensing object.
  • step S48 may be implemented as: when it is detected that the candidate respiratory frequency in the frequency domain is close to the candidate respiratory frequency in the time domain, and one or more of the respiratory energy ratio and the first autocorrelation coefficient is greater than the corresponding threshold, The frequency domain candidate respiratory frequency or the time domain candidate respiratory frequency is output as respiratory detection information.
  • one or more of the respiratory energy ratio and the first autocorrelation coefficient is greater than the corresponding threshold, including: the respiratory energy ratio is greater than the corresponding first threshold, the first autocorrelation coefficient is greater than the corresponding second threshold, or the respiratory energy ratio is greater than the corresponding The first threshold and the first autocorrelation coefficient are greater than the corresponding second threshold.
  • the frequency domain candidate respiratory frequency is detected to be close to the time domain candidate respiratory frequency and the respiratory energy ratio is greater than the corresponding first threshold, or when the frequency domain candidate respiratory frequency is detected to be close to the time domain candidate respiratory frequency and the first self-
  • the correlation coefficient is greater than the corresponding second threshold, or when it is detected that the candidate respiratory frequency in the frequency domain is close to the candidate respiratory frequency in the time domain, the respiratory energy ratio is greater than the corresponding first threshold, and the first autocorrelation coefficient is greater than the corresponding second threshold.
  • Step S48 can be specifically implemented as:
  • Step S111 Determine whether the difference between the frequency domain candidate respiratory frequency and the time domain candidate respiratory frequency is less than a third threshold
  • the third threshold can be set according to the actual situation, and the value range of the third threshold can be 0.1-2 beats per minute (BPM). If it is determined that the difference between the frequency domain candidate respiratory frequency and the time domain candidate respiratory frequency is less than the third threshold, it means that the frequency domain candidate respiratory frequency and the time domain candidate respiratory frequency are close. By verifying that the candidate respiratory frequency in the frequency domain is similar to the candidate respiratory frequency in the time domain, in order to avoid false detections caused by separate time domain or frequency domain analysis and calculation, improve accuracy and ensure that the target signal set has respiratory information of the sensing object.
  • BPM beats per minute
  • step S112 delete this target signal set.
  • the difference between the candidate respiratory frequency in the frequency domain and the candidate respiratory frequency in the time domain is not less than the third threshold, it means that the candidate respiratory frequency in the frequency domain is far from the candidate respiratory frequency in the time domain, and there is a high probability that the candidate respiratory frequency cannot be obtained from the target.
  • the signal collection obtains the breathing information of the sensing object.
  • deleting the target signal set means not processing the target signal set, but processing the target signal set of the next subcarrier.
  • the No. 6 subcarrier is processed, and in step S81 it is determined that the difference between the frequency domain candidate respiratory frequency and the time domain candidate respiratory frequency of the No. 6 subcarrier is greater than or equal to the third threshold, then the No. 6 subcarrier is Target signal set deletion of subcarriers. That is, the No. 6 subcarrier is not detected and the next subcarrier is continued to be processed.
  • step S113 Determine whether one or more of the respiratory energy ratio and the first autocorrelation coefficient are greater than the corresponding threshold
  • step S113 includes: judging whether the respiratory energy ratio is greater than a first threshold, or judging whether the first autocorrelation coefficient is greater than a second threshold, or judging whether the respiratory energy ratio is greater than the first threshold, and at the same time the first Whether the autocorrelation coefficient is greater than the second threshold. If any of the above determinations are yes, step S114 is executed. If the above three judgments are all negative, step S112 is executed.
  • Deleting the target signal set means not processing the subcarriers corresponding to the target signal set, but processing the next subcarrier.
  • step S114 Output the frequency domain candidate respiratory frequency or the time domain candidate respiratory frequency as respiratory detection information.
  • step S83 can be implemented as the following steps:
  • Step S121 When it is detected that the candidate respiratory frequency in the frequency domain is close to the candidate respiratory frequency in the time domain, and one or more of the respiratory energy ratio and the first autocorrelation coefficient are greater than the corresponding threshold, record the respiratory energy ratio and the first autocorrelation coefficient. Sum.
  • the first device performs steps S111 and S113 on the frequency domain candidate respiratory frequency, respiratory energy ratio, time domain candidate respiratory frequency, and first autocorrelation coefficient of the subcarrier. Determine, if the judgment result is yes, record the sum of the respiratory energy ratio of the subcarrier and the first autocorrelation coefficient, and use the sum of the respiratory energy ratio of the subcarrier and the first autocorrelation coefficient as the weight of the subcarrier. If the judgment result is no, the sum of the respiratory energy ratio and the first autocorrelation coefficient of the subcarrier is not recorded.
  • Step S122 Add the sum of all recorded respiratory energy ratios and the first autocorrelation coefficient as the denominator.
  • the first device performs the judgments of steps S111 and S113 on all subcarriers, and can obtain the breathing energy ratio and the breathing energy ratio corresponding to the subcarriers that meet the conditions (that is, the judgment results of steps S111 and S113 are yes).
  • the sum of the first autocorrelation coefficients is the sum of the recorded respiratory energy ratios of all qualified subcarriers and the sum of the first autocorrelation coefficients as the denominator. That is, the weights of all eligible subcarriers are summed as the denominator.
  • Step S123 Use the maximum value of the sum of the recorded respiratory energy ratio and the first autocorrelation coefficient as the numerator.
  • all recorded values are obtained, that is, the weights of all subcarriers that meet the conditions (that is, the judgment results of step S111 and step S113 are yes) are obtained, and the largest weight is used as the numerator.
  • Step S124 It is detected that the ratio of the numerator to the denominator is less than the third threshold, and the frequency domain candidate respiratory frequency or the time domain candidate respiratory frequency is output as respiratory detection information.
  • the third threshold can be set according to the actual situation, for example, it can be 50%, which is not specifically limited in this application.
  • the first device determines whether the ratio of the maximum weight to the sum of the weights of all eligible subcarriers is less than a third threshold. If it is less than The candidate respiratory frequency in the time domain can be used as respiratory detection information. If it is not less than, it means that there may be a special situation of a single subcarrier at this time, and the frequency domain candidate respiratory frequency or time domain candidate respiratory frequency at this time cannot be output as a respiratory detection node. For this purpose, breathing detection information can be output It means that the breath cannot be detected, or there is no detectable breath in the current environment.
  • the existing breathing detection method cannot process hierarchical channel state information, and the existing breathing detection method can only process noise-free or slightly noisy channel state information, which generally processes the channel state information alone.
  • the breathing detection method provided by this application extracts the sub-signals of the channel state information of each sub-carrier layer to obtain the target signal set of each sub-carrier. And perform fast Fourier transform and autocorrelation function calculation on the target signal set of each subcarrier to simultaneously obtain the frequency domain candidate respiratory frequency, respiratory energy ratio, time domain candidate respiratory frequency and first autocorrelation coefficient of the subcarrier.
  • the possibility of the subcarrier carrying the sensing object's respiratory information is determined to verify the frequency domain candidate Whether the respiratory frequency or time-domain candidate respiratory frequency is the respiratory frequency of the sensing object, double verification ensures the accuracy of the respiratory detection information. Furthermore, the frequency domain candidate respiratory frequencies or time domain candidate respiratory frequencies of all subcarriers are verified to ensure the consistency between the data of each target signal set, and no false detection will be caused by the special circumstances of a single carrier, further providing respiratory detection. Accuracy of information.
  • Figure 13 is a schematic diagram of the main interface of the user equipment provided by the embodiment of the present application.
  • the breath detection method provided by the embodiment of the present application can be applied to breath detection applications, or implemented as breath detection applications.
  • the respiration detection application is installed on the user device 104.
  • the user clicks the respiration detection application on the main interface.
  • the user equipment 104 establishes communication with the second device 101.
  • the user equipment 104 can obtain the channel state information of the Wi-Fi signal from the second device 101 or the end server 105, and then according to the Wi-Fi signal
  • the channel state information of the Fi signal executes the breathing detection method provided by the embodiment of the present application to obtain the user's breathing detection information.
  • the user device 104 obtains the environment change information from the second device 101 or the end server 105, and then analyzes the user's breathing detection information based on the environment change information. Please refer to Figure 14 as well.
  • the user device 104 enters the breath detection information interface.
  • the corresponding breathing detection information is displayed on the breathing detection information interface, including body movement, deep sleep state, and night breathing detected within the night sleep recording time of 8 hours and 9 minutes. Changes in frequency (respiration detection information).
  • the user device 104, the first device 102 or the cloud server 105 can also have information analysis and processing energy, further analyze and process the environmental change information, and output more detailed breathing detection information including: sleep score, night sleep duration, wake time, and wake times. , respiratory frequency, respiratory variability index, breathing instructions, and detailed respiratory selected data collected during the day's sleep.
  • FIG. 16 is a schematic flowchart of another breath detection method provided by an embodiment of the present application.
  • Step S161 The first device receives a Wi-Fi signal from at least one second device.
  • Step S162 The first device determines N signal sets based on the phase difference or amplitude of the channel state information of the Wi-Fi signal of at least one second device.
  • the phase differences or amplitudes in the N signal sets respectively belong to N different numerical ranges. , where N is an integer greater than or equal to 2.
  • Step S163 The first device determines environmental change information based on N signal sets, where the environmental change information includes breathing detection information.
  • N signal sets are determined based on the phase difference or amplitude of the channel state information, so that when processing layered channel state information, the layered channel state information is processed according to the phase difference of the channel state information. Or the amplitude is divided into N signal sets, and each signal set is extracted. Respiration detection can be performed based on the N signal sets to achieve information based on this layer. Perform respiratory detection based on channel status information to obtain accurate respiratory detection information.
  • step S162 may specifically include: extracting the phase difference of the channel state information for each subcarrier; dividing phase differences with similar values into a signal set to obtain N signal sets; or, for each subcarrier, Extract the amplitude of the channel state information; divide the amplitudes with similar values into a signal set to obtain N signal sets.
  • step S163 may specifically include: filtering N signal sets to obtain a target signal set; performing frequency domain analysis and processing on the target signal set to obtain the frequency domain candidate respiratory frequency and the respiratory energy corresponding to the frequency domain candidate respiratory frequency. ratio; analyze and process the target signal set in the time domain to obtain the time domain candidate respiratory frequency and the first autocorrelation coefficient corresponding to the time domain candidate respiratory frequency; according to the frequency domain candidate respiratory frequency, respiratory energy ratio, time domain candidate respiratory frequency And the first autocorrelation coefficient outputs environmental change information.
  • the target signal set of each subcarrier is obtained by extracting the sub-signals of the channel state information of each subcarrier layer. And perform fast Fourier transform and autocorrelation function calculation on the target signal set of each subcarrier to simultaneously obtain the frequency domain candidate respiratory frequency, respiratory energy ratio, time domain candidate respiratory frequency and first autocorrelation coefficient of the subcarrier.
  • the possibility of the subcarrier carrying the sensing object's respiratory information is determined to verify the frequency domain candidate Whether the respiratory frequency or time-domain candidate respiratory frequency is the respiratory frequency of the sensing object, double verification ensures the accuracy of the respiratory detection information.
  • outputting the environmental change information according to the frequency domain candidate respiratory frequency, the respiratory energy ratio, the time domain candidate respiratory frequency and the first autocorrelation coefficient includes: when it is detected that the respiratory energy ratio is greater than the corresponding first threshold and the first autocorrelation coefficient is greater than When the corresponding second threshold is reached, the frequency domain candidate respiratory frequency or the time domain candidate respiratory frequency is output as respiratory detection information.
  • outputting the environmental change information based on the frequency domain candidate respiratory frequency, the respiratory energy ratio, the time domain candidate respiratory frequency and the first autocorrelation coefficient includes: when it is detected that the frequency domain candidate respiratory frequency is close to the time domain candidate respiratory frequency, and the respiratory energy When one or more of the ratio and the first autocorrelation coefficient are greater than the corresponding threshold, the frequency domain candidate respiratory frequency or the time domain candidate respiratory frequency is output as respiratory detection information.
  • the frequency domain candidate respiratory frequency or the time domain candidate is Outputting the respiratory frequency as respiratory detection information includes: when it is detected that the candidate respiratory frequency in the frequency domain is close to the candidate respiratory frequency in the time domain, and one or more of the respiratory energy ratio and the first autocorrelation coefficient is greater than the corresponding threshold, recording the respiratory energy ratio and the sum of the first autocorrelation coefficient; add the sum of all recorded respiratory energy ratios and the first autocorrelation coefficient as the denominator; take the maximum value of the sum of the recorded respiratory energy ratio and the first autocorrelation coefficient as Numerator; when it is detected that the ratio of the numerator to the denominator is less than the third threshold, the frequency domain candidate respiratory frequency or the time domain candidate respiratory frequency is output as respiratory detection information.
  • filtering the N signal sets to obtain the target signal set includes: for each subcarrier, deleting signal sets whose phase differences or amplitudes are less than the preset threshold in the N signal sets to obtain the remaining signal sets. ; Obtain the target signal set based on the remaining signal sets.
  • obtaining the target signal set according to the remaining signal sets includes: performing one or more of the following operations on each remaining signal set, and using the final remaining signal set as the target signal set: judging the remaining Whether the phase difference or amplitude within the preset time period is missing in the signal set; if so, delete the remaining signal set; or, determine the first missing amount and the second missing amount in the remaining signal set every preset time interval Whether the difference is higher than the preset missing threshold; wherein, the first missing amount corresponds to the highest value of the missing amount of phase difference or the missing amount of amplitude within the preset time interval, and the second missing amount corresponds to the missing amount of phase difference within the preset time interval or the minimum value of the missing amount of amplitude; if so, delete the remaining signal set.
  • performing analysis and processing on the target signal set in the frequency domain to obtain the frequency domain candidate respiratory frequency and the respiratory energy ratio corresponding to the frequency domain candidate respiratory frequency includes: performing fast Fourier transform on the target signal set to obtain the target signal set The corresponding frequency domain signal; according to the respiratory frequency band and the frequency domain signal, the frequency domain candidate respiratory frequency and the respiratory energy ratio corresponding to the frequency domain candidate respiratory frequency are obtained.
  • obtaining the frequency domain candidate respiratory frequency and the respiratory energy ratio corresponding to the frequency domain candidate respiratory frequency according to the respiratory frequency band and the frequency domain signal includes obtaining a sub-frequency domain signal in the frequency domain signal with a frequency located in the respiratory frequency band; converting the amplitude of the sub-frequency domain signal The ratio of the sum of the values to the sum of the amplitudes of the frequency domain signals is used as the respiratory energy ratio; the frequency corresponding to the maximum amplitude in the sub-frequency domain signal is used as the frequency domain candidate respiratory frequency.
  • performing analysis and processing on the target signal set in the time domain to obtain the time domain candidate respiratory frequency and the first autocorrelation coefficient corresponding to the time domain candidate respiratory frequency includes: performing autocorrelation calculation on the target signal set to obtain the corresponding The autocorrelation coefficient; according to the respiratory frequency band and the autocorrelation coefficient, the time domain candidate respiratory frequency and the first autocorrelation coefficient corresponding to the time domain candidate respiratory frequency are obtained.
  • obtaining the time domain candidate respiratory frequency and the first autocorrelation coefficient corresponding to the time domain candidate respiratory frequency based on the respiratory frequency band and the autocorrelation coefficient includes: using the frequency corresponding to the maximum autocorrelation coefficient in the respiratory frequency band as the time domain candidate respiratory frequency; The autocorrelation coefficient corresponding to the candidate respiratory frequency in the time domain is used as the first autocorrelation coefficient.
  • the following takes processing the phase difference of channel state information as an example to provide a schematic flow chart of another breathing detection method.
  • Figure 17 is a schematic flow chart of another breath detection method provided by an embodiment of the present application.
  • the breath detection method is applied to the first device shown in Figure 1 as an example.
  • Step S170 Start.
  • the first device in response to performing the operation of the breathing detection method of the present application, the first device begins to receive data packets sent by the second device through M subcarriers, and measures each data packet transmitted by each subcarrier. CSI raw information.
  • the breath detection application is installed on the first device, the operation of executing the breath detection method of the present application is for the user to click on the breath detection application on the first device.
  • the first device may execute the breath detection method of the present application after being powered on. This application does not specifically limit this.
  • Step S170 corresponds to step S41 in Figure 4.
  • Step S171 Calculate the phase difference of the channel state information of all subcarriers.
  • the first device receives the channel state information of M subcarriers, and for each subcarrier, calculates the phase difference of the channel state information of the subcarrier, so that all subcarriers can be calculated. The phase difference of the channel state information of the carrier.
  • Step S172 Extract sub-signals.
  • the phase difference of the channel state information of the subcarriers is obtained, and the first device extracts the sub-signals according to the phase difference of the channel state information of the subcarriers.
  • the sub-signal is the above-mentioned signal set.
  • Step S171 and step S172 correspond to step S42 in Figure 4.
  • step S172 is specifically the following steps:
  • Step S1721 Count the bins in the histogram.
  • the phase differences of the channel state information of all subcarriers are used as bins in the histogram, and the bins in the histogram are counted. For example, the number of each phase difference (such as 5.5 degrees, 5 degrees, 4.5 degrees, etc.) is counted. As shown in Figure 6, the number corresponding to each bin can be obtained.
  • Step S1722 Remove sampling points corresponding to counts that account for too small a proportion.
  • the first device may delete sampling points (ie, phase differences) corresponding to counts that account for too little proportion by deleting sampling points whose counted number is less than a preset threshold (for example, 10).
  • a preset threshold for example, 10
  • the statistical number of phase differences of 5.1 degrees is less than 10
  • the phase difference of 5.1 degrees can be deleted, that is, among the phase differences of the channel state information of all subcarriers, the phase difference is 5.1 degrees
  • the total quantity is less than 10.
  • Step S1722 corresponds to a specific implementation of step S43 in Figure 4. For example, for each subcarrier, delete the signal sets whose phase differences or amplitudes are less than the preset threshold in the N signal sets to obtain the remaining The signal collection below will not be described again here.
  • Step S1723 Merge numerically consecutive bins into a bin group.
  • the first device combines numerically continuous bins into a bin group, that is, combines numerically continuous phase differences into a bin group, that is, combines to obtain a sub-signal.
  • a bin group that is, combines numerically continuous phase differences into a bin group, that is, combines to obtain a sub-signal.
  • four bin groups are obtained.
  • the group distances of the bin groups are (2.5-2.8] degrees, (3.4-4.1] degrees, (4.5-4.9] degrees and (5.1-5.4] degrees respectively.
  • the four bin groups are Each group corresponds to four signal sets.
  • step S1723 may be performed first and then step S1722, which will not be described again.
  • Step S1724 Determine for each bin group whether there is data loss exceeding 2 seconds.
  • the first device determines the bins in each bin group and determines whether the bin in the bin group has data loss exceeding 2 seconds.
  • Step S1724 corresponds to the first operation in Figure 4.
  • step S1725 delete this bin group.
  • the first device deletes this bin group, it deletes the signal set.
  • step S1726 Determine for each bin group whether the fluctuation of the amount of missing data per second is higher than 50%.
  • the first device determines each bin group to determine whether there is data loss exceeding 2 seconds in the bin group. If not, step S1726 is executed. That is, after executing step S1724, the first device obtains the remaining signal set. Step S1726 judges the remaining signal set to determine whether the difference between the highest value of the phase difference missing amount per second and the lowest value of the phase difference missing amount per second in the remaining signal set is higher than 50%. For example, for the first second of the signal set A, it is detected that the highest value of the phase difference missing amount in the first second is 60%, and the highest value 60% corresponds to the sampling point a. It is detected that the lowest value of the phase difference missing amount in the first second is 20%. The lowest value 20% corresponds to the sampling point b. Then the difference between 60% and 20% is less than 50%. Then, when judging the signal set to the At 1 second, the signal set is retained.
  • Step S1724 corresponds to the second operation in Figure 4.
  • step S1722, step S1724, step S1725 and step S1726 correspond to step S43 in Figure 4.
  • step S43 For specific content, please refer to step S43.
  • the first device determines for each bin group whether the fluctuation of the amount of missing data per second is higher than 50%. If so, step S1725 is executed. If not, step S1727 is executed: linear interpolation to complete the data.
  • the first device obtains the target signal set after executing steps S1722 to S1726.
  • the first device can perform data processing on the target signal set, such as linear interpolation to complete the data, and then use the processed target signal set for processing in the following steps.
  • Step S1728 Obtain the breathing sub-signal of the current carrier.
  • the first device performs linear interpolation on the target signal set to complete the data, and obtains the breathing sub-signal of the current carrier, that is, the target signal set of the current carrier. After the first device obtains the target signal set of all carriers, it will be used in subsequent steps.
  • Step S173 Perform joint calculation in time domain and frequency domain.
  • step S173 corresponds to step S44, step S45, step S46 and step S47 in Figure 4.
  • step S44, step S45, step S46 and step S47 please refer to step S44, step S45, step S46 and step S47.
  • step S173 is specifically the following steps:
  • Step S1731 Perform fast Fourier transform.
  • the first device performs fast Fourier transform on the target signal set (that is, the respiratory sub-signal obtained in step S1728) to obtain a frequency domain signal corresponding to the target signal set.
  • step S1731 specifically corresponds to step S44 in Figure 4.
  • Step S1732 Calculate the respiratory energy ratio. Specifically, calculate the proportion of the sum of the amplitudes of the respiratory frequency band to the sum of the amplitudes of all frequencies.
  • the first device takes the sum of the amplitudes of all signals in the frequency domain signal with frequencies located in the respiratory frequency band as the numerator, and the sum of all the amplitudes of the frequency domain signals obtained after the fast Fourier calculation in step S173 as the denominator, and the numerator
  • the ratio to the denominator is the respiratory energy ratio.
  • Step S1732 corresponds to steps S81 to S82 in FIG. 8 .
  • steps S81 to S82 please refer to steps S81 to S82 .
  • Step S1733 Calculate the respiratory signal-to-noise ratio. Specifically, calculate the proportion of the maximum amplitude of the respiratory frequency band to the sum of all frequency amplitudes.
  • the first device calculates the ratio of the maximum amplitude in the sub-frequency domain signal to the sum of the amplitudes of the frequency domain signal to obtain the respiratory signal-to-noise ratio.
  • Step S1734 Frequency domain analysis: Calculate the frequency domain candidate respiratory frequency and calculate the respiratory energy ratio. Specifically, the frequency corresponding to the highest respiratory signal-to-noise ratio is used as the frequency domain candidate respiratory frequency.
  • step S1732 to step S1734 correspond to step S45 in Figure 4
  • step S1734 corresponds to S83 in Figure 8.
  • Step S1735 Perform autocorrelation calculation.
  • the first device performs autocorrelation calculation on the target signal set (that is, the respiratory sub-signal obtained in step S1728), and obtains the autocorrelation coefficient corresponding to the target signal set.
  • step S1735 corresponds to step S46 in Figure 4.
  • step S46 For specific content, please refer to step S46.
  • Step S1736 Time domain analysis: Calculate the time domain candidate respiratory frequency and use the autocorrelation coefficient corresponding to the time domain candidate respiratory frequency as the first autocorrelation coefficient. Specifically, use the frequency corresponding to the highest autocorrelation coefficient in the respiratory frequency band as the time domain Candidate respiratory rate.
  • step S1736 corresponds to step S47 in Figure 4 .
  • the details of step S1736 may include steps S101 to S102 as shown in FIG. 10 , which will not be described again here.
  • Step S174 Perform joint judgment in time domain and frequency domain.
  • the first device jointly determines the candidate respiratory frequency in the frequency domain, the respiratory energy ratio, the candidate respiratory frequency in the time domain and the first autocorrelation coefficient obtained in the above step S173, and outputs the detection result.
  • step S174 corresponds to step S48 in Figure 4.
  • step S48 For specific content, please refer to step S48.
  • step S174 is specifically the following steps:
  • Step S1741 Determine whether the error between the time domain candidate respiratory frequency and the frequency domain candidate respiratory frequency is less than a threshold.
  • the first device determines whether the difference between the frequency domain candidate respiratory frequency and the time domain candidate respiratory frequency is less than a third threshold.
  • step S1741 corresponds to step S111 in Figure 11.
  • step S111 For specific content, please refer to step S111.
  • step S1742 Delete this breathing sub-signal.
  • the first device determines that the difference between the frequency domain candidate respiratory frequency and the time domain candidate respiratory frequency is greater than or equal to the third threshold, and then determines that the target signal set (ie, respiratory sub-signal) performs time domain sum frequency
  • the target signal set ie, respiratory sub-signal
  • the results obtained by domain analysis are quite different.
  • the target signal set is not suitable for breath detection, and the target signal set will be used for breath detection.
  • step S1742 corresponds to step S112 in Figure 11.
  • step S112 For specific content, please refer to step S112.
  • step S1743 Determine whether the respiratory energy ratio and the first autocorrelation coefficient are greater than the threshold.
  • the first device after the first device executes step S1741 and determines that the error between the time domain candidate respiratory frequency and the frequency domain candidate respiratory frequency is less than the third threshold, the first device continues to determine the respiratory energy ratio and the first autocorrelation coefficient. whether one or more is greater than The corresponding threshold value, for example, determines whether the respiratory energy ratio is greater than the first threshold, or determines whether the first autocorrelation coefficient is greater than the second threshold, or determines whether the respiratory energy ratio is greater than the first threshold and at the same time whether the first autocorrelation coefficient is greater than the second threshold. Two thresholds.
  • step S1743 corresponds to step S113 in Figure 11.
  • step S113 For specific content, please refer to step S113.
  • Step S1744 Count all respiratory sub-signals that meet the above conditions, and use the sum of the respiratory energy ratio and the first autocorrelation coefficient of the respiratory sub-signals that meet the above conditions as the weight.
  • the first device counts the respiratory sub-signals whose judgment results in steps S1741 and S1743 are both "yes", records the respiratory energy ratio and the first autocorrelation coefficient of the respiratory sub-signals that meet the above conditions, and adds each The sum of the breathing energy ratio and the first autocorrelation coefficient of a subcarrier is used as the weight of the subcarrier.
  • Step S1745 Determine whether the weight of the item with the highest weight among the candidate respiratory frequencies accounts for less than the threshold among all candidate respiratory frequencies.
  • the first device adds the sum of all recorded respiratory energy ratios and the first autocorrelation coefficient as the denominator, and calculates the maximum value of the sum of the recorded respiratory energy ratio and the first autocorrelation coefficient. as molecules. Then it is judged whether the weight of the item with the highest weight among the candidate respiratory frequencies is less than the threshold value among all the candidate respiratory frequencies, that is, it is judged that the maximum value of the sum of the respiratory energy ratio and the first autocorrelation coefficient is among all the recorded breaths. Whether the proportion of the sum of the energy ratio and the first autocorrelation coefficient is less than the third threshold.
  • step S1746 use the candidate respiratory frequency as the detection result.
  • the first device detects that the judgment result in step S1745 is yes, and outputs the frequency domain candidate respiratory frequency or the time domain candidate respiratory frequency as respiratory detection information.
  • steps S1744 to S1746 correspond to steps S121 to S124 in FIG. 12 , and for specific content, reference can be made to steps S121 to S124.
  • step S1747 There is no detectable breath in the environment.
  • the first device detects that the judgment result in step S1745 is no, and the first device judges that there is no detectable breathing in the current environment.
  • Step S175 Output the respiratory detection results.
  • the first device outputs a respiration detection result including respiration detection information, as shown in Figures 14 and 15.
  • Figure 18 is a schematic flow chart of another breath detection method provided by an embodiment of the present application.
  • step S1728 in Figure 17 is different from step S1828 in Figure 18.
  • Step S1828 uses the breath sub-signal with a smaller current range as the carrier breath signal.
  • the first device uses the breathing sub-signals with a smaller group distance as the carrier breathing signal.
  • the numerical range of the first signal set is 0.3 (2.8-2.5), which is smaller than the corresponding numerical range of the second signal set, which is 0.7 (4-3.4.1).
  • the first signal set is taken as the target. Signal collection. In this way, the user's calculation amount in subsequent steps S173 and S174 is reduced, and the inventor found that the data of respiratory sub-signals with smaller group distances are more concentrated and the analysis effect is better.
  • Figure 19 is a schematic diagram of the software architecture of the electronic device provided by the embodiment of the present application.
  • the electronic device (such as the first device, the second device or the user device) may also adopt other architectures.
  • the embodiment of the present invention may also take the Harmony system as an example to illustrate the software architecture of the electronic device.
  • Harmony includes four layers, from bottom to top: kernel layer, system basic capability layer, framework layer and application layer.
  • the Harmony system uses a multi-core design, optionally including the Linux kernel, Harmony microkernel and LiteOS. With this design, devices with different device capabilities can choose the appropriate system core.
  • the kernel layer also includes the Kernel Abstract Layer, which provides basic kernel capabilities to other Harmony layers, such as process management, thread management, memory management, File system management, network management, peripheral management, etc.
  • the system basic service layer is the core capability set of the Harmony system, which supports the Harmony system to provide application services through the framework layer in multi-device deployment scenarios.
  • This layer optionally includes the following parts:
  • System basic capability subsystem set Provides basic capabilities for the operation, scheduling, migration and other operations of distributed applications on multiple devices of the Harmony system. It consists of distributed soft bus, distributed data management and file management, distributed task scheduling, and Ark It consists of runtime, distributed security and privacy protection. Among them, the Ark runtime provides C/C++/JavaScript multi-language runtime and basic system class libraries, and also provides static Java programs using the Ark compiler (that is, the parts of the application or framework layer developed using the Java language). Runtime.
  • Basic software service subsystem set Provides public and general software services for the Harmony system, consisting of graphics and images, distributed media, distributed AI, multi-mode input, MSDP&DV, event notification, telephone service, distributed DFX and other subsystems .
  • the basic software service subsystem set can be deployed according to the deployment environment of different device forms, and each subsystem can be tailored according to the functional granularity.
  • Enhanced software service subsystem set Provides differentiated capability-enhanced software services for the Harmony system for different devices, consisting of tablet business software, smart screen business software, vehicle and machine business software, IoT business software and other subsystems.
  • the enhanced software service subsystem set can be tailored according to the deployment environment of different device forms, and can be tailored according to the subsystem granularity, and each subsystem can be tailored according to the functional granularity.
  • Harmony Driver Framework and Hardware Abstraction Adaptation Layer (HAL): are the basis for the openness of the Harmony system hardware ecosystem. They provide hardware capability abstraction for hardware upwards and development frameworks and operating environments for various peripheral drivers downwards.
  • Hardware service subsystem set Provides public and adapted hardware services for the Harmony system, consisting of pan-Sensor, location, power, USB, biometric and other hardware service subsystems.
  • the hardware service subsystem set can be deployed according to the deployment environment of different device forms, and each subsystem can be tailored according to functional granularity.
  • Proprietary hardware service subsystem Provides differentiated hardware services for the Harmony system for different devices, optionally including tablet-specific hardware services, car-specific hardware services, wearable-specific hardware services, and IoT-specific hardware services and other subsystems.
  • the proprietary hardware service subsystem can be tailored according to the subsystem granularity, and each subsystem can be tailored according to the functional granularity.
  • the framework layer provides multi-language user program frameworks and meta-capability frameworks such as Java/C/C++/JavaScript for Harmony system applications, as well as multi-language framework APIs for various software and hardware services that are open to the outside world.
  • the application layer includes system applications and third-party applications, which can include camera, gallery, calendar, call, map, navigation, WLAN, Bluetooth, music, video, short message and other applications.
  • Applications in the Harmony system are built based on AA and FA.
  • Figure 20 is a schematic structural diagram of an embodiment of the second device provided in this application. It can be used to perform the breathing detection methods in the embodiments corresponding to Figures 4, 8, 10, 11, 12, 16, 17, and 18.
  • the second device 101 may include components such as a processor 201 and a memory 202. These components connect and communicate via one or more buses.
  • the processor 201 is the control center of the second device, using various interfaces and lines to connect various parts of the entire base station, by running or executing software programs and/or modules stored in the memory 202, and calling the software programs and/or modules stored in the memory 202. data to perform various functions of the base station and/or process data.
  • the processor 201 may be composed of an integrated circuit (IC for short), for example, it may be composed of a single packaged IC, or it may be composed of multiple packaged ICs connected with the same function or different functions.
  • the processor 201 may be a communication processor (CP for short).
  • the memory 202 can be used to store software programs and modules.
  • the processor 201 executes the software programs stored in the memory 202. programs and modules to execute various functional applications of the second device and implement data processing.
  • the memory 202 may include volatile memory, such as nonvolatile dynamic random access memory (NVRAM), phase change random access memory (phase change RAM, PRAM) , magnetoresistive random access memory (mageto-resistive RAM, MRAM), etc., and may also include non-volatile memory, such as at least one disk storage device, electronically erasable programmable read-only memory (Electrically erasable programmable read- only memory (EEPROM for short), flash memory devices, such as NOR flash memory or NAND flash memory.
  • NVRAM nonvolatile dynamic random access memory
  • PRAM phase change random access memory
  • MRAM magnetoresistive random access memory
  • EEPROM electronically erasable programmable read-only memory
  • flash memory devices such as NOR flash memory or NAND flash memory.
  • FIG 21 is a schematic structural diagram of an embodiment of the first device provided in this application. It can be used to perform the breath detection method in the corresponding embodiment of the present application.
  • the first device 102 may include: a processor 211, a memory 212 and other components. In addition, these components can also be connected and communicated through one or more buses, etc.
  • the processor 211 is the control center of the first device, using various interfaces and lines to connect various parts of the entire user device, by running or executing software programs and/or modules stored in the memory 212, and calling software programs stored in the memory 212. Data to perform various functions of the terminal and/or to process data.
  • the processor 211 may be composed of an integrated circuit (Integrated Circuit, IC for short), for example, it may be composed of a single packaged IC, or it may be composed of multiple packaged ICs connected with the same function or different functions.
  • the processor 211 may be a CP.
  • the present application also provides a computer storage medium, wherein the computer storage medium can store a program, and when executed, the program can include some or all of the steps in each embodiment of the breath detection method provided by the present application.
  • the storage media can be disks, optical disks, read-only memory (ROM) or random access memory (RAM), etc.
  • non-volatile memory may be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory.
  • non-volatile memory can be read-only memory (ROM), programmable ROM (PROM), erasable programmable read-only memory (erasable PROM, EPROM), electrically removable memory. Erase electrically programmable read-only memory (EPROM, EEPROM) or flash memory.
  • Volatile memory can be random access memory (RAM), which is used as an external cache.
  • RAM static random access memory
  • DRAM dynamic random access memory
  • SDRAM synchronous dynamic random access memory
  • double data rate SDRAM double data rate SDRAM
  • DDR SDRAM double data rate SDRAM
  • ESDRAM enhanced synchronous dynamic random access memory
  • synchronous link dynamic random access memory direct memory bus random access memory
  • direct ram bus RAM direct ram bus RAM
  • FIG 22 is a schematic structural diagram of an embodiment of user equipment provided by this application. It can be used to execute the breath detection method in the embodiment of the present application and present corresponding breath detection information.
  • the user equipment 104 may include components such as a processor 221 , a memory 222 , and a display screen 223 .
  • these components can also be connected and communicated through one or more buses, etc.
  • the processor 221 is the control center of the first device, using various interfaces and lines to connect various parts of the entire user device, by running or executing software programs and/or modules stored in the memory 222, and calling software programs stored in the memory 222. Data to perform various functions of the terminal and/or to process data.
  • the processor 221 may be composed of an integrated circuit (IC for short), for example, it may be composed of a single packaged IC, or it may be composed of multiple packaged ICs connected with the same function or different functions.
  • the processor 221 may be a CP.
  • the display screen 223 is used to display corresponding environmental change information, such as the breathing detection information shown in Figures 14 and 15.
  • the present application also provides a computer storage medium, wherein the computer storage medium can store a program, and when executed, the program can include some or all of the steps in each embodiment of the breath detection method provided by the present application.
  • the storage media can be disks, optical disks, read-only memory (ROM) or random access memory (RAM), etc.
  • non-volatile memory may be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory.
  • non-volatile memory can be read-only memory (ROM), programmable ROM (PROM), erasable programmable read-only memory (erasable PROM, EPROM), electrically removable memory. Erase electrically programmable read-only memory (EPROM, EEPROM) or flash memory.
  • Volatile memory can be random access memory (RAM), which is used as an external cache.
  • RAM static random access memory
  • DRAM dynamic random access memory
  • SDRAM synchronous dynamic random access memory
  • double data rate SDRAM double data rate SDRAM
  • DDR SDRAM double data rate SDRAM
  • ESDRAM enhanced synchronous dynamic random access memory
  • synchronous link dynamic random access memory direct memory bus random access memory
  • direct ram bus RAM direct ram bus RAM
  • the correspondence between A and B can be understood as the association between A and B, or the association between A and B.
  • the disclosed systems, devices and methods can be implemented in other ways.
  • the device embodiments described above are only illustrative.
  • the division of units is only a logical function division. In actual implementation, there may be other division methods.
  • multiple units or components may be combined or integrated. to another system, or some features can be ignored, or not implemented.
  • the coupling or direct coupling or communication connection between each other shown or discussed may be through some interfaces, and the indirect coupling or communication connection of the devices or units may be in electrical, mechanical or other forms.
  • a unit described as a separate component may or may not be physically separate.
  • a component shown as a unit may or may not be a physical unit, that is, it may be located in one place, or it may be distributed to multiple network units. can be based on actual It is actually necessary to select some or all of the units to achieve the purpose of this embodiment.
  • each functional unit in each embodiment of the present application can be integrated into one processing unit, each unit can exist physically alone, or two or more units can be integrated into one unit.
  • Functions may be stored in a computer-readable storage medium when implemented in the form of software functional units and sold or used as independent products.
  • the technical solution of the present application is essentially or the part that contributes to the existing technology or the part of the technical solution can be embodied in the form of a software product.
  • the computer software product is stored in a storage medium, including Several instructions are used to cause a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods of various embodiments of the present application.
  • the aforementioned storage media include: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), magnetic disk or optical disk and other media that can store program code. .

Landscapes

  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Physics & Mathematics (AREA)
  • Public Health (AREA)
  • Medical Informatics (AREA)
  • Pulmonology (AREA)
  • Animal Behavior & Ethology (AREA)
  • Business, Economics & Management (AREA)
  • Quality & Reliability (AREA)
  • Biophysics (AREA)
  • Molecular Biology (AREA)
  • Surgery (AREA)
  • Electromagnetism (AREA)
  • Physiology (AREA)
  • Pathology (AREA)
  • Veterinary Medicine (AREA)
  • Heart & Thoracic Surgery (AREA)
  • General Business, Economics & Management (AREA)
  • Epidemiology (AREA)
  • Primary Health Care (AREA)
  • Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)

Abstract

本申请提供了呼吸检测方法、电子设备、存储介质及程序产品。该方法包括第一设备接收至少一个第二设备的Wi-Fi信号；所述第一设备根据所述至少一个第二设备的Wi-Fi信号的信道状态信息的相位差或幅度确定N个信号集合，所述N个信号集合内的所述相位差或所述幅度分别属于N个不同的数值范围，其中N为大于或等于2的整数；所述第一设备根据所述N个信号集合确定环境变化信息，所述环境变化信息包括呼吸检测信息。可以基于该分层的信道状态信息进行呼吸检测，得到准确的呼吸检测信息。

Description

呼吸检测方法、电子设备、存储介质及程序产品
本申请要求于2022年6月30日提交中国专利局、申请号为202210771443.5，发明名称为“呼吸检测方法、电子设备、存储介质及程序产品”的中国专利的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及通信技术领域，尤其涉及了呼吸检测方法、电子设备、存储介质及程序产品。
背景技术
长期的呼吸频率监测具有非常高的医疗指导价值，使用电子设备对呼吸频率进行长期的监测，对一些疾病的提前发现有非常大的帮助。当前呼吸频率检测主要有两种方式：1)使用专业的医疗设备进行短期的检测；2)使用穿戴设备进行呼吸监测。为了测量睡眠呼吸频率，需要购买苹果手表。而接触式的方法都存在一定的侵扰性。用户不易长期坚持使用监测。
近年来，人们开始研究使用无线网路(Wi-Fi)感知技术进行非接触式呼吸频率检测。Wi-Fi设备能够以信道状态信息(Channel State Information,CSI)的形式获得子载波上的幅度和相位信息，信道状态信息描述了信号由信号发送端沿多径到达信号接收端所经历的衰减和相位旋转，任一信号传播路径的改变都会影响信号接收端获得的信道状态信息，如人体呼吸导致的胸腔起伏会使得信道状态信息的测量值呈现周期的变化，这为利用信道状态信息对人体呼吸进行检测提供了可能。
发明人在实施本申请实施例过程中发现，从Wi-Fi信号中提取出的信道状态信息存在噪声，如提取出的信道状态信息存在分层，基于该分层的信道状态信息无法准确检测呼吸，甚至无法实现呼吸检测。
发明内容
有鉴于此，有必要提供呼吸检测方法、电子设备、存储介质及程序产品，可以处理存在分层的信道状态信息，并可以基于该分层的信道状态信息进行呼吸检测，得到准确的呼吸检测信息。
第一方面，本申请实施例提供一种呼吸检测方法，该方法包括：第一设备接收Wi-Fi信号的信道状态信息，第一设备根据信道状态信息的相位差或幅度确定N个信号集合，N个信号集合内的相位差或幅度分别属于N个不同的数值范围，其中N为大于或等于2的整数，第一设备根据N个信号集合确定环境变化信息，环境变化信息包括呼吸检测信息。
其中，每个信号集合中的数据包括信道状态信息的相位差或幅度。若根据信道状态信息的相位差确定N个信号集合，则该N个信号集合内的相位差分别属于N个不同的数值范围。若根据信道状态信息的幅度确定N个信号集合，则该N个信号集合内的幅度分别属于N个不同的数值范围。
在本申请实施例中，根据信道状态信息的相位差或幅度确定N个信号集合，用以在处理存在分层的信道状态信息时，将该分层的信道状态信息根据信道状态信息的相位差或幅度划分成N个信号集合，提取出每个信号集合，基于该N个信号集合可以进行呼吸检测，实现基于该分层的信道状态信息进行呼吸检测，得到准确的呼吸检测信息。
优选地，根据信道状态信息的相位差或幅度确定N个信号集合包括：确定每一子载波的信道状态信息的相位差；将相位差划分为N个信号集合；或，确定每一子载波的信道状态信息的幅度；将幅度划分为一个信号集合，得到N个信号集合。
优选地，根据N个信号集合确定环境变化信息包括：根据N个信号集合得到目标信号集合；根据目标信号集合得到频域候选呼吸频率和与频域候选呼吸频率对应的呼吸能量比；根据目标信号集合得到时域候选呼吸频率和与时域候选呼吸频率对应的第一自相关系数；根据频域候选呼吸频率、呼吸能量比、时域候选呼吸频率以及第一自相关系数输出环境变化信息。
在本申请实施例中，通过将各个子载波分层的信道状态信息的子信号提取出来，得到各个子载波的目标信号集合。并对各个子载波的目标信号集合进行快速傅里叶变换以及自相关函数计算，以同时获得该子载波的频域候选呼吸频率、呼吸能量比、时域候选呼吸频率以及第一自相关系数。通过对该子载波的频域候选呼吸频率、呼吸能量比、时域候选呼吸频率以及第一自相关系数同时联合判断，判断该子载波携带感知对象呼吸信息的可能性，以此验证频域候选呼吸频率或时域候选呼吸频率是否为感知对象的呼吸频率，双重验证保证呼吸检测信息准确性。
优选地，根据频域候选呼吸频率、呼吸能量比、时域候选呼吸频率以及第一自相关系数输出环境变化信息包括：当检测到呼吸能量比大于对应的第一阈值且第一自相关系数大于对应的第二阈值时，将频域候选呼吸频率或时域候选呼吸频率作为呼吸检测信息输出。
优选地，根据频域候选呼吸频率、呼吸能量比、时域候选呼吸频率以及第一自相关系数输出环境变化信息包括：当检测到频域候选呼吸频率与时域候选呼吸频率相近，且呼吸能量比和第一自相关系数中一个或多个大于对应的阈值时，将频域候选呼吸频率或时域候选呼吸频率作为呼吸检测信息输出。
优选地，当检测到频域候选呼吸频率与时域候选呼吸频率相近，且呼吸能量比和第一自相关系数中一个或多个大于对应的阈值时，将频域候选呼吸频率或时域候选呼吸频率作为呼吸检测信息输出包括：当检测到频域候选呼吸频率与时域候选呼吸频率相近，且呼吸能量比和第一自相关系数中一个或多个大于对应的阈值时，记录呼吸能量比与第一自相关系数之和；将所记录的所有的呼吸能量比与第一自相关系数之和加和作为分母；将所记录的呼吸能量比与第一自相关系数之和的最大值作为分子；检测到分子与分母的比值小于第三阈值时，将频域候选呼吸频率或时域候选呼吸频率作为呼吸检测信息输出。
优选地，对N个信号集合进行筛选，得到目标信号集合包括：针对每一子载波，删除N个信号集合中相位差或幅度的数量少于预设阈值的信号集合，得到剩下的信号集合；根据剩下的信号集合得到目标信号集合。
优选地，根据剩下的信号集合得到目标信号集合包括：对每一剩下的信号集合进行如下操作中的一种或多种，将最终剩下的信号集合作为目标信号集合：判断剩下的信号集合内是否缺少预设时间段内的相位差或幅度；若是，删除剩下的信号集合；或，判断剩下的信号集合每隔预设时间间隔内第一缺失量与第二缺失量的差是否高于预设缺失阈值；其中，第一缺失量对应预设时间间隔内相位差的缺失量或幅度的缺失量的最高值，第二缺失量对应预设时间间隔内相位差的缺失量或幅度的缺失量的最低值；若是，删除剩下的信号集合。
优选地，对目标信号集合进行频域上分析处理，得到频域候选呼吸频率和与频域候选呼吸频率对应的呼吸能量比包括：将目标信号集合进行快速傅里叶变换，得到与目标信号集合对应的频域信号；根据呼吸频段与频域信号得到频域候选呼吸频率和与频域候选呼吸频率对应的呼吸能量比。
优选地，根据呼吸频段与频域信号得到频域候选呼吸频率和频域候选呼吸频率对应的呼吸能量比包括获取频域信号中频率位于呼吸频段的子频域信号；将子频域信号的幅值总和与频域信号 的幅值总和的比值作为呼吸能量比；将子频域信号中最大幅值所对应的频率作为频域候选呼吸频率。
优选地，对目标信号集合进行时域上分析处理，得到时域候选呼吸频率和与时域候选呼吸频率对应的第一自相关系数包括：对目标信号集合进行自相关计算，得到目标信号集合对应的自相关系数；根据呼吸频段与自相关系数得到时域候选呼吸频率和与时域候选呼吸频率对应的第一自相关系数。
优选地，根据呼吸频段与自相关系数得到时域候选呼吸频率和时域候选呼吸频率对应的第一自相关系数包括：将呼吸频段内最大自相关系数所对应的频率作为时域候选呼吸频率；将时域候选呼吸频率对应的自相关系数作为第一自相关系数。
第二方面，本申请实施例提供一种电子设备，电子设备包括至少一个处理器、存储器，其中存储器用于存储指令，处理器用于执行指令实现如上任一项的方法。
在一种可能实现方式中，电子设备包括终端设备或无线访问节点。
在一种可能实现方式中，电子设备用于根据呼吸检测信息呈界面或输出语音。
第三方面，本申请实施例提供一种计算机可读存储介质，计算机可读存储介质存储有程序，程序使得电子设备执行如上任一项的方法。
第四方面，本申请实施例提供一种计算机程序产品，计算机程序产品包括计算机可读指令，当计算机可读指令被一个或多个处理器执行时实现如上任一项的方法。
可以理解地，上述第二方面提供的电子设备、第三方面提供的计算机可读存储介质以及第四方面提供的计算机程序产品，与上述第一方面或第二方面提供的方法对应，因此，其所能达到的有益效果或各种实现方式可参考上文，此处不再赘述。
附图说明
图1为本申请提供的呼吸检测方法的应用场景示意图；
图2A为本申请提供的一种信道状态信息的示意图；
图2B为本申请提供的另一种信道状态信息的示意图；
图2C为对图2B中的信道状态信息进行处理的结果示意图；
图3A为本申请提供的呼吸检测方法对图2B的信道状态信息进行信号提取后的结果示意图；
图3B为本申请提供的呼吸检测方法对图3A的信道状态信息进行快速傅里叶变换和自相关计算后的结果示意图；
图4为本申请实施例提供的一种呼吸检测方法流程示意图；
图5A为本申请提供的第6号子载波、第16号子载波以及第26号子载波的示意图；
图5B为将图5A中第6号子载波的子信号提取出来的示意图；
图6为本申请提供的对子载波的相位差进行计数统得到的直方图；
图7为所有子载波的目标信号集合进行快速傅里叶变换得到的结果示意图；
图8为本申请提供的获取频域候选呼吸频率和呼吸能量比的方法流程示意图；
图9为对一个子载波的目标信号集合进行自相关计算后得到的结果示意图；
图10为本申请提供的获取时域候选呼吸频率和第一自相关系数的方法流程示意图；
图11为本申请提供的一种根据频域候选呼吸频率、呼吸能量比、时域候选呼吸频率以及第 一自相关系数输出呼吸检测信息方法流程示意图；
图12为本申请提供的另一种根据频域候选呼吸频率、呼吸能量比、时域候选呼吸频率以及第一自相关系数输出呼吸检测信息方法流程示意图；
图13为本申请实施例提供的用户设备的主界面示意图；
图14为本申请实施例提供的用户设备的呼吸检测信息界面示意图；
图15为本申请实施例提供的用户设备的另一呼吸检测信息界面示意图；
图16为本申请实施例提供的另一种呼吸检测方法流程示意图；
图17为本申请实施例提供的另一种呼吸检测方法流程示意图；
图18为本申请实施例提供的另一种呼吸检测方法流程示意图；
图19为本申请实施例提供的电子设备的软件架构示意图；
图20为本申请实施例提供的第二设备的一个结构示意图；
图21为本申请实施例提供的第一设备的一个结构示意图；
图22为本申请实施例提供的用户设备的一个结构示意图。
具体实施方式
为使本申请的目的、技术方案和优点更加清楚，下面将结合附图对本申请实施方式作进一步地详细描述。
在对本申请实施例进行详细的解释说明之前，先对本申请实施例涉及的应用场景予以介绍。
请一并参阅图1，图1为本申请实施例提供一种应用场景的示意图。该应用场景包括第二设备101、第一设备102和感知对象103。图1以应用场景包括一个第二设备101、一个第一设备102和一个感知对象103为例，实际应用中可以包括两个或两个以上的第二设备101，两个或两个以上的第一设备102，图1所示的设备数量和形态、对象数量和形态举例并不构成对本申请实施例的限定。
其中，第二设备101可以包括一根或一根以上的发送天线。第一设备102可以包括两根或两根以上的接收天线，本申请对此不作具体限定。第二设备101的一个发送天线和第一设备102的一个接收天线为一对天线对，该天线对对应一条路径，从频域出发的话，一条路径包括多条子载波，每条子载波是一个频域信道；若从时域出发，一条路径包括多条径，有直线传播的径，也有其他的发射径(即多径传播)。
第二设备101向第一设备102发送Wi-Fi信号，该Wi-Fi信号经过感知对象103后到达第一设备102。第一设备102基于Wi-Fi信号得到信道状态信息，并根据信道状态信息确定第一设备102所处环境的环境变化信息，包括对感知对象103进行感知的感知结果。该感知结果可以用于指示感知对象103的呼吸状态，如感知对象103是否呼吸或感知对象103的呼吸频率等。
在一些实施例中，应用场景中还可以包括用户设备104。用户设备104用于接收第一设备102传输的环境变化信息，并将环境变化信息通知用户。其中，通知方式可以为界面显示或语音播放等。用户设备104与第一设备102通信，第一设备102与用户设备104的连接方式可能为有线连接(网线、PLC等有线连接方式)或无线连接(Wi-Fi、蓝牙等无线传输方式，本申请对通信方式不限定。其中，环境变化信息可以为感知对象103的呼吸检测信息，如图14和图15所示。
在一些实施例中，应用场景中的用户设备104与第二设备101为同一设备，也即第一设备102可以将感知到的环境变化信息发送至第二设备101(用户设备104)。在另一些实施例中，应用场景中的用户设备104与第一设备102为同一设备，第一设备102将计算出的环境变化信息通知用户。 也即，用户设备104可以执行本申请实施例提供的呼吸检测方法，在用户携带(如手持)用户设备104移动至布置有第一设备102或第二设备101的新环境中时，用户设备104依然可以与该环境中的第二设备101或第一设备102进行通信。用户设备104可以根据接收到的由第二设备101发送的Wi-Fi信号进行呼吸检测，感知所处环境的环境变化信息，或用户设备104可以接收新环境中第一设备102传输的环境变化信息。
需要说明的是，该感知对象103和第一设备102均处于第二设备101的覆盖范围内。
在一些实施例中，应用场景中还可以包括云端服务器105。云端服务器105可以与第一设备102通信，用于接收并存储第一设备102传输的Wi-Fi信号的信道状态信息或环境变化信息。云端服务器105可以与用户设备104通信，用于将Wi-Fi信号的信道状态信息或环境变化信息传输给用户设备104。
在一些实施例中，第一设备102、云端服务器105或用户设备104接收到Wi-Fi信号的信道状态信息或环境变化信息后，第一设备102、云端服务器105或用户设备104可以对接收到的信息(Wi-Fi信号的信道状态信息或环境变化信息)进行分析处理，得到感知对象103的呼吸检测信息，用户设备104从云端服务器105或第一设备102处获得呼吸检测信息或用户设备104计算出呼吸检测信息后，用户设备104将呼吸检测信息呈现出来，如图14和图15所示。
在另一些实施例中，第一设备102、云端服务器105或用户设备104可以对接收到的信息(Wi-Fi信号的信道状态信息或环境变化信息)进行融合处理，如根据接收到的信息以及其他融合信息进行融合计算，输出如作息检测、久坐提醒、房间内存在呼吸、呼吸频率等分析结果。用户设备104获得该分析结果后将该分析结果呈现出来，如图14和图15所示。
在另一些实施例中，本申请实施例的呼吸检测方法可以被分割成一个或多个模块，该一个或多个模块可以是能够完成特定功能的一系列计算机程序指令段，所述指令段用于描述本申请实施例的呼吸检测方法的执行过程。该一个或多个模块可存储第一设备102和/或用户设备104和/或云端服务105中。也即，本申请实施例中呼吸检测方法中某些步骤可以由第一设备102执行，某些步骤可以由用户设备104执行，某些步骤可以由云端服务器105执行，本申请对此不作具体限定。
第二设备101是一种用于发射Wi-Fi信号的实体。第二设备101例如可以是蓝牙眼镜、蓝牙手表、蓝牙耳机、蓝牙音箱、蓝牙智慧屏、平板电脑、蓝牙台灯、蓝牙门锁、蓝牙插座、蓝牙电子秤、手机、穿戴式设备、平板、带无线收发功能的电脑、路由器、无线接入点AP、LTE室内信号发送基站、虚拟现实(virtual reality，VR)终端设备、增强现实(augmented reality，AR)终端设备、无人驾驶(self-driving)中的无线终端等。
第一设备102是一种用于接收Wi-Fi信号的实体。第一设备102例如可以是无线路由器或无线接入点AP、LTE室内信号发送基站、手机、穿戴式设备、平板、带无线收发功能的电脑等。
用户设备104是一种用于将感知结果通知用户的终端设备，用户设备104例如可以是手机、穿戴式设备、平板、带无线收发功能的电脑、虚拟现实(virtual reality，VR)终端设备、增强现实(augmented reality，AR)终端设等。
其中，上述应用场景可以为上述设备之间可以通过Wi-Fi进行通信的环境，如居家环境、办公环境等。
下面以图1中应用场景为居家环境为例，在这居家环境中第二设备101发送Wi-Fi信号，第一设备102接收Wi-Fi信号，根据Wi-Fi信号的信道状态信息执行本申请实施例提供的呼吸检测方 法，获得居家环境中感知对象103的呼吸检测信息。或，第一设备102接收Wi-Fi信号，第一设备102将Wi-Fi信号的信道状态信息发送至用户设备104或云端服务105，由用户设备104或云端服务105执行本申请实施例提供的呼吸检测方法，根据Wi-Fi信号的信道状态信息获得居家环境中感知对象103的呼吸检测信息。若由第一设备102执行本申请实施例提供的呼吸检测方法，用户设备104从第一设备102处获得感知对象103的呼吸检测信息。若由云端服务器105执行本申请实施例提供的呼吸检测方法，用户设备104从云端服务器105处获得感知对象103的呼吸检测信息。
具体地，如上述居家环境，用户独自居住，在用户睡眠时，第一设备102可以监测居家环境中感知对象103(即用户)睡眠过程中的环境变化信息，进而根据环境变化信息得到呼吸检测信息。第一设备102、第二设备101、云端服务器105或用户设备104通过所获得的呼吸检测信息可以帮助用户获知在睡眠过程中是否存在呼吸暂停障碍，由此可知避免由于患有睡眠呼吸障碍的用户不易自主察觉发生呼吸暂停的时间和轻重程度，因而可能会导致该症状无法得到及时的治疗和改善的问题。通过本申请实施例的呼吸检测方法所获得的呼吸检测信息准确性高，进而提高了用户进行人体健康状况检测的准确性和成功率。
需要说明的是，图1中感知对象103为一个人物仅用于举例并不构成对本申请实施例的限定。感知对象103还可以为动物等，本申请实施例对第二设备101、第一设备102、用户设备104所采用的具体技术和具体设备形态以及感知对象103不做限定。
可以理解的是，本申请实施例描述的应用场景是为了更加清楚的说明本申请实施例的技术方案，并不构成对本申请实施例提供的技术方案的限定，本领域技术人员可知，随着系统架构的演变和新业务场景的出现，本申请实施例提供的技术方案对于类似的技术问题，同样适用。
发明人发现目前的Wi-Fi呼吸检测方法仅可处理无噪声或轻微噪声信道状态信息，如图2A所示，信道状态信息不存在分层。发明人还发现从Wi-Fi信号中提取出的信道状态信息普遍存在噪声，导致信道状态信息存在噪声的原因包括但不限于：为了通信效率最大化，设备(如路由器)启用自动增益控制。而自动增益控制会影响信道状态信息，使得信道状态信息存在噪声。存在噪声的信道状态信息存在分层，信道状态信息包括多层子信号。如图2B所示，信道状态信息包括四层子信号，该四层子信号分别对应图2B的曲线S1、S2、S3以及S4。每层子信号之间存在间距，每层子信号随着时间具有一定连续性。如曲线S1，在0-3秒之间有信号，6-8秒之间有信号，11-15秒之间有信号。
请一并参阅图2C，图2C中曲线S5指示以现有的呼吸检测方法对图2B的信道状态信息进行快速傅里叶变换(Fast Fourier Transform,FFT)后，各个频率对应的幅度均为0。图2C中曲线S6指示以现有的呼吸检测方法对图2B中的信道状态信息进行自相关函数(Autocorrelation Function，ACF)计算后，呼吸频段所对应的自相关系数为负数。由此可知，目前呼吸检测方法无法在图2B中分层的信道状态信息中获得频率信息，导致无法基于该多层信道状态信息进行呼吸检测，距离商用落地差距较大。
需要说明的是，现有的呼吸检测方法均是单独对信道状态信息进行快速傅里叶变换，或是，单独对信道状态信息进行自相关函数计算。图2C并非指示现有的呼吸检测方法对信道状态信息进行快速傅里叶变换也进行自相关函数计算。
基于此，本申请实施例提供的呼吸检测方法，可以用于处理无噪声或轻微噪声信道状态信息，还可以用于处理存在分层的信道状态信息，并可以基于该分层的信道状态信息进行呼吸检测，得到 准确的呼吸检测信息。
请一并参阅图3A，图3A展示了使用本申请实施例提供的呼吸检测方法对图2B的信道状态信息进行处理后得到的信道状态信息，即为提取出的子信号。处理后所得到的信道状态信息数据更均匀，且数据分布更清晰。图3B展示了使用本申请实施例提供的呼吸检测方法对图3A的信道状态信息进行快速傅里叶变换和自相关计算后得到的结果。使用本申请实施例提供的呼吸检测方法提取出子信号后，对子信号进行快速傅里叶变换后，在呼吸频段可以获得较高的幅度特性值。对子信号进行自相关计算后，在呼吸频段获得较高的自相关系数。也就是说，基于本申请实施例提供的呼吸检测方法可以在图2B中分层的信道状态信息中获得频率信息，进而可以基于该多层信道状态信息进行呼吸检测。
接下来对本申请实施例提供的呼吸检测方法进行介绍。本申请实施例提供的一种呼吸检测方法可以应用于第一设备102、用户设备104和云端服务器105，本申请对此不作具体限定。
本申请中使用了流程图用来说明根据本申请的实施例的装置所执行的操作。应当理解的是，前面或下面操作不一定按照顺序来精确地执行。相反，根据需要，可以按照倒序或同时处理各种步骤。同时，也可以将其他操作添加到这些过程中，或从这些过程移除某一步或数步操作。
请一并参阅图4，图4为本申请实施例提供的一种呼吸检测方法流程示意图。下述以该呼吸检测方法应用图1中所示的第一设备为例。
步骤S41：获取每一子载波的信道状态信息。
在本申请实施例中，第二设备以特定速率(如每秒200Hz)通过M个子载波发出数据包，第一设备的两根天线同时接收多个子载波上的数据包，并从中测量出由每个子载波传输的每个数据包的信道状态信息。在第一设备接收到来自第二设备的报文之后，第一设备提取报文中的前导部分，将接收到的前导部分除以其本地端存储的已知序列，得到相应的信道状态信息。其中，报文中的前导部分是第二设备和第一设备双方均已知的序列。
其中，M(也即子载波的数量)与Wi-Fi协议、带宽及使用的天线数量相关，例如802.11a/g在20MHz模式下子载波数量为52个，802.11n在20MHz模式下有56个子载波。本申请实施例对子载波的数量不作具体限定。
示例性的，报文可以是携带特殊的训练符号的数据报文，也可以是空数据报文(null data packet，NDP)，还可以是物理层协议数据单元(physical layer protocol date units，PPDU)。
在一些实施例中，第一设备以每个报文为一次采样，以特定采样频率(如50Hz)进行采样。在另一些实施例中，第一设备每接收到一个报文进行一次采样，随着时间不断采样，如采集15秒，得到以时间为顺序的一段信道状态信息序列。
在本申请实施例中，信道状态信息用于反映当前无线信道的状况。在802.11n协议中，针对每一个正交频分复用(orthogonal frequency division multiplexing，OFDM)子载波组进行测量，获取该OFDM子载波组对应的CSI矩阵。CSI矩阵的行数为发送天线数，CSI矩阵的列数为接收天线数。每个CSI矩阵的元素值表征一个子载波上的信道响应。
第一设备包括K根接收天线，第二设备包括J根发送天线，子载波的数量为M，则对于每个通信链路，CSI矩阵具体展开为：
H＝[h1,h2,…hi]，i∈[1,M]
其中，H为CSI矩阵，hi表示第i子载波的信道状态信息，M为子载波的数量。
因此要分析的每个数据包索引最终都会获得K(发送天线的数量)×J(接收天线的数量)×M(子载波的数量)个信道状态信息。子载波是一个包含实部与虚部的复数，实部对应信道状态信息的幅度，实部对应信道状态信息的相位，第i子载波的信道状态信息hi的具体计算公式为：
hi＝|h|ej∠θ
其中，|h|表示子载波的幅度，θ表示子载波的相位。
步骤S42：针对每一子载波，对信道状态信息的相位差或幅度进行处理，得到N个信号集合，N个信号集合内的相位差或幅度分别属于N个不同的数值范围，其中N为大于或等于2的整数。
其中，信道状态信息的所述相位差或所述幅度包括信道状态信息的幅度、相位以及相位差。
在步骤S41中针对每一子载波，随着时间不断采样，获得每一子载波的信道状态信息序列。第一设备在步骤S41中所获得的信道状态信息存在分层，为此在步骤S42中，第一设备将该信道状态信息中的子信号提取出来，得到对应的信号集合。基于子载波是复数，针对每一子载波，根据时间顺序，提取该子载波的信道状态信息中的相位或幅度，得到该子载波信道状态信息的相位序列或幅度序列，可以以相位序列或幅度序列表示子信号。第一设备可以依据相位序列或幅度序列实现对信道状态信息中子信号的提取。
以第二设备包括一根发送天线，第一设备包括两根接收天线(天线A和天线B)，采样时间为15秒，对第6号子载波为例进行说明。
天线A所接收到的第6号子载波的相位序列为：
Q1＝[a1,a2,…ai]，i∈[1,T]
其中，Q1为天线A所接收到的第6号子载波的相位序列，ai表示15秒内第i个采样点采集到的第6号子载波所对应的相位，T为15秒内的采样点数量。
天线B所接收到的第6号子载波的相位序列为：
Q2＝[b1,b2,…bi]，i∈[1,T]
其中，Q2为天线B所接收到的第6号子载波的相位序列，bi表示15秒内第i个采样点采集到的第6号子载波所对应的相位，T为15秒内的采样点数量。
计算第6号子载波的相位差，将天线A所接收到的第6号子载波的相位序列与天线B所接收到的第6号子载波的相位序列作差，即将同一时刻(或采样点)的两个相位作差，得到相位差序列：
C＝A-B＝[c1,c2,…ci]＝[a1-b1,a2-b2,…ai-bi]，i∈[1,T]
其中，C为第6号子载波的相位差序列，ci表示15秒内第i个采样点采集到的第6号子载波所对应的相位差，T为15秒内的采样点数量。
在本申请实施例中，单个子载波包含呼吸信息的信道状态信息中可能包括多个子信号，其中子信号可以理解为噪声严重(分层)的信道状态信息中包含呼吸频率信息的最小子集，导致信道状态信息存在分层的原因可以是受自动增益控制影响。发明人发现子信号也同样具备反映当前无线信道的状况的能力。对于存在分层的信道状态信息，可以将各个层的子信号提取出来，以此可以实现对分层的信道状态信息的处理。
以子载波的数量为52为例，请一并参阅图5A，分别展示了采样时间为15秒，52个子载波中的第6号子载波、第16号子载波以及第26号子载波对应的相位差序列。由图5A可以看到，每一子载波的信道状态信息均包括两层，该两层彼此间隔，每层代表一子信号。以图5A中的第6 号子载波进行说明，第6号子载波包括两条曲线S7和S8，曲线S7和S8彼此间隔，曲线S7代表一子信号，曲线S8代表另一子信号。
在本申请实施例中，针对每一子载波，将其信道状态信息中所包括的多个子信号分别提取出来，得到多个信号集合。若子载波的信道状态信息中包括N个子信号，则可以提取出N个信号集合。换句话说，从图5A所示的子载波的相位差序列，子载波有N个分层，则可以提取出N个信号集合。请一并参阅图5B，图5B展示了将图5A中第6号子载波的子信号提取出来，如提取出曲线S7，获取曲线S7上的各个相位差序列，得到一信号集合。相应地，提取出曲线S8，获取曲线S8上的各个相位差序列，得到另一信号集合。该两个信号集合的采样时间均为15秒。以第6号子载波的子信号的一次上下起伏代表感知对象的一次呼吸。如图5B所示，窗口51截取出的是第6号子载波的子信号一个呼吸周期，该呼吸周期可以是先呼气再吸气，也可以是先吸气再呼气。也即窗口51中的曲线51a对应呼气时的相位差序列，则窗口51中的曲线51b对应吸气时的相位差序列。或者，窗口51中的曲线51a对应吸气时的相位差序列，则窗口51中的曲线51b对应呼气时的相位差序列。
在本申请实施例中，针对每一子载波，提取信道状态信息的所述相位差或所述幅度后，可以得到以时间为顺序的所述相位差或所述幅度序列，如上述的相位差序列。将所述相位差或所述幅度序列中数值相近的所述相位差或所述幅度划分为同一信号集合，以此通过不同信号集合划分出数值相远的所述相位差或所述幅度。以图5A进行说明，曲线S7上相位差均与该曲线S7上其他相位差在数值上更接近，而曲线S7上的相位差与曲线S8上的相位差在数值上更远，由此可以将曲线S7上的相位差划分至同一信号集合，曲线S7上的相位差属于同一个数值范围，如为(-4.5至-4.6]度。将曲线S8上的相位差划分至另一信号集合，曲线S8上的相位差属于同一个数值范围，如为(-5至-4.9]度。
可以理解，不同信号集合其对应的数值范围不同。也就是说，对于每个子载波来说，若其包括N个子信号，则经过步骤S42可以得到N个信号集合，其对应有N个不同的数值范围。
在一些实施例中，可以将采样时间内所有采样点对应的相位差进行计数统计，统计出在每个相位差对应的采样点数量，然后将数值相近的数值合并为一个集合。示例性地，以图2中的B为例，对15秒内所有采样点对应的相位差进行计数统计，将数值相近的相位差合并为一组，可以得到如图6所示的直方图。如图6所示，可以得到四个信号集合(分组)，其中第一个信号集合对应的数值范围为(2.5-2.8]度。第二个信号集合对应的数值范围为(3.4-4.1]度。第三个信号集合对应的数值范围为(4.5-4.9]度。第四个信号集合对应的数值范围为(5.1-5.4]度。
同理，若根据时间顺序，提取该子载波的信道状态信息中的幅度，天线A可以得到52个子载波信道状态信息幅度序列，其中每个子载波的信道状态信息包括多个子信号。对于每一子载波，将数值上相近的幅度划分至同一信号集合。相应地，天线B可以得到52个子载波信道状态信息幅度序列，对天线B所获得的每个子载波进行同样操作，提取出每个子载波的信号集合，在此不再赘述。
在本申请实施例中，可以依据相位差、或依据幅度，还可以同时依据相位差和幅度，将每个子载波的信道状态信息的多个子信号分别提取出来，本申请对此不作具体限定。
可以理解，在本申请实施例提供的将存在噪声的信道状态信息中的子信号提取出来的思想下，还可以有其他方式将信道状态信息中所包括的多个子信号分别提取出来以得到多个信号集 合，本申请实施例对此不作具体限定。
步骤S43：对N个信号集合进行筛选，得到目标信号集合。
在本申请实施例中，针对每一子载波，第一设备对该子载波中的N个信号集合进行筛选，以获得可以代表该子载波的目标信号集合，也即可以根据该目标信号集合获得该子载波所携带的感知对象的呼吸频率信息，该目标信号集合的数据分布均匀。
在一些实施例中，步骤S43可以实现为：针对每一子载波，删除N个信号集合中所述相位差或所述幅度的数量少于预设阈值的信号集合，得到剩下的信号集合。即针对每一子载波，对该子载波的每个信号集合进行筛选，判断每个信号集合中所述相位差或所述幅度的数量是否少于预设阈值，若是，将该信号集合删除，若否，保留该信号集合。然后再根据筛选后所剩下的信号集合得到目标信号集合。
示例性地，针对每一子载波，将预设阈值设置为采样量(如2000)的10％，其中采样量即为针对每一子载波，在采样时间内，第一设备所获得的该子载波的信道状态信息的数量。判断每个信号集合中所包括的所述相位差或所述幅度的总共数量是否少于200，若是，则删除该信号集合。如图6所示，假设预设阈值为100，则第三个信号集合和第四信号集合的所述相位差或所述幅度总量均少于预设阈值，删除第三个信号集合和第四信号集合，剩下第一信号集合和第二信号集合，也即保留图2B中曲线S3和曲线S4的数据，将数据分布稀疏的曲线S1和曲线S2的数据删除。
可以理解，预设阈值可以根据实际情况设置，其也可以设置为所有采样量的20％，在另一些实施例中，预设阈值可以设置为采样点的10％或20％等，本申请对此不作具体限定。
在另一些实施例中，在删除N个信号集合中所述相位差或所述幅度的数量少于预设阈值的信号集合后，得到剩下的信号集合，对每一剩下的信号集合进行如下操作(第一操作和第二操作)中的一种或多种，将最终剩下的信号集合作为目标信号集合。
第一操作：对每一剩下的信号集合，判断该信号集合内是否缺少预设时间段内的所述相位差或所述幅度，若是，删除该信号集合。若否，保留该信号集合。
其中，预设时间段可以为2秒、5秒或6秒，预设时间段可以根据采样时间设置，如设置为采样时间的10％或20％，本申请对此不作具体限定。
示例性地，以预设时间段为2秒为例，在第一操作中，针对每一剩下的信号集合，判断该信号集合是否存在连续2秒所述相位差或所述幅度的缺失。如图2B所示，曲线S1对应的信号集合中，缺少3秒至6秒的信号，则曲线S1存在连续2秒相位差的缺失，该2秒内所缺少的相位差可能在曲线S3上也可能在曲线S4上。
第二操作：对每一剩下的信号集合，判断该信号集合每隔预设时间间隔内第一缺失量与第二缺失量的差是否高于预设缺失阈值；其中，第一缺失量对应该预设时间间隔内所述相位差或所述幅度缺失量的最高值，第二缺失量对应该预设时间间隔内所述相位差或所述幅度缺失量的最低值，若是，删除该信号集合，若否，保留该信号集合。
其中，预设时间间隔可以为一个定值，如可以为1秒、2秒或3秒，预设缺失阈值可以为40％、50％或60％，本申请对此不作具体限定。
在本申请实施例中，所述相位差或所述幅度缺失量可以根据采样点或采样量确定。示例性地，若设置每一采样点所采集到的所述相位差或所述幅度的数量为100，若信号集合上某一采样点的所述相位差或所述幅度数量为60，则可以确定该信号集合对应该采样点上的所述相位差或所述幅度 缺失量为40％。
示例性地，以预设时间间隔为1秒，预设缺失阈值为50％为例，在第二操作中，针对每一剩下的信号集合，获取该信号集合每秒的第一缺失量与第二缺失量。如在第1秒内，获取该信号集合于该秒所述相位差或所述幅度缺失量的最高值(第一缺失量)为60％。获取该信号集合于该秒所述相位差或所述幅度缺失量的最低值(第二缺失量)为30％，则60％-30％的差小于50％，则在对该信号集合判断到第1秒时，保留该信号集合。同理，继续判断第2秒。若在2秒判断该信号集合中第一缺失量与第二缺失量的差高于预设缺失阈值，则删除该信号集合，继续判断该子载波的下一信号集合。
第一设备可以同时对所有的子载波的信号集合进行上述判断，也可以依子载波顺序对各个子载波判断，获得各个子载波的目标信号集合。
在本申请实施例中，第一设备将每一子载波的多个信号集合进行第一操作和第二操作中的一个或多个操作后，最终所剩下的信号集合作为该子载波的目标信号集合。针对每一子载波，目标信号集合的数量可以为多个，如可以为2个。
在一些实施例中，第一设备将每一子载波的多个信号集合进行第一操作和第二操作中的一个或多个操作后，检测最终所剩下的信号集合的数量是否为2或大于2。若是，对最终所剩下的信号集合进一步筛选，选出一个信号集合作为目标信号集合。若否，则直接将最终剩下的信号集合作为目标信号集合。第一设备将最终所剩下的信号集合中数值范围区间最小的信号集合作为目标信号集合。示例性地，如图6所示，最终所剩下的信号集合包括第一信号集合和第二信号集合，其中，第一信号集合的数值范围的区间为0.3(2.8-2.5)小于第二个信号集合对应的数值范围的区间0.7(4-3.4.1)，将第一信号集合作为目标信号集合。以图2B进行说明，剩下的第一信号集合和第二信号集合分别对应曲线S3和S4，也即选择曲线线宽最小的S4作为目标信号集合。
在一些实施例中，第一设备选出目标信号集合后，对该目标信号集合进行数据处理，如线性插值补齐数据，之后将处理后的目标信号集合用于下述步骤的处理。
步骤S44：将目标信号集合进行快速傅里叶变换，得到与目标信号集合对应的频域信号。
在本申请实施例中，针对每一子载波，第一设备将该子载波的目标信号集合进行离散傅里叶变换，提取出目标信号集合的频域信息，得到目标信号集合对应的频域信号。转换为频域信号之后可以很方便地分析出目标信号集合中的频率成分，以便在频域上进行处理。
经过步骤S43后，第一设备可以得到每一子载波对应的目标信号集合，第一设备对所有子载波的目标信号集合进行快速傅里叶变换，得到各个目标信号集合对应的频域信号。图7为所有子载波的目标信号集合进行快速傅里叶变换得到的结果，其中每条曲线代表对一个子载波对应的目标信号集合进行快速傅里叶变换后得到的结果，可以看到在频率为10次/每分钟-20次/每分钟之间具良好的振幅特性。
步骤S45：根据呼吸频段与频域信号得到频域候选呼吸频率和与频域候选呼吸频率对应的呼吸能量比。
在本申请实施例中，第一设备获得各个子载波的目标信号集合对应的频域信号后，根据各个频域信号得到各个子载波对应的频域候选呼吸频率和与频域候选呼吸频率对应的呼吸能量比。
其中，呼吸频段是人体正常情况下的呼吸功率数值。例如，呼吸频段可以是0.1-0.6Hz。
请一并参阅图8，步骤S45具体可以实现为如下步骤：
步骤S81：获取频域信号中频率位于呼吸频段的子频域信号。
示例性地，以呼吸频段为0.1-0.6Hz为例，针对每一子载波的频域信号，第一设备将频域信号中频率位于0.1-0.6Hz内的信号作为该子载波的子频域信号，得到该子载波的子频域信号。
步骤S82：将子频域信号的幅值总和与频域信号的幅值总和的比值作为呼吸能量比。
在本申请实施例中，针对每一子载波，第一设备将该子载波的子频域信号的幅值总和与频域信号的幅值总和的比值作为呼吸能量比，具体地，第一设备将该子载波的子频域信号内各个幅值加和得到子频域信号的幅值总和，将该子载波的频域信号内各个幅值加和得到频域信号的幅值总和，计算子频域信号的幅值总和与频域信号的幅值总和的比值，将该比值作为呼吸能量比，该呼吸能量比对应该子载波。
在一些实施例中，计算呼吸能量比为：将频域信号中频率位于呼吸频段的所有信号的幅值加和作为分子，将频域信号的所有幅值加和作为分母，分子与分母的比值即为呼吸能量比。
步骤S83：将子频域信号中最大幅值所对应的频率作为频域候选呼吸频率。
在本申请实施例中，针对每一子载波，第一设备获取该子载波的子频域信号中幅值最大的信号，并将该信号所对应的频率作为频域候选呼吸频率。在另一些实施例中，第一设备计算每一子载波的呼吸信噪比(breath-to-noise ratio，BNR)，将该子载波呼吸信噪比最大时所对应的频率作为该子载波的频域候选呼吸频率。计算呼吸信噪比为即计算子频域信号中最大的幅值与频域信号的幅值总和的比值。
步骤S46：对目标信号集合进行自相关计算，得到目标信号集合对应的自相关系数。
在本申请实施例中，第一设备对各个子载波的目标信号集合进行自相关计算，即对目标信号集合计算自相关函数(Autocorrelation Function，ACF)，得到各个子载波的目标信号集合对应的自相关系数(Autocorrelation Coefficient，AC)。
请一并参阅图9，图9为对一个子载波的目标信号集合进行自相关计算后得到的结果，其中曲线代表对该子载波的自相关分析结果，可以看到在频率为12次/每分钟-16次/每分钟之间自相关系数更大。
步骤S47：根据呼吸频段与自相关系数得到时域候选呼吸频率和与时域候选呼吸频率对应的第一自相关系数。
在本申请实施例中，第一设备获得每一子载波的目标信号集合对应的自相关系数后，根据各个自相关系数得到各个子载波对应的时域候选呼吸频率和与时域候选呼吸频率对应的第一自相关系数。
请一并参阅图10，步骤S47具体可以实现为：
步骤S101：将呼吸频段内最大自相关系数所对应的频率作为时域候选呼吸频率。
在本申请实施例中，以呼吸频段为0.1-0.6Hz，将目标信号集合中频率位于0.1-0.6Hz内的信号划分为第一集合，获取第一集合。第一信号集合中各个信道状态信息均有其对应的自相关系数，将第一信号集合中自相关系数最大的信号所对应的频率作为时域候选呼吸频率。换句话说，在得到各个子载波的目标信号集合对应的自相关系数后，针对每一子载波，寻找频率位于呼吸频段的自相关系数，得到呼吸频段的自相关系数。将呼吸频段的自相关系数中数值最大的自相关系数所对应的频率作为时域候选呼吸频率。
步骤S102：将时域候选呼吸频率对应的自相关系数作为第一自相关系数。
也即将第一信号集合中最大的自相关系数作为第一自相关系数。换句话说，将呼吸频段的自相关系数中数值最大的自相关系数作为第一自相关系数。
步骤S48：根据频域候选呼吸频率、呼吸能量比、时域候选呼吸频率以及第一自相关系数输出呼吸检测信息。
在本申请实施例中，第一设备经过上述步骤S44-S47，对每个子载波的目标信号集合均进行了频域和时域计算，分别得到各个子载波的频域候选呼吸频率、频域候选呼吸频率对应的呼吸能量比、时域候选呼吸频率以及时域候选呼吸频率对应的第一自相关系数，第一设备对得到的频域候选呼吸频率、呼吸能量比、时域候选呼吸频率以及第一自相关系数进行时域和频域上联合判断，以得到可靠的呼吸检测信息。
在本申请实施例中，步骤S48可以实现为当检测到呼吸能量比大于对应的第一阈值且第一自相关系数大于对应的第二阈值时，将频域候选呼吸频率或时域候选呼吸频率作为呼吸检测信息输出。
其中，第一阈值与第二阈值可相同也可不同，第一阈值与第二阈值根据实际情况确定，如可以设置第一阈值以及第二阈值的取值范围为0.1-10。
当检测到呼吸能量比大于对应的第一阈值时，则说明频域候选呼吸频率为感知对象的呼吸频率的可能性高。当检测到第一自相关系数大于对应的第二阈值时，则说明时域候选呼吸频率为感知对象的呼吸频率的可能性高。通过同时判断呼吸能量比是否大于对应的第一阈值以及第一自相关系数是否大于对应的第二阈值，实现时域和频域上联合判断呼吸频率。当呼吸能量比大于对应的第一阈值且第一自相关系数大于对应的第二阈值时，则说明频域候选呼吸频率与时域候选呼吸频率为感知对象的呼吸频率的可能性高，也说明时域和频域的一致性，更能保证频域候选呼吸频率或时域候选呼吸频率为感知对象的呼吸频率。
在另一些实施例中，步骤S48可以实现为当检测到频域候选呼吸频率与时域候选呼吸频率相近，且呼吸能量比和第一自相关系数中一个或多个大于对应的阈值时，将频域候选呼吸频率或时域候选呼吸频率作为呼吸检测信息输出。
其中，呼吸能量比和第一自相关系数中一个或多个大于对应的阈值包括：呼吸能量比大于对应的第一阈值、第一自相关系数大于对应的第二阈值或呼吸能量比大于对应的第一阈值且第一自相关系数大于对应的第二阈值。也即，当检测频域候选呼吸频率与时域候选呼吸频率相近且呼吸能量比大于对应的第一阈值时，或者，当检测到频域候选呼吸频率与时域候选呼吸频率相近且第一自相关系数大于对应的第二阈值时，或者，当检测到频域候选呼吸频率与时域候选呼吸频率相近，呼吸能量比大于对应的第一阈值且第一自相关系数大于对应的第二阈值时，将频域候选呼吸频率或时域候选呼吸频率作为呼吸检测信息输出。
请一并参阅图11，步骤S48具体可以实现为：
步骤S111：判断频域候选呼吸频率与时域候选呼吸频率的差值是否小于第三阈值；
其中，第三阈值可以根据实际情况设置，第三阈值的取值范围可以为0.1-2次/分钟(Beat Per Minute，BPM)。若判断频域候选呼吸频率与时域候选呼吸频率的差值小于第三阈值，则说明频域候选呼吸频率与时域候选呼吸频率相近。通过验证频域候选呼吸频率与时域候选呼吸频率相近，以此为了避免单独的时域或频域分析计算造成的误检，提高准确性，保证目标信号集合中有感知对象的呼吸信息。
若否，步骤S112，删除此目标信号集合。
在一些实施例中，若频域候选呼吸频率与时域候选呼吸频率的差值不小于第三阈值，则说明频域候选呼吸频率与时域候选呼吸频率相差较远，极大概率无法从目标信号集合获得感知对象的呼吸信息。
在本申请实施例中，删除此目标信号集合即不处理该目标信号集合，而是处理下一子载波的目标信号集合。示例性地，对第6号子载波进行处理，在进行步骤S81判断第6号子载波的频域候选呼吸频率与时域候选呼吸频率的差值大于或等于第三阈值，则将第6号子载波的目标信号集合删除。即不对第6号子载波进行检测，继续处理下一子载波。
若是，步骤S113：判断呼吸能量比和第一自相关系数中一个或多个是否大于对应的阈值；
在本申请实施例中，步骤S113包括：判断呼吸能量比是否大于第一阈值，或，判断第一自相关系数是否大于第二阈值，或，判断呼吸能量比是否大于第一阈值，同时第一自相关系数是否大于第二阈值。若上述任一判断为是，则执行步骤S114。若上述三个判断都为否，则执行步骤S112。
其中，删除此目标信号集合即不对该目标信号集合对应的子载波处理，而是处理下一子载波。
若是，步骤S114：将频域候选呼吸频率或时域候选呼吸频率作为呼吸检测信息输出。
上述步骤S111和步骤S113的执行顺序可以改变。
在另一些实施例中，请一并参阅图12，步骤S83可以实现为如下步骤：
步骤S121：当检测到频域候选呼吸频率与时域候选呼吸频率相近，且呼吸能量比和第一自相关系数中一个或多个大于对应的阈值时，记录呼吸能量比与第一自相关系数之和。
在本申请实施例中，针对每一子载波，第一设备对该子载波的频域候选呼吸频率、呼吸能量比、时域候选呼吸频率以及第一自相关系数均执行步骤S111和步骤S113的判断，若判断结果为是，记录该子载波的呼吸能量比与第一自相关系数之和，将该子载波的呼吸能量比与第一自相关系数之和作为该子载波的权值。若判断结果为否，则不记录该子载波的呼吸能量比与第一自相关系数之和。
步骤S122：将所记录的所有的呼吸能量比与第一自相关系数之和加和作为分母。
在本申请实施例中，第一设备对所有子载波均进行步骤S111和步骤S113的判断，可以得到符合条件(即步骤S111和步骤S113的判断结果为是)的子载波对应的呼吸能量比与第一自相关系数之和，将所记录的所有符合条件的子载波的呼吸能量比与第一自相关系数之和加和作为分母。也即将所有符合条件的子载波的权值加和作为分母。
步骤S123：将所记录的呼吸能量比与第一自相关系数之和的最大值作为分子。
在本申请实施例中，获取所记录的所有权值，即获取所有符合条件(即步骤S111和步骤S113的判断结果为是)的子载波的权值，将最大的权值作为分子。
步骤S124：检测到分子与分母的比值小于第三阈值，将频域候选呼吸频率或时域候选呼吸频率作为呼吸检测信息输出。
其中，第三阈值可以根据实际情况设置，如可以为50％，本申请对此不作具体限定。
在本申请实施例中，第一设备判断最大值的权值与所有符合条件的子载波的权值之和的比值是否小于第三阈值。若小于，即说明所有符合条件的子载波之间具有一致性，保证各目标信号集合之间的一致性，不会因为单个子载波的特殊情况导致误检，此时的频域候选呼吸频率或时域候选呼吸频率可以作为呼吸检测信息。若不小于，说明此时可能存在单个子载波的特殊情况，无法将此时的频域候选呼吸频率或时域候选呼吸频率可以作为呼吸检测结输出。为此可以输出呼吸检测信息 为无法检测呼吸，或当前环境中无可检测的呼吸。
在本申请实施例中，现有的呼吸检测方法无法对分层的信道状态信息进行处理，且现有的呼吸检测方法仅可处理无噪声或轻微噪声信道状态信息，其一般单独对信道状态信息进行快速傅里叶变换，或是，单独对信道状态信息进行自相关函数计算。而本申请提供的呼吸检测方法通过将各个子载波分层的信道状态信息的子信号提取出来，得到各个子载波的目标信号集合。并对各个子载波的目标信号集合进行快速傅里叶变换以及自相关函数计算，以同时获得该子载波的频域候选呼吸频率、呼吸能量比、时域候选呼吸频率以及第一自相关系数。通过对该子载波的频域候选呼吸频率、呼吸能量比、时域候选呼吸频率以及第一自相关系数同时联合判断，判断该子载波携带感知对象呼吸信息的可能性，以此验证频域候选呼吸频率或时域候选呼吸频率是否为感知对象的呼吸频率，双重验证保证呼吸检测信息准确性。进一步地，对所有子载波的频域候选呼吸频率或时域候选呼吸频率进行验证，保证各个目标信号集合数据之间的一致性，不会因为单个载波的特殊情况导致误检，进一步提供呼吸检测信息准确性。
请一并参阅图13，图13为本申请实施例提供的用户设备的主界面示意图。
可以理解，本申请实施例提供的呼吸检测方法可以应用于呼吸检测应用，或实现为呼吸检测应用。以图1所示的用户设备104上安装呼吸检测应用为例。用户设备104上安装呼吸检测应用，在进入睡眠前，用户(如感知对象103)点击主界面上的呼吸检测应用。用户设备104响应于用户的点击操作，用户设备104基于与第二设备101建立通信，用户设备104可以从第二设备101或端服务器105处获取Wi-Fi信号的信道状态信息，进而根据Wi-Fi信号的信道状态信息执行本申请实施例提供的呼吸检测方法，得到用户的呼吸检测信息。或者，用户设备104从第二设备101或端服务器105处获取环境变化信息，进而根据环境变化信息分析出用户的呼吸检测信息。请一并参阅图14，用户设备104响应于用户点击呼吸检测应用，进入呼吸检测信息界面。在用户设备104获得环境变化信息或呼吸检测信息后，在呼吸检测信息界面上显示出相应的呼吸检测信息，包括在夜间睡眠记录时间8小时9分钟内检测到的体动、熟睡状态以及夜间呼吸频率的变化(呼吸检测信息)。用户设备104、第一设备102或云端服务器105还可以具备信息分析处理能量，根据环境变化信息进一步分析处理，输出更为详细的呼吸检测信息包括：睡眠得分、夜间睡眠时长、清醒时长、清醒次数、呼吸频次、呼吸变异性指数、呼吸指令、以及详细的当日睡眠所采集到的呼吸精选数据。如图14，用户点击呼吸检测界面中下一页控件，用户设备104显示该更为详细呼吸检测信息如图15所示。
请一并参阅图16，图16为本申请实施例提供的另一种呼吸检测方法流程示意图。
步骤S161：第一设备接收至少一个第二设备的Wi-Fi信号。
步骤S162：第一设备根据至少一个第二设备的Wi-Fi信号的信道状态信息的相位差或幅度确定N个信号集合，N个信号集合内的相位差或幅度分别属于N个不同的数值范围，其中N为大于或等于2的整数。
步骤S163：第一设备根据N个信号集合确定环境变化信息，环境变化信息包括呼吸检测信息。
在本申请实施例中，根据信道状态信息的相位差或幅度确定N个信号集合，用以在处理存在分层的信道状态信息时，将该分层的信道状态信息根据信道状态信息的相位差或幅度划分成N个信号集合，提取出每个信号集合，基于该N个信号集合可以进行呼吸检测，实现基于该分层的信 道状态信息进行呼吸检测，得到准确的呼吸检测信息。
优选地，步骤S162具体可以包括：针对每一子载波，提取出信道状态信息的相位差；将数值相近的相位差划分为一个信号集合，得到N个信号集合；或，针对每一子载波，提取出信道状态信息的幅度；将数值相近的幅度划分为一个信号集合，得到N个信号集合。
优选地，步骤S163具体可以包括：对N个信号集合进行筛选，得到目标信号集合；对目标信号集合进行频域上分析处理，得到频域候选呼吸频率和与频域候选呼吸频率对应的呼吸能量比；对目标信号集合进行时域上分析处理，得到时域候选呼吸频率和与时域候选呼吸频率对应的第一自相关系数；根据频域候选呼吸频率、呼吸能量比、时域候选呼吸频率以及第一自相关系数输出环境变化信息。
在本申请实施例中，通过将各个子载波分层的信道状态信息的子信号提取出来，得到各个子载波的目标信号集合。并对各个子载波的目标信号集合进行快速傅里叶变换以及自相关函数计算，以同时获得该子载波的频域候选呼吸频率、呼吸能量比、时域候选呼吸频率以及第一自相关系数。通过对该子载波的频域候选呼吸频率、呼吸能量比、时域候选呼吸频率以及第一自相关系数同时联合判断，判断该子载波携带感知对象呼吸信息的可能性，以此验证频域候选呼吸频率或时域候选呼吸频率是否为感知对象的呼吸频率，双重验证保证呼吸检测信息准确性。
优选地，根据频域候选呼吸频率、呼吸能量比、时域候选呼吸频率以及第一自相关系数输出环境变化信息包括：当检测到呼吸能量比大于对应的第一阈值且第一自相关系数大于对应的第二阈值时，将频域候选呼吸频率或时域候选呼吸频率作为呼吸检测信息输出。
优选地，根据频域候选呼吸频率、呼吸能量比、时域候选呼吸频率以及第一自相关系数输出环境变化信息包括：当检测到频域候选呼吸频率与时域候选呼吸频率相近，且呼吸能量比和第一自相关系数中一个或多个大于对应的阈值时，将频域候选呼吸频率或时域候选呼吸频率作为呼吸检测信息输出。
优选地，当检测到频域候选呼吸频率与时域候选呼吸频率相近，且呼吸能量比和第一自相关系数中一个或多个大于对应的阈值时，将频域候选呼吸频率或时域候选呼吸频率作为呼吸检测信息输出包括：当检测到频域候选呼吸频率与时域候选呼吸频率相近，且呼吸能量比和第一自相关系数中一个或多个大于对应的阈值时，记录呼吸能量比与第一自相关系数之和；将所记录的所有的呼吸能量比与第一自相关系数之和加和作为分母；将所记录的呼吸能量比与第一自相关系数之和的最大值作为分子；检测到分子与分母的比值小于第三阈值时，将频域候选呼吸频率或时域候选呼吸频率作为呼吸检测信息输出。
优选地，对N个信号集合进行筛选，得到目标信号集合包括：针对每一子载波，删除N个信号集合中相位差或幅度的数量少于预设阈值的信号集合，得到剩下的信号集合；根据剩下的信号集合得到目标信号集合。
优选地，根据剩下的信号集合得到目标信号集合包括：对每一剩下的信号集合进行如下操作中的一种或多种，将最终剩下的信号集合作为目标信号集合：判断剩下的信号集合内是否缺少预设时间段内的相位差或幅度；若是，删除剩下的信号集合；或，判断剩下的信号集合每隔预设时间间隔内第一缺失量与第二缺失量的差是否高于预设缺失阈值；其中，第一缺失量对应预设时间间隔内相位差的缺失量或幅度的缺失量的最高值，第二缺失量对应预设时间间隔内相位差的缺失量或幅度的缺失量的最低值；若是，删除剩下的信号集合。
优选地，对目标信号集合进行频域上分析处理，得到频域候选呼吸频率和与频域候选呼吸频率对应的呼吸能量比包括：将目标信号集合进行快速傅里叶变换，得到与目标信号集合对应的频域信号；根据呼吸频段与频域信号得到频域候选呼吸频率和与频域候选呼吸频率对应的呼吸能量比。
优选地，根据呼吸频段与频域信号得到频域候选呼吸频率和频域候选呼吸频率对应的呼吸能量比包括获取频域信号中频率位于呼吸频段的子频域信号；将子频域信号的幅值总和与频域信号的幅值总和的比值作为呼吸能量比；将子频域信号中最大幅值所对应的频率作为频域候选呼吸频率。
优选地，对目标信号集合进行时域上分析处理，得到时域候选呼吸频率和与时域候选呼吸频率对应的第一自相关系数包括：对目标信号集合进行自相关计算，得到目标信号集合对应的自相关系数；根据呼吸频段与自相关系数得到时域候选呼吸频率和与时域候选呼吸频率对应的第一自相关系数。
优选地，根据呼吸频段与自相关系数得到时域候选呼吸频率和时域候选呼吸频率对应的第一自相关系数包括：将呼吸频段内最大自相关系数所对应的频率作为时域候选呼吸频率；将时域候选呼吸频率对应的自相关系数作为第一自相关系数。
下述以对信道状态信息的相位差进行处理为例子给出另一呼吸检测方法的流程示意图。
请参见图17，图17为本申请实施例提供的另一呼吸检测方法流程示意图。下述以该呼吸检测方法应用图1中所示的第一设备为例。
步骤S170：开始。在本申请实施例中，第一设备响应于执行本申请呼吸检测方法的操作，开始接收第二设备通过M个子载波所发出的数据包，并从中测量出由每个子载波传输的每个数据包的CSI原始信息。在第一设备上安装呼吸检测应用时，执行本申请呼吸检测方法的操作为用户点击第一设备上的呼吸检测应用。在另一些实施例中，第一设备可以在开机启动后就执行本申请呼吸检测方法。本申请对此不作具体限定。
步骤S170对应图4中的步骤S41，具体内容可以参考步骤S41。
步骤S171：计算所有子载波的信道状态信息的相位差。在本申请实施例中，在步骤S170中，第一设备接收到M个子载波的信道状态信息，针对每一子载波，计算该子载波的信道状态信息的相位差，由此可以计算得到所有子载波的信道状态信息的相位差。
步骤S172：提取子信号。在本申请实施例中，在步骤S171中获得子载波的信道状态信息的相位差，第一设备根据子载波的信道状态信息的相位差提取出子信号。其中，子信号即上述信号集合。
步骤S171和步骤S172对应图4中的步骤S42，具体内容可以参考步骤S42。
在一些实施例中，步骤S172具体为如下步骤：
步骤S1721：对直方图中的bin进行计数统计。
在本申请实施例中，将所有子载波的信道状态信息的相位差作为直方图中的bin，对直方图中的bin进行计数统计。示例性地，分别统计出各个相位差(如5.5度、5度、4.5度等)的个数。如图6所示，可以得到各个bin对应的个数数量。
步骤S1722：去掉占比过于少的计数所对应的采样点。
在本申请实施例中，第一设备去掉占比过于少的计数所对应的采样点(即相位差)可以为对统计的个数数量少于预设阈值(例如10)的采样点删除。如图6所示，相位差为5.1度的统计个数小于10，则可以删除相位差5.1度，也即在所有子载波的信道状态信息的相位差中，相位差为5.1度 的总共数量小于10个。
其中步骤S1722对应图4中步骤S43的一种具体实现，如，针对每一子载波，删除N个信号集合中所述相位差或所述幅度的数量少于预设阈值的信号集合，得到剩下的信号集合，在此不再赘述。
步骤S1723：合并数值上连续的bin为一个bin分组。
在本申请实施例中，第一设备合并数值上连续的bin为一个bin分组也即将数值上连续的相位差合并为一个bin分组，也即合并得到一子信号。如图6所示，得到四个bin分组，该bin分组的组距分别为(2.5-2.8]度、(3.4-4.1]度、(4.5-4.9]度以及(5.1-5.4]度，该四个分组即对应四个信号集合。
其中，可以先执行步骤S1723再执行步骤S1722，在此不再赘述。
步骤S1724：对每一bin分组进行判断，是否存在超过2秒的数据丢失。
在本申请实施例中，第一设备对每一bin分组中的bin进行判断，判断该bin分组中的bin是否存在超过2秒的数据丢失。
步骤S1724对应图4中的第一操作，具体内容可以参考第一操作。
若是，步骤S1725：删除此bin分组。第一设备删除此bin分组即删除信号集合。
若否，步骤S1726：对每一bin分组进行判断，每秒内缺失的数据量波动是否高于50％。
在本申请实施例中，第一设备对每一bin分组进行判断，判断该bin分组中是否存在超过2秒的数据丢失，若否，执行步骤S1726。也即第一设备在执行步骤S1724后，得到剩余的信号集合。步骤S1726对剩余的信号集合判断，判断该剩余的信号集合中，每秒内相位差缺失量的最高值与相位差缺失量的最低值之间的差是否高于50％。例如，针对信号集合A的第一秒，检测到在该第一秒内相位差缺失量的最高值为60％，该最高值60％对应采样点a。检测到在该第一秒内相位差缺失量的最低值为20％，该最低值20％对应采样点b，则60％-20％的差小于50％，则在对该信号集合判断到第1秒时，保留该信号集合。
步骤S1724对应图4中的第二操作，具体内容可以参考第二操作。
在本申请实施例中，步骤S1722、步骤S1724、步骤S1725以及步骤S1726对应图4中的步骤S43，具体内容可以参考步骤S43。
第一设备对每一bin分组进行判断，每秒内缺失的数据量波动是否高于50％，若是，执行步骤S1725，若否，执行步骤S1727：线性插值补齐数据。
在本申请实施例中，第一设备执行完步骤S1722至步骤S1726后得到目标信号集合。第一设备可以对目标信号集合进行数据处理，如线性插值补齐数据，之后将处理后的目标信号集合用于下述步骤的处理。
步骤S1728：获取当前载波的呼吸子信号。
在本申请实施例中，第一设备对目标信号集合进行线性插值补齐数据后，得到当前载波的呼吸子信号，也即当前载波的目标信号集合。第一设备获得所有载波的目标信号集合后将用于后续步骤。
步骤S173：进行时域和频域上联合计算。
其中，步骤S173即对应图4中的步骤S44、步骤S45、步骤S46以及步骤S47，具体内容可以参考步骤S44、步骤S45、步骤S46以及步骤S47。
在一些实施例中，步骤S173具体为如下步骤：
步骤S1731：进行快速傅里叶变换。
在本申请实施例中，第一设备对目标信号集合(也即步骤S1728所得到的呼吸子信号)进行快速傅里叶变换，得到与目标信号集合对应的频域信号。
其中，步骤S1731具体对应图4中的步骤S44。
步骤S1732：计算呼吸能量比，具体地，计算呼吸频段的幅值总和在全部频率幅值总和的占比。
其中，第一设备将频域信号中频率位于呼吸频段的所有信号的幅值加和作为分子，将步骤S173快速傅里叶计算后所得到的频域信号的所有幅值加和作为分母，分子与分母的比值即为呼吸能量比。
步骤S1732对应到图8中的步骤S81至步骤S82，具体内容可以参考步骤S81至步骤S82。
步骤S1733：计算呼吸信噪比，具体地，计算呼吸频段最大的幅值在全部频率幅值总和的占比。
在本申请实施例中，第一设备计算子频域信号中最大的幅值与频域信号的幅值总和的比值得到呼吸信噪比。
步骤S1734：频域分析：计算频域候选呼吸频率并计算呼吸能量比，具体地，将呼吸信噪比最高时对应的频率作为频域候选呼吸频率。
其中，步骤S1732至步骤S1734对应图4中的步骤S45，步骤S1734对应到图8中的S83。
步骤S1735：进行自相关计算。在本申请实施例中，第一设备对目标信号集合(也即步骤S1728所得到的呼吸子信号)进行自相关计算，得到目标信号集合对应的自相关系数。
其中，步骤S1735对应图4中的步骤S46，具体内容可以参考步骤S46。
步骤S1736：时域分析：计算时域候选呼吸频率并以时域候选呼吸频率对应的自相关系数作为第一自相关系数，具体地，将呼吸频段内自相关系数最高处对应的频率作为时域候选呼吸频率。
在本申请实施例中，步骤S1736对应到图4中的步骤S47。步骤S1736的具体可以包括如图10所示的步骤S101至步骤S102，在此不再赘述。
步骤S174：进行时域和频域上联合判断。
在本申请实施例中，第一设备对上述步骤S173所获得频域候选呼吸频率、呼吸能量比、时域候选呼吸频率以及第一自相关系数进行联合判断，输出检测结果。
其中，步骤S174对应图4中的步骤S48，具体内容可以参考步骤S48。
在一些实施例中，步骤S174具体为如下步骤：
步骤S1741：判断时域候选呼吸频率与频域候选呼吸频率的误差是否小于阈值。在本申请实施例中，第一设备判断频域候选呼吸频率与时域候选呼吸频率之间的差值是否小于第三阈值。
其中，步骤S1741对应图11中的步骤S111，具体内容可以参考步骤S111。
若否，步骤S1742：删除此呼吸子信号。在本申请实施例中，第一设备判断频域候选呼吸频率与时域候选呼吸频率之间的差值大于或等于第三阈值，则判断目标信号集合(即呼吸子信号)进行时域和频域分析所得到的结果差距较大，该目标信号集合不适于进行呼吸检测，该目标信号集合将被用于呼吸检测。
其中，步骤S1742对应图11中的步骤S112，具体内容可以参考步骤S112。
若是，步骤S1743：判断呼吸能量比和第一自相关系数是否大于阈值。
在本申请实施例中，第一设备执行步骤S1741后，判断时域候选呼吸频率与频域候选呼吸频率的误差小于第三阈值，则第一设备继续判断呼吸能量比和第一自相关系数中一个或多个是否大于 对应的阈值，如判断呼吸能量比是否大于第一阈值，或，判断第一自相关系数是否大于第二阈值，或，判断呼吸能量比是否大于第一阈值且同时第一自相关系数是否大于第二阈值。
其中，步骤S1743对应图11中的步骤S113，具体内容可以参考步骤S113。
步骤S1744：统计所有满足上述条件的呼吸子信号，将满足上述条件的呼吸子信号的呼吸能量比和第一自相关系数之和作为权值。
在本申请实施例中，第一设备统计步骤S1741与步骤S1743的判断结果均为“是”的呼吸子信号，记录满足上述条件的呼吸子信号的呼吸能量比和第一自相关系数，将每一子载波的呼吸能量比和第一自相关系数之和作为该子载波的权值。
步骤S1745：判断候选呼吸频率中权值最高项的权值在全部候选呼吸频率的权值占比是否小于阈值。
在本申请实施例中，第一设备将所记录的所有的呼吸能量比和第一自相关系数之和加和作为分母，将所记录的呼吸能量比和第一自相关系数之和的最大值作为分子。则判断候选呼吸频率中权值最高项的权值在全部候选呼吸频率的权值占比是否小于阈值，即判断呼吸能量比和第一自相关系数之和的最大值在所记录的所有的呼吸能量比和第一自相关系数之和的占比是否小于第三阈值。
是，步骤S1746：将候选呼吸频率作为检测结果。第一设备检测到步骤S1745的判断结果为是，将频域候选呼吸频率或时域候选呼吸频率作为呼吸检测信息输出。
其中，步骤S1744至步骤S1746对应图12的步骤S121至步骤S124，具体内容可以参考的步骤S121至步骤S124。
否，步骤S1747：环境中无可检测呼吸。
在本申请实施例中，第一设备检测到步骤S1745的判断结果为否，第一设备判断当前环境中无可检测的呼吸。
步骤S175：输出呼吸检测结果。
在本申请实施例中，第一设备输出呼吸检测结果包括呼吸检测信息，如图14、15所示。
请参见图18，图18为本申请实施例提供的另一呼吸检测方法流程示意图。
图18与图17的区别在于：图17的步骤S1728与图18中步骤S1828不同，步骤S1828为将当前范围更小的呼吸子信号作为载波呼吸信号。在本申请实施例中，第一设备将组距更小的呼吸子信号作为载波呼吸信号。如图6所示，第一信号集合的数值范围的区间为0.3(2.8-2.5)小于第二个信号集合对应的数值范围的区间0.7(4-3.4.1)，将第一信号集合作为目标信号集合。以此，减少用户后续步骤S173和步骤S174的计算量，且发明人发现组距更小的呼吸子信号的数据更集中，分析效果更好。
请参见图19，图19为本申请实施例提供的电子设备的软件架构示意图。
该电子设备(例如第一设备、第二设备或用户设备)还可以采用其他架构，本发明实施例还可以Harmony系统为例，说明电子设备的软件架构。
在一些实施方案中，Harmony包括四层，从下往上分别为内核层、系统基础能力层、框架层以及应用层。
Harmony系统采用多内核设计，可选地包括Linux内核、Harmony微内核和LiteOS。通过这种设计，具有不同设备能力的设备能够选择合适的系统内核。内核层还包括内核抽象层(Kernal Abstract Layer)，对其他Harmony层提供基础的内核能力，例如进程管理、线程管理、内存管理、 文件系统管理、网络管理和外设管理等。
系统基础服务层是Harmony系统的核心能力集合，支持Harmony系统在多设备部署的场景下通过框架层对应用服务提供服务。该层可选地包括以下几个部分：
系统基本能力子系统集：为分布式应用在Harmony系统多设备上的运行、调度、迁移等操作提供了基础能力，由分布式软总线、分布式数据管理和文件管理、分布式任务调度、方舟运行时、分布式安全和隐私保护等组成。其中，方舟运行时提供了C/C++/JavaScript多语言运行时和基础的系统类库，也为使用方舟编译器静态化的Java程序(即应用程序或框架层中使用Java语言开发的部分)提供运行时。
基础软件服务子系统集：为Harmony系统提供公共的、通用的软件服务，由图形图像、分布式媒体、分布式AI、多模输入、MSDP&DV、事件通知、电话服务、分布式DFX等子系统组成。基础软件服务子系统集可以根据不同设备形态的部署环境，每个子系统内部可以按照功能粒度进行裁剪。
增强软件服务子系统集：为Harmony系统提供针对不同设备的、差异化的能力增强型软件服务，由平板业务软件、智慧屏业务软件、车机业务软件、IoT业务软件等子系统组成。增强软件服务子系统集可以根据不同设备形态的部署环境，可以按子系统粒度进行裁剪，每个子系统内部又可以按功能粒度进行裁剪。
鸿蒙驱动框架(HDF)和硬件抽象适配层(HAL):是Harmony系统硬件生态开放的基础，向上对硬件提供硬件能力抽象，向下提供各种外设驱动的开发框架和运行环境。
硬件服务子系统集：为Harmony系统提供公共的、适配的硬件服务，由泛Sensor、位置、电源、USB、生物识别等硬件服务子系统组成。硬件服务子系统集能够根据不同设备形态的部署环境，每个子系统内部可以按功能粒度进行裁剪。
专有硬件服务子系统：为Harmony系统提供针对不同设备的、差异化的硬件服务，可选地包括平板专有硬件服务、车机专有硬件服务、穿戴专有硬件服务、IoT专有硬件服务等子系统。专有硬件服务子系统可以按子系统粒度进行裁剪，每个子系统颞部可以按功能粒度裁剪。
框架层为Harmony系统的应用程序提供了Java/C/C++/JavaScript等多语言的用户程序框架和元能力框架，以及各种软硬件服务对外开放的多语言框架API。
应用层包括系统应用和三方应用，可以包括相机，图库，日历，通话，地图，导航，WLAN，蓝牙，音乐，视频，短信息等应用。Harmony系统中应用以AA和FA为基础构建应用。
参见图20，为本申请提供的第二设备的一个实施例结构示意图。可以用于执行图4、8、10、11、12、16、17、18所对应实施例中的呼吸检测方法。
如图20所示，第二设备101可以包括：处理器201及存储器202等组件。这些组件通过一条或多条总线进行连接及通信。
其中，处理器201为第二设备的控制中心，利用各种接口和线路连接整个基站的各个部分，通过运行或执行存储在存储器202内的软件程序和/或模块，以及调用存储在存储器202内的数据，以执行基站的各种功能和/或处理数据。处理器201可以由集成电路(integrated circuit，简称IC)组成，例如可以由单颗封装的IC所组成，也可以由连接多颗相同功能或不同功能的封装IC而组成。举例来说，处理器201可以为通信处理器(communication processor，简称CP)。
存储器202可用于存储软件程序以及模块，处理器201通过运行存储在存储器202的软件程 序以及模块，从而执行第二设备的各种功能应用以及实现数据处理。在本申请具体实施方式中，存储器202可以包括易失性存储器，例如非挥发性动态随机存取内存(nonvolatile random access memory，简称NVRAM)、相变化随机存取内存(phase change RAM，简称PRAM)、磁阻式随机存取内存(mageto-resistive RAM，简称MRAM)等，还可以包括非易失性存储器，例如至少一个磁盘存储器件、电子可擦除可编程只读存储器(Electrically erasable programmable read-only memory，简称EEPROM)、闪存器件，例如反或闪存(NOR flash memory)或是反及闪存(NAND flash memory)。
参见图21，为本申请提供的第一设备一个实施例的结构示意图。可以用于执行本申请对应实施例中的呼吸检测方法。
如图21所示，第一设备102可以包括：处理器211、存储器212等组件。除此之外，这些组件也可以通过一条或多条总线等进行连接及通信。
处理器211为第一设备的控制中心，利用各种接口和线路连接整个用户设备的各个部分，通过运行或执行存储在存储器212内的软件程序和/或模块，以及调用存储在存储器212内的数据，以执行终端的各种功能和/或处理数据。处理器211可以由集成电路(Integrated Circuit，简称IC)组成，例如可以由单颗封装的IC所组成，也可以由连接多颗相同功能或不同功能的封装IC而组成。举例来说，处理器211可以为CP。
具体实现中，本申请还提供一种计算机存储介质，其中，该计算机存储介质可存储有程序，该程序执行时可包括本申请提供的呼吸检测方法各实施例中的部分或全部步骤。的存储介质可为磁碟、光盘、只读存储记忆体(read-only memory，ROM)或随机存储记忆体(random access memory，RAM)等。
还应理解，本申请实施例中的存储器可以是易失性存储器或非易失性存储器，或可包括易失性和非易失性存储器两者。其中，非易失性存储器可以是只读存储器(read-only memory，ROM)、可编程只读存储器(programmable ROM，PROM)、可擦除可编程只读存储器(erasable PROM，EPROM)、电可擦除可编程只读存储器(electrically EPROM，EEPROM)或闪存。易失性存储器可以是随机存取存储器(random access memory，RAM)，其用作外部高速缓存。通过示例性但不是限制性说明，许多形式的RAM可用，例如静态随机存取存储器(static RAM，SRAM)、动态随机存取存储器(dynamic RAM，DRAM)、同步动态随机存取存储器(synchronous DRAM，SDRAM)、双倍数据速率同步动态随机存取存储器(double data rate SDRAM，DDR SDRAM)、增强型同步动态随机存取存储器(enhanced SDRAM，ESDRAM)、同步连接动态随机存取存储器(synch link DRAM，SLDRAM)和直接内存总线随机存取存储器(direct ram bus RAM，DRRAM)。应注意，本文描述的系统和方法的存储器旨在包括但不限于这些和任意其它适合类型的存储器。
参见图22，为本申请提供的用户设备的一个实施例结构示意图。可以用于执行本申请实施例中的呼吸检测方法，并呈现出相应的呼吸检测信息。
如图22所示，用户设备104可以包括：处理器221、存储器222以及显示屏223等组件。除此之外，这些组件也可以通过一条或多条总线等进行连接及通信。
处理器221为第一设备的控制中心，利用各种接口和线路连接整个用户设备的各个部分，通过运行或执行存储在存储器222内的软件程序和/或模块，以及调用存储在存储器222内的数据，以执行终端的各种功能和/或处理数据。处理器221可以由集成电路(Integrated Circuit，简称IC)组成，例如可以由单颗封装的IC所组成，也可以由连接多颗相同功能或不同功能的封装IC而组成。举例 来说，处理器221可以为CP。
显示屏223用于显示相应的环境变化信息，如图14、图15所示的呼吸检测信息。
具体实现中，本申请还提供一种计算机存储介质，其中，该计算机存储介质可存储有程序，该程序执行时可包括本申请提供的呼吸检测方法各实施例中的部分或全部步骤。的存储介质可为磁碟、光盘、只读存储记忆体(read-only memory，ROM)或随机存储记忆体(random access memory，RAM)等。
还应理解，本申请实施例中的存储器可以是易失性存储器或非易失性存储器，或可包括易失性和非易失性存储器两者。其中，非易失性存储器可以是只读存储器(read-only memory，ROM)、可编程只读存储器(programmable ROM，PROM)、可擦除可编程只读存储器(erasable PROM，EPROM)、电可擦除可编程只读存储器(electrically EPROM，EEPROM)或闪存。易失性存储器可以是随机存取存储器(random access memory，RAM)，其用作外部高速缓存。通过示例性但不是限制性说明，许多形式的RAM可用，例如静态随机存取存储器(static RAM，SRAM)、动态随机存取存储器(dynamic RAM，DRAM)、同步动态随机存取存储器(synchronous DRAM，SDRAM)、双倍数据速率同步动态随机存取存储器(double data rate SDRAM，DDR SDRAM)、增强型同步动态随机存取存储器(enhanced SDRAM，ESDRAM)、同步连接动态随机存取存储器(synch link DRAM，SLDRAM)和直接内存总线随机存取存储器(direct ram bus RAM，DRRAM)。应注意，本文描述的系统和方法的存储器旨在包括但不限于这些和任意其它适合类型的存储器。
本申请中，A与B对应可以理解为A与B关联，或者A与B具有关联关系。
应理解，本申请实施例中的方式、情况、类别以及实施例的划分仅是为了描述的方便，不应构成特别的限定，各种方式、类别、情况以及实施例中的特征在不矛盾的情况下可以相结合。
还应理解，申请实施例中的“第一”和“第二”仅为了区分，不应对本申请构成任何限定。
应理解，本文中术语“和/或”，仅仅是一种描述关联对象的关联关系，表示可以存在三种关系，例如，A和/或B，可以表示：单独存在A，同时存在A和B，单独存在B这三种情况。另外，本文中字符“/”，一般表示前后关联对象是一种“或”的关系。
本领域普通技术人员可以意识到，结合本文中所公开的实施例描述的各示例的单元及算法步骤，能够以电子硬件、或者计算机软件和电子硬件的结合来实现。这些功能究竟以硬件还是软件方式来执行，取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能，但是这种实现不应认为超出本申请的范围。
所属领域的技术人员可以清楚地了解到，为描述的方便和简洁，上述描述的系统、装置和单元的具体工作过程，可以参考前述方法实施例中的对应过程，在此不再赘述。
在本申请所提供的几个实施例中，应该理解到，所揭露的系统、装置和方法，可以通过其它的方式实现。例如，以上所描述的装置实施例仅仅是示意性的，例如，单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，例如多个单元或组件可以结合或者可以集成到另一个系统，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口，装置或单元的间接耦合或通信连接，可以是电性，机械或其它的形式。
作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实 际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外，在本申请各个实施例中的各功能单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。
功能如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读取存储介质中。基于这样的理解，本申请的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质中，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)执行本申请各个实施例方法的全部或部分步骤。而前述的存储介质包括：U盘、移动硬盘、只读存储器(read-only memory，ROM)、随机存取存储器(random access memory，RAM)、磁碟或者光盘等各种可以存储程序代码的介质。
可以理解的是，在本申请的实施例中涉及的各种数字编号仅为描述方便进行的区分，并不用来限制本申请的实施例的范围。以上为本申请提供的实施例，并不用以限制本申请，凡在本申请的精神和原则之内，所作的任何修改、等同替换、改进等，均应包含在本申请的保护范围之内。

Claims (16)

  1. 一种呼吸检测方法，其特征在于，所述方法包括：
    第一设备接收至少一个第二设备的Wi-Fi信号；
    所述第一设备根据所述至少一个第二设备的Wi-Fi信号的信道状态信息的相位差或幅度确定N个信号集合，所述N个信号集合内的所述相位差或所述幅度分别属于N个不同的数值范围，其中N为大于或等于2的整数；
    所述第一设备根据所述N个信号集合确定环境变化信息，所述环境变化信息包括呼吸检测信息。
  2. 根据权利要求1所述的方法，其特征在于，所述根据所述至少一个第二设备的Wi-Fi信号的信道状态信息的相位差或幅度确定N个信号集合包括：
    确定每一子载波的所述信道状态信息的相位差；
    根据所述相位差得到N个所述信号集合；
    或，确定每一所述子载波的所述信道状态信息的幅度；
    根据所述幅度得到N个所述信号集合。
  3. 根据权利要求1或2所述的方法，其特征在于，所述根据所述N个信号集合确定环境变化信息包括：
    根据所述N个信号集合得到目标信号集合；
    根据所述目标信号集合得到频域候选呼吸频率和与所述频域候选呼吸频率对应的呼吸能量比；
    根据所述目标信号集合得到时域候选呼吸频率和与所述时域候选呼吸频率对应的第一自相关系数；
    根据所述频域候选呼吸频率、所述呼吸能量比、所述时域候选呼吸频率以及所述第一自相关系数输出环境变化信息。
  4. 根据权利要求1至3任一项所述的方法，其特征在于，所述根据所述频域候选呼吸频率、所述呼吸能量比、所述时域候选呼吸频率以及所述第一自相关系数输出环境变化信息包括：
    当检测到所述呼吸能量比大于对应的第一阈值且所述第一自相关系数大于对应的第二阈值时，将所述频域候选呼吸频率或所述时域候选呼吸频率作为呼吸检测信息输出。
  5. 根据权利要求1至4任一项所述的方法，其特征在于，所述根据所述频域候选呼吸频率、所述呼吸能量比、所述时域候选呼吸频率以及所述第一自相关系数输出环境变化信息包括：
    当检测到所述频域候选呼吸频率与所述时域候选呼吸频率相近，且所述呼吸能量比和所述第一自相关系数中一个或多个大于对应的阈值时，将所述频域候选呼吸频率或所述时域候选呼吸频率作为呼吸检测信息输出。
  6. 根据权利要求3所述的方法，其特征在于，所述根据所述N个信号集合得到目标信号集合包括：
    针对每一所述子载波，获取剩下的信号集合，其中所述剩下的信号集合为所述N个信号集合中所述相位差或所述幅度的数量少于预设阈值的信号集合；
    根据所述剩下的信号集合得到目标信号集合。
  7. 根据权利要求6所述的方法，其特征在于，所述根据所述剩下的信号集合得到目标信号集合包括：
    判断所述剩下的信号集合内是否缺少预设时间段内的所述相位差或所述幅度；
    若否，将所述剩下的信号集合作为目标信号集合；
    或，判断所述剩下的信号集合每隔预设时间间隔内第一缺失量与第二缺失量的差是否高于预设缺失阈值；其中，第一缺失量对应所述预设时间间隔内所述相位差的缺失量或所述幅度的缺失量的最高值，第二缺失量对应所述预设时间间隔内所述相位差的缺失量或所述幅度的缺失量的最低值；
    若否，将所述剩下的信号集合作为目标信号集合。
  8. 根据权利要求6或7所述的方法，其特征在于，所述根据所述目标信号集合得到频域候选呼吸频率和与所述频域候选呼吸频率对应的呼吸能量比包括：
    将所述目标信号集合进行快速傅里叶变换，得到与所述目标信号集合对应的频域信号；
    根据呼吸频段与所述频域信号得到频域候选呼吸频率和与所述频域候选呼吸频率对应的呼吸能量比。
  9. 根据权利要求8所述的方法，其特征在于，所述根据呼吸频段与所述频域信号得到频域候选呼吸频率和所述频域候选呼吸频率对应的呼吸能量比包括：
    获取所述频域信号中频率位于呼吸频段的子频域信号；
    确定所述子频域信号的幅值总和与所述频域信号的幅值总和的比值为呼吸能量比；
    确定所述子频域信号中最大幅值所对应的频率为所述频域候选呼吸频率。
  10. 根据权利要求3、6或7所述的方法，其特征在于，所述根据所述目标信号集合得到时域候选呼吸频率和与所述时域候选呼吸频率对应的第一自相关系数包括：
    对所述目标信号集合进行自相关计算，得到所述目标信号集合对应的自相关系数；
    根据呼吸频段与所述自相关系数得到时域候选呼吸频率和与所述时域候选呼吸频率对应的第一自相关系数。
  11. 根据权利要求10所述的方法，其特征在于，所述根据所述呼吸频段与所述自相关系数得到时域候选呼吸频率和所述时域候选呼吸频率对应的第一自相关系数包括：
    确定所述呼吸频段内最大自相关系数所对应的频率为时域候选呼吸频率；
    确定所述时域候选呼吸频率对应的自相关系数为第一自相关系数。
  12. 一种电子设备，其特征在于，所述电子设备包括至少一个处理器、存储器，其中所述存储器用于存储指令，所述处理器用于执行所述指令实现如权利要求1至11中任一项所述的方法。
  13. 根据权利要求12所述的电子设备，其特征在于，所述电子设备包括终端设备或无线访问节点。
  14. 根据权利要求12或13所述的电子设备，其特征在于，所述电子设备用于根据呼吸检测信息呈界面或输出语音。
  15. 一种计算机可读存储介质，其特征在于，所述计算机可读存储介质存储有程序，所述程序使得电子设备执行如权利要求1至11任一项所述的方法。
  16. 一种计算机程序产品，其特征在于，所述计算机程序产品包括计算机可读指令，当所述计算机可读指令被一个或多个处理器执行时实现如权利要求1至11中任一项所述的方法。
PCT/CN2023/102515 2022-06-30 2023-06-26 呼吸检测方法、电子设备、存储介质及程序产品 WO2024002029A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202210771443.5A CN117375741A (zh) 2022-06-30 2022-06-30 呼吸检测方法、电子设备、存储介质及程序产品
CN202210771443.5 2022-06-30

Publications (1)

Publication Number Publication Date
WO2024002029A1 true WO2024002029A1 (zh) 2024-01-04

Family

ID=89382926

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2023/102515 WO2024002029A1 (zh) 2022-06-30 2023-06-26 呼吸检测方法、电子设备、存储介质及程序产品

Country Status (2)

Country Link
CN (1) CN117375741A (zh)
WO (1) WO2024002029A1 (zh)

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106175767A (zh) * 2016-07-01 2016-12-07 华中科技大学 一种非接触式的多人呼吸参数实时检测方法及系统
CN106618497A (zh) * 2016-12-13 2017-05-10 北京理工大学 复杂环境下基于信道状态信息监测睡眠的方法
CN108123765A (zh) * 2017-12-25 2018-06-05 儒安科技有限公司 一种人员实时检测方法及系统
US20210186369A1 (en) * 2019-12-18 2021-06-24 Wistron Neweb Corporation Breath detection device and method thereof
CN113743374A (zh) * 2021-10-11 2021-12-03 南京邮电大学 一种基于信道状态信息呼吸感知的人员身份识别方法
WO2022095869A1 (zh) * 2020-11-04 2022-05-12 蓬托森思股份有限公司 一种无接触式呼吸或心跳检测方法

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106175767A (zh) * 2016-07-01 2016-12-07 华中科技大学 一种非接触式的多人呼吸参数实时检测方法及系统
CN106618497A (zh) * 2016-12-13 2017-05-10 北京理工大学 复杂环境下基于信道状态信息监测睡眠的方法
CN108123765A (zh) * 2017-12-25 2018-06-05 儒安科技有限公司 一种人员实时检测方法及系统
US20210186369A1 (en) * 2019-12-18 2021-06-24 Wistron Neweb Corporation Breath detection device and method thereof
WO2022095869A1 (zh) * 2020-11-04 2022-05-12 蓬托森思股份有限公司 一种无接触式呼吸或心跳检测方法
CN113743374A (zh) * 2021-10-11 2021-12-03 南京邮电大学 一种基于信道状态信息呼吸感知的人员身份识别方法

Also Published As

Publication number Publication date
CN117375741A (zh) 2024-01-09

Similar Documents

Publication Publication Date Title
Zhang et al. Anti-fall: A non-intrusive and real-time fall detector leveraging CSI from commodity WiFi devices
KR102446811B1 (ko) 복수의 디바이스들로부터 수집된 데이터 통합 및 제공 방법 및 이를 구현한 전자 장치
Canzian et al. Trajectories of depression: unobtrusive monitoring of depressive states by means of smartphone mobility traces analysis
Zhou et al. Omnidirectional coverage for device-free passive human detection
Zou et al. Wi-Fi radar: Recognizing human behavior with commodity Wi-Fi
KR101510860B1 (ko) 사용자 의도 파악 어플리케이션 서비스 방법 및 서버와 이를 이용한 사용자 의도 파악 어플리케이션 서비스 시스템
Xu et al. WiStep: Device-free step counting with WiFi signals
CN101938691B (zh) 多模态邻近度检测
WO2018133264A1 (zh) 一种人体室内定位自动检测方法及系统
CN105874771A (zh) 嗅探智能电话
Bi et al. Familylog: A mobile system for monitoring family mealtime activities
Xu et al. Attention-based gait recognition and walking direction estimation in wi-fi networks
WO2022109712A1 (en) Filtering channel responses for motion detection
CN103295584B (zh) 音声数据检测装置、音声监控系统及其方法
WO2022148477A1 (zh) 一种Wi-Fi人体检测的方法和智能设备
US20160334437A1 (en) Mobile terminal, computer-readable recording medium, and activity recognition device
CN103873171B (zh) 基于多用户互协作的协作频谱感知决策融合方法
Zhu et al. NotiFi: A ubiquitous WiFi-based abnormal activity detection system
Sun et al. Coughloc: Location-aware indoor acoustic sensing for non-intrusive cough detection
Hao et al. CSI‐HC: A WiFi‐Based Indoor Complex Human Motion Recognition Method
CN114222202B (zh) 一种基于WiFi CSI的环境自适应活动检测方法及系统
CN109448705A (zh) 一种语音切分方法及装置、计算机装置和可读存储介质
Man et al. PWiG: A phase-based wireless gesture recognition system
CN110220554A (zh) 用于输出信息的方法和装置
WO2024002029A1 (zh) 呼吸检测方法、电子设备、存储介质及程序产品

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 23830206

Country of ref document: EP

Kind code of ref document: A1