WO2022022391A1 - 基于无线信号感知打喷嚏的方法及相关装置 - Google Patents

基于无线信号感知打喷嚏的方法及相关装置 Download PDF

Info

Publication number
WO2022022391A1
WO2022022391A1 PCT/CN2021/107962 CN2021107962W WO2022022391A1 WO 2022022391 A1 WO2022022391 A1 WO 2022022391A1 CN 2021107962 W CN2021107962 W CN 2021107962W WO 2022022391 A1 WO2022022391 A1 WO 2022022391A1
Authority
WO
WIPO (PCT)
Prior art keywords
wireless signal
sneeze
doppler
information
signal
Prior art date
Application number
PCT/CN2021/107962
Other languages
English (en)
French (fr)
Inventor
杜瑞
童文
韩霄
董明杰
陈凯彬
孙滢翔
彭晓辉
杨讯
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
Priority to EP21849813.7A priority Critical patent/EP4184074A4/en
Publication of WO2022022391A1 publication Critical patent/WO2022022391A1/zh
Priority to US18/160,466 priority patent/US20230168372A1/en

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
    • G01S13/02—Systems using reflection of radio waves, e.g. primary radar systems; Analogous systems
    • G01S13/50—Systems of measurement based on relative movement of target
    • G01S13/52—Discriminating between fixed and moving objects or between objects moving at different speeds
    • G01S13/56—Discriminating between fixed and moving objects or between objects moving at different speeds for presence detection
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S15/00—Systems using the reflection or reradiation of acoustic waves, e.g. sonar systems
    • G01S15/88—Sonar systems specially adapted for specific applications
    • G01S15/89—Sonar systems specially adapted for specific applications for mapping or imaging
    • G01S15/8906—Short-range imaging systems; Acoustic microscope systems using pulse-echo techniques
    • G01S15/8979—Combined Doppler and pulse-echo imaging systems
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/02—Services making use of location information
    • H04W4/023—Services making use of location information using mutual or relative location information between multiple location based services [LBS] targets or of distance thresholds
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
    • G01S13/003—Bistatic radar systems; Multistatic radar systems
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
    • G01S13/88—Radar or analogous systems specially adapted for specific applications
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
    • G01S7/02—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00
    • G01S7/41—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00 using analysis of echo signal for target characterisation; Target signature; Target cross-section
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
    • G01S7/02—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00
    • G01S7/41—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00 using analysis of echo signal for target characterisation; Target signature; Target cross-section
    • G01S7/414—Discriminating targets with respect to background clutter
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/0464—Convolutional networks [CNN, ConvNet]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G06N3/09—Supervised learning
    • 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
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F24—HEATING; RANGES; VENTILATING
    • F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
    • F24F11/00—Control or safety arrangements
    • F24F11/62—Control or safety arrangements characterised by the type of control or by internal processing, e.g. using fuzzy logic, adaptive control or estimation of values
    • F24F11/63—Electronic processing
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F24—HEATING; RANGES; VENTILATING
    • F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
    • F24F2120/00—Control inputs relating to users or occupants
    • F24F2120/20—Feedback from users
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/02—Services making use of location information
    • H04W4/021—Services related to particular areas, e.g. point of interest [POI] services, venue services or geofences
    • 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
    • H04W4/33—Services specially adapted for particular environments, situations or purposes for indoor environments, e.g. buildings

Definitions

  • the present application relates to the field of wireless communication technologies, and in particular, to a method and related device for sensing sneeze based on a wireless signal.
  • WLAN sensing is a technology with broad application prospects. It uses radio frequency (RF) signals sent by WLAN devices to sense the surrounding environment, and extracts the The corresponding parameters in the received signal are analyzed to obtain relevant information in the surrounding environment.
  • WLAN sensing can borrow the WLAN infrastructure (such as WLAN equipment) that has been widely deployed now to realize environmental awareness.
  • the WLAN devices deployed in the environment can be used to continuously detect and obtain the channel status information of the environment where they are located, and through a large amount of data analysis and comparison, real-time monitoring of whether there are abnormal conditions in the environment, so as to ensure the safety of the family home. Safety.
  • WLAN devices can also be deployed in a specific monitoring area to sense and measure the patient's heartbeat, temperature and other biometric information through WLAN signals or specific data signals, so as to monitor the area where the patient is located in real time.
  • sneezing or coughing is an important mode of transmission of the virus, especially in environments with poor air mobility such as indoors, the droplets produced by sneezing/coughing and the subsequent virus-carrying Aerosols can stay for more than a few hours and are potentially dangerous. Therefore, by identifying and locating actions such as sneezing/coughing, identifying the location of the potentially infected person sneezing/coughing, and judging the range of high-concentration virus-containing aerosols that may be generated around, it can help avoid potential infections. Risk areas can hinder the spread of the virus to a certain extent.
  • sneezing can be detected by sound signals or cameras. Specifically, since a sneeze has a relatively special sound feature, the sneeze can be identified through the collected sound signal.
  • the method of identifying sneeze by sound signal is susceptible to the influence of environmental noise, and cannot detect another characteristic of sneezing, "droplets".
  • the sneeze is recognized by the camera. With the help of the high-definition camera, the sneeze detection based on vision can be realized through artificial intelligence, and the detection of droplets can also be realized to a certain extent.
  • the method of sneezing recognition through cameras is easily affected by external lighting conditions and obstructions, and cannot achieve all-weather perception.
  • Embodiments of the present application provide a method and a related device for sensing sneeze based on a wireless signal, which can utilize existing WLAN equipment and perform signal processing on wireless signals to realize sneeze droplet detection of sneezing, without being affected by light and air in the environment. effect of noise.
  • the present application provides a method for sensing sneeze based on a wireless signal, and the method can be applied to an access point/site in a WLAN or a cloud computing center.
  • the method includes: acquiring a wireless signal, and performing Doppler estimation on the wireless signal to obtain Doppler information of the wireless signal; and determining whether the first object is a sneeze droplet based on the Doppler information of the wireless signal.
  • the wireless signal propagates in the space including the first object.
  • the Doppler information of the wireless signal can be used to reflect the influence of the first object on the frequency of the wireless signal.
  • the first object may be droplets or small water droplets in space, or the like.
  • the above-mentioned Doppler information may include a time-Doppler spectrum, a range-Doppler-time spectrum, or Doppler information of other dimensions.
  • This scheme uses the special effect of sneeze droplets on the Doppler information of wireless signals to identify sneeze droplets, thereby identifying whether to sneeze or not. This process is not affected by light, noise and obstructions in the environment, and can Improve the applicability of droplet detection.
  • the above-mentioned wireless signal is also propagated in the space containing the second object, and the Doppler information of the wireless signal can also be used to reflect the frequency of the wireless signal by the second object
  • the influence produced by the first object is different from the influence produced by the second object on the frequency of the wireless signal.
  • the method may further include: determining whether the second object is a sneeze action based on the Doppler information of the wireless signal.
  • the second object may be a human body movement.
  • This scheme uses the Doppler information of multiple objects in the space to have different effects on the wireless signal to identify these multiple objects, which can not only realize the identification of sneezing droplets, but also realize the identification of sneezing. Combining sneeze droplets and sneezing actions for recognition can reduce misjudgments.
  • the method further includes: performing angle-of-arrival estimation, distance estimation and Doppler estimation on the wireless signal to obtain spatial position information of the first object and the spatial location information of the second object.
  • the spatial position information of the first object includes the first angle of arrival of the wireless signal reflected by the first object relative to the receiving device, the first distance between the first object and the receiving device, and the space of the second object.
  • the location information includes a second angle of arrival of the wireless signal reflected by the second object relative to the receiving device, and a second distance between the second object and the receiving device.
  • the method further includes: outputting one or more of the following information: whether the first object is a sneeze droplet, whether the second object is a sneeze action, the space of the first object Location information or spatial location information of the second object.
  • the way of outputting information can be directly sending the information to the relevant mobile device, or uploading the information to the cloud, and the cloud can remind the cleaning staff to clean according to the number of people sneezing, the scope of the sneeze droplets and other information. .
  • this program can remind relevant personnel of the occurrence area and influence scope of sneeze droplets and avoid potential infection risks.
  • the method further includes: acquiring an attenuation spectrum or a wideband spectrum of the wireless signal, which may be based on the Doppler information of the wireless signal and the wireless signal to determine whether the first object is a sneeze droplet or not.
  • the attenuation spectrum of the wireless signal can be used to reflect the influence of the first object on the amplitude attenuation of the wireless signal
  • the broadband spectrum of the wireless signal can be used to reflect the broadband spectrum energy generated by the first object on the wireless signal Impact.
  • This scheme not only considers the influence of sneeze droplets on the Doppler information of wireless signals, but also combines the influence of sneeze droplets on the attenuation spectrum/broadband spectrum of wireless signals to comprehensively judge whether they are sneeze droplets, which can further improve the accuracy sex.
  • determining whether the first object is a sneeze droplet based on the Doppler information of the wireless signal specifically includes: firstly performing feature extraction on the Doppler information of the wireless signal , obtain the first input feature; then input the first input feature into the classification model for processing, and output a classification result, the classification result being whether the first object is sneezing droplets.
  • determining whether the first object is a sneeze droplet based on the Doppler information of the wireless signal specifically includes: directly inputting the Doppler information of the wireless signal into a classification model process, and output a classification result, where the classification result is whether the first object is a sneeze droplet.
  • determining whether the first object is a sneeze droplet based on the Doppler information of the wireless signal specifically includes: firstly dividing the Doppler information of the wireless signal into the first One Doppler information and second Doppler information, the extension of the first Doppler information in the Doppler frequency domain is smaller than the extension of the second Doppler information in the Doppler frequency domain;
  • the first Doppler information is input into the first recognizer for recognition, to recognize whether the second object is a sneeze action; then the second Doppler information is input into the second recognizer for recognition, and the second object is recognized Whether there is a Doppler feature of the first object in the Doppler information; finally, whether the second object is a sneeze action and whether there is a Doppler feature of the first object in the second Doppler information is input into a judgment device to determine whether the first object is a sneeze droplet.
  • the present application provides an electronic device, comprising: a first acquisition module for acquiring a wireless signal, the wireless signal propagating in a space containing a first object; a first processing module for processing the wireless signal Doppler estimation to obtain the Doppler information of the wireless signal, the Doppler information of the wireless signal is used to reflect the influence of the first object on the frequency of the wireless signal; a first determination module is used to determine the frequency of the wireless signal based on the wireless signal.
  • the Doppler information of the signal determines whether the first object is a sneeze droplet.
  • the above-mentioned wireless signal can also be propagated in a space containing a second object, and the Doppler information of the wireless signal is also used to reflect the frequency of the wireless signal by the second object
  • the influence produced by the first object is different from the influence produced by the second object on the frequency of the wireless signal.
  • the above electronic device further includes a second determination module.
  • the second determining module is configured to determine whether the second object is a sneeze action based on the Doppler information of the wireless signal.
  • the above-mentioned first determining module and second determining module may be the same module or different modules.
  • the second object may be a human body movement.
  • the above electronic device may further include a second processing module.
  • the second processing module is used for performing angle of arrival estimation, distance estimation and Doppler estimation on the wireless signal to obtain the spatial position information of the first object and the spatial position information of the second object.
  • the spatial position information of the first object includes the first angle of arrival of the wireless signal reflected by the first object relative to the receiving device, the first distance between the first object and the receiving device, and the space of the second object.
  • the location information includes a second angle of arrival of the wireless signal reflected by the second object relative to the receiving device, and a second distance between the second object and the receiving device.
  • the above electronic device may further include an output module.
  • the output module is used to output one or more of the following information: whether the first object is a sneeze droplet, whether the second object is a sneeze action, the spatial position information of the first object or the space of the second object location information.
  • the above electronic device may further include a second acquisition module and a third determination module.
  • the second acquisition module is configured to acquire an attenuation spectrum or a wideband spectrum of the wireless signal, where the attenuation spectrum of the wireless signal is used to reflect the influence of the first object on the amplitude attenuation of the wireless signal, and the wideband spectrum of the wireless signal is is used to reflect the influence of the first object on the wideband spectrum energy of the wireless signal;
  • the third determining module is used to determine the first object based on the Doppler information of the wireless signal and the attenuation spectrum or wideband spectrum of the wireless signal Whether the subject is a sneeze droplet.
  • the above-mentioned first determining module 30 is specifically configured to: perform feature extraction on the Doppler information of the wireless signal to obtain the first input feature; input the first input feature into the classification Processing is performed in the model, and a classification result is output, and the classification result is whether the first object is a sneeze droplet.
  • the above-mentioned first determining module 30 is specifically configured to: input the Doppler information of the wireless signal into a classification model for processing, and output a classification result, where the classification result is the first Whether the subject is a sneeze droplet.
  • the above-mentioned first determining module 30 is specifically configured to: divide the Doppler information of the wireless signal into the first Doppler information and the second Doppler information, the first Doppler information
  • the extension of one Doppler information in the Doppler frequency domain is smaller than the extension of the second Doppler information in the Doppler frequency domain;
  • the first Doppler information is input into the first identifier for identification, and the identification find out whether the second object is a sneezing action;
  • Whether the second object is a sneeze action and whether there is a Doppler feature of the first object in the second Doppler information are input into a determiner to determine whether the first object is a sneeze droplet.
  • the present application provides another electronic device including a processor.
  • the processor is used for: acquiring a wireless signal, and performing Doppler estimation on the wireless signal to obtain Doppler information of the wireless signal; and determining whether the first object is a sneeze droplet based on the Doppler information of the wireless signal .
  • the wireless signal propagates in the space including the first object.
  • the Doppler information of the wireless signal can be used to reflect the influence of the first object on the frequency of the wireless signal.
  • the electronic device may further include a memory, which is coupled to the processor and stores necessary program instructions and data of the electronic device.
  • the present application provides a computer-readable storage medium storing instructions in the computer-readable storage medium, the instructions being executable by one or more processors on a processing circuit.
  • the computer When running on a computer, the computer is made to execute the method for sensing sneeze based on a wireless signal described in any of the above aspects.
  • the computer-readable storage medium may be a non-volatile readable storage medium.
  • the present application provides a computer program product containing instructions, which, when executed on a computer, cause the computer to execute the method for sensing sneeze based on a wireless signal described in any one of the above aspects.
  • the present application provides a chip or a chip system, including a processing circuit.
  • the processing circuit can be configured to perform the following operations: acquiring a wireless signal, and performing Doppler estimation on the wireless signal to obtain Doppler information of the wireless signal; and determining whether the first object is not based on the Doppler information of the wireless signal For sneeze droplets.
  • the wireless signal propagates in the space including the first object.
  • the Doppler information of the wireless signal can be used to reflect the influence of the first object on the frequency of the wireless signal.
  • the chip or chip system may further include an input and output interface.
  • the input and output interface can be used to output one or more of the following information: whether the first object is a sneeze droplet, whether the second object is a sneeze action, the spatial position information of the first object or the spatial position information of the second object.
  • existing WLAN devices can be used to realize sneeze droplet detection by performing signal processing on wireless signals, which is not affected by light and noise in the environment.
  • FIG. 1 is a system architecture diagram provided by an embodiment of the present application.
  • FIG. 2 is a schematic structural diagram of an AP or STA provided by an embodiment of the present application.
  • FIG. 3 is a schematic diagram of an application scenario provided by an embodiment of the present application.
  • FIG. 4 is a schematic flowchart of a method for sensing sneeze based on a wireless signal provided by an embodiment of the present application
  • FIG. 5 is a schematic diagram of the angle of arrival estimation provided by an embodiment of the present application.
  • FIG. 6 is a schematic diagram of distance estimation provided by an embodiment of the present application.
  • Fig. 7 is the schematic diagram of the distance-Doppler spectrum of sneeze droplet provided by the embodiment of the present application.
  • FIG. 8 is another schematic flowchart of the method for sensing sneeze based on a wireless signal provided by an embodiment of the present application
  • FIG. 9 is a schematic diagram of a signal attenuation spectrum provided by an embodiment of the present application.
  • FIG. 10 is a schematic diagram of a real sneeze Doppler measurement result provided by an embodiment of the present application.
  • 11 is a schematic diagram of the relationship between Doppler and bistatic angle provided by an embodiment of the present application.
  • FIG. 12 is a schematic diagram of a scenario of information output provided by an embodiment of the present application.
  • FIG. 13 is a schematic diagram of an AR-based related information notification provided by an embodiment of the present application.
  • FIG. 14 is a schematic diagram of the synthesis of the true velocity of sneeze droplets provided in the embodiment of the present application.
  • FIG. 15 is an example flow chart provided by an embodiment of the present application.
  • 16 is a schematic diagram of a multi-base joint sensing scenario provided by an embodiment of the present application.
  • FIG. 17 is another example flow diagram provided by an embodiment of the present application.
  • FIG. 19 is another schematic structural diagram of an electronic device provided by an embodiment of the present application.
  • the following describes the system architecture and/or application scenarios of the method for sneezing based on a wireless signal provided by the embodiments of the present application. It is understandable that the scenarios described in the embodiments of the present application are for the purpose of illustrating the technical solutions of the embodiments of the present application more clearly, and do not constitute a limitation on the technical solutions provided by the embodiments of the present application.
  • the embodiment of the present application provides a method for sensing sneeze based on a wireless signal, which can utilize existing WLAN equipment without additional microphones, cameras and other equipment, and perform signal processing on radio frequency signals (or wireless signals) to realize sneezing. Localization, identification and droplet detection can be unaffected by light and noise in the environment.
  • the method can be applied to a wireless communication system, which can be a wireless local area network or a cellular network; the method can be implemented by a communication device in the wireless communication system or a chip or processor in the communication device.
  • the communication device may be an access point (access point, AP) device or a station (station, STA) device.
  • the access point device and the station device can be either single-link devices or multi-link devices.
  • FIG. 1 is a system architecture diagram provided by an embodiment of the present application.
  • the system architecture includes at least two WLAN devices (such as AP1 and STA2 in FIG. 1 ), wherein one WLAN device (such as STA2 ) transmits RF signals, and the other WLAN devices (such as AP1 ) receive RF signals.
  • the system architecture may further include a cloud computing center.
  • the WLAN device may support a WLAN communication protocol, and the communication protocol may include IEEE 802.11be (or called Wi-Fi 7, EHT protocol), and may also include IEEE 802.11ax, IEEE 802.11ac and other protocols.
  • the communication protocol may also include the next-generation protocol of IEEE 802.11be, and the like.
  • the device implementing the method of the present application may be an AP or STA in a WLAN, or a chip or a processing system installed in the AP or STA, or a cloud computing center.
  • WLAN devices can realize sneeze location, recognition and sneeze droplet detection by sending and receiving Wi-Fi signals, and related calculations can be completed in the WLAN AP; they can also be uploaded to the cloud computing center for processing using powerful cloud computing capabilities. Understandably, compared with AP-side processing, cloud-based processing has powerful computing power and greater flexibility. It can adapt corresponding perception algorithms to different situations to maximize perception performance.
  • radio frequency signal can be used interchangeably, and all refer to signals propagated in a wireless manner.
  • An access point (such as AP1) is a device with wireless communication function, supports communication using WLAN protocol, and has the function of communicating with other devices (such as stations or other access points) in the WLAN network. The ability to communicate with other devices.
  • an access point may be referred to as an access point station (AP STA).
  • the device with wireless communication function can be a complete device, or a chip or a processing system installed in the complete device. The device with these chips or processing system installed can be controlled by the chip or the processing system.
  • the AP in this embodiment of the present application is a device that provides services for the STA, and can support the 802.11 series of protocols.
  • the AP can be a communication entity such as a communication server, router, switch, and bridge; the AP can include various forms of macro base stations, micro base stations, relay stations, etc.
  • the AP can also be the chips and processing devices in these various forms of equipment. system, so as to implement the methods and functions of the embodiments of the present application.
  • a station for example, STA2 is a device with wireless communication function, supports communication using the WLAN protocol, and has the ability to communicate with other stations or access points in the WLAN network.
  • a station can be referred to as a non-access point station (non-access point station, non-AP STA).
  • STA is any user communication device that allows the user to communicate with the AP and then communicate with the WLAN.
  • the device can be a complete device, or a chip or processing system installed in the complete device. These chips or processing systems are installed. Devices of the system may implement the methods and functions of the embodiments of the present application under the control of a chip or a processing system.
  • the STA may be a tablet computer, a desktop computer, a laptop computer, a notebook computer, an Ultra-mobile Personal Computer (UMPC), a handheld computer, a netbook, a Personal Digital Assistant (PDA), a mobile phone, etc.
  • UMPC Ultra-mobile Personal Computer
  • PDA Personal Digital Assistant
  • FIG. 2 is a schematic structural diagram of an AP or a STA provided by an embodiment of the present application. As shown in FIG.
  • an AP or STA may include: an application (application) layer module, a transmission control protocol (transmission control protocol, TCP)/user datagram protocol (user datagram protocol, UDP) processing module, an internet protocol (internet protocol) protocol, IP) processing module, logical link control (logical link control, LLC) module, media access control (media access control, MAC) layer module, physical (physical, PHY) layer baseband module, radio frequency radio and antenna.
  • the AP or STA shown in FIG. 2 may be either a single-antenna structure or a multi-antenna structure, which is not limited in this embodiment of the present application.
  • the WLAN system can provide high-speed and low-latency transmission.
  • the WLAN system will be applied in more scenarios or industries, such as the Internet of Things industry, the Internet of Vehicles industry, or the Banking industry, used in corporate offices, stadiums and exhibition halls, concert halls, hotel rooms, dormitories, wards, classrooms, supermarkets, squares, streets, production workshops and warehousing, etc.
  • devices that support WLAN communication can be sensor nodes in smart cities (such as smart water meters, smart electricity meters, and smart air detection nodes), smart devices in smart homes (such as smart cameras, projectors, etc.) devices, display screens, TV sets, stereos, refrigerators, washing machines, etc.), nodes in the Internet of Things, entertainment terminals (such as AR, VR and other wearable devices), smart devices in smart office (such as printers, projectors, Amplifiers, stereos, etc.), IoV devices in the Internet of Vehicles, infrastructure in daily life scenarios (such as vending machines, self-service navigation desks in supermarkets, self-service cash registers, self-service ordering machines, etc.), and large-scale sports Or equipment for music venues, etc.
  • the specific forms of the STA and the AP are not particularly limited in the embodiments of the present application, which are only exemplary descriptions herein.
  • Embodiment 1 of the present application describes a method for sensing sneeze based on a wireless signal provided by the present application in combination with a possible application scenario.
  • the application scenario of the embodiment of the present application takes two devices as an example, one of which is a sending device and the other is a receiving device.
  • the "transmitting device” mentioned in this application refers to a WLAN device that transmits wireless signals
  • the "receiving device” refers to a WLAN device that receives wireless signals.
  • the transmitting device in this application can both transmit signals and receive signals. Because it is mainly used to transmit wireless signals in this application, it is called a transmitting device; the receiving device in this application can both receive A signal can also transmit a signal. In this application, it is mainly a function of receiving wireless signals, so it is called a receiving device.
  • FIG. 3 is a schematic diagram of an application scenario provided by an embodiment of the present application.
  • there are at least two WLAN devices in the space namely WLAN device 1 and WLAN device 2, and a target person (target human in Figure 3) is sneezing in the space, and there is a sneeze and a Generated sneeze droplets.
  • the WLAN device 1 sends a wireless signal, and the wireless signal can reach the WLAN device 2 through a direct path/direct path, or can reach the WLAN device 2 after being reflected by a target human, and can also reach the WLAN device 2 after being reflected by sneeze droplets.
  • the wireless signal received by the WLAN device 2 is a superposition of multipath signals (including the direct path in FIG. 3 , two reflection paths and other multipath signals not shown in FIG. 3 ).
  • the WLAN device 2 can perform various signal processing on the received wireless signal, through the influence of the sneezing action of the target object (such as the target person in FIG.
  • the wireless signal for example, the influence on the frequency of the wireless signal
  • the influence of sneeze droplets on wireless signals for example, the effect on the frequency, amplitude attenuation or broadband spectrum energy of wireless signals, etc.
  • sneeze droplets on wireless signals for example, the effect on the frequency, amplitude attenuation or broadband spectrum energy of wireless signals, etc.
  • the WLAN device 2 may also transmit the received wireless signal to the cloud computing center for processing, and the cloud computing center performs signal processing on the wireless signal, and the like.
  • WLAN device 1 and WLAN device 2 may be both APs and STAs, and one WLAN device may be an AP and the other WLAN device may be a STA.
  • WLAN device 1 may be a STA and WLAN device 2 may be an AP.
  • FIG. 4 is a schematic flowchart of a method for sensing sneeze based on a wireless signal provided by an embodiment of the present application.
  • the method for sensing sneeze based on wireless signals includes but is not limited to the following steps:
  • a first device acquires a wireless signal, and the wireless signal propagates in a space containing a first object.
  • the first device when the first device is a receiving device, the first device receives/collects wireless signals, and performs preprocessing on the received/collected wireless signals, such as filtering.
  • the receiving device receives/collects wireless signals, and can send the received/collected wireless signals to the cloud computing center, and the cloud computing center performs subsequent signal processing processes/steps. Understandably, the cloud computing center has powerful computing power and greater flexibility, and can adapt the corresponding perception algorithm to different situations to maximize the perception performance.
  • a large bandwidth is required to support data transmission.
  • the wireless signal received/collected by the above-mentioned receiving device may be a superposition of multi-path/multi-path signals, and at least one signal in the wireless signal is obtained after being reflected by the first object. Understandably, after the sending device sends the original wireless signal, the original wireless signal propagates in the space containing the first object, so the original wireless signal reaches the receiving device after being reflected by at least the first object. Optionally, the original wireless signal may also reach the receiving device through a direct path.
  • the first object may be droplets, small water droplets in space, or the like.
  • step S102 Doppler estimation may be performed on the preprocessed wireless signal.
  • the first device performs Doppler estimation on the wireless signal to obtain Doppler information of the wireless signal, where the Doppler information of the wireless signal is used to reflect the influence of the first object on the frequency of the wireless signal.
  • the Doppler information of the wireless signal may be used to reflect the influence of the first object on the frequency of the wireless signal.
  • the above-mentioned Doppler information of the wireless signal may include time-Doppler spectrum, range-Doppler-time spectrum, or Doppler information of other dimensions.
  • the first device may perform Doppler estimation on the above wireless signal to obtain a time-Doppler spectrum of the wireless signal.
  • the first device may perform signal processing such as angle-of-arrival estimation, distance estimation, Doppler estimation, or multi-dimensional joint processing on the above-mentioned wireless signal to obtain the distance-Doppler-time spectrum of the wireless signal or Doppler of other dimensions. information.
  • the Doppler information of the wireless signal can respectively reflect the influence of the multiple objects on the frequency of the wireless signal, that is, a The motion characteristics of multiple objects can be reflected on the Doppler spectrum.
  • the moving droplets will produce Doppler modulation on the signals propagating in the space, so the received/collected wireless signals are analyzed accordingly, such as time- Frequency analysis can find Doppler time-varying features in the time-Doppler dimension, these features can reflect the real movement of droplets, so Doppler information can be used to identify droplets.
  • moving droplets not only have unique characteristics in the time-Doppler dimension; the movement of droplets can also generate corresponding information with corresponding droplet movement characteristics in other dimensions, for example, in the time-distance dimension. On the other hand, it can reflect the information that the distance changes with time, and this information can also assist the subsequent identification of sneeze droplets.
  • time-angle information, or other joint dimension information which can also assist in the identification of subsequent sneeze droplets.
  • the angle of arrival estimation can be estimated by beamforming algorithm, subspace method or other array signal processing methods
  • the distance estimation can be estimated by signal processing algorithms such as matched filtering
  • the Doppler estimation can be estimated by using signals such as Fourier transform Processing algorithm to estimate.
  • the principles of angle of arrival estimation, distance estimation and Doppler estimation are introduced below.
  • FIG. 5 is a schematic diagram of the angle of arrival estimation provided by the embodiment of the present application.
  • the spacing between the array elements is d.
  • the receiving array is in the far field of the signal (that is, when the signal arrives at the antenna array in the form of a plane wave)
  • the path difference of the plane wave reaching different array elements is dsin ⁇
  • the corresponding received signal produces a corresponding phase difference between the receiving array elements.
  • the beamforming algorithm and the subspace class method as an example. It is understandable that the "received signal” mentioned in this application refers to a wireless signal received by a receiving device.
  • Beamforming Use an antenna to form a beam and scan the space. When the energy in a certain direction is strong, it can be determined that there is a signal in this direction.
  • the beam width (ie, beam resolution) formed by such methods is limited by the aperture of the antenna array, and the larger the aperture, the higher the resolution.
  • MUSIC multiple signal classification method
  • x represents the signal received by the receiving array
  • s represents the source signal, that is, the signal sent by the transmitter
  • A represents the steering vector matrix
  • n represents the noise.
  • L represents the number of snapshots for joint processing, Represents the estimated covariance matrix based on L snapshots.
  • the steering vector of the received signal is orthogonal to the decomposed signal-noise subspace. Based on this conclusion, the following formula (1-5) can be used to search for spectral peaks and to estimate the angle of arrival.
  • a represents the steering vector of the receiving array.
  • C represents the complex number set
  • M represents the number of array elements of the receiving array
  • K represents the number of beam directions
  • d represents the distance between the array elements
  • ⁇ represents the incident angle of the signal
  • ⁇ is the wavelength
  • the subspace class method also includes methods such as estimation of signal parameters via rotation invariance techniques (ESPRIT).
  • ESPRIT rotation invariance techniques
  • the angle of arrival estimation can also be performed by the sparse class method.
  • FIG. 6 is a schematic diagram of distance estimation provided by an embodiment of the present application.
  • the device 1, the target object and the device 2 constitute a typical WLAN perception scene.
  • the device 1 sends a wireless signal, and the wireless signal is received by the device 2 after being reflected by the target object, and the device 2 can also receive the direct path signal directly sent by the device 1. Therefore, when the device 1 sends a wireless signal, the signal received by the device 2 is the superposition of the direct path signal and the reflection path signal (for ease of description, FIG. 6 takes one reflection path as an example for description).
  • the device 2 Using time-domain matched filter processing (ie autocorrelation) on the reference signal and the received signal, two peaks can be obtained on the time-energy spectrum.
  • the first peak represents the direct path signal
  • the second peak represents the reflected path signal.
  • the time difference between these two peaks can represent the propagation delay difference between the reflection path and the direct path, which corresponds to the propagation distance difference in the real environment.
  • the target object is focusing on the transceiver device, and the target object and device 1 and The sum of the distances between the devices 2 is on the ellipse of the sum of the direct path distance and the difference of the calculated propagation distance. Combined with the angle information of the angle of arrival obtained earlier, the positioning of the target object can be realized.
  • matched filtering can be processed in both the time domain and the frequency domain.
  • Matched filtering is a processing method in radar, but the method can also be applied to channel estimation in communications. It is also understandable that in addition to distance estimation based on matched filtering, distance estimation may also be performed by other signal processing methods, which will not be described here.
  • the distance in the above-mentioned distance-Doppler-time spectrum may refer to the difference in propagation distance between the reflection path and the direct path.
  • the Doppler estimate can be processed directly using the Fourier transform. Specifically, a time domain signal can be transformed into the frequency domain through Fourier transform, and its spectral energy can be analyzed to estimate its Doppler component. Generally, due to the uncertainty principle, the longer the duration of the time domain signal, the higher the resolution in the frequency domain.
  • T represents the duration of the time domain signal
  • ⁇ f represents the frequency interval
  • the first device determines whether the first object is a sneeze droplet based on the Doppler information of the wireless signal.
  • the Doppler information of the above wireless signal is a time-Doppler spectrum
  • the time-Doppler spectrum is a two-dimensional feature
  • the time-Doppler spectrum can be used as an image, and the neural network for training and recognition. That is, the first device can directly input the Doppler information of the wireless signal into the classification model for processing, and obtain the classification result output by the classification model, and the classification result can be whether the first object is sneeze droplets.
  • the wireless signal can also be analyzed in other dimensions, such as: the three-dimensional information of the distance-Doppler-time spectrum is used as a further input for classification information about the model.
  • the classification model may be a convolutional neural network model. Assuming that the input layer of the convolutional neural network is two-dimensional information (that is, a picture or a two-dimensional matrix), the first device can expand the Doppler information of the above wireless signal into a two-dimensional matrix, input it into the neural network, and pass the subsequent The convolution layer performs feature extraction, and then the classification layer is used to classify, and finally determine whether the first object is sneeze droplets.
  • the input layer of the convolutional neural network is two-dimensional information (that is, a picture or a two-dimensional matrix)
  • the first device can expand the Doppler information of the above wireless signal into a two-dimensional matrix, input it into the neural network, and pass the subsequent
  • the convolution layer performs feature extraction, and then the classification layer is used to classify, and finally determine whether the first object is sneeze droplets.
  • the first device may perform feature extraction on the Doppler information of the wireless signal to obtain a first input feature, and then input the first input feature into a classification model for processing to obtain a classification result output by the classification model.
  • the classification result may be whether the first object is a sneeze droplet.
  • the classification model can be a deep neural network or another classifier.
  • FIG. 7 is a schematic diagram of a distance-Doppler spectrum of sneeze droplets provided in an embodiment of the present application. Because there is a corresponding relationship between velocity and Doppler component, velocity can be used to reflect Doppler information. As shown in FIG. 7 , the horizontal axis represents speed (unit m/s), and the vertical axis represents distance (unit m). The rightmost rectangular bar (gray bar) in Figure 7 represents the size of the energy dimension, and different gray levels represent different energy sizes. Sneeze droplets usually have higher velocity (Doppler characteristics of sneeze droplets generally appear at higher Doppler positions), and sneeze droplets have a larger spread in the Doppler frequency domain.
  • the first device can extract the Doppler information corresponding to the first object from the Doppler information of the wireless signal, and can input the extracted Doppler information corresponding to the first object into a two-dimensional neural network identification in the network.
  • the first device can obtain the recognition result output by the recognizer, that is, whether there is a Doppler feature of sneeze droplets in the Doppler information corresponding to the first object, and can input the recognition result into the decision device for judgment. Whether the first object is sneeze droplets.
  • the Doppler information corresponding to the first object if there is a Doppler feature of sneeze droplets in the Doppler information corresponding to the first object, it means that the first object is sneeze droplets; otherwise, if the Doppler information corresponding to the first object contains There is no Doppler feature of sneeze droplets, indicating that the first object is not sneeze droplets.
  • the first device may input the entire Doppler information of the wireless signal into a two-dimensional neural network identifier for identification, without extracting part of the Doppler information.
  • the first device can obtain the recognition result output by the recognizer, that is, whether there is a Doppler feature of sneeze droplets in the Doppler information of the wireless signal, and can input the recognition result into a judger for judgment, and judge the first device. Whether an object is a sneeze droplet.
  • the first device may perform template matching on the Doppler information of the above wireless signal to determine whether the first object is a sneeze droplet. For example, the first device may match the Doppler information of the wireless signal with the Doppler information of the sneeze droplets. If the degree of matching/similarity between the Doppler information of the wireless signal and the Doppler information of the sneeze droplets is greater than a threshold, it can be determined that the first object is the sneeze droplets.
  • the first object is sneeze droplets
  • the special influence of the sneeze droplets on the Doppler information of the wireless signal is used to identify the sneeze droplets, so as to identify whether the sneeze is a sneeze. impact and improve the applicability of droplet detection.
  • the receiving device may receive/collect wireless signals, and perform preprocessing on the received/collected wireless signals, such as filtering, etc.; and may perform Doppler estimation on the preprocessed wireless signals to obtain Doppler information of the wireless signal.
  • the receiving device sends the Doppler information of the wireless signal to the cloud computing center, and the cloud computing center determines whether the first object is a sneeze droplet based on the Doppler information of the wireless signal.
  • the Doppler information of the wireless signal is sent to the cloud computing center for processing, and the powerful computing capability of the cloud computing center can be utilized, and the computing complexity of the receiving device can be reduced.
  • the moving limbs when a person performs a corresponding action, the moving limbs will produce Doppler modulation on the signal in space, so corresponding analysis is performed on the received/collected wireless signal, such as time-frequency analysis , the time-dependent Doppler features can be found in the time-Doppler dimension, which correspond to the real body movements of the human body and can be used for further action recognition. Therefore, the method for sensing sneeze based on a wireless signal provided in the embodiments of the present application can be used not only for sensing sneezing, but also for sensing coughing. Specifically, the first device may acquire a wireless signal, and the wireless signal propagates in a space containing a third object, and the third object may be a coughing action.
  • the first device may perform signal processing such as angle of arrival estimation, distance estimation, Doppler estimation, or multi-dimensional joint processing on the wireless signal to obtain Doppler information of the wireless signal.
  • the Doppler information of the wireless signal can be used to reflect the influence of the third object on the frequency of the wireless signal.
  • the first device may determine whether the third object is coughing based on the Doppler information of the wireless signal.
  • the second embodiment of the present application combines sneezing droplets and sneezing actions to identify, which can reduce misjudgment;
  • the influence on the attenuation spectrum/broadband spectrum of the wireless signal is used to comprehensively judge whether it is sneeze droplets, and further improve the accuracy.
  • the sneeze droplet can be identified through the attenuation spectrum/broadband spectrum of the wireless signal, realizing the sneeze droplet identification/sneeze perception in extreme cases.
  • FIG. 8 is another schematic flowchart of the method for sensing sneeze based on a wireless signal provided by an embodiment of the present application.
  • the method for sensing sneeze based on wireless signals includes but is not limited to the following steps:
  • the first device acquires a wireless signal, and the wireless signal propagates in a space including a first object and a second object.
  • the above-mentioned first object may be droplets, small water droplets, etc. in the space; the above-mentioned second object may be human body movements.
  • step S201 in this embodiment of the present application, reference may be made to the implementation of step S101 in the embodiment shown in FIG. 4 , and details are not described herein again.
  • the first device may perform signal preprocessing, such as filtering, on the wireless signal.
  • the preprocessed wireless signal can be processed in three ways, such as the following steps S202, S203 and S204. It is understandable that the wireless signals in the following steps S202, S203 and S204 may be preprocessed wireless signals.
  • the first device performs angle of arrival estimation, distance estimation, and Doppler estimation on the wireless signal, and determines the spatial position information of the first object and the spatial position information of the second object.
  • the first device performs angle of arrival estimation (such as beamforming algorithm, subspace method or other array signal processing methods, etc.), distance estimation (such as matched filtering and other signal processing algorithms), and Doppler estimation for the above-mentioned wireless signal (such as Fourier transform and other signal processing algorithms) and other signal processing, the sneeze (ie the second object) and the droplet (ie the first object) are located in the relevant signal dimensions and regions, that is: determine the first object and the spatial position information of the second object.
  • the angle of arrival estimation, the distance estimation, and the Doppler estimation may refer to the corresponding descriptions in the foregoing Embodiment 1, which will not be repeated here.
  • the spatial position information of the first object may include the first angle of arrival of the wireless signal reflected by the first object relative to the receiving device, and the first distance between the first object and the receiving device; the spatial position information of the second object It may include a second angle of arrival of the wireless signal reflected by the second object relative to the receiving device, and a second distance between the second object and the receiving device.
  • the location of moving objects (such as the first object and the second object) generally needs the help of Doppler for detection.
  • the sneezing action usually has a lower velocity (the Doppler feature of the sneezing action generally appears at a lower Doppler position), and the extension of the sneezing action in the Doppler frequency domain Relatively small; sneeze droplets usually have higher velocity (Doppler characteristics of sneeze droplets generally appear at higher Doppler positions), and the spread of sneeze droplets in the Doppler frequency domain is larger.
  • the position of the sneeze action (here refers to the spatial coordinates or distance) is not much different from the starting position of the sneeze droplets (here refers to the spatial coordinates or distance). Therefore, in the initial stage, the position of the sneeze action and the starting position of the sneeze droplets can be equivalent to one position. However, with the change of time, the difference between the position of the sneeze action and the position of the sneeze droplets gradually increased.
  • a possible locating process for the first object and the second object includes: (a) performing range-Doppler processing on the received signal of each antenna to generate the result shown in FIG. 7 above.
  • the horizontal axis is Doppler or velocity, and the vertical axis is distance.
  • the distance here refers to the propagation distance difference obtained after the distance estimation process.
  • (b) Perform incoherent accumulation (direct matrix superposition) of the range-Doppler maps obtained by multiple antennas to improve the signal-to-noise ratio. After stacking, a range-Doppler map is obtained.
  • the first device performs angle of arrival estimation, distance estimation and Doppler estimation on the wireless signal to obtain Doppler information of the wireless signal, where the Doppler information of the wireless signal is used to reflect the first object and the second object effect on the frequency of the wireless signal.
  • the first device performs signal processing such as angle-of-arrival estimation, distance estimation, Doppler estimation, or multi-dimensional joint processing on the wireless signal to obtain Doppler information of the wireless signal. Subsequently, based on the Doppler information of the wireless signal, the sneeze action and sneeze droplet can be identified in the relevant signal dimension.
  • the Doppler information of the wireless signal can be used to reflect the effects of the first object and the second object on the frequency of the wireless signal.
  • the influence of the first object on the frequency of the wireless signal is not the same as the influence of the second object on the frequency of the wireless signal, that is, the Doppler feature corresponding to the first object is more than that of the second object. Puller characteristics are not the same.
  • the Doppler information of the wireless signal may include time-Doppler dimension, distance-Doppler-time dimension, or Doppler information of other dimensions. Therefore, in the process of obtaining the Doppler information of the wireless signal, angle information (here, the angle of arrival) and/or distance information (here, the propagation distance difference) may also be required. Taking people as an example, when people walk or perform corresponding actions, the moving limbs will produce Doppler modulation of the signal, so the corresponding analysis of the received/collected signal, such as time-frequency analysis, can be done in time-Doppler. In the Le dimension, the time-varying features of Doppler are found, which can correspond to the real body movements of the human body and can be used for further action recognition.
  • the moving target (here refers to the first object or the second object) not only has unique characteristics in the time-Doppler dimension; the moving target can also generate corresponding information with corresponding target movement characteristics in other dimensions, For example, in the time-distance dimension, it can reflect the information that the distance changes with time, and this information can also assist in the identification of subsequent sneezing actions and sneezing droplets.
  • the moving target also has time-angle information, or other joint dimension information, which can also assist in the identification of subsequent sneezing actions and sneezing droplets.
  • the first device acquires an attenuation spectrum or a broadband spectrum of the wireless signal, where the attenuation spectrum of the wireless signal is used to reflect the influence of the first object on the amplitude attenuation of the wireless signal, and the broadband spectrum of the wireless signal is used to reflect the The effect of the first object on the broadband spectral energy of the wireless signal.
  • the first device may perform signal processing such as moving average on the wireless signal, and obtain an attenuation spectrum or a wideband spectrum of the wireless signal after signal processing such as the moving average.
  • the attenuation spectrum of the wireless signal can be used to reflect the influence of the first object on the amplitude attenuation of the wireless signal
  • the broadband spectrum of the wireless signal can be used to reflect the influence of the first object on the broadband spectrum energy of the wireless signal.
  • the attenuation spectrum may be information in the time-energy dimension
  • the broadband spectrum may be information in the frequency-energy dimension or information in the frequency-energy-time dimension.
  • FIG. 9 is a schematic diagram of a signal attenuation spectrum provided by an embodiment of the present application. As shown in Figure 9, the horizontal axis is time, and the vertical axis is received signal energy.
  • the broadband spectrum of a signal refers to the broadband spectrum characteristics of the signal, that is, the horizontal axis is the frequency, and the vertical axis is the energy of the corresponding frequency. Under the condition of large bandwidth, different signals/electromagnetic waves pass through different media.
  • the broadband spectrum characteristics formed by the signal passing through the sneeze droplets also have specific characteristics, so these information can be used as the input information for subsequent sneeze droplet identification, and the sneeze droplet identification and detection can be carried out.
  • the above-mentioned steps S202, S203 and S204 may be performed in parallel in three steps; or two of the steps may be performed in parallel, and the other step may be performed before the two steps performed in parallel, or may be performed in parallel.
  • the first device determines whether the first object is a sneeze droplet and the second object is a sneezing action based on the Doppler information of the wireless signal and the attenuation spectrum/broadband spectrum of the wireless signal.
  • the information in dimensions such as time-Doppler or distance-time-Doppler of the wireless signal can be obtained, Combine the attenuation spectrum (information in the time-energy dimension) or broadband spectrum (information in the dimension of frequency-energy or frequency-energy-time) of the wireless signal, and input it as a whole to a recognizer (which can be a neural network, or It is a non-neural network) for joint identification, to identify whether the second object is a sneeze action and whether the first object is a sneeze droplet.
  • a recognizer which can be a neural network, or It is a non-neural network
  • the input layer of the convolutional neural network is two-dimensional information (that is, a picture or a two-dimensional matrix)
  • all the feature information here refers to the Doppler information, attenuation spectrum or broadband spectrum of the wireless signal
  • the dimensional matrix is input to the convolutional neural network, and the feature extraction is carried out through the subsequent convolutional layer, and then the classification layer is used to classify, and finally determine whether there is a sneeze action and produce sneeze droplets.
  • the first device may also input the recognition result of the sneeze action (that is, whether the second object is a sneeze action) and the recognition result of the sneeze droplets (whether the first object is a sneeze droplet) to the subsequent determiner. , to provide information for the final judgment.
  • the first device may perform identification based on a two-dimensional neural network for various information (including Doppler information, attenuation spectrum or broadband spectrum information of wireless signals, etc.)
  • the sneezing action, whether there is the Doppler feature of the sneeze droplet, whether there is the attenuation feature of the sneeze droplet) are combined to judge.
  • the first device can input the above-mentioned Doppler information of the wireless signal into a two-dimensional neural network recognizer for recognition, and output recognition results, such as whether there is a sneeze action and whether there is Doppler of sneezing droplets. feature.
  • the first device can input the attenuation spectrum or broadband spectrum of the wireless signal into another two-dimensional neural network identifier for identification, and output identification results, such as whether there is an attenuation characteristic of sneeze droplets.
  • FIG. 10 is a schematic diagram of a real sneeze Doppler measurement result provided by an embodiment of the present application.
  • the target person sits relaxedly in front of the experimental equipment (here refers to the sending and receiving equipment), sneezes and produces sneeze droplets.
  • the horizontal axis is time
  • the vertical axis is speed. It can be seen that there is a relatively obvious speed component at the position where the speed is 5 m/s or more.
  • These velocity components are much larger than the normal motion velocity of the human body, so it can be judged whether there are sneeze droplets by detecting at high-speed positions.
  • these velocity components also have unique time-velocity characteristics, which can be used as valid input information for identification to further determine whether sneeze droplets appear.
  • the recognition result output by the recognizer may be a 2-bit binary, for example, the high bit indicates whether the second object is a sneeze action, and the low bit indicates whether the first object is sneezing droplets. For example, “10” indicates that the second object is a sneeze motion and the first object is not a sneeze droplet, "11” indicates that the second object is a sneeze action and the first object is a sneeze droplet, and "00” indicates that the second object is not Sneeze action and the first object is not sneeze droplets.
  • the recognition result output by the recognizer can have three values, namely "00”, “10” and “11”, and only when the value is "11", the result output by the decider is "1" ”, indicating that there are both sneezing actions and sneezing droplets.
  • the first device may perform feature extraction and feature fusion on the Doppler information of the wireless signal and the attenuation spectrum/broadband spectrum of the wireless signal to obtain the input feature.
  • the first device may input the input feature into a classifier for processing, and obtain a classification result output by the classifier, such as whether the first object is sneezing droplets and whether the second object is a sneezing action.
  • the first device can perform feature extraction separately for various information (including information such as Doppler information, attenuation spectrum, or broadband spectrum of wireless signals), and jointly input the extracted signal features into the subsequent classification network. Among them, sneezing action recognition and sneezing droplet identification are performed, and finally it is determined whether a sneezing action occurs and sneezing droplets are generated.
  • the embodiments of the present application combine sneezing droplets and sneezing actions for identification, which can reduce misjudgments caused by other external reasons, such as when measuring Doppler, the droplets sprayed from a watering can and the like are different from each other.
  • the characteristics of sneeze droplets have certain similarities. Combined with the action recognition results of sneezing, the misjudgment caused by such situations can be reduced.
  • identifying with the action of sneezing can reduce misjudgments caused by similar attenuation spectra caused by other reasons.
  • the embodiment of the present application not only considers the influence of the sneeze droplets on the Doppler information of the wireless signal, but also combines the influence of the sneeze droplets on the attenuation spectrum/broadband spectrum of the wireless signal, that is, using Doppler and attenuation It can comprehensively judge whether it is a sneeze droplet, and further improve the accuracy.
  • FIG. 11 is a schematic diagram of the relationship between the Doppler and the bistatic angle provided by the embodiment of the present application. As shown in FIG. 11 , FIG. 11 shows the relationship between the Doppler information that can be perceived through the wireless signal and the bistatic angle ⁇ formed by the transmitting device, the target, and the receiving device. Among them, the Doppler f d perceived by the wireless signal and the bistatic angle ⁇ satisfy the following formula:
  • v represents the moving speed of the target (such as the first object or the second object)
  • ⁇ represents the wavelength of the carrier wave
  • ⁇ represents the movement direction of the target and the angle bisector of the bistatic angle ⁇ .
  • this embodiment of the present application further includes step S206: the first device outputs one or more of the following information: whether the first object is a sneeze droplet, and whether the second object is a sneezing action , the spatial position information of the first object or the spatial position information of the second object.
  • the first device can send one or more of the following information to the mobile device associated with the first device: whether the first object is a sneeze droplet, whether the second object is a sneeze action, the spatial position of the first object information or spatial location information of the second object.
  • the first device can also use the identification results and positioning results of the first device (such as whether the first object is a sneeze droplet, whether the second object is a sneeze action, the spatial position information of the first object or the second object).
  • FIG. 12 is a schematic diagram of a scenario of information output provided by an embodiment of the present application.
  • the first device is a WLAN AP
  • the WLAN AP can directly send the information (the information determined in the above steps S202 and S205) to the mobile device, and can also transmit the information to the background server, and the background server can send the information according to the Information such as the number of people sneezing and the scope of the sneeze droplets are notified to the mobile terminal.
  • the notification of relevant information may be in the form of text push, or may be combined with technologies such as augmented reality (AR)/virtual reality (VR) to mark the range of sneeze droplets for reminder.
  • AR augmented reality
  • VR virtual reality
  • FIG. 13 is a schematic diagram of an AR-based related information notification provided by an embodiment of the present application. As shown in Figure 13, the size and position of the virtual target person and the sneeze droplets in the AR environment and the movement information (such as the movement trajectory) of the sneeze droplets can be realized on the mobile device through AR and other related technologies.
  • the embodiment of the present application can remind relevant personnel of the occurrence area and influence range of sneeze droplets, and avoid potential infection risks.
  • FIG. 14 is a schematic diagram of the real velocity synthesis of sneeze droplets provided in the embodiment of the present application.
  • device 1 is a device that sends wireless signals
  • device 2 and device 3 are devices that receive/collect wireless signals.
  • the velocity v 2 observed by the device 2 can be obtained as:
  • f d represents the Doppler perceived by device 2 through wireless signals
  • ⁇ represents the bistatic angle formed by device 1, sneeze droplets and device 2
  • ⁇ represents the movement of sneeze droplets The angle between the direction and the bisector of the bistatic angle ⁇ .
  • the velocity v3 observed by device 3 is:
  • f' d represents the Doppler perceived by device 3 through wireless signals
  • ⁇ ' represents the bistatic angle formed by device 1
  • ⁇ ' represents the sneeze fly The angle between the movement direction of the foam and the bisector of the bi-base angle ⁇ '.
  • the real speed and direction of sneeze droplets can be synthesized. If further combined with the motion model of the sneeze droplet itself, the diffusion model of the sneeze droplet can be better estimated, so as to determine the approximate scope of influence of the sneeze droplet.
  • FIG. 15 is an exemplary flowchart provided by an embodiment of the present application.
  • the preprocessing of the signal is performed first, and the preprocessing includes steps such as filtering.
  • the preprocessed signal is processed in three different ways. The first processing: After Doppler estimation, angle of arrival estimation, distance estimation, joint estimation and other signal processing steps, sneeze (action) and sneeze droplets are located.
  • the second processing through time-frequency joint processing, distance-frequency-time joint processing, etc., time-Doppler spectrum and distance-Doppler-time spectrum of sneezing (action) and sneeze droplets are obtained.
  • the third processing After signal processing such as moving average, signal attenuation spectrum/broadband spectrum analysis is performed.
  • the received/collected wireless signals are first preprocessed, and then three types of processing are performed on the preprocessed wireless signals; Estimation and Doppler estimation to realize sneezing action and sneezing droplet localization; second processing: perform angle of arrival estimation, distance estimation, Doppler estimation and multi-dimensional joint processing on the preprocessed wireless signal to realize sneezing Detection/recognition of motion and sneeze droplets in the Doppler dimension or other combined dimensions; the third processing: obtaining the attenuation spectrum or broadband spectrum of the preprocessed wireless signal; finally combining the results of the three processing to achieve sneezing Movement and location of sneeze droplets, and judging whether there are sneeze droplets. It can reduce misjudgments, improve accuracy, and realize sneeze droplet recognition/sneeze perception from all angles.
  • the method for sensing sneezing based on wireless signals provided in this application may also be applied to a multi-base joint sensing scenario or a multi-transmission and multi-receive sensing scenario. It is understandable that the "multi-base” mentioned in this application may refer to multiple receiving devices.
  • FIG. 16 is a schematic diagram of a multi-base joint sensing scenario provided by an embodiment of the present application.
  • WLAN device 1 or node 1
  • WLAN device 2 or node 2
  • WLAN device 3 or node 3
  • the target person is sneezing, and there are sneeze droplets generated by the sneeze.
  • the WLAN device 1 sends a wireless signal, and the wireless signal can reach the WLAN device 2 and the WLAN device 3 through the direct path, or can reach the WLAN device 2 and the WLAN device 3 after being reflected by the target human, and can also be reflected by the sneeze droplets (droplets).
  • WLAN device 2 and WLAN device 3 are reached.
  • the wireless signal received by the WLAN device 2 is the superposition of the direct path signal from the WLAN device 1 to the WLAN device 2 and multiple reflected path signals.
  • the wireless signal received by the WLAN device 3 is the superposition of the direct path signal from the WLAN device 1 to the WLAN device 3 and multiple reflected path signals.
  • the WLAN device 2 and the WLAN device 3 can respectively perform various signal processing on the wireless signals received by themselves, and combine the processed results of multiple nodes (such as the WLAN device 2 and the WLAN device 3) to locate and identify the sneezing action of the target object. And the sneeze droplets produced by sneezing can increase space gain and improve perception efficiency.
  • the WLAN device 2 and the WLAN device 3 can also transmit the wireless signals received by themselves to the cloud computing center for processing, and the cloud computing center performs signal processing on these wireless signals.
  • the WLAN device 1 may be a STA, and the WLAN device 2 and the WLAN device 3 may be APs.
  • each node in the multiple nodes processes wireless signals in the same way as the aforementioned single node processes wireless signals, the difference is that after each node obtains the Doppler information and attenuation spectrum/broadband spectrum of the wireless signal, it can Send this information to a node, and the node will combine the Doppler information and attenuation spectrum/broadband spectrum of the wireless signal sent by each node to identify and detect, and determine whether someone sneezes and whether there are sneeze droplets in the space.
  • the processing node is used as an example for description, that is, each node sends the obtained Doppler information and attenuation spectrum/broadband spectrum of the wireless signal to the processing node.
  • the possible processing procedures of the processing nodes are described below.
  • the processing node may be any node.
  • the processing node After the processing node receives the Doppler information and attenuation spectrum/broadband spectrum of the wireless signals sent by all nodes, it can integrate the relevant feature information from all nodes into an information matrix, input it into the neural network, and pass the subsequent convolution layers. Perform feature extraction, and then classify through the classification layer, and finally determine whether there is a sneeze action and generate sneeze droplets. Optionally, combined with the positioning information of all nodes, the complete information of the sneeze action and sneeze droplets can be sensed.
  • the processing node After the processing node receives the Doppler information and attenuation spectrum/broadband spectrum of the wireless signal sent by all nodes, it can first perform feature extraction for each information, and jointly input the features from all nodes to the subsequent classification network. In the middle, sneeze action recognition and sneeze droplet identification are performed, and finally it is determined whether a sneeze action occurs and sneeze droplets are generated. Optionally, combined with the positioning information of all nodes, the complete information of the sneeze action and sneeze droplets can be sensed.
  • the processing node After the processing node receives the Doppler information and attenuation spectrum/broadband spectrum of the wireless signals sent by all nodes, it can identify the information based on the two-dimensional neural network, and identify the results obtained by all nodes (whether there is a Sneeze action and produce sneeze droplets) for fusion judgment.
  • the embodiment of the present application can not only realize the location and identification/detection of sneeze action and sneeze droplets, but also realize sneeze perception in all directions; it can also improve spatial gain and improve perception efficiency. This is because if the sending node, the target (person), and a certain receiving node are in a straight line, the bistatic angle ⁇ is equal to 180 degrees, then the Doppler information cannot be measured on this receiving node, so one or more other Each receiving node can detect Doppler information, so as to realize the recognition of sneezing action and sneeze droplets, thereby improving spatial gain and improving perception efficiency.
  • the embodiment of the present application can still realize the location and identification/detection of sneeze action and sneeze droplets in all directions by combining multiple nodes.
  • FIG. 17 is another exemplary flowchart provided by the embodiment of the present application.
  • node 1 and node 2 collect wireless signals respectively, and perform signal preprocessing on the respectively collected wireless signals, and the preprocessing includes steps such as filtering.
  • the preprocessing includes steps such as filtering.
  • node 1 and node 2 respectively perform three kinds of processing on the preprocessed signal.
  • the first processing After Doppler estimation, angle of arrival estimation, distance estimation, joint estimation and other signal processing steps, sneeze (action) and sneeze droplets are located.
  • the second processing through time-frequency joint processing, distance-frequency-time joint processing, etc., time-Doppler spectrum and distance-Doppler-time spectrum of sneezing (action) and sneeze droplets are obtained.
  • the third processing After signal processing such as moving average, signal attenuation spectrum/broadband spectrum analysis is performed. Node 1 and/or Node 2 send the results of the second and third processes respectively performed to the processing node.
  • the processing node can be either node 1 or node 2.
  • the processing node performs feature extraction and feature fusion (including time-frequency/time-space/time-distance features, attenuation features, etc.) on the results of the second and third processing performed by node 1 and node 2, and then The results of feature extraction and feature fusion are input into the classifier (neural network/non-neural network) to identify sneeze (action) and sneeze droplets, and finally output the first processing location and sneeze droplet detection results, namely Sneeze localization and sneeze droplet detection.
  • feature extraction and feature fusion including time-frequency/time-space/time-distance features, attenuation features, etc.
  • FIG. 18 is a schematic structural diagram of an electronic device provided by an embodiment of the present application.
  • the electronic device may be an AP or STA in the WLAN, or a chip or a processing system or circuit installed in the AP or STA, or may be a cloud computing center.
  • the electronic device 100 may include:
  • the first acquisition module 10 is used to acquire a wireless signal, and the wireless signal propagates in the space containing the first object; the first processing module 20 is used to perform Doppler estimation on the wireless signal to obtain the Doppler of the wireless signal The Doppler information of the wireless signal is used to reflect the influence of the first object on the frequency of the wireless signal; the first determination module 30 is used to determine the first object based on the Doppler information of the wireless signal. Whether the subject is a sneeze droplet.
  • the above-mentioned wireless signal can also be propagated in a space containing a second object, and the Doppler information of the wireless signal is also used to reflect the influence of the second object on the frequency of the wireless signal, and the first object has no effect on the frequency of the wireless signal.
  • the effect of the frequency of the wireless signal is different from the effect of the second object on the frequency of the wireless signal.
  • the electronic device 100 further includes a second determination module 40 .
  • the second determining module 40 is configured to determine whether the second object is a sneeze action based on the Doppler information of the wireless signal.
  • the first determination module 30 and the second determination module 40 may be the same module or different modules.
  • the above electronic device 100 may further include a second processing module 50 .
  • the second processing module 50 is configured to perform angle of arrival estimation, distance estimation and Doppler estimation on the wireless signal to obtain the spatial position information of the first object and the spatial position information of the second object.
  • the spatial position information of the first object includes the first angle of arrival of the wireless signal reflected by the first object relative to the receiving device, the first distance between the first object and the receiving device, and the space of the second object.
  • the location information includes a second angle of arrival of the wireless signal reflected by the second object relative to the receiving device, and a second distance between the second object and the receiving device.
  • the above electronic device 100 may further include an output module 60 .
  • the output module 60 is used to output one or more of the following information: whether the first object is a sneeze droplet, whether the second object is a sneeze action, the spatial position information of the first object or the information of the second object Spatial location information.
  • the above electronic device 100 may further include a second acquiring module 70 and a third determining module 80 .
  • the second acquisition module 70 is configured to acquire an attenuation spectrum or a wideband spectrum of the wireless signal, where the attenuation spectrum of the wireless signal is used to reflect the influence of the first object on the amplitude attenuation of the wireless signal, and the wideband spectrum of the wireless signal is used to reflect the influence of the first object on the broadband spectral energy of the wireless signal;
  • the third determination module 80 is used to determine the wireless signal based on the Doppler information of the wireless signal and the attenuation spectrum or broadband spectrum of the wireless signal. Whether the first object is sneeze droplets.
  • the above-mentioned first determining module 30 is specifically configured to: perform feature extraction on the Doppler information of the wireless signal to obtain a first input feature; input the first input feature into a classification model for processing, and output a classification result,
  • the classification result is whether the first object is a sneeze droplet.
  • the first determining module 30 is specifically configured to: input the Doppler information of the wireless signal into a classification model for processing, and output a classification result, where the classification result is whether the first object is a sneeze droplet.
  • the above-mentioned first determining module 30 is specifically configured to: divide the Doppler information of the wireless signal into the first Doppler information and the second Doppler information, and the first Doppler information is in the Doppler information.
  • the expansion in the frequency domain is smaller than the expansion of the second Doppler information in the Doppler frequency domain;
  • the first Doppler information is input into the first identifier for identification, and whether the second object is a sneeze is identified Action; input the second Doppler information into a second identifier for identification, and identify whether there is a Doppler feature of the first object in the second Doppler information; whether the second object is a sneeze
  • the action and whether there is a Doppler feature of the first object in the second Doppler information are input into a decision device to determine whether the first object is a sneeze droplet.
  • the above-mentioned first acquisition module 10 , the above-mentioned first processing module 20 , the above-mentioned first determination module 30 , the above-mentioned second determination module 40 , the above-mentioned second processing module 50 , the above-mentioned second acquisition module 70 , and the above-mentioned third determination module 80 Can be integrated into a module, such as a processing module.
  • the above-mentioned output module 60 may also be a transceiver module.
  • each module or unit shown above may also refer to the corresponding description of the first device in the embodiment shown in FIG. 4 or FIG. 8 to execute the method executed by the first device in any of the above embodiments. and function.
  • the electronic device 100 provided in this embodiment of the present application can execute the method for sensing sneeze based on a wireless signal performed by the first device.
  • the electronic device 100 provided in this embodiment of the present application can execute the method for sensing sneeze based on a wireless signal performed by the first device.
  • FIG. 19 is another schematic structural diagram of an electronic device provided by an embodiment of the present application.
  • the electronic device 1000 provided in this embodiment of the present application includes a processor 1001 , a memory 1002 , and a bus system 1004 .
  • the electronic device 1000 may further include a transceiver 1003 .
  • the processor 1001 , the memory 1002 and the transceiver 1003 are connected through a bus system 1004 .
  • the above-mentioned processor 1001 is configured to acquire a wireless signal, which propagates in a space containing the first object; perform Doppler estimation on the wireless signal to obtain Doppler information of the wireless signal, and the Doppler information of the wireless signal is obtained.
  • the Le information is used to reflect the influence of the first object on the frequency of the wireless signal; based on the Doppler information of the wireless signal, it is determined whether the first object is a sneeze droplet.
  • the above transceiver 1003 can be used to output one or more of the following information: whether the first object is a sneeze droplet, whether the second object is a sneeze action, the spatial position information of the first object or the Spatial location information of the second object.
  • the above-mentioned memory 1002 is used for storing programs.
  • the program may include program code, and the program code includes computer operation instructions.
  • the memory 1002 includes, but is not limited to, random access memory (RAM), read-only memory (ROM), erasable programmable read only memory (EPROM), or Portable read-only memory (compact disc read-only memory, CD-ROM). Only one memory is shown in FIG. 19 , of course, the number of memories may also be set as multiple as required.
  • the memory 1002 may also be the memory in the processor 1001, which is not limited here.
  • the memory 1002 stores the following elements, executable units or data structures, or a subset thereof, or an extended set thereof:
  • Operation instructions including various operation instructions, which are used to realize various operations.
  • Operating System Includes various system programs for implementing various basic services and handling hardware-based tasks.
  • the above-mentioned processor 1001 controls the operation of the electronic device 1000.
  • the processor 1001 may be one or more central processing units (CPUs).
  • CPUs central processing units
  • the CPU may be a single-core CPU. It can also be a multi-core CPU.
  • bus system 1004 various components of the electronic device 1000 are coupled together through a bus system 1004, where the bus system 1004 may include a power bus, a control bus, a status signal bus, and the like in addition to a data bus.
  • bus system 1004 may include a power bus, a control bus, a status signal bus, and the like in addition to a data bus.
  • the various buses are labeled as bus system 1004 in FIG. 19 .
  • FIG. 19 only a schematic drawing is shown in FIG. 19 .
  • the processor 1001 , the memory 1002 and the transceiver 1003 may also cooperate to execute the method for sneezing based on a wireless signal performed by the first device.
  • an embodiment of the present application also provides a computer program product, the computer program product includes computer program code, when the computer program code is run on a computer, the computer is made to execute the first described in FIG. 4 or FIG. 8 . method steps of the device.
  • the computer program code in the computer program product can be executed, for example, by the processor 1001 in the electronic device 1000 shown in FIG. 19 , to control the transceiver 1003 to cooperate with the wireless signal-based sensing performed in any of the foregoing embodiments.
  • the functions of the computer program product can be implemented by hardware or software. When implemented by software, these functions can be stored in a computer-readable storage medium or executed as one or more instructions or codes on the computer-readable storage medium. transmission.
  • an embodiment of the present application further provides a computer-readable storage medium, where computer program code is stored in the computer-readable storage medium, and when the above-mentioned processor executes the computer program code, the electronic device executes any of the foregoing embodiments.
  • the computer-readable storage medium may be the internal memory in the electronic device 1000 shown in FIG. 19 , or may be an external memory connected to the electronic device 1000 described above.
  • an embodiment of the present application further provides a device, the device may exist in the form of a chip product, the structure of the device includes a processing circuit and an interface circuit, and the processing circuit is used to execute the method of any of the foregoing embodiments , the interface circuit is used to communicate with other devices.
  • the electronic device, computer-readable storage medium, computer program product, and chip of the embodiments of the present application can execute the method for sensing sneeze based on a wireless signal in any of the foregoing embodiments.
  • the specific implementation process and beneficial effects refer to the foregoing inverse embodiment. , and will not be repeated here.
  • the steps of the methods or algorithms described in conjunction with the disclosure of the present application may be implemented in a hardware manner, or may be implemented in a manner in which a processor executes software instructions.
  • the software instructions can be composed of corresponding software modules, and the software modules can be stored in random access memory (Random Access Memory, RAM), flash memory, Erasable Programmable Read-Only Memory (Erasable Programmable ROM, EPROM), electrically erasable programmable Programmable read-only memory (Electrically EPROM, EEPROM), registers, hard disk, removable hard disk, compact disk read only (CD-ROM), or any other form of storage medium known in the art.
  • An exemplary storage medium is coupled to the processor, such that the processor can read information from, and write information to, the storage medium.
  • the storage medium can also be an integral part of the processor.
  • the processor and storage medium may reside in an ASIC.
  • the ASIC may be located in the core network interface device.
  • the processor and the storage medium may also exist in the core network interface device as discrete components.
  • the functions described in this application may be implemented in hardware, software, firmware, or any combination thereof.
  • the functions When implemented in software, the functions may be stored on or transmitted over as one or more instructions or code on a computer-readable medium.
  • Computer-readable media includes both computer-readable storage media and communication media including any medium that facilitates transfer of a computer program from one place to another.
  • a storage medium can be any available medium that can be accessed by a general purpose or special purpose computer.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Remote Sensing (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Signal Processing (AREA)
  • Computing Systems (AREA)
  • Biophysics (AREA)
  • General Health & Medical Sciences (AREA)
  • Molecular Biology (AREA)
  • Data Mining & Analysis (AREA)
  • General Engineering & Computer Science (AREA)
  • Computational Linguistics (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Evolutionary Computation (AREA)
  • Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • Artificial Intelligence (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Acoustics & Sound (AREA)
  • Electromagnetism (AREA)
  • Radar Systems Or Details Thereof (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

本申请涉及无线通信技术领域，尤其涉及一种基于无线信号感知打喷嚏的方法及相关装置，比如应用于支持802.11ax或802.11be等标准的无线局域网中。该方法包括：获取无线信号，该无线信号在包括第一对象的空间内传播；对该无线信号进行多普勒估计，得到该无线信号的多普勒信息，该无线信号的多普勒信息可以用于反映第一对象对该无线信号的频率产生的影响；基于该无线信号的多普勒信息，确定该第一对象是否为喷嚏飞沫。采用本申请实施例，可以利用现有的WLAN设备，通过对无线信号进行信号处理实现打喷嚏的喷嚏飞沫检测，不受环境中的光照和噪声的影响。

Description

基于无线信号感知打喷嚏的方法及相关装置
本申请要求于2020年7月29日提交中国国家知识产权局、申请号为202010746355.0、申请名称为“基于无线信号感知打喷嚏的方法及相关装置”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及无线通信技术领域，尤其涉及一种基于无线信号感知打喷嚏的方法及相关装置。
背景技术
无线局域网(wireless local area networks，WLAN)感知(WLAN Sensing)是一项具有广阔应用前景的技术，它利用WLAN设备发送的射频(radio frequency，RF)信号对周围环境进行感知，通过一定的算法提取接收信号中的相应参数进行分析，获取周围环境中的相关信息。WLAN sensing可以借用现在已经广泛部署的WLAN基础设施(如WLAN设备)，实现环境感知。例如，在家庭安防方面，可以通过环境中所部署的WLAN设备来不断的探测、获取所在环境的信道状态信息，通过大量的数据分析对比，实时监控环境中是否存在异常情况，从而确保家庭住宅的安全。又如，在医院医疗方面，也可以在特定的监控区域部署一定量的WLAN设备，通过WLAN信号或特定的数据信号感知测量病人的心跳、温度等生物特征信息，以便对病人所在区域进行实时监控。
打喷嚏或咳嗽作为呼吸道疾病的常见伴随症状，是病毒的重要传播方式，特别是在室内等空气流动性较差的环境中，打喷嚏/咳嗽所产生的飞沫及随后产生的带有病毒的气溶胶可以停留数个小时以上，具有潜在的危险性。因此，通过对打喷嚏/咳嗽等动作的识别和定位，识别出潜在感染者打喷嚏/咳嗽的位置，并对周围可能产生的高浓度含有病毒的气溶胶范围进行判定，可以有助于规避潜在风险区域，在一定程度上可以阻碍病毒的传播。
目前，可以通过声音信号或摄像头来对打喷嚏进行识别。具体地，由于打喷嚏具有比较特殊的声音特征，所以可以通过采集到的声音信号来对打喷嚏进行识别。但通过声音信号对打喷嚏进行识别的方法容易受到环境噪声的影响，并且无法对打喷嚏的另一特征“飞沫”进行检测。通过摄像头来对打喷嚏进行识别，在高清晰摄像头帮助下，可以通过人工智能实现基于视觉的打喷嚏检测，也可以在一定程度上实现飞沫的检测。但是通过摄像头来对打喷嚏进行识别的方法容易受到外部光照条件的影响和遮挡物影响，无法实现全天候的感知。
发明内容
本申请实施例提供一种基于无线信号感知打喷嚏的方法及相关装置，可以利用现有的WLAN设备，通过对无线信号进行信号处理实现打喷嚏的喷嚏飞沫检测，不受环境中的光照和噪声的影响。
下面从不同的方面介绍本申请，应理解的是，下面的不同方面的实施方式和有益效果 可以互相参考。
第一方面，本申请提供一种基于无线信号感知打喷嚏的方法，该方法可以应用于WLAN中的接入点/站点，或者云端计算中心中。该方法包括：获取无线信号，并对该无线信号进行多普勒估计，得到该无线信号的多普勒信息；基于该无线信号的多普勒信息，确定该第一对象是否为喷嚏飞沫。其中，该无线信号在包括第一对象的空间内传播。该无线信号的多普勒信息可以用于反映第一对象对该无线信号的频率产生的影响。该第一对象可以是飞沫或空间中的小水滴等。
可选的，上述多普勒信息可以包括时间-多普勒谱、距离-多普勒-时间谱、或其他维度的多普勒信息。
本方案利用喷嚏飞沫对无线信号的多普勒信息产生的特殊影响，来识别喷嚏飞沫，从而识别出是否打喷嚏，该过程可以不受环境中的光照、噪声以及遮挡物的影响，可以提高飞沫检测的适用性。
结合第一方面，在一种可能的设计中，上述无线信号还在包含第二对象的空间内传播，该无线信号的多普勒信息还可以用于反映该第二对象对该无线信号的频率产生的影响，该第一对象对该无线信号的频率产生的影响与该第二对象对该无线信号的频率产生的影响不同。该方法还可以包括：基于上述无线信号的多普勒信息，确定该第二对象是否为打喷嚏动作。其中，第二对象可以是人的肢体动作。
本方案利用空间中多个对象对无线信号的多普勒信息可以产生不同的影响，来识别这多个对象，不仅可以实现喷嚏飞沫的识别，还可以实现打喷嚏动作的识别。联合喷嚏飞沫和打喷嚏的动作进行识别，可以减少误判的情况。
结合第一方面，在一种可能的设计中，获取到无线信号之后，该方法还包括：对该无线信号进行到达角估计、距离估计以及多普勒估计，得到该第一对象的空间位置信息和该第二对象的空间位置信息。其中，该第一对象的空间位置信息包括经过该第一对象反射的无线信号相对于接收设备的第一到达角、和该第一对象与该接收设备的第一距离，该第二对象的空间位置信息包括经过该第二对象反射的无线信号相对于该接收设备之间的第二到达角、和该第二对象与该接收设备的第二距离。
结合第一方面，在一种可能的设计中，该方法还包括：输出以下一种或多种信息：第一对象是否为喷嚏飞沫、第二对象是否为打喷嚏动作、第一对象的空间位置信息或第二对象的空间位置信息。可选的，输出信息的方式可以是直接将信息发送给相关的移动设备，也可以是将信息上传至云端，由云端根据打喷嚏的人数、喷嚏飞沫的范围等信息，提醒保洁人员进行清洁。
本方案通过输出各种信息，可以提醒相关人员喷嚏飞沫的出现区域和影响范围，规避潜在的传染风险。
结合第一方面，在一种可能的设计中，获取到无线信号之后，该方法还包括：获取该无线信号的衰减谱或宽带谱，并可以基于该无线信号的多普勒信息和该无线信号的衰减谱或宽带谱，确定该第一对象是否为喷嚏飞沫。其中，该无线信号的衰减谱可以用于反映该第一对象对该无线信号的幅度衰减产生的影响，该无线信号的宽带谱可以用于反映该第一对象对该无线信号的宽带频谱能量产生的影响。
本方案不仅考虑喷嚏飞沫对无线信号的多普勒信息产生的影响，还结合喷嚏飞沫对无线信号的衰减谱/宽带谱产生的影响，来综合判断是否为喷嚏飞沫，可以进一步提高准确性。
结合第一方面，在一种可能的设计中，基于上述无线信号的多普勒信息，确定上述第一对象是否为喷嚏飞沫，具体包括：先将该无线信号的多普勒信息进行特征提取，得到第一输入特征；再将该第一输入特征输入分类模型中进行处理，输出分类结果，该分类结果为该第一对象是否为喷嚏飞沫。
结合第一方面，在一种可能的设计中，基于上述无线信号的多普勒信息，确定上述第一对象是否为喷嚏飞沫，具体包括：直接将该无线信号的多普勒信息输入分类模型中进行处理，输出分类结果，该分类结果为该第一对象是否为喷嚏飞沫。
结合第一方面，在一种可能的设计中，基于上述无线信号的多普勒信息，确定上述第一对象是否为喷嚏飞沫，具体包括：先将该无线信号的多普勒信息分为第一多普勒信息和第二多普勒信息，该第一多普勒信息在多普勒频域上的扩展小于该第二多普勒信息在多普勒频域上的扩展；再将该第一多普勒信息输入第一识别器中进行识别，识别出该第二对象是否为打喷嚏动作；然后将该第二多普勒信息输入第二识别器中进行识别，识别出该第二多普勒信息中是否存在该第一对象的多普勒特征；最后将该第二对象是否为打喷嚏动作以及该第二多普勒信息中是否存在该第一对象的多普勒特征输入判决器，判断该第一对象是否为喷嚏飞沫。
本方案提供多种飞沫检测的方法，在实际应用中，可以根据不同情况选择不同的飞沫检测方法。
第二方面，本申请提供一种电子设备，包括：第一获取模块，用于获取无线信号，该无线信号在包含第一对象的空间内传播；第一处理模块，用于对该无线信号进行多普勒估计，得到该无线信号的多普勒信息，该无线信号的多普勒信息用于反映该第一对象对该无线信号的频率产生的影响；第一确定模块，用于基于该无线信号的多普勒信息，确定该第一对象是否为喷嚏飞沫。
结合第二方面，在一种可能的设计中，上述无线信号还可以在包含第二对象的空间内传播，该无线信号的多普勒信息还用于反映该第二对象对该无线信号的频率产生的影响，该第一对象对该无线信号的频率产生的影响与该第二对象对该无线信号的频率产生的影响不同。上述电子设备还包括第二确定模块。该第二确定模块，用于基于该无线信号的多普勒信息，确定该第二对象是否为打喷嚏动作。其中，上述第一确定模块和第二确定模块，可以是同一模块，也可以是不同模块。第二对象可以是人的肢体动作。
结合第二方面，在一种可能的设计中，上述电子设备还可以包括第二处理模块。该第二处理模块，用于对该无线信号进行到达角估计、距离估计以及多普勒估计，得到该第一对象的空间位置信息和该第二对象的空间位置信息。其中，该第一对象的空间位置信息包括经过该第一对象反射的无线信号相对于接收设备的第一到达角、和该第一对象与该接收设备的第一距离，该第二对象的空间位置信息包括经过该第二对象反射的无线信号相对于该接收设备之间的第二到达角、和该第二对象与该接收设备的第二距离。
结合第二方面，在一种可能的设计中，上述电子设备还可以包括输出模块。该输出模块，用于输出以下一种或多种信息：该第一对象是否为喷嚏飞沫、该第二对象是否为打喷 嚏动作、该第一对象的空间位置信息或该第二对象的空间位置信息。
结合第二方面，在一种可能的设计中，上述电子设备还可以包括第二获取模块和第三确定模块。该第二获取模块，用于获取该无线信号的衰减谱或宽带谱，该无线信号的衰减谱用于反映该第一对象对该无线信号的幅度衰减产生的影响，该无线信号的宽带谱用于反映该第一对象对该无线信号的宽带频谱能量产生的影响；该第三确定模块，用于基于该无线信号的多普勒信息和该无线信号的衰减谱或宽带谱，确定该第一对象是否为喷嚏飞沫。
结合第二方面，在一种可能的设计中，上述第一确定模块30具体用于：将该无线信号的多普勒信息进行特征提取，得到第一输入特征；将该第一输入特征输入分类模型中进行处理，输出分类结果，该分类结果为该第一对象是否为喷嚏飞沫。
结合第二方面，在一种可能的设计中，上述第一确定模块30具体用于：将该无线信号的多普勒信息输入分类模型中进行处理，输出分类结果，该分类结果为该第一对象是否为喷嚏飞沫。
结合第二方面，在一种可能的设计中，上述第一确定模块30具体用于：将该无线信号的多普勒信息分为第一多普勒信息和第二多普勒信息，该第一多普勒信息在多普勒频域上的扩展小于该第二多普勒信息在多普勒频域上的扩展；将该第一多普勒信息输入第一识别器中进行识别，识别出该第二对象是否为打喷嚏动作；将该第二多普勒信息输入第二识别器中进行识别，识别出该第二多普勒信息中是否存在该第一对象的多普勒特征；将该第二对象是否为打喷嚏动作以及该第二多普勒信息中是否存在该第一对象的多普勒特征输入判决器，判断该第一对象是否为喷嚏飞沫。
第三方面，本申请提供另一种电子设备，包括处理器。该处理器用于：获取无线信号，并对该无线信号进行多普勒估计，得到该无线信号的多普勒信息；基于该无线信号的多普勒信息，确定该第一对象是否为喷嚏飞沫。其中，该无线信号在包括第一对象的空间内传播。该无线信号的多普勒信息可以用于反映第一对象对该无线信号的频率产生的影响。
可选的，该电子设备还可以包括存储器，该存储器用于与处理器耦合，其保存电子设备必要的程序指令和数据。
第四方面，本申请提供一种计算机可读存储介质，该计算机可读存储介质中存储有指令，该指令可以由处理电路上的一个或多个处理器执行。当其在计算机上运行时，使得计算机执行上述任一方面所述的基于无线信号感知打喷嚏的方法。可选的，该计算机可读存储介质可以是非易失性可读存储介质。
第五方面，本申请提供了一种包含指令的计算机程序产品，当其在计算机上运行时，使得计算机执行上述任一方面所述的基于无线信号感知打喷嚏的方法。
第六方面，本申请提供一种芯片或芯片系统，包括处理电路。该处理电路可用于执行以下操作：获取无线信号，并对该无线信号进行多普勒估计，得到该无线信号的多普勒信息；基于该无线信号的多普勒信息，确定该第一对象是否为喷嚏飞沫。其中，该无线信号在包括第一对象的空间内传播。该无线信号的多普勒信息可以用于反映第一对象对该无线信号的频率产生的影响。
可选的，该芯片或芯片系统还可以包括输入输出接口。该输入输出接口可以用于输出以下一种或多种信息：第一对象是否为喷嚏飞沫、第二对象是否为打喷嚏动作、第一对象 的空间位置信息或第二对象的空间位置信息。
实施本申请实施例，可以利用现有的WLAN设备，通过对无线信号进行信号处理实现打喷嚏的喷嚏飞沫检测，不受环境中的光照和噪声的影响。
附图说明
为了更清楚地说明本申请实施例的技术方案，下面将对实施例描述中所需要使用的附图作简单地介绍。
图1是本申请实施例提供的一种系统架构图；
图2是本申请实施例提供的AP或STA的结构示意图；
图3是本申请实施例提供的一种应用场景的示意图；
图4是本申请实施例提供的基于无线信号感知打喷嚏的方法的一示意流程图；
图5是本申请实施例提供的到达角估计的一示意图；
图6是本申请实施例提供的距离估计的一示意图；
图7是本申请实施例提供的喷嚏飞沫的距离-多普勒谱的示意图；
图8是本申请实施例提供的基于无线信号感知打喷嚏的方法的另一示意流程图；
图9是本申请实施例提供的信号衰减谱示意图；
图10是本申请实施例提供的真实喷嚏多普勒测量结果示意图；
图11是本申请实施例提供的多普勒与双基地夹角的关系示意图；
图12是本申请实施例提供的信息输出的场景示意图；
图13是本申请实施例提供的基于AR的相关信息通知示意图；
图14是本申请实施例提供的喷嚏飞沫的真实速度合成示意图；
图15是本申请实施例提供的一种流程示例图；
图16是本申请实施例提供的多基地联合感知场景的示意图；
图17是本申请实施例提供的另一种流程示例图；
图18是本申请实施例提供的电子设备的一结构示意图；
图19是本申请实施例提供的电子设备的另一结构示意图。
具体实施方式
下面将结合本申请实施例中的附图，对本申请实施例中的技术方案进行清楚、完整地描述。
为便于理解本申请实施例的技术方案，下面将对本申请实施例提供的基于无线信号感知打喷嚏的方法的系统架构和/或应用场景进行说明。可理解的，本申请实施例描述的场景是为了更加清楚的说明本申请实施例的技术方案，并不构成对于本申请实施例提供的技术方案的限定。
本申请实施例提供一种基于无线信号感知打喷嚏的方法，可以利用现有的WLAN设备，无需额外的麦克风、摄像头等设备，通过对射频信号(或无线信号)进行信号处理，实现打喷嚏的定位、识别和飞沫检测，可以不受环境中的光照和噪声的影响。该方法可以 应用于无线通信系统中，该无线通信系统可以为无线局域网或蜂窝网；该方法可以由无线通信系统中的通信设备或通信设备中的芯片或处理器实现。该通信设备可以是接入点(access point，AP)设备或站点(station，STA)设备。该接入点设备和站点设备既可以是单链路设备，也可以是多链路设备。
参见图1，图1是本申请实施例提供的一种系统架构图。如图1所示，该系统架构包括至少2个WLAN设备(如图1中的AP1和STA2)，其中一个WLAN设备(如STA2)发送RF信号，其他WLAN设备(如AP1)接收RF信号。可选的，如图1所示，该系统架构还可以包括云端计算中心。其中，WLAN设备可以支持WLAN通信协议，该通信协议可以包括IEEE 802.11be(或称为Wi-Fi 7，EHT协议)，还可以包括IEEE 802.11ax，IEEE 802.11ac等协议。当然，随着通信技术的不断演进和发展，该通信协议还可以包括IEEE 802.11be的下一代协议等。
本申请中，实现本申请方法的装置可以是WLAN中的AP或STA，或者是，安装在AP或STA中的芯片或处理系统，还可以是云端计算中心。WLAN设备可以通过收发Wi-Fi信号，实现打喷嚏定位、识别和喷嚏飞沫检测，相关计算可以放在WLAN AP中完成；也可以上传至云端计算中心，利用强大的云端计算能力进行处理。可理解的，相较于AP侧处理，云端处理具有强大的计算能力，还具有更大的灵活性，可以针对不同情况适配相应的感知算法，最大的发挥感知性能。
可理解的，本申请中的“射频信号”、“WLAN信号”、“Wi-Fi信号”以及“无线信号”之间可以相互替换使用，均指采用无线方式传播的信号。
接入点(例如AP1)是一种具有无线通信功能的装置，支持采用WLAN协议进行通信，具有与WLAN网络中其他设备(比如站点或其他接入点)通信的功能，当然，还可以具有与其他设备通信的功能。在WLAN系统中，接入点可以称为接入点站点(AP STA)。该具有无线通信功能的装置可以为一个整机的设备，还可以是安装在整机设备中的芯片或处理系统等，安装这些芯片或处理系统的设备可以在芯片或处理系统的控制下，实现本申请实施例的方法和功能。本申请实施例中的AP是为STA提供服务的装置，可以支持802.11系列协议。例如，AP可以为通信服务器、路由器、交换机、网桥等通信实体；AP可以包括各种形式的宏基站，微基站，中继站等，当然AP还可以为这些各种形式的设备中的芯片和处理系统，从而实现本申请实施例的方法和功能。
站点(例如STA2)是一种具有无线通信功能的装置，支持采用WLAN协议进行通信，具有与WLAN网络中的其他站点或接入点通信的能力。在WLAN系统中，站点可以称为非接入点站点(non-access point station,non-AP STA)。例如，STA是允许用户与AP通信进而与WLAN通信的任何用户通信设备，该装置可以为一个整机的设备，还可以是安装在整机设备中的芯片或处理系统等，安装这些芯片或处理系统的设备可以在芯片或处理系统的控制下，实现本申请实施例的方法和功能。例如，STA可以为平板电脑、桌面型、膝上型、笔记本电脑、超级移动个人计算机(Ultra-mobile Personal Computer，UMPC)、手持计算机、上网本、个人数字助理(Personal Digital Assistant，PDA)、手机等可以联网的用户设备，或物联网中的物联网节点，或车联网中的车载通信装置或，娱乐设备，游戏设备或系统，全球定位系统设备等，STA还可以为上述这些终端中的芯片或处理系统。
具体地，本申请关注利用无线信号/Wi-Fi信号感知打喷嚏的方法。下面对AP和STA的结构作简要的说明。参见图2，图2是本申请实施例提供的AP或STA的结构示意图。如图2所示，AP或STA可以包括：应用(application)层模块、传输控制协议(transmission control protocol，TCP)/用户数据报协议(user datagram protocol，UDP)处理模块、网际互连协议(internet protocol，IP)处理模块、逻辑链路控制(logical link control，LLC)模块、媒体接入控制(media access control，MAC)层模块、物理(physical，PHY)层基带模块、射频radio以及天线等。其中，图2所示的AP或STA既可以是单天线结构，也可以是多天线结构，本申请实施例对此不作限定。
WLAN系统可以提供高速率低时延的传输，随着WLAN应用场景的不断演进，WLAN系统将会应用于更多场景或产业中，比如，应用于物联网产业，应用于车联网产业或应用于银行业，应用于企业办公，体育场馆展馆，音乐厅，酒店客房，宿舍，病房，教室，商超，广场，街道，生成车间和仓储等。当然，支持WLAN通信的设备(比如接入点或站点)可以是智慧城市中的传感器节点(比如，智能水表，智能电表，智能空气检测节点)，智慧家居中的智能设备(比如智能摄像头，投影仪，显示屏，电视机，音响，电冰箱，洗衣机等)，物联网中的节点，娱乐终端(比如AR，VR等可穿戴设备)，智能办公中的智能设备(比如，打印机，投影仪，扩音器，音响等)，车联网中的车联网设备，日常生活场景中的基础设施(比如自动售货机，商超的自助导航台，自助收银设备，自助点餐机等)，以及大型体育或音乐场馆的设备等。本申请实施例中对于STA和AP的具体形式不做特殊限制，在此仅是示例性说明。
上述内容简述了本申请实施例提供的系统架构，下面将结合本申请实施例提供的可能的应用场景以及更多的附图，对本申请实施例提供的基于无线信号感知打喷嚏的方法进行详细说明。
实施例一
本申请实施例一结合一种可能的应用场景对本申请提供的基于无线信号感知打喷嚏的方法进行说明。本申请实施例的应用场景以2个设备为例，其中一个为发送设备，另一个为接收设备。可理解的，本申请提及的“发送设备”是指发送无线信号的WLAN设备，“接收设备”是指接收无线信号的WLAN设备。还可理解的，本申请中的发送设备既可以发送信号，也可以接收信号，因其在本申请中主要是发送无线信号的功能，所以称为发送设备；本申请中的接收设备既可以接收信号，也可以发送信号，因其在本申请中主要是接收无线信号的功能，所以称为接收设备。
具体地，参见图3，图3是本申请实施例提供的一种应用场景的示意图。如图3所示，空间内存在至少2个WLAN设备，分别为WLAN设备1和WLAN设备2，并且该空间内有一目标人(如图3中的target human)正在打喷嚏，且存在打喷嚏而产生的喷嚏飞沫(droplets)。WLAN设备1发送无线信号，该无线信号可以通过直达径/直射径到达WLAN设备2，也可以经过target human反射后到达WLAN设备2，还可以经过喷嚏飞沫(droplets)反射后到达WLAN设备2。WLAN设备2接收到的无线信号为多路信号(包括图3中的直达径、两条反射径及图3中其他未示出的多径信号)的叠加。WLAN设备2可以对接收到 的无线信号进行多种信号处理，通过目标对象(如图3的目标人)的打喷嚏动作对无线信号的影响(比如，对无线信号的频率产生的影响)，以及喷嚏飞沫对无线信号的影响(比如，对无线信号的频率、幅度衰减或宽带频谱能量等产生的影响)，来定位和识别目标对象的打喷嚏动作和打喷嚏所产生的喷嚏飞沫，从而实现打喷嚏的定位、识别和喷嚏飞沫检测，并且可以不受环境中的光照和噪声的影响。
可选的，WLAN设备2也可以将接收到的无线信号传送至云端计算中心进行处理，由云端计算中心对这个无线信号进行信号处理等。
可理解的，WLAN设备1和WLAN设备2既可以为AP，也可以为STA，还可以一个WLAN设备为AP，另一个WLAN设备为STA，如WLAN设备1为STA，WLAN设备2为AP。
基于上述图3所示的应用场景，本申请实施例提供一种基于无线信号感知打喷嚏的方法。参见图4，图4是本申请实施例提供的基于无线信号感知打喷嚏的方法的一示意流程图。如图4所示，该基于无线信号感知打喷嚏的方法包括但不限于以下步骤：
S101，第一设备获取无线信号，该无线信号在包含第一对象的空间内传播。
具体地，在第一设备为接收设备的情况下，第一设备接收/采集无线信号，并对接收/采集到的无线信号进行预处理，比如滤波处理等。在第一设备为云端计算中心的情况下，接收设备接收/采集无线信号，并可以将接收/采集到的无线信号发送至云端计算中心，由云端计算中心进行后续信号处理过程/步骤。可理解的，云端计算中心具有强大的计算能力，还具有更大的灵活性，可以针对不同情况适配相应的感知算法，最大的发挥感知性能。但是，由于接收设备接收/采集到的无线信号的数据量大，需要较大的带宽支持数据的传输。
其中，上述接收设备接收/采集到的无线信号可以为多路/多径信号的叠加，该无线信号中的至少一路信号经过第一对象反射后得到。可理解的，发送设备发送原始无线信号后，该原始无线信号在包含第一对象的空间内传播，故该原始无线信号至少经过第一对象反射后到达接收设备。可选的，该原始无线信号还可以通过直射径到达接收设备。该第一对象可以为飞沫、空间中的小水滴等。
可理解的，后续步骤可以基于预处理后的无线信号进行其他信号处理，如步骤S102可以针对预处理后的无线信号进行多普勒估计。
S102，第一设备对该无线信号进行多普勒估计，得到该无线信号的多普勒信息，该无线信号的多普勒信息用于反映第一对象对该无线信号的频率产生的影响。
具体地，上述无线信号的多普勒信息可以用于反映上述第一对象对该无线信号的频率产生的影响。上述无线信号的多普勒信息可以包括时间-多普勒谱、距离-多普勒-时间谱、或其他维度的多普勒信息。第一设备可以对上述无线信号进行多普勒估计，得到该无线信号的时间-多普勒谱。或者，第一设备可以对上述无线信号进行到达角估计、距离估计、多普勒估计或多维联合处理等信号处理，得到该无线信号的距离-多普勒-时间谱或其他维度的多普勒信息。可理解的，如果发送设备发送的原始无线信号在包含多个对象的空间内传播，则上述无线信号的多普勒信息可以分别反映这多个对象对该无线信号的频率产生的影响，即一张多普勒谱上可以反映多个对象的运动特征。
以飞沫为例，由于飞沫在空间中运动，运动的飞沫会对在该空间中传播的信号产生多 普勒调制，所以对接收/采集到的无线信号进行相应的分析，如时间-频率分析，可以在时间-多普勒维度上发现多普勒随时间变化的特征，这些特征可以反映飞沫的真实运动，因此可以利用多普勒信息来识别飞沫。同理，运动的飞沫不仅在时间-多普勒维度上具有独特的特征；飞沫的运动也可以在其他维度上产生相应的具有对应飞沫运动特性的信息，如，在时间-距离维度上，可以反映出距离随时间变化的信息，该信息同样可以辅助后续的喷嚏飞沫识别。还存在时间-角度信息，或者其他联合维度的信息，也可以辅助后续的喷嚏飞沫的识别。
其中，到达角估计可以采用波束赋形算法、子空间类方法或其他阵列信号处理方法进行估计，距离估计可以采用匹配滤波等信号处理算法进行估计，多普勒估计可以采用傅里叶变换等信号处理算法进行估计。下面分别对到达角估计、距离估计以及多普勒估计的原理进行介绍。
(1)到达角估计
参见图5，图5是本申请实施例提供的到达角估计的一示意图。如图5所示，假设一个位于远场的阵列天线，以四个阵元的一维均匀线阵为例，阵元之间的间距为d。当接收阵列处于信号的远场(即信号以平面波的形式抵达天线阵列时)，平面波抵达不同阵元的波程差为dsinθ，对应的接收信号在接收阵元之间产生相应的相位差。下面以波束赋形算法和子空间类方法为例进行简单说明。可理解的，本申请提及的“接收信号”是指接收设备接收到的无线信号。
波束赋形(beamforming)：利用天线形成一个波束，对空间进行扫描，当某个方向上的能量较强时，可以判定此方向上存在信号。这类方法形成的波束宽度(即波束分辨率)受制于天线阵列的孔径，孔径越大，分辨率越高。
这里以多重信号分类方法(multiple signal classification，MUSIC)为例对子空间类方法进行简单说明。
x＝As+n………………………………………………(1-1)
公式(1-1)中，x表示接收阵列接收到的信号，s表示源信号，即发送端发送的信号，A表示导向矢量矩阵，n为噪声。
首先求取接收信号的协方差矩阵(一般使用多个快拍)：
Figure PCTCN2021107962-appb-000001
公式(1-2)中，L表示联合处理的快拍个数，
Figure PCTCN2021107962-appb-000002
表示基于L个快拍、估计得到的协方差矩阵。
通过对上述协方差矩阵
Figure PCTCN2021107962-appb-000003
进行特征分解，分为信号子空间和噪声子空间：
Figure PCTCN2021107962-appb-000004
公式(1-3)中，
Figure PCTCN2021107962-appb-000005
为信号子空间，
Figure PCTCN2021107962-appb-000006
为噪声子空间。
其中，相关推导可以证明：
A HU n＝0………………………………………………(1-4)
即，接收信号的导向矢量与分解得到的信号噪声子空间正交。基于这个结论，可以利用下述公式(1-5)进行谱峰搜索，进行到达角估计。
Figure PCTCN2021107962-appb-000007
公式(1-5)中，a表示接收阵列的导向矢量(steering vector)。
A＝[a(θ 1),a(θ 2),...,a(θ K)]∈C M×K…………………………(1-6)
Figure PCTCN2021107962-appb-000008
公式(1-6)和公式(1-7)中，C表示复数集合，M表示接收阵列的阵元个数，K表示波束方向的个数，d表示阵元间距，θ表示信号的入射角，λ表示波长。
可选的，子空间类方法还包括旋转不变信号参数估计技术(estimation of signal parameters via rotational invariance techniques，ESPRIT)等方法。除了子空间类方法，还可以通过稀疏类的方法进行到达角估计。
(2)距离估计
参见图6，图6是本申请实施例提供的距离估计的一示意图。如图6所示，设备1、目标对象以及设备2三者组成一个典型的WLAN感知场景。设备1发送无线信号，无线信号经过目标对象反射之后由设备2接收，并且设备2也可以接收到由设备1直接发送的直达径信号。因此当设备1发送无线信号时，设备2接收到的信号是直达径信号和反射径信号(为便于描述，图6以1条反射径为例进行说明)的叠加。如果设备1发送的无线信号具有较好的自相关特性(即当两个信号在时间上完全对齐时，自相关能量最大；当两个信号存在时延时，自相关能量很小)，设备2对参考信号和接收信号采用时域匹配滤波处理(即自相关)，可以在时间-能量频谱上得到两个尖峰。其中第一个尖峰表示直达径信号，第二个尖峰表示反射径信号。这两个尖峰之间的时间差可以表示反射径和直达径之间的传播时延差，对应真实环境中的传播距离差。
因此，得到反射径和直达径之间的传播距离差、以及设备1和设备2(即收发设备)的位置信息，可以推导出目标对象是在以收发设备为焦点，且目标对象与设备1和设备2之间的距离和为：直达径距离与计算得到的传播距离差之和的椭圆上。再结合前面得到的到达角的角度信息，便可以实现目标对象的定位。
可理解的，匹配滤波既可以在时域处理，也可以在频域进行处理。匹配滤波属于雷达中的处理方法，但是该方法也可以应用于通信中的信道估计。还可理解的，除了基于匹配滤波的距离估计，还可以通过别的信号处理方式进行距离估计，此处不再说明。
可理解的，上述距离-多普勒-时间谱中的距离可以指反射径和直达径之间的传播距离差。
(3)多普勒估计
多普勒估计可以直接采用傅里叶变换进行处理。具体地，可以将一段时域信号，通过傅里叶变换变换到频域，分析其频谱能量，从而估计其多普勒分量。通常，由于测不准原则，时域信号的时长越长，频域上的分辨率越高。
Figure PCTCN2021107962-appb-000009
其中，T表示时域信号的时长，Δf表示频率间隔。
S103，第一设备基于该无线信号的多普勒信息，确定第一对象是否为喷嚏飞沫。
具体地，如果上述无线信号的多普勒信息为时间-多普勒谱，由于时间-多普勒谱是二维特征，所以可以将时间-多普勒谱作为一个图像，直接输入针对图像的神经网络中进行训练和识别。即：第一设备可以直接将上述无线信号的多普勒信息输入分类模型中进行处理，获取该分类模型输出的分类结果，该分类结果可以是第一对象是否为喷嚏飞沫。可理解的，除了上面提到的时间-多普勒谱之外，也可以将无线信号在别的维度上进行分析，如：将距离-多普勒-时间谱的三维信息，作为进一步输入分类模型的信息。其中，该分类模型可以为卷积神经网络模型。假设卷积神经网络的输入层为二维信息(即图片或二维矩阵)，第一设备可以将上述无线信号的多普勒信息都展开为一个二维的矩阵，输入神经网络，通过后续的卷积层进行特征提取，再通过分类层进行分类，最终判定第一对象是否为喷嚏飞沫。
可选的，第一设备可以对上述无线信号的多普勒信息进行特征提取，得到第一输入特征，再可以将该第一输入特征输入分类模型中进行处理，获取该分类模型输出的分类结果，该分类结果可以是第一对象是否为喷嚏飞沫。该分类模型可以为深度神经网络，也可以为别的分类器。
可选的，参见图7，图7是本申请实施例提供的喷嚏飞沫的距离-多普勒谱的示意图。因为速度和多普勒分量存在对应关系，所以可以利用速度来反映多普勒信息。如图7所示，横轴表示速度(单位m/s)，纵轴表示距离(单位m)。图7中最右边的矩形长条(灰度长条)表示能量维度的大小，不同灰度表示不同能量大小。喷嚏飞沫通常具有较高的速度(喷嚏飞沫的多普勒特征一般出现在较高的多普勒位置)，且喷嚏飞沫在多普勒频域上扩展较大。所以，第一设备可以从上述无线信号的多普勒信息中提取出该第一对象对应的多普勒信息，并可以将提取出来的该第一对象对应的多普勒信息输入一个二维神经网络的识别器中进行识别。第一设备可以获取该识别器输出的识别结果，即该第一对象对应的多普勒信息中是否存在喷嚏飞沫的多普勒特征，并可以将该识别结果输入判决器中进行判决，判断该第一对象是否为喷嚏飞沫。可理解的，如果该第一对象对应的多普勒信息中存在喷嚏飞沫的多普勒特征，说明该第一对象是喷嚏飞沫；反之，如果该第一对象对应的多普勒信息中不存在喷嚏飞沫的多普勒特征，说明该第一对象不是喷嚏飞沫。
可选的，第一设备可以将上述无线信号的整个多普勒信息输入一个二维神经网络的识别器中进行识别，无需提取其中部分多普勒信息。第一设备可以获取该识别器输出的识别结果，即该无线信号的多普勒信息中是否存在喷嚏飞沫的多普勒特征，并可以将该识别结果输入判决器中进行判决，判断该第一对象是否为喷嚏飞沫。
可选的，第一设备可以将上述无线信号的多普勒信息进行模板匹配，确定第一对象是否为喷嚏飞沫。例如，第一设备可以将该无线信号的多普勒信息与喷嚏飞沫的多普勒信息进行匹配。如果该无线信号的多普勒信息与喷嚏飞沫的多普勒信息之间的匹配度/相似度大于一个阈值，则可以确定第一对象是喷嚏飞沫。
可理解的，如果第一对象是喷嚏飞沫，则说明空间中存在打喷嚏。本申请实施例利用喷嚏飞沫对无线信号的多普勒信息产生的特殊影响，来识别喷嚏飞沫，从而识别出是否打喷嚏，该识别过程可以不受环境中的光照、噪声以及遮挡物的影响，提高飞沫检测的适用 性。
作为一个可选实施例，接收设备可以接收/采集无线信号，对接收/采集到的无线信号进行预处理，比如滤波处理等；并可以对预处理后的该无线信号进行多普勒估计，得到该无线信号的多普勒信息。接收设备将该无线信号的多普勒信息发送至云端计算中心，云端计算中心基于该无线信号的多普勒信息，确定第一对象是否为喷嚏飞沫。本申请实施例将无线信号的多普勒信息发送至云端计算中心进行处理，可以利用云端计算中心强大的计算能力，可以减少接收设备的计算复杂度。
作为另一个可选实施例，因为人在进行相应动作时，运动的肢体会对空间内的信号产生多普勒调制，所以对接收/采集到的无线信号进行相应的分析，如时间-频率分析，可以在时间-多普勒维度上发现多普勒随时间变化的特征，这些特征对应于人体真实的肢体运动，可以被用来进行进一步的动作识别。因此，本申请实施例提供的基于无线信号感知打喷嚏的方法不仅可以用于感知打喷嚏，还可以用于感知咳嗽。具体地，第一设备可以获取无线信号，该无线信号在包含第三对象的空间内传播，该第三对象可以为咳嗽动作。第一设备可以对该无线信号进行到达角估计、距离估计、多普勒估计或多维联合处理等信号处理，得到该无线信号的多普勒信息。该无线信号的多普勒信息可以用于反映该第三对象对该无线信号的频率产生的影响。第一设备可以基于该无线信号的多普勒信息，确定该第三对象是否为咳嗽动作。
实施例二
本申请实施例二一方面联合喷嚏飞沫和打喷嚏的动作进行识别，可以减少误判；另一方面，不仅考虑喷嚏飞沫对无线信号的多普勒信息产生的影响，还结合喷嚏飞沫对无线信号的衰减谱/宽带谱产生的影响，来综合判断是否为喷嚏飞沫，进一步提高准确性。此外，在无法获得无线信号的多普勒信息的情况下，可通过无线信号的衰减谱/宽带谱来识别喷嚏飞沫，实现了极端情况下的喷嚏飞沫识别/打喷嚏感知。
参见图8，图8是本申请实施例提供的基于无线信号感知打喷嚏的方法的另一示意流程图。如图8所示，该基于无线信号感知打喷嚏的方法包括但不限于以下步骤：
S201，第一设备获取无线信号，该无线信号在包含第一对象和第二对象的空间内传播。
其中，上述第一对象可以为空间中的飞沫、小水滴等；上述第二对象可以为人的肢体动作。
具体地，本申请实施例中步骤S201的实现方式可以参考图4所示实施例的步骤S101的实现方式，在此不再赘述。
可选的，第一设备获取到无线信号后，可以对该无线信号进行信号预处理，比如滤波。预处理后的无线信号可以分别进行三种处理，如下述步骤S202、步骤S203以及步骤S204。可理解的，下述步骤S202、步骤S203以及步骤S204中的无线信号可以是预处理后的无线信号。
S202，第一设备对该无线信号进行到达角估计、距离估计以及多普勒估计，确定第一对象的空间位置信息和第二对象的空间位置信息。
具体地，第一设备对上述无线信号进行到达角估计(如波束赋形算法、子空间类方法或其他阵列信号处理方法等)，距离估计(如匹配滤波等信号处理算法)以及多普勒估计(如傅里叶变换等信号处理算法)等信号处理，在相关的信号维度和区域内进行打喷嚏(即第二对象)和飞沫(即第一对象)的定位，即：确定第一对象的空间位置信息和第二对象的空间位置信息。其中，到达角估计、距离估计以及多普勒估计可以参考前述实施例一中的相应描述，在此不再赘述。该第一对象的空间位置信息可以包括经过该第一对象反射的无线信号相对于接收设备的第一到达角、和该第一对象与接收设备的第一距离；该第二对象的空间位置信息可以包括经过该第二对象反射的无线信号相对于接收设备的第二到达角、和该第二对象与接收设备的第二距离。
可理解的，运动目标(如第一对象和第二对象)的定位一般需要多普勒帮助进行检测。如上述图7所示，打喷嚏动作通常具有较低的速度(打喷嚏动作的多普勒特征一般出现在较低的多普勒位置)，且打喷嚏动作在多普勒频域上的扩展相对较小；喷嚏飞沫通常具有较高的速度(喷嚏飞沫的多普勒特征一般出现在较高的多普勒位置)，且喷嚏飞沫在多普勒频域上的扩展较大。通过诸如此类的特征，可以判断出检测的目标是打喷嚏动作还是喷嚏飞沫。
还可理解的，打喷嚏动作的位置(这里指空间坐标或距离)和喷嚏飞沫的起始位置(这里指空间坐标或距离)相差不大。所以在初始阶段，打喷嚏动作的位置和喷嚏飞沫的起始位置可以等效为一个位置。但随着时间的变化，打喷嚏动作的位置与喷嚏飞沫的位置相差逐渐增大。
可选的，第一对象和第二对象的一种可能定位流程包括：(a)在每个天线的接收信号上进行距离-多普勒处理，产生如上述图7所示的结果。横轴为多普勒或速度，纵轴为距离。这里的距离是指距离估计处理后得到的传播距离差。(b)将多个天线得到的距离-多普勒图进行非相干积累(直接矩阵叠加)，提升信噪比。叠加之后得到一张距离-多普勒图。(c)对叠加之后的距离-多普勒图进行检测，判断运动目标(如第一对象、第二对象)在该距离-多普勒的二维图上对应的距离。(d)联合所有天线进行到达角估计，具体为选取每个天线的距离-多普勒二维图中运动目标对应的距离，针对每个天线进行达到角估计(这是因为平面波抵达不同阵元的波程差dsinθ与传播距离差存在换算关系，所以求解出到达角θ)。(e)根据估计到的到达角角度和距离信息，对运动目标(如第一对象、第二对象)的位置(这里指空间坐标或运动目标相对收发设备的距离和角度信息)进行判断。
S203，第一设备对该无线信号进行到达角估计、距离估计以及多普勒估计，得到该无线信号的多普勒信息，该无线信号的多普勒信息用于反映第一对象和第二对象对该无线信号的频率产生的影响。
具体地，第一设备对上述无线信号进行到达角估计，距离估计，多普勒估计或多维联合处理等信号处理，得到该无线信号的多普勒信息。后续可基于该无线信号的多普勒信息在相关的信号维度上进行打喷嚏动作和喷嚏飞沫的识别。该无线信号的多普勒信息可以用于反映上述第一对象和上述第二对象对该无线信号的频率产生的影响。该第一对象对该无线信号的频率产生的影响与该第二对象对该无线信号的频率产生的影响不相同，即：该第一对象对应的多普勒特征与该第二对象对应的多普勒特征不相同。
其中，该无线信号的多普勒信息可以包括时间-多普勒维度、距离-多普勒-时间维度、或其他维度的多普勒信息。因此在获得无线信号的多普勒信息的过程中，也可能需要角度信息(这里指到达角)和/或距离信息(这里指传播距离差)。以人为例，人在行走或者进行相应动作时，运动的肢体会对信号产生多普勒调制，所以对接收/采集到的信号进行相应的分析，如时间-频率分析，可以在时间-多普勒维度上发现多普勒随时间变化的特征，这些特征可以对应于人体真实的肢体运动，可以被用来进行进一步的动作识别。同理，飞沫也会产生独特的特征，对应飞沫的真实运动。因此，运动的目标(这里指第一对象或第二对象)不仅在时间-多普勒维度上具有独特的特征；运动的目标也可以在其他维度上产生相应的具有对应目标运动特性的信息，如在时间-距离维度上，可以反映出距离随时间变化的信息，该信息同样可以辅助后续打喷嚏动作和喷嚏飞沫的识别。运动的目标还存在时间-角度信息，或者其他联合维度的信息，也可以辅助后续打喷嚏动作和喷嚏飞沫的识别。
S204，第一设备获取该无线信号的衰减谱或宽带谱，该无线信号的衰减谱用于反映该第一对象对该无线信号的幅度衰减产生的影响，该无线信号的宽带谱用于反映该第一对象对该无线信号的宽带频谱能量产生的影响。
具体地，第一设备可以对上述无线信号进行滑动平均等信号处理，并获取该滑动平均等信号处理后的无线信号的衰减谱或宽带谱。该无线信号的衰减谱可以用于反映上述第一对象对该无线信号的幅度衰减产生的影响，该无线信号的宽带谱用于反映该第一对象对该无线信号的宽带频谱能量产生的影响。其中，衰减谱可以为时间-能量维度的信息，宽带谱可以为频率-能量维度的信息或频率-能量-时间维度的信息。
可理解的，信号的衰减谱是指信号的幅度随时间的变化，当存在喷嚏飞沫/小水滴时，喷嚏飞沫/小水滴会对信号产生相应的衰减，从而在幅度上形成相应的低谷，这些特征可以作为后续喷嚏飞沫识别的输入信息。参见图9，图9是本申请实施例提供的信号衰减谱示意图。如图9所示，横轴为时间，纵轴为接收到的信号能量。由图9可以看出，在喷嚏飞沫/小水滴的出现时间点如1.3s至2.5s之间产生了对应的衰减特征。信号的宽带谱是指信号的宽带谱特征，即横轴为频率，纵轴为对应频率的能量，在大带宽条件下，不同信号/电磁波穿过不同介质所形成的宽带谱特征不同，所以无线信号穿过喷嚏飞沫所形成的宽带谱特征也具有特定的特性，故这些信息可以作为后续喷嚏飞沫识别的输入信息，进行喷嚏飞沫的识别检测。
可选的，上述步骤S202、步骤S203以及步骤S204，可以三个步骤并行执行；也可以是其中两个步骤并行执行，另外一个步骤既可以在并行执行的两个步骤之前，也可以在并行执行的两个步骤之后；还可以三个步骤顺序执行。顺序执行时，可以是步骤S202、步骤S203、步骤S204，也可以是步骤S202、步骤S204、步骤S203，还可以是步骤S204、步骤S202、步骤S203；还可以是步骤S204、步骤S203、步骤S202；还可以是步骤S203、步骤S202、步骤S204；还可以是步骤S203、步骤S204、步骤S202。
S205，第一设备基于该无线信号的多普勒信息和该无线信号的衰减谱/宽带谱，确定第一对象是否为喷嚏飞沫和第二对象是否为打喷嚏动作。
具体地，在获得上述无线信号的多普勒信息和上述无线信号的衰减谱/宽带谱后，可以将该无线信号的时间-多普勒或距离-时间-多普勒等维度上的信息，联合该无线信号的衰减 谱(时间-能量维度上的信息)或宽带谱(频率-能量或频率-能量-时间等维度上的信息)，整体输入到一个识别器(可以是神经网络，也可以是非神经网络)中进行联合识别，对第二对象是否为打喷嚏动作和第一对象是否为喷嚏飞沫进行识别。换句话说，这里以卷积神经网络为例，对此过程进行说明。假设卷积神经网络的输入层为二维信息(即图片或二维矩阵)，则可以将所有的特征信息(这里指无线信号的多普勒信息、衰减谱或宽带谱)都展开为一个二维的矩阵，输入该卷积神经网络，通过后续的卷积层进行特征提取，再通过分类层进行分类，最终判定是否出现打喷嚏动作并产生喷嚏飞沫。
可理解的，结合上述步骤S202确定的空间位置信息，便可以感知到打喷嚏动作和喷嚏飞沫的完整信息。
可选的，第一设备也可以将打喷嚏动作的识别结果(即第二对象是否是打喷嚏动作)和喷嚏飞沫的识别结果(第一对象是否是喷嚏飞沫)输入到后面的判决器中，为最终的判断提供信息。换句话说，第一设备可以分别针对各种信息(包括无线信号的多普勒信息、衰减谱或宽带谱信息等)进行基于二维神经网络的识别，对得到的识别结果(即：是否存在打喷嚏动作，是否出现喷嚏飞沫的多普勒特征，是否出现喷嚏飞沫的衰减特征)进行融合判断。可理解的，第一设备可以将上述无线信号的多普勒信息输入一个二维神经网络的识别器中进行识别，输出识别结果，如是否存在打喷嚏动作以及是否出现喷嚏飞沫的多普勒特征。第一设备可以将该无线信号的衰减谱或宽带谱输入另一个二维神经网络的识别器中进行识别，输出识别结果，如是否出现喷嚏飞沫的衰减特征。
参见图10，图10是本申请实施例提供的真实喷嚏多普勒测量结果示意图。本次测量过程中，目标人放松的坐在实验设备(这里指发送设备和接收设备)前，打喷嚏并且产生喷嚏飞沫。图10中横轴为时间，纵轴为速度，可以看出在速度为5米/秒甚至以上的位置存在比较明显的速度分量。这些速度分量远大于人体的正常运动速度，所以可以通过在高速度位置进行检测判断是否出现喷嚏飞沫。另一方面，这些速度分量还具有独特的时间-速度特征，可以作为识别的有效输入信息，进一步判断是否出现喷嚏飞沫。
其中，识别器输出的识别结果可以为2bit的二进制，比如，高bit位表示第二对象是否是打喷嚏动作，低bit位表示第一对象是否是喷嚏飞沫。例如，“10”表示第二对象是打喷嚏动作且第一对象不是喷嚏飞沫，“11”表示第二对象是打喷嚏动作且第一对象是喷嚏飞沫，“00”表示第二对象不是打喷嚏动作且第一对象不是喷嚏飞沫。本申请中识别器输出的识别结果可以有3种取值，即“00”、“10”以及“11”，只有在取值为“11”的情况下，判决器输出的结果才是“1”，说明既有打喷嚏动作也存在喷嚏飞沫。
可选的，第一设备可以对上述无线信号的多普勒信息和上述无线信号的衰减谱/宽带谱进行特征提取和特征融合，得到输入特征。第一设备可以将该输入特征输入分类器中进行处理，并获取该分类器输出的分类结果，如第一对象是否是喷嚏飞沫且第二对象是否是打喷嚏动作等。换句话说，第一设备可以针对各种信息(包括无线信号的多普勒信息、衰减谱或宽带谱等信息)，分别进行特征提取，将提取到的信号特征，联合输入到后续的分类网络之中，进行打喷嚏动作识别和喷嚏飞沫识别，最终判定是否出现打喷嚏动作并产生喷嚏飞沫。
可理解的，本申请实施例，一方面联合喷嚏飞沫和打喷嚏的动作进行识别，可以减少 因为其他外部原因引起的误判，如测量多普勒时，喷壶等所喷出的飞沫与喷嚏飞沫的特征具有一定相似性，结合打喷嚏的动作识别结果，可以降低此类情况引起的误判。又如测量衰减时，结合打喷嚏的动作进行识别，可以降低其他原因导致的类似衰减谱所引起的误判。另一方面，本申请实施例不仅考虑喷嚏飞沫对无线信号的多普勒信息产生的影响，还结合喷嚏飞沫对无线信号的衰减谱/宽带谱产生的影响，即利用多普勒和衰减的双检测，来综合判断是否为喷嚏飞沫，进一步提高准确性。
此外，在无法获得无线信号的多普勒信息的情况下，可通过无线信号的衰减谱/宽带谱来识别喷嚏飞沫，可以实现全方位角度的喷嚏飞沫识别/打喷嚏感知。具体原因如下：参见图11，图11是本申请实施例提供的多普勒与双基地夹角的关系示意图。如图11所示，图11示出了通过无线信号可以感知到的多普勒信息与发送设备、目标、接收设备三者构成的双基地夹角β的关系。其中，通过无线信号感知到的多普勒f d与双基地夹角β满足下述公式：
Figure PCTCN2021107962-appb-000010
其中，公式(2-1)中v表示目标(如第一对象或第二对象)的运动速度，λ表示载波的波长，δ表示目标的运动方向与双基地夹角β的角平分线的夹角。
可理解的，当双基地夹角β等于0°时，图11退化为一个单基地场景，通过无线信号可以感知的多普勒达到最大。而当β趋近180°时，多普勒衰减最小，无法通过多普勒对目标(即第一对象或第二对象)进行感知。所以，可以通过对无线信号的衰减谱或者宽带谱分析来进行喷嚏飞沫识别/检测。故结合无线信号的多普勒信息和衰减谱，可以实现全方位角度的喷嚏飞沫识别/打喷嚏感知。
作为一个可选实施例，本申请实施例在步骤S205之后，还包括步骤S206：第一设备输出以下一种或多种信息：第一对象是否为喷嚏飞沫、第二对象是否为打喷嚏动作、第一对象的空间位置信息或第二对象的空间位置信息。
具体地，第一设备可以向与这个第一设备关联的移动设备发送以下一种或多种信息：第一对象是否为喷嚏飞沫、第二对象是否为打喷嚏动作、第一对象的空间位置信息或第二对象的空间位置信息。可选的，第一设备也可以将第一设备的识别结果和定位结果(如第一对象是否为喷嚏飞沫、第二对象是否为打喷嚏动作、第一对象的空间位置信息或第二对象的空间位置信息)上传至云端，由云端根据打喷嚏的人数、喷嚏飞沫的范围等信息，提醒相关人员进行回避或处理，例如提供周围的人员回避喷嚏飞沫区域，或者提供保洁人员对喷嚏飞沫区域进行清洁。参见图12，图12是本申请实施例提供的信息输出的场景示意图。如图12所示，第一设备为WLAN AP，WLAN AP可以直接将信息(上述步骤S202和步骤S205确定出的信息)发送给移动设备，还可以将信息传输至后台服务器，由后台服务器根据打喷嚏的人数、喷嚏飞沫的范围等信息通知移动终端相关信息。
可选的，相关信息的通知可以采用文字推送的方式，也可以结合增强现实(augmented reality，AR)/虚拟现实(virtual reality，VR)等技术标记喷嚏飞沫的范围进行提醒。参见图13，图13是本申请实施例提供的基于AR的相关信息通知示意图。如图13所示，可以 通过AR等相关技术，在移动设备上现实虚拟目标人物和喷嚏飞沫在AR环境中的大小、位置以及喷嚏飞沫的移动信息(比如移动轨迹)。
本申请实施例通过输出各种信息，可以提醒相关人员喷嚏飞沫的出现区域和影响范围，规避潜在的传染风险。
可理解的，单个设备观测到的喷嚏飞沫的速度只是喷嚏飞沫的真实速度在某个观测角度上的投影，并不是喷嚏飞沫的真是速度。参见图14，图14是本申请实施例提供的喷嚏飞沫的真实速度合成示意图。如图14所示，以2个观测节点为例，如设备2和设备3，设备1为发送无线信号的设备，设备2和设备3为接收/采集无线信号的设备。如图14所示，根据上述公式(2-1)，可以得到设备2观测到的速度v 2为：
Figure PCTCN2021107962-appb-000011
公式(2-2)中，f d表示设备2通过无线信号感知到的多普勒，β表示设备1、喷嚏飞沫以及设备2三者构成的双基地夹角，δ表示喷嚏飞沫的运动方向与双基地夹角β的角平分线的夹角。
设备3观测到的速度v 3为：
Figure PCTCN2021107962-appb-000012
公式(2-3)中，f' d表示设备3通过无线信号感知到的多普勒，β’表示设备1、喷嚏飞沫以及设备3三者构成的双基地夹角，δ’表示喷嚏飞沫的运动方向与双基地夹角β’的角平分线的夹角。
因此，基于多个设备的部署位置关系和设备2和设备3分别观测得到的速度，可以合成喷嚏飞沫的真实速度和方向。如果进一步结合喷嚏飞沫自身的运动模型，就可以较好的估计出喷嚏飞沫的扩散模型，从而判断出喷嚏飞沫的大致影响范围。
作为另一个可选实施例，为了更好地理解本申请实施例的技术方案，下面结合本申请实施例的一种可能的流程示例对本申请实施例的技术方案进行说明。参见图15，图15是本申请实施例提供的一种流程示例图。如图15所示，当WLAN设备采集到无线信号之后，先进行信号的预处理，预处理包括滤波等步骤。信号预处理之后，分别将这个预处理后的信号进行三种处理。第一种处理：经过多普勒估计、到达角估计、距离估计、联合估计等信号处理步骤，对打喷嚏(动作)和喷嚏飞沫进行定位。第二种处理：经过时间-频率联合处理、距离-频率-时间联合处理等，获得打喷嚏(动作)和喷嚏飞沫的时间-多普勒谱、距离-多普勒-时间谱。第三种处理：经过滑动平均等信号处理，进行信号衰减谱/宽带谱分析。将第二种处理和第三种处理的结果进行特征提取和特征融合(包括时-频/时-空/时-距特征、衰减特征等)，再将特征提取和特征融合的结果输入分类器(神经网络/非神经网络)中对打喷嚏(动作)且出现喷嚏飞沫进行识别，最后输出第一种处理的定位和喷嚏飞沫的检测结果，即喷嚏定位和喷嚏飞沫检测。
本申请实施例，首先对接收/采集到的无线信号进行预处理，然后对预处理后的无线信号分别进行三种处理；第一种处理：对预处理后的无线信号进行到达角估计、距离估计以及多普勒估计，实现打喷嚏动作和喷嚏飞沫的定位；第二种处理：对预处理后的无线信号进行到达角估计、距离估计、多普勒估计以及多维联合处理，实现打喷嚏动作和喷嚏飞沫在多普勒维度或其他联合维度上的检测/识别；第三种处理：获取预处理后的无线信号的衰减谱或宽带谱；最后结合三种处理的结果，实现打喷嚏动作和喷嚏飞沫的定位，以及判断是否出现喷嚏飞沫。可以减少误判，提高准确性以及实现全方位角度的喷嚏飞沫识别/打喷嚏感知。
作为又一个可选实施例，本申请提供的基于无线信号感知打喷嚏的方法还可以应用于多基地联合感知的场景或者多发多收的感知场景中。可理解的，本申请中提及的“多基地”可以指多个接收设备。
具体地，参见图16，图16是本申请实施例提供的多基地联合感知场景的示意图。如图16所示，以3个WLAN设备/节点为例，分别为WLAN设备1(或节点1)、WLAN设备2(或节点2)以及WLAN设备3(或节点3)，并且该空间内有一目标人(如图16中的target human)正在打喷嚏，且存在打喷嚏而产生的喷嚏飞沫(droplets)。WLAN设备1发送无线信号，该无线信号可以通过直达径到达WLAN设备2和WLAN设备3，也可以经过target human反射后到达WLAN设备2和WLAN设备3，还可以经过喷嚏飞沫(droplets)反射后到达WLAN设备2和WLAN设备3。其中，WLAN设备2接收的无线信号是WLAN设备1到WLAN设备2的直射径信号与多条反射径信号的叠加。同理，WLAN设备3接收的无线信号是WLAN设备1到WLAN设备3的直射径信号与多条反射径信号的叠加。WLAN设备2和WLAN设备3可以分别对自己接收到的无线信号进行多种信号处理，联合多个节点(如WLAN设备2和WLAN设备3)处理后的结果来定位和识别目标对象的打喷嚏动作和打喷嚏所产生的喷嚏飞沫，可以提升空间增益，提升感知效率。
可选的，WLAN设备2和WLAN设备3也可以将自己接收到的无线信号传送至云端计算中心进行处理，由云端计算中心对这些无线信号进行信号处理等。
可理解的，WLAN设备1可以STA，WLAN设备2和WLAN设备3可以为AP。
因为多个节点中每个节点对无线信号的处理方式与前述单个节点对无线信号的处理方式相同，区别在于：每个节点在得到无线信号的多普勒信息和衰减谱/宽带谱后，可以将这些信息发送给某个节点，由这个节点结合每个节点发送的无线信号的多普勒信息和衰减谱/宽带谱，进行识别和检测，确定空间中是否有人打喷嚏且是否出现喷嚏飞沫。为便于描述，下面以处理节点为例进行说明，即：每个节点将得到无线信号的多普勒信息和衰减谱/宽带谱发送给处理节点。下面对处理节点的可能处理过程进行说明。其中，处理节点可以是任一节点。
1、处理节点接收到所有节点发送的无线信号的多普勒信息和衰减谱/宽带谱后，可以将来自所有节点的相关特征信息整合为一个信息矩阵，输入神经网络，通过后续的卷积层进行特征提取，再通过分类层进行分类，最终判定是否出现打喷嚏动作并产生喷嚏飞沫。可选的，结合所有节点的定位信息，便可以感知到打喷嚏动作和喷嚏飞沫的完整信息。
2、处理节点接收到所有节点发送的无线信号的多普勒信息和衰减谱/宽带谱后，可以先针对各项信息，分别进行特征提取，联合来自所有节点的特征输入到后续的分类网络之中，进行打喷嚏动作识别和喷嚏飞沫识别，最终判定是否出现打喷嚏动作并产生喷嚏飞沫。可选的，结合所有节点的定位信息，便可以感知到打喷嚏动作和喷嚏飞沫的完整信息。
3、处理节点接收到所有节点发送的无线信号的多普勒信息和衰减谱/宽带谱后，可以分别对来自这些信息进行基于二维神经网络的识别，对所有节点得到的识别结果(是否出现打喷嚏动作并产生喷嚏飞沫)进行融合判断。
本申请实施例通过联合多个节点，不仅可以实现打喷嚏动作和喷嚏飞沫的定位和识别/检测，以及实现全方位角度的打喷嚏感知；还可以提升空间增益，提升感知效率。这是因为如果发送节点、目标(人)、某个接收节点在一条直线上，则双基地夹角β等于180度，则在这个接收节点上无法测得多普勒信息，因此其他一个或多个的接收节点可以检测到多普勒信息，从而实现打喷嚏动作和喷嚏飞沫的识别，进而提升空间增益，提升感知效率。另一方面，如果在某些实际情况下，无需获取无线信号的衰减谱(比如某些实际场景中，WLAN设备都布设在屋顶)，说明双基地夹角β不会等于180度的情况，则本申请实施例联合多个节点仍然可以实现全方位角度的打喷嚏动作和喷嚏飞沫的定位和识别/检测。
为更好地理解多基地感知场景下本申请实施例提供的基于无线信号感知打喷嚏的方法，参见图17，图17是本申请实施例提供的另一种流程示例图。如图17所示，节点1和节点2分别采集无线信号，并分别对各自采集到的无线信号进行信号的预处理，预处理包括滤波等步骤。信号预处理之后，节点1和节点2分别将这个预处理后的信号进行三种处理。第一种处理：经过多普勒估计、到达角估计、距离估计、联合估计等信号处理步骤，对打喷嚏(动作)和喷嚏飞沫进行定位。第二种处理：经过时间-频率联合处理、距离-频率-时间联合处理等，获得打喷嚏(动作)和喷嚏飞沫的时间-多普勒谱、距离-多普勒-时间谱。第三种处理：经过滑动平均等信号处理，进行信号衰减谱/宽带谱分析。节点1和/或节点2将各自进行的第二种处理和第三种处理的结果发送给处理节点。处理节点可以是节点1和节点2中的任一个节点。处理节点将节点1和节点2各自进行的第二种处理和第三种处理的结果进行特征提取和特征融合(包括时-频/时-空/时-距特征、衰减特征等)，再将特征提取和特征融合的结果输入分类器(神经网络/非神经网络)中对打喷嚏(动作)且出现喷嚏飞沫进行识别，最后输出第一种处理的定位和喷嚏飞沫的检测结果，即喷嚏定位和喷嚏飞沫检测。
上述详细阐述了本申请实施例的基于无线信号感知打喷嚏的方法，为了便于更好地实施本申请实施例的上述方案，本申请实施例还提供了相应的装置或设备。
参见图18，图18是本申请实施例提供的电子设备的一结构示意图。该电子设备可以为WLAN中的AP或STA，或者是，安装在AP或STA中的芯片或处理系统或电路，还可以是云端计算中心。如图18所示，该电子设备100可包括：
第一获取模块10，用于获取无线信号，该无线信号在包含第一对象的空间内传播；第一处理模块20，用于对该无线信号进行多普勒估计，得到该无线信号的多普勒信息，该无线信号的多普勒信息用于反映该第一对象对该无线信号的频率产生的影响；第一确定模块 30，用于基于该无线信号的多普勒信息，确定该第一对象是否为喷嚏飞沫。
可选的，上述无线信号还可以在包含第二对象的空间内传播，该无线信号的多普勒信息还用于反映该第二对象对该无线信号的频率产生的影响，该第一对象对该无线信号的频率产生的影响与该第二对象对该无线信号的频率产生的影响不同。该电子设备100还包括第二确定模块40。该第二确定模块40，用于基于该无线信号的多普勒信息，确定该第二对象是否为打喷嚏动作。其中，第一确定模块30和第二确定模块40，可以是同一模块，也可以是不同模块。
可选的，上述电子设备100还可以包括第二处理模块50。该第二处理模块50，用于对该无线信号进行到达角估计、距离估计以及多普勒估计，得到该第一对象的空间位置信息和该第二对象的空间位置信息。其中，该第一对象的空间位置信息包括经过该第一对象反射的无线信号相对于接收设备的第一到达角、和该第一对象与该接收设备的第一距离，该第二对象的空间位置信息包括经过该第二对象反射的无线信号相对于该接收设备之间的第二到达角、和该第二对象与该接收设备的第二距离。
可选的，上述电子设备100还可以包括输出模块60。该输出模块60，用于输出以下一种或多种信息：该第一对象是否为喷嚏飞沫、该第二对象是否为打喷嚏动作、该第一对象的空间位置信息或该第二对象的空间位置信息。
可选的，上述电子设备100还可以包括第二获取模块70和第三确定模块80。该第二获取模块70，用于获取该无线信号的衰减谱或宽带谱，该无线信号的衰减谱用于反映该第一对象对该无线信号的幅度衰减产生的影响，该无线信号的宽带谱用于反映该第一对象对该无线信号的宽带频谱能量产生的影响；该第三确定模块80，用于基于该无线信号的多普勒信息和该无线信号的衰减谱或宽带谱，确定该第一对象是否为喷嚏飞沫。
可选的，上述第一确定模块30具体用于：将该无线信号的多普勒信息进行特征提取，得到第一输入特征；将该第一输入特征输入分类模型中进行处理，输出分类结果，该分类结果为该第一对象是否为喷嚏飞沫。
可选的，上述第一确定模块30具体用于：将该无线信号的多普勒信息输入分类模型中进行处理，输出分类结果，该分类结果为该第一对象是否为喷嚏飞沫。
可选的，上述第一确定模块30具体用于：将该无线信号的多普勒信息分为第一多普勒信息和第二多普勒信息，该第一多普勒信息在多普勒频域上的扩展小于该第二多普勒信息在多普勒频域上的扩展；将该第一多普勒信息输入第一识别器中进行识别，识别出该第二对象是否为打喷嚏动作；将该第二多普勒信息输入第二识别器中进行识别，识别出该第二多普勒信息中是否存在该第一对象的多普勒特征；将该第二对象是否为打喷嚏动作以及该第二多普勒信息中是否存在该第一对象的多普勒特征输入判决器，判断该第一对象是否为喷嚏飞沫。
其中，上述第一获取模块10、上述第一处理模块20、上述第一确定模块30、上述第二确定模块40、上述第二处理模块50、上述第二获取模块70以及上述第三确定模块80可以集成为一个模块，如处理模块。上述输出模块60也可以是收发模块。
具体实现中，如上所示的各个模块或单元的实现还可以对应参照图4或图8所示的实施例中第一设备的相应描述，执行上述任一实施例中第一设备所执行的方法和功能。
本申请实施例提供的电子设备100可执行上述第一设备执行的基于无线信号感知打喷嚏的方法，其具体的实现过程及有益效果参见前述任一实施例的描述，在此不再赘述。
参见图19，图19是本申请实施例提供的电子设备的另一结构示意图。如图19所示，本申请实施例提供的电子设备1000包括处理器1001、存储器1002和总线系统1004。可选的，该电子设备1000还可以包括收发器1003。其中，处理器1001、存储器1002和收发器1003通过总线系统1004连接。
上述处理器1001，用于获取无线信号，该无线信号在包含第一对象的空间内传播；对该无线信号进行多普勒估计，得到该无线信号的多普勒信息，该无线信号的多普勒信息用于反映该第一对象对该无线信号的频率产生的影响；基于该无线信号的多普勒信息，确定该第一对象是否为喷嚏飞沫。
可选的，上述收发器1003可以用于输出以下一种或多种信息：该第一对象是否为喷嚏飞沫、该第二对象是否为打喷嚏动作、该第一对象的空间位置信息或该第二对象的空间位置信息。
上述存储器1002用于存放程序。具体地，程序可以包括程序代码，程序代码包括计算机操作指令。存储器1002包括但不限于是随机存储记忆体(random access memory，RAM)、只读存储器(read-only memory，ROM)、可擦除可编程只读存储器(erasable programmable read only memory，EPROM)、或便携式只读存储器(compact disc read-only memory，CD-ROM)。图19中仅示出了一个存储器，当然，存储器也可以根据需要，设置为多个。存储器1002也可以是处理器1001中的存储器，在此不做限制。
存储器1002存储了如下的元素，可执行单元或者数据结构，或者它们的子集，或者它们的扩展集：
操作指令：包括各种操作指令，用于实现各种操作。
操作系统：包括各种系统程序，用于实现各种基础业务以及处理基于硬件的任务。
上述处理器1001控制电子设备1000的操作，处理器1001可以是一个或多个中央处理器(central processing unit，CPU)，在处理器1001是一个CPU的情况下，该CPU可以是单核CPU，也可以是多核CPU。
具体的应用中，电子设备1000的各个组件通过总线系统1004耦合在一起，其中总线系统1004除包括数据总线之外，还可以包括电源总线、控制总线和状态信号总线等。但是为了清楚说明起见，在图19中将各种总线都标为总线系统1004。为便于表示，图19中仅是示意性画出。
可选的，如上所示的电子设备1000中还可通过处理器1001、存储器1002和收发器1003相配合，以执行上述第一设备执行的基于无线信号感知打喷嚏的方法。
可选的，本申请实施例还提供一种计算机程序产品，该计算机程序产品包括计算机程序代码，当该计算机程序代码在计算机上运行时，使得该计算机执行图4或图8所描述的第一设备的方法步骤。
该计算机程序产品中的计算机程序代码，例如可由上述图19所示的电子设备1000中的处理器1001执行，用以控制收发器1003，使其配合执行前述任一实施例执行的基于无 线信号感知打喷嚏的方法。
该计算机程序产品的各功能可以通过硬件或软件来实现，当通过软件实现时，可以将这些功能存储在计算机可读存储介质中或者作为计算机可读存储介质上的一个或多个指令或代码进行传输。
可选的，本申请实施例还提供一种计算机可读存储介质，该计算机可读存储介质中存储有计算机程序代码，当上述处理器执行该计算机程序代码时，电子设备执行前述任一实施例的方法。该计算机可读存储介质可以为上述图19所示的电子设备1000中的内部存储器，也可以为与上述电子设备1000连接的外部存储器。
可选的，本申请实施例还提供了一种装置，该装置可以以芯片的产品形态存在，该装置的结构中包括处理电路和接口电路，该处理电路用于执行前述任一实施例的方法，该接口电路用于与其它装置通信。
本申请实施例的电子设备、计算机可读存储介质、计算机程序产品以及芯片，可执行前述任一实施例的基于无线信号感知打喷嚏的方法，其具体的实现过程及有益效果参见前述反方实施例，在此不再赘述。
结合本申请公开内容所描述的方法或者算法的步骤可以硬件的方式来实现，也可以是由处理器执行软件指令的方式来实现。软件指令可以由相应的软件模块组成，软件模块可以被存放于随机存取存储器(Random Access Memory，RAM)、闪存、可擦除可编程只读存储器(Erasable Programmable ROM，EPROM)、电可擦可编程只读存储器(Electrically EPROM，EEPROM)、寄存器、硬盘、移动硬盘、只读光盘(CD-ROM)或者本领域熟知的任何其它形式的存储介质中。一种示例性的存储介质耦合至处理器，从而使处理器能够从该存储介质读取信息，且可向该存储介质写入信息。当然，存储介质也可以是处理器的组成部分。处理器和存储介质可以位于ASIC中。另外，该ASIC可以位于核心网接口设备中。当然，处理器和存储介质也可以作为分立组件存在于核心网接口设备中。
本领域技术人员应该可以意识到，在上述一个或多个示例中，本申请所描述的功能可以用硬件、软件、固件或它们的任意组合来实现。当使用软件实现时，可以将这些功能存储在计算机可读介质中或者作为计算机可读介质上的一个或多个指令或代码进行传输。计算机可读介质包括计算机可读存储介质和通信介质，其中通信介质包括便于从一个地方向另一个地方传送计算机程序的任何介质。存储介质可以是通用或专用计算机能够存取的任何可用介质。
以上所述的具体实施方式，对本申请的目的、技术方案和有益效果进行了进一步详细说明，所应理解的是，以上所述仅为本申请的具体实施方式而已，并不用于限定本申请的保护范围，凡在本申请的技术方案的基础之上，所做的任何修改、等同替换、改进等，均应包括在本申请的保护范围之内。

Claims (14)

  1. 一种基于无线信号感知打喷嚏的方法，其特征在于，包括：
    获取无线信号，所述无线信号在包含第一对象的空间内传播；
    对所述无线信号进行多普勒估计，得到所述无线信号的多普勒信息，所述无线信号的多普勒信息用于反映所述第一对象对所述无线信号的频率产生的影响；
    基于所述无线信号的多普勒信息，确定所述第一对象是否为喷嚏飞沫。
  2. 根据权利要求1所述的方法，其特征在于，所述无线信号还在包含第二对象的空间内传播，所述无线信号的多普勒信息还用于反映所述第二对象对所述无线信号的频率产生的影响，所述第一对象对所述无线信号的频率产生的影响与所述第二对象对所述无线信号的频率产生的影响不同；
    所述方法还包括：
    基于所述无线信号的多普勒信息，确定所述第二对象是否为打喷嚏动作。
  3. 根据权利要求1或2所述的方法，其特征在于，所述获取无线信号之后，所述方法还包括：
    对所述无线信号进行到达角估计、距离估计以及多普勒估计，得到所述第一对象的空间位置信息和所述第二对象的空间位置信息；
    其中，所述第一对象的空间位置信息包括经过所述第一对象反射的无线信号相对于接收设备的第一到达角、和所述第一对象与所述接收设备的第一距离，所述第二对象的空间位置信息包括经过所述第二对象反射的无线信号相对于所述接收设备之间的第二到达角、和所述第二对象与所述接收设备的第二距离。
  4. 根据权利要求1-3任一项所述的方法，其特征在于，所述方法还包括：
    输出以下一种或多种信息：所述第一对象是否为喷嚏飞沫、所述第二对象是否为打喷嚏动作、所述第一对象的空间位置信息或所述第二对象的空间位置信息。
  5. 根据权利要求1-4任一项所述的方法，其特征在于，所述获取无线信号之后，所述方法还包括：
    获取所述无线信号的衰减谱或宽带谱，所述无线信号的衰减谱用于反映所述第一对象对所述无线信号的幅度衰减产生的影响，所述无线信号的宽带谱用于反映所述第一对象对所述无线信号的宽带频谱能量产生的影响；
    基于所述无线信号的多普勒信息和所述无线信号的衰减谱或宽带谱，确定所述第一对象是否为喷嚏飞沫。
  6. 一种电子设备，其特征在于，包括：
    第一获取模块，用于获取无线信号，所述无线信号在包含第一对象的空间内传播；
    第一处理模块，用于对所述无线信号进行多普勒估计，得到所述无线信号的多普勒信息，所述无线信号的多普勒信息用于反映所述第一对象对所述无线信号的频率产生的影响；
    第一确定模块，用于基于所述无线信号的多普勒信息，确定所述第一对象是否为喷嚏飞沫。
  7. 根据权利要求6所述的电子设备，其特征在于，所述无线信号还在包含第二对象的空间内传播，所述无线信号的多普勒信息还用于反映所述第二对象对所述无线信号的频率产生的影响，所述第一对象对所述无线信号的频率产生的影响与所述第二对象对所述无线信号的频率产生的影响不同；
    所述电子设备还包括：
    第二确定模块，用于基于所述无线信号的多普勒信息，确定所述第二对象是否为打喷嚏动作。
  8. 根据权利要求6或7所述的电子设备，其特征在于，所述电子设备还包括：
    第二处理模块，用于对所述无线信号进行到达角估计、距离估计以及多普勒估计，得到所述第一对象的空间位置信息和所述第二对象的空间位置信息；
    其中，所述第一对象的空间位置信息包括经过所述第一对象反射的无线信号相对于接收设备的第一到达角、和所述第一对象与所述接收设备的第一距离，所述第二对象的空间位置信息包括经过所述第二对象反射的无线信号相对于所述接收设备之间的第二到达角、和所述第二对象与所述接收设备的第二距离。
  9. 根据权利要求6-8任一项所述的电子设备，其特征在于，所述电子设备还包括：
    输出模块，用于输出以下一种或多种信息：所述第一对象是否为喷嚏飞沫、所述第二对象是否为打喷嚏动作、所述第一对象的空间位置信息或所述第二对象的空间位置信息。
  10. 根据权利要求6-9任一项所述的电子设备，其特征在于，所述电子设备还包括：
    第二获取模块，用于获取所述无线信号的衰减谱或宽带谱，所述无线信号的衰减谱用于反映所述第一对象对所述无线信号的幅度衰减产生的影响，所述无线信号的宽带谱用于反映所述第一对象对所述无线信号的宽带频谱能量产生的影响；
    第三确定单元，用于基于所述无线信号的多普勒信息和所述无线信号的衰减谱或宽带谱，确定所述第一对象是否为喷嚏飞沫。
  11. 一种电子设备，其特征在于，包括处理器，所述处理器用于执行以下操作：
    获取无线信号，所述无线信号在包含第一对象的空间内传播；
    对所述无线信号进行多普勒估计，得到所述无线信号的多普勒信息，所述无线信号的多普勒信息用于反映所述第一对象对所述无线信号的频率产生的影响；
    基于所述无线信号的多普勒信息，确定所述第一对象是否为喷嚏飞沫。
  12. 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中存储有指令，当所述指令在计算机上运行时，使得所述计算机执行如权利要求1-5任一项所述的方法。
  13. 一种包含指令的计算机程序产品，其特征在于，当所述指令在计算机上运行时，使得所述计算机执行如权利要求1-5任一项所述的方法。
  14. 一种芯片或芯片系统，其特征在于，包括处理电路，所述处理电路用于执行以下操作：
    获取无线信号，所述无线信号在包含第一对象的空间内传播；
    对所述无线信号进行多普勒估计，得到所述无线信号的多普勒信息，所述无线信号的多普勒信息用于反映所述第一对象对所述无线信号的频率产生的影响；
    基于所述无线信号的多普勒信息，确定所述第一对象是否为喷嚏飞沫。
PCT/CN2021/107962 2020-07-29 2021-07-22 基于无线信号感知打喷嚏的方法及相关装置 WO2022022391A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
EP21849813.7A EP4184074A4 (en) 2020-07-29 2021-07-22 WIRELESS SIGNAL-BASED SNEEZE DETECTION METHOD AND RELATED APPARATUS
US18/160,466 US20230168372A1 (en) 2020-07-29 2023-01-27 Method for Sensing Sneezing Based on Wireless Signal, and Related Apparatus

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010746355.0A CN114079851B (zh) 2020-07-29 2020-07-29 基于无线信号感知打喷嚏的方法及相关装置
CN202010746355.0 2020-07-29

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US18/160,466 Continuation US20230168372A1 (en) 2020-07-29 2023-01-27 Method for Sensing Sneezing Based on Wireless Signal, and Related Apparatus

Publications (1)

Publication Number Publication Date
WO2022022391A1 true WO2022022391A1 (zh) 2022-02-03

Family

ID=80037133

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/107962 WO2022022391A1 (zh) 2020-07-29 2021-07-22 基于无线信号感知打喷嚏的方法及相关装置

Country Status (4)

Country Link
US (1) US20230168372A1 (zh)
EP (1) EP4184074A4 (zh)
CN (1) CN114079851B (zh)
WO (1) WO2022022391A1 (zh)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105517493A (zh) * 2013-07-22 2016-04-20 奎伟伍姆英国有限公司 咳嗽检测、分析以及通信平台
JP2017117416A (ja) * 2015-12-25 2017-06-29 富士通株式会社 被感染候補者特定装置、被感染候補者特定方法、および被感染候補者特定プログラム
WO2019239812A1 (ja) * 2018-06-14 2019-12-19 パナソニックＩｐマネジメント株式会社 情報処理方法、情報処理プログラム及び情報処理システム
CN111246892A (zh) * 2018-03-02 2020-06-05 松下知识产权经营株式会社 净化方法、净化装置以及净化系统

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
GB0817570D0 (en) * 2008-09-25 2008-11-05 Secr Defence Improved weather radar
CN103308934B (zh) * 2013-06-20 2015-05-13 上海无线电设备研究所 一种利用wifi反射信号实现室内移动人员定位的方法
CN107343315B (zh) * 2017-08-09 2020-04-10 上海大学 基于无线局域网络多频点联合测量的定位系统及方法
CN110113116B (zh) * 2019-05-14 2021-06-04 山东科技大学 基于wifi信道信息的人体行为识别方法

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105517493A (zh) * 2013-07-22 2016-04-20 奎伟伍姆英国有限公司 咳嗽检测、分析以及通信平台
JP2017117416A (ja) * 2015-12-25 2017-06-29 富士通株式会社 被感染候補者特定装置、被感染候補者特定方法、および被感染候補者特定プログラム
CN111246892A (zh) * 2018-03-02 2020-06-05 松下知识产权经营株式会社 净化方法、净化装置以及净化系统
WO2019239812A1 (ja) * 2018-06-14 2019-12-19 パナソニックＩｐマネジメント株式会社 情報処理方法、情報処理プログラム及び情報処理システム

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
LIU PENG, HUALING ZHANG, DAN LI: "Dynamic Characteristics of Human Droplet Indoor Transmission", REFRIGERATION AND AIR-CONDITIONING, vol. 30, no. 4, 31 August 2016 (2016-08-31), pages 371 - 376, XP055891002, ISSN: 1671-6612 *
See also references of EP4184074A4

Also Published As

Publication number Publication date
EP4184074A4 (en) 2023-11-08
CN114079851B (zh) 2023-02-03
EP4184074A1 (en) 2023-05-24
US20230168372A1 (en) 2023-06-01
CN114079851A (zh) 2022-02-22

Similar Documents

Publication Publication Date Title
Liu et al. Survey on CSI-based indoor positioning systems and recent advances
Liu et al. Survey on WiFi‐based indoor positioning techniques
Koike-Akino et al. Fingerprinting-based indoor localization with commercial MMWave WiFi: A deep learning approach
Milioris et al. Low-dimensional signal-strength fingerprint-based positioning in wireless LANs
Song et al. A survey on indoor positioning technologies
Shi et al. Detection of LSSUAV using hash fingerprint based SVDD
WO2017219603A1 (zh) 运动物体反射无线信号的识别方法
Soltanaghaei et al. Tagfi: Locating ultra-low power wifi tags using unmodified wifi infrastructure
Korany et al. Multiple people identification through walls using off-the-shelf WiFi
Li et al. Wireless localisation in WiFi using novel deep architectures
CN111707986B (zh) 一种基于稀疏面阵的三维参数估计方法
Cinefra An adaptive indoor positioning system based on Bluetooth Low Energy RSSI
Hao et al. A method for improving UWB indoor positioning
US20240053464A1 (en) Radar Detection and Tracking
Liu et al. Deep-learning-based wireless human motion tracking for mobile ship environments
Shih et al. EasyAPPos: Positioning Wi-Fi access points by using a mobile phone
WO2022022391A1 (zh) 基于无线信号感知打喷嚏的方法及相关装置
Kanaris et al. Combining smart lighting and radio fingerprinting for improved indoor localization
US20220201434A1 (en) Coverage extension for device localization through collaborative ranging
Rajaee et al. Energy efficient localization in wireless ad-hoc sensor networks using probabilistic neural network and independent component analysis
Mostafa et al. A survey of indoor localization systems in multi-floor environments
Sun et al. Moving target localization and activity/gesture recognition for indoor radio frequency sensing applications
Chen et al. Toward wide-area contactless wireless sensing
Tian et al. SAPIL: single access point based indoor localisation using Wi‐Fi L‐shaped antenna array
Bhat et al. Gesture Recognition with mmWave Wi-Fi Access Points: Lessons Learned

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: 21849813

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2021849813

Country of ref document: EP

Effective date: 20230216

NENP Non-entry into the national phase

Ref country code: DE