WO2024021824A1 - 无线网络WiFi感知方法、系统及计算机设备 - Google Patents

无线网络WiFi感知方法、系统及计算机设备 Download PDF

Info

Publication number
WO2024021824A1
WO2024021824A1 PCT/CN2023/096573 CN2023096573W WO2024021824A1 WO 2024021824 A1 WO2024021824 A1 WO 2024021824A1 CN 2023096573 W CN2023096573 W CN 2023096573W WO 2024021824 A1 WO2024021824 A1 WO 2024021824A1
Authority
WO
WIPO (PCT)
Prior art keywords
signal
wifi
signals
path
generate
Prior art date
Application number
PCT/CN2023/096573
Other languages
English (en)
French (fr)
Inventor
高静
Original Assignee
中兴通讯股份有限公司
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 中兴通讯股份有限公司 filed Critical 中兴通讯股份有限公司
Publication of WO2024021824A1 publication Critical patent/WO2024021824A1/zh

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B17/00—Monitoring; Testing
    • H04B17/30—Monitoring; Testing of propagation channels
    • H04B17/309—Measuring or estimating channel quality parameters
    • H04B17/318—Received signal strength
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W16/00—Network planning, e.g. coverage or traffic planning tools; Network deployment, e.g. resource partitioning or cells structures
    • H04W16/18—Network planning tools
    • H04W16/20—Network planning tools for indoor coverage or short range network deployment
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W64/00—Locating users or terminals or network equipment for network management purposes, e.g. mobility management
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D30/00—Reducing energy consumption in communication networks
    • Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks

Definitions

  • the present application relates to the field of communication technology, and in particular to a wireless network WiFi sensing method, system and computer equipment.
  • RSSI Received Signal Strength Indicator
  • the WiFi signal will be affected by multiple obstacles. The influence will reach the receiver along multiple paths including refraction and transmission. Among them, the signals on different paths will have different degrees of attenuation and time delay and other distortion phenomena.
  • the signals received by the receiving end are distorted signals of different paths.
  • the superposition result is the so-called multipath effect. Affected by the multipath effect, the stability of the RSSI received indoors is poor, and there will be large fluctuations even in indoor static scenes. Therefore, the indoor multipath effect greatly limits the perception ability of RSSI, making it only available for implementation. Some coarse-grained indoor positioning and other perception tasks.
  • This application provides a wireless network WiFi sensing method, system and computer equipment.
  • the present application provides a wireless network WiFi sensing method.
  • the method includes: obtaining WiFi signals transmitted along a path; detecting the scene status of the WiFi signal coverage area in real time; and analyzing the WiFi signal according to the scene status. Perform signal processing to generate a spectrogram.
  • this application provides a wireless network WiFi sensing system, including: a signal acquisition module configured to acquire WiFi signals transmitted along a path; a detection module configured to detect the scene status of the WiFi signal coverage area in real time; signal processing A module configured to perform signal processing on the WiFi signal according to the scene state and generate a spectrum diagram.
  • the present application provides a computer device.
  • the computer device includes a memory and a processor.
  • Computer-readable instructions are stored in the memory, and the computer-readable instructions are executed by one or more of the processors.
  • one or more of the processors are caused to perform the steps of any of the methods described in the first aspect above.
  • the application also provides a computer-readable storage medium that can be read and written by a processor.
  • the storage medium stores computer instructions, and the computer-readable instructions are read and written by one or more processors. When executed, one or more processors are caused to perform the steps of any of the methods described in the first aspect above.
  • Figure 1 is a schematic structural diagram of a wireless network WiFi sensing system provided by an embodiment of the present application
  • FIG. 2 is a schematic flowchart of a wireless network WiFi sensing method provided by an embodiment of the present application
  • FIG. 3 is a schematic flowchart of the sub-steps of step S300 in Figure 2;
  • FIG. 4 is a schematic flowchart of the sub-steps of step S310 in Figure 3;
  • FIG. 5 is a schematic flowchart of the sub-steps of step S311 in Figure 4;
  • FIG. 6 is a schematic flowchart of the sub-steps of step S313 in Figure 4;
  • FIG. 7 is a schematic flowchart of the sub-steps of step S320 in Figure 3;
  • Figure 8 is an overall flow diagram of a wireless network WiFi sensing method provided by another embodiment of the present application.
  • Figure 9 is a schematic structural diagram of a computer device provided by an embodiment of the present application.
  • the embodiments of the present application provide a wireless network WiFi sensing method, system and computer equipment.
  • the embodiments of the present application first obtain the WiFi signal transmitted by the path, which is beneficial to subsequent processing of the WiFi signal transmitted by the path; by detecting WiFi signal coverage in real time
  • the scene status of the area can obtain the dynamic changes in the area at different times; the WiFi signal is processed according to the scene status and a spectrogram is generated, realizing multi-dimensional and high-precision WiFi perception.
  • the spectrogram can reflect the relationship between spatial changes and signal changes. The mapping relationship between them can identify WiFi signals based on the spectrogram, which can be used to implement coarse-grained and fine-grained sensing tasks.
  • the solution of the embodiment of the present application processes multi-path WiFi signals through real-time detection of scene states to obtain spectrum diagrams, achieving multi-dimensional and high-precision WiFi perception, and thus can be used to achieve coarse-grained and fine-grained perception tasks.
  • it can alleviate the problem of WiFi perception degradation caused by multipath effects and achieve multi-dimensional and high-precision WiFi perception. It can not only be used to implement coarse-grained perception tasks, but also can be used to achieve fine-grained perception. Task.
  • FIG. 1 shows a schematic structural diagram of a wireless network WiFi sensing system provided by an embodiment of the present application.
  • the wireless network WiFi sensing system includes first obtaining the WiFi signal transmitted by the path through the signal acquisition module, which is beneficial to subsequent processing of the WiFi signal transmitted by the path; detecting the scene status of the WiFi signal coverage area in real time through the detection module, The dynamic changes in the area at different times can be obtained; the signal processing module is used to process the WiFi signal according to the scene state, and generate a spectrum diagram to identify the WiFi signal based on the spectrum diagram, which can alleviate the problem of WiFi perception degradation caused by the multipath effect, and achieve It provides multi-dimensional and high-precision WiFi perception, which can not only be used to implement coarse-grained perception tasks, but also can be used to implement fine-grained perception tasks.
  • the signal acquisition module is connected to the detection module, and the detection module is connected to the signal processing module.
  • the signal acquisition module can use the WiFi chip to collect signals, and it only needs to support baseband signal collection, which will not be described in detail here.
  • the wireless network WiFi sensing system can be applied to the WiFi chip of the transceiver to support all projects that can obtain channel state information (Channel State Information, CSI) or passive WiFi radar (Passive WiFi Radar, PWR) information collection.
  • CSI Channel State Information
  • PWR Passive WiFi Radar
  • the WiFi signal may be a multi-path transmission WiFi signal or a single-path transmission WiFi signal.
  • the wireless network WiFi sensing system is suitable for wireless transmission using orthogonal frequency division multiplexing.
  • Signal The processing is also applicable to the processing of wireless signals transmitted using time division multiplexing, and is also applicable to the processing of other wireless signals transmitted using modulation and demodulation technology, which will not be described in detail here.
  • the wireless network WiFi sensing system shown in Figure 1 does not limit the embodiments of the present application, and may include more or fewer modules than shown in the figure, or combine certain components, or Different component arrangements.
  • Figure 2 shows a schematic flow chart of a wireless network WiFi sensing method provided by an embodiment of the present application.
  • the wireless network WiFi sensing method is applied to a wireless network WiFi sensing system.
  • the wireless network WiFi sensing method includes but is not limited to step S100, step S200 and step S300.
  • Step S100 Obtain the WiFi signal transmitted by the path.
  • WiFi signals can be transmitted via multi-path or single-path.
  • multiple antennas can transmit multi-path WiFi signals according to frequency division multiplexing or time division.
  • WiFi signals can be transmitted on various paths without interfering with each other, improving channel utilization.
  • the front-end of the hardware collects the WiFi signal transmitted by the path or the front-end of the radio frequency device collects the WiFi radar signal transmitted by the path as a WiFi signal, thereby obtaining the WiFi signal transmitted by the path, which is conducive to subsequent signal processing based on the obtained WiFi signal transmitted by the path.
  • Step S200 Detect the scene status of the WiFi signal coverage area in real time.
  • WiFi signals in an actual signal propagation environment, wireless signals will be affected by multiple obstacles and reach the receiver along multiple paths including refraction and transmission. Signals on different paths will experience different Therefore, WiFi signals have distance limitations and accuracy limitations. By detecting the scene status of the WiFi signal coverage area in real time, the dynamic changes in the area at different times, that is, the scene status, can be obtained, which is beneficial to subsequent based on Scene states perform signal processing to alleviate distance limitations and accuracy limitations.
  • Step S300 Perform signal processing on the WiFi signal according to the scene state to generate a spectrum diagram.
  • the scene state may be an activity recognition, gesture recognition or fall detection scene, or may be a change in the relative distance and speed of the object and the antenna.
  • signal processing is performed on the WiFi signal to generate a spectrogram, to achieve
  • the spectrogram can reflect the mapping relationship between spatial changes and signal changes, and identify WiFi signals based on the spectrogram, which can be used to implement coarse-grained and fine-grained sensing tasks.
  • CSI is an extension of WiFi communication.
  • CSI is used to estimate the communication channel between transmitter and receiver, while providing amplitude and phase information. The phase angle and amplitude difference can be improved through appropriate wireless hardware.
  • PWR information is based on the principle of radar. PWR information associates the transmitted signal from the access point with the transmitted signal from the monitoring area, and WiFi radar positioning is relatively rough. Therefore, CSI has better performance in line-of-sight configuration and can handle larger data volumes offline; while PWR information has better performance in spatial configuration of WiFi access points and radar receivers and is real-time. By combining the two and selecting appropriate one of CSI and PWR information for signal processing according to different detection scenarios, the real-time performance and accuracy of WiFi perception can be improved.
  • the WiFi signal is a multi-path WiFi signal.
  • signal processing is performed on the WiFi signal according to the scene state to generate a spectrum diagram, including but not limited to step S310 and step S320.
  • Step S310 When the scene state is a line-of-sight configuration, signal processing is performed on the WiFi signal according to the channel state information to generate a spectrum diagram.
  • line-of-sight configuration includes human activity recognition, gesture recognition, fall detection, etc. that appear in the scene. Due to the physical isolation between transmitting and receiving, there are limitations in the sensing range and sensing accuracy. Therefore, using channel state information to perform signal processing on WiFi signals can achieve a wider detection range and higher detection accuracy, and can also process large amounts of data.
  • the channel state information can calculate the channel information of each subcarrier in frequency division multiplexing, or can also calculate the channel information of each subcarrier in time division multiplexing. Taking orthogonal frequency division multiplexing as an example below, the channel state information is used for The information performs signal processing on WiFi signals and generates a spectrum diagram.
  • Step S311 Perform noise reduction processing on the WiFi signals of each path to obtain multiple noise reduction signals.
  • the data signals collected through the hardware front-end usually carry noise.
  • the peaks on all 56 orthogonal frequency division multiplexing subcarriers are uniformly sampled, and then the sampled WiFi signals of each path are denoised. After processing, multiple noise-reduced signals are obtained, which can provide high signal-to-noise ratio basic signals for detection.
  • noise reduction processing is performed on the WiFi signals of each path to obtain multiple noise reduction signals, including but not limited to the following steps:
  • Step S3111 Obtain channel state information signals based on the WiFi signals of each path.
  • Orthogonal Frequency Division Multiplexing is widely used in multiple WiFi standards, and the bandwidth in the OFDM system is shared among multiple overlapping orthogonal subcarriers.
  • the OFDM signal is defined according to the following formula:
  • x(t) is the OFDM signal
  • N is the number of subcarriers
  • a n is the n-th symbol in the rectangular symbol sequence
  • T s is the symbol period of OFDM
  • t is time
  • j is the coefficient.
  • the matrix symbol sequence may be Quadrature Phase Shift Keying (QPSK) or Quadrature Amplitude Modulation (QAM).
  • QPSK Quadrature Phase Shift Keying
  • QAM Quadrature Amplitude Modulation
  • j is a coefficient that can be adjusted according to the actual situation.
  • the received signal includes multi-path reflection and direct signal.
  • the reflected signal includes delay and phase offset signals from moving people and stationary objects.
  • the received signal is defined according to the following formula:
  • y(t) is the received signal
  • a p is the attenuation factor of the p-th path
  • ⁇ is the delay
  • f d is the Doppler shift
  • n(t) is Gassian white noise
  • j is an adjustable coefficient.
  • the received signal includes WiFi signals of multiple paths, and a channel state information signal, that is, a CSI signal, is obtained based on the superposition of the WiFi signals of each path, which is beneficial to subsequent noise reduction processing of the signal.
  • a channel state information signal that is, a CSI signal
  • Step S3112 Calculate the ratio of the channel state information signals of two adjacent antennas to obtain multiple noise reduction signals.
  • the channel status signal obtained according to step S3111 is highly noisy.
  • the CSI signal ratio is still a complex value
  • the resulting amplitude is the quotient of its CSI signal amplitude
  • the phase is the difference between the phases of two adjacent CSI signals.
  • impulse noise is a scaled noise that amplifies the power of each antenna at the same level on the same receiver, i.e. although the power scaling changes over time, it is between different antennas on the same receiver. Consistently, the noise can be eliminated by calculating the amplitude coefficients of the two antennas.
  • the phase offset (such as the carrier) is the same as the frequency offset and the sampling frequency offset.
  • the phase frequency offset is random and time-varying, but they are two identical antennas and can be effectively canceled by calculating the phase difference between the two antennas. Therefore, by calculating the ratio of the channel state information signals of two adjacent antennas, multiple noise reduction signals are obtained, which can eliminate most of the noise in the CSI signal amplitude and time-varying phase offset, and provide a high signal-to-noise ratio basic signal for detection ; Moreover, multiple CSI signals are obtained based on Multi-Input Multi-Output (MIMO) technology, which can achieve longer detection range and higher detection accuracy.
  • MIMO Multi-Input Multi-Output
  • the CSI signal at the carrier frequency f c is expressed as H(f c ,t).
  • (f,t) represents the noise reduction signal
  • H 1 (f c , t) represents the CSI signal of one antenna
  • H 2 (f c , t) represents the CSI signal of another antenna.
  • (f,t) represents the noise reduction signal
  • H s,1 is the CSI signal of the first antenna
  • H s,2 is the CSI signal of the second antenna
  • S(t) represents the amplitude pulse noise
  • d 1 (t) represents the target reflection path length of the first antenna
  • d 2 (t) represents the target reflection path length of the second antenna
  • a 1 represents the attenuation factor of the first antenna
  • a 2 represents the attenuation factor of the second antenna
  • j and ⁇ are both adjustable coefficients.
  • (f,t) represents the noise reduction signal
  • d 1 (t) represents the target reflection path length of the first antenna
  • d 2 (t) represents the target reflection path length of the second antenna
  • a 1 represents the first
  • a 2 represents the attenuation factor of the second antenna
  • j and ⁇ are both adjustable coefficients.
  • Step S312 Perform time-varying correlation analysis on each noise reduction signal to obtain multiple principal component signals.
  • the data amount of the obtained channel state information signal is also large.
  • the Principal Components Analysis (PCA) algorithm is used to identify the time-varying correlations between the CSI signal streams of each denoising signal, and then these time-varying correlations are combined to extract the components that represent the measurement changes of the CSI signals; other methods can also be used.
  • the dimensionality reduction algorithm can extract the main components from a larger amount of data, so I will not go into details here.
  • the number of principal components is chosen as a trade-off between classification performance and computational complexity. Illustratively, 70% of the signal variance is captured from two or three principal components, but also 60%. According to this variance ratio, the CSI signal is extracted. of six principal parts, if the first principal part contains noise due to reflections from stationary objects, discard the first principal part and use only five main part.
  • the above algorithm can obtain the principal component signal of the CSI signal from a large amount of data, which can not only ensure the classification performance, but also reduce the computational complexity and facilitate subsequent processing based on the principal component signal.
  • Step S313 Generate a spectrum diagram based on multiple principal component signals.
  • generating a spectrum diagram based on multiple principal component signals includes but is not limited to the following steps:
  • Step S3131 Divide each principal component signal into multiple signal segments of the same length, and perform Fourier transform on each signal segment to obtain a spectrogram corresponding to each principal component signal.
  • the CSI signal is highly sensitive to the surrounding environment. Since the radio frequency reflection of the human body shows different frequencies when performing different activities, a calibration process of scanning the background is usually required.
  • the embodiment of the present application uses short-time Fourier transform (short-term Fourier transform, STFT) performs spectral conversion on each principal component signal.
  • STFT applies a sliding window to divide the principal component signal into multiple signal segments of the same length, and then performs a fast Fourier transform (FFT) on the samples in each segment to obtain the corresponding signal of each principal component.
  • FFT fast Fourier transform
  • spectrogram The wavelet transform algorithm can also be used to convert the spectral image, as long as the spectral image conversion can be performed, I will not go into details here.
  • the spectrogram contains the three dimensions of FFT: time, frequency and amplitude. Obtaining the spectrogram is helpful for subsequent calculation of the spectrogram based on the spectrogram.
  • the STFT algorithm is expressed by the following formula:
  • X(t,k) represents the Fourier transform result
  • k represents the frequency index
  • x[n] represents the time domain input signal
  • w[n-t] represents the window function
  • j represents the adjustable coefficient
  • Step S3132 Perform arithmetic calculation processing on each spectrogram to generate a spectrogram.
  • Spectrograms can reflect the mapping relationship between spatial changes and signal changes, thereby perceiving and identifying WiFi signals. Among them, arithmetic calculations can also be used to calculate the maximum value or minimum value, which can improve the perception accuracy.
  • Step S320 When the scene state is spatial configuration, signal processing is performed on the WiFi signal according to the passive WiFi radar information to generate a spectrum diagram.
  • the channel state information can calculate the channel information of each subcarrier in frequency division multiplexing, and can also calculate the channel information of each subcarrier in time division multiplexing. Taking orthogonal frequency division multiplexing as an example below, for passive WiFi The radar information performs signal processing on WiFi signals and generates a spectrum diagram.
  • PWR information usually captures a signal with a much longer duration than CSI to ensure that a sufficient number of WiFi signals can be captured. Therefore, the PWR information uses the entire WiFi signal, that is, the PWR information does not process the signal of each subcarrier. Instead, the entire OFDM is regarded as a signal, and PWR information cannot access the information within each subcarrier.
  • the spatial configuration includes the distance between the transceiver and objects in the area, or a rough calculation of the movement of the object. Therefore, the PWR information is used to perform real-time signal processing on the entire WiFi signal without paying attention to the information within each subcarrier to generate a spectrum diagram. Spectrograms can reflect the mapping relationship between spatial changes and overall signal changes, which is beneficial to the perception and identification of WiFi signals.
  • the WiFi signal is a multi-path WiFi radar signal.
  • signal processing is performed on the WiFi signal according to the passive WiFi radar information to generate a spectrum diagram, including but Not limited to the following steps:
  • Step S321 Perform ambiguity measurement processing on the WiFi radar signals of each path to obtain each measurement signal.
  • a cross-ambiguity function is used to perform ambiguity measurement processing on the WiFi signals of each path to obtain each measurement signal, which can effectively extract the signals collected by the radar, which is beneficial to subsequent measurement Fast signal processing.
  • the mutual ambiguity function can be a low-complexity mutual ambiguity function, or other versions of the mutual ambiguity function can be used to effectively extract distance and Doppler information, which will not be described here.
  • CAF is an effective tool for extracting target range and Doppler information in the field of passive radar. It requires two channels, the monitoring channel collects signals from the monitoring area, and the reference channel measures signals directly from the transmitter.
  • Step S322 Perform interference signal elimination processing on each metric signal to obtain each interference-free signal.
  • the main interference source of PWR information is signals from other WiFi hotspots that directly enter the monitoring channel, that is, direct interference information.
  • This direct interference information has higher energy than the signal reflected from the moving target and can be shielded Doppler pulses at the CAF surface.
  • the improved CLEAN algorithm is used to perform direct interference signal elimination processing on each measurement signal to obtain a de-interference signal, which can improve the signal-to-noise ratio of the target signal.
  • the improved CLEAN algorithm can be the CLEAN-PSF algorithm or the CLEAN-SC algorithm.
  • the improved CLEAN algorithm shares a similar structure to the CAF process and only generates a self-fuzzy surface from the reference channel, expressed by the following formula:
  • CAF self represents the self-blurred surface
  • CAF k ( ⁇ ,f d ) represents k metric signals
  • a k represents the maximum amplitude of the k-th CAF surface
  • T k represents the maximum phase shift of the k-th CAF surface
  • ⁇ represents the time delay
  • Step S323 Perform noise elimination processing on each de-interference signal to obtain each Doppler pulse.
  • the noise on the CAF surface is further reduced.
  • the false alarm rate (Constant False-Alarm Rate, CFAR) is used to estimate the background noise distribution of the interference signal, and is applied to the CAF surface to obtain each Doppler pulse. By obtaining the Doppler pulse, it is beneficial to subsequently generate a spectrogram based on the Doppler pulse.
  • ⁇ represents the threshold of the CAF surface
  • i represents the distance index
  • j represents the Doppler index
  • N ⁇ represents the processing time of the Doppler index.
  • R ⁇ represents the upper limit of the processing time
  • strong pulses above the threshold represent human activity, otherwise it is inferred that no movement has occurred, and therefore human activity can be measured.
  • Step S324 Select the largest Doppler pulse to generate a spectrogram.
  • multiple Doppler pulses are obtained according to step S323, and a spectrogram is generated by selecting the maximum Doppler pulse from each Doppler within the CAF surface and combining a series of measurement results.
  • the spectrum diagram is the Doppler spectrum diagram of PWR information.
  • the Doppler spectrum can reflect the mapping relationship between spatial changes and overall signal changes, thereby perceiving and identifying WiFi signals.
  • the overall process is to first obtain the WiFi signal transmitted by the path. It can be a multi-path WiFi signal or a single-path WiFi signal. Taking the multi-path WiFi signal as an example, the WiFi signal can be obtained through the front end of the hardware. Multi-path WiFi signals or multi-path WiFi radar signals are obtained through the front end of radio frequency equipment. According to the detected scene conditions, state, when the scene state is a line-of-sight configuration, for example, gesture recognition and fall recognition are performed, and the channel state information is used to process the multi-path WiFi signal obtained through the front end of the hardware. First, each antenna of the same transceiver is calculated.
  • the phase and amplitude of the component signals are obtained by obtaining the spectrum diagram corresponding to each principal component, and then averaging each spectrum diagram to generate the final spectrum diagram to observe changes in WiFi signals and identify WiFi signals to achieve high-precision WiFi sensing tasks. .
  • the passive WiFi radar information is used to obtain multi-path WiFi through the front end of the radio frequency device.
  • To process the radar signal first obtain each metric signal through the fuzzy function, eliminate the interference signal on the surface of the metric signal, and obtain the de-interference signal, and then perform noise elimination processing on the de-interference signal to obtain multiple Doppler pulses. Select the maximum value in the Doppler pulse to generate a spectrogram to observe changes in WiFi signals, identify WiFi signals, and combine with channel status information to achieve multi-dimensional WiFi sensing tasks.
  • the above-mentioned real-time detection of scene status correspondingly selects one of the channel status information and passive WiFi radar information to process the WiFi signal, generates a spectrogram to reflect the changing status of the signal, and uses channels across different scenarios.
  • Status information and passive WiFi radar information process signals to achieve multi-dimensional and high-precision WiFi perception.
  • the above wireless network WiFi sensing method can be applied to fine-grained sensing tasks such as fall detection tasks, gesture recognition, and motion detection, and can also be applied to coarse-grained tasks such as distance detection, and has a wide range of applications.
  • Figure 9 shows a computer device 900 provided by an embodiment of the present application.
  • the computer device 900 may be a server or a terminal.
  • the internal structure of the computer device 900 includes but is not limited to:
  • Memory 910 is configured to store programs
  • the processor 920 is configured to execute the program stored in the memory 910.
  • the processor 920 executes the program stored in the memory 910, the processor 920 is configured to execute the above-mentioned wireless network WiFi sensing method.
  • the processor 920 and the memory 910 may be connected through a bus or other means.
  • the memory 910 can be configured to store non-transitory software programs and non-transitory computer executable programs, such as the wireless network WiFi sensing method described in any embodiment of this application.
  • the processor 920 implements the above wireless network WiFi sensing method by running non-transient software programs and instructions stored in the memory 910 .
  • the memory 910 may include a program storage area and a data storage area, where the program storage area may store an operating system and an application program required for at least one function; the storage data area may store the implementation of the above wireless network WiFi sensing method.
  • memory 910 may include high-speed random access memory, and may also include non-transitory memory, such as at least one magnetic disk storage device, flash memory device, or other non-transitory solid-state storage device.
  • the memory 910 may include memory located remotely relative to the processor 920, and these remote memories may be connected to the processor 920 through a network. Examples of the above-mentioned networks include but are not limited to the Internet, intranets, local area networks, mobile communication networks and combinations thereof.
  • the non-transitory software programs and instructions required to implement the above wireless network WiFi sensing method are stored in the memory 910. When executed by one or more processors 920, the wireless network WiFi sensing method provided by any embodiment of the present application is executed.
  • Embodiments of the present application also provide a computer-readable storage medium that stores computer-executable instructions, and the computer-executable instructions are used to execute the above wireless network WiFi sensing method.
  • the storage medium stores computer-executable instructions, which are executed by one or more control processors 920, such as by one processor 920 in the above-mentioned computer device 900, so that the above The one or more processors 920 execute the wireless network WiFi sensing method provided by any embodiment of this application.
  • Embodiments of this application include: first obtaining the WiFi signal transmitted by the path, which is beneficial to subsequent processing of the WiFi signal transmitted by the path; by detecting the scene status of the WiFi signal coverage area in real time, the dynamic changes in the area at different times can be obtained; according to the scene status Perform signal processing on WiFi signals and generate spectrograms, achieving multi-dimensional and high-precision WiFi perception.
  • Spectrograms can reflect the mapping relationship between spatial changes and signal changes.
  • WiFi signals can be identified based on the spectrograms, which can be used to implement coarse Granular and fine-grained perception tasks.
  • the solution of the embodiment of the present application processes multi-path WiFi signals through real-time detection of scene states to obtain spectrum diagrams, achieving multi-dimensional and high-precision WiFi perception, and thus can be used to achieve coarse-grained and fine-grained perception tasks.
  • it can alleviate the problem of WiFi perception degradation caused by multipath effects and achieve multi-dimensional and high-precision WiFi perception. It can not only be used to implement coarse-grained perception tasks, but also can be used to achieve fine-grained perception. Task.
  • Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, Digital Versatile Disk (DVD) or other optical disk storage, magnetic cassettes, tapes, disk storage or other magnetic storage devices, or may Any other medium set up to store the desired information and that can be accessed by the computer.
  • communication media typically embodies computer readable instructions, data structures, program modules or other data in a modulated data signal such as or other transport mechanisms and may include any information delivery media.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Quality & Reliability (AREA)
  • Physics & Mathematics (AREA)
  • Electromagnetism (AREA)
  • Radar Systems Or Details Thereof (AREA)

Abstract

本申请提供了一种无线网络WiFi感知方法、系统及计算机设备，涉及通信领域，无线网络WiFi感知方法包括：获取路径传输的WiFi信号(S100)；实时检测所述WiFi信号覆盖区域的场景状态(S200)；根据所述场景状态对所述WiFi信号进行信号处理，生成频谱图(S300)。

Description

无线网络WiFi感知方法、系统及计算机设备
相关申请的交叉引用
本申请基于申请号为202210875975.3、申请日为2022年07月25日的中国专利申请提出，并要求该中国专利申请的优先权，该中国专利申请的全部内容在此引入本申请作为参考。
技术领域
本申请涉及通信技术领域，特别涉及一种无线网络WiFi感知方法、系统及计算机设备。
背景技术
基于无线网络wifi的监测感知都是通过接收信号强度指示(Received Signal Strength Indicator，RSSI)来进行的，在现实场景中，无线网络WiFi信号在室内环境传播时，WiFi信号会受到多个障碍物的影响，会沿着包括折射、发射在内的多条路径到达接收机，其中，不同路径上的信号会出现不同程度的衰减和时延等失真现象，接收端收到的信号为不同路径失真信号的叠加结果，即所谓的多径效应。受多径效应的影响，室内接收到的RSSI稳定性较差，即使在室内静态场景下也会有较大波动，因此室内多径效应大大限制了RSSI的感知能力，使其只能用于实现一些粗粒度的室内定位等感知任务。
发明内容
本申请提供一种无线网络WiFi感知方法、系统及计算机设备。
第一方面，本申请提供了一种无线网络WiFi感知方法，所述方法包括：获取路径传输的WiFi信号；实时检测所述WiFi信号覆盖区域的场景状态；根据所述场景状态对所述WiFi信号进行信号处理，生成频谱图。
第二方面，本申请提供了一种无线网络WiFi感知系统，包括：信号获取模块，设置为获取路径传输的WiFi信号；检测模块，设置为实时检测所述WiFi信号覆盖区域的场景状态；信号处理模块，设置为根据所述场景状态对所述WiFi信号进行信号处理，生成频谱图。
第三方面，本申请提供了一种计算机设备，所述计算机设备包括存储器和处理器，所述存储器中存储有计算机可读指令，所述计算机可读指令被一个或多个所述处理器执行时，使得一个或多个所述处理器执行如上第一方面描述的任一项所述方法的步骤。
第四方面，本申请还提供了一种计算机可读存储介质，所述存储介质可被处理器读写，所述存储介质存储有计算机指令，所述计算机可读指令被一个或多个处理器执行时，使得一个或多个处理器执行如上第一方面描述的任一项所述方法的步骤。
附图说明
图1是本申请的一个实施例提供的无线网络WiFi感知系统的结构示意图；
图2是本申请的一个实施例提供的无线网络WiFi感知方法的流程示意图；
图3是图2中步骤S300的子步骤流程示意图；
图4是图3中步骤S310的子步骤流程示意图；
图5是图4中步骤S311的子步骤流程示意图；
图6是图4中步骤S313的子步骤流程示意图；
图7是图3中步骤S320的子步骤流程示意图；
图8是本申请的另一个实施例提供的无线网络WiFi感知方法的整体流程示意图；
图9是本申请实施例提供的计算机设备的结构示意图。
具体实施方式
为了使本申请的目的、技术方案及优点更加清楚明白，以下结合附图及实施例，对本申请进行进一步详细说明。应当理解，此处所描述的具体实施例仅仅用以解释本申请，并不用于限定本申请。
需要说明的是，虽然在流程图中示出了逻辑顺序，但是在某些情况下，可以以不同于流程图中的顺序执行所示出或描述的步骤。说明书和权利要求书及上述附图中的术语“第一”、“第二”等是用于区别类似的对象，而不必用于描述特定的顺序或先后次序。
本申请实施例提供了一种无线网络WiFi感知方法、系统及计算机设备，本申请实施例通过首先获取路径传输的WiFi信号，有利于后续对路径传输的WiFi信号进行处理；通过实时检测WiFi信号覆盖区域的场景状态，能够得到不同时刻区域内的动态变化；根据场景状态对WiFi信号进行信号处理，生成频谱图，实现了WiFi感知的多维度和高精度，频谱图能够反映空间变化与信号变化之间的映射关系，根据频谱图识别WiFi信号，从而能够用于实现粗粒度和精细粒度的感知任务。即是说，本申请实施例的方案通过实时检测的场景状态对多路径的WiFi信号进行处理，得到频谱图，实现了WiFi感知的多维度和高精度，从而能够用于实现粗粒度和精细粒度的感知任务。与相关技术相比，能够缓解多径效应造成的WiFi感知能力下降问题，实现了多维度和高精度的WiFi感知，不仅能够用于实现粗粒度的感知任务，还能够用于实现精细粒度的感知任务。
下面结合附图，对本申请实施例作进一步阐述。
参见图1，图1示出了本申请实施例提供的无线网络WiFi感知系统的结构示意图。在图1的示例中，无线网络WiFi感知系统包括首先通过信号获取模块获取路径传输的WiFi信号，有利于后续对路径传输的WiFi信号进行处理；通过检测模块实时检测WiFi信号覆盖区域的场景状态，能够得到不同时刻区域内的动态变化；利用信号处理模块根据场景状态对WiFi信号进行信号处理，生成频谱图，以根据频谱图识别WiFi信号，能够缓解多径效应造成的WiFi感知能力下降问题，实现了多维度和高精度的WiFi感知，不仅能够用于实现粗粒度的感知任务，还能够用于实现精细粒度的感知任务。
在一实施例中，信号获取模块与检测模块连接，检测模块与信号处理模块连接。其中，信号获取模块可以利用WiFi芯片采集信号，能够支持基带的信号采集即可，这里不作赘述。无线网络WiFi感知系统能够应用于收发器的WiFi芯片支持可以获取到信道状态信息(Channel State Information，CSI)或者被动WiFi雷达(Passive WiFi Radar，PWR)信息采集的所有项目。
在一实施例中，可以为多路径传输的WiFi信号，也可以为单路径传输的WiFi信号，对于传输的WiFi信号，该无线网络WiFi感知系统适用于对利用正交频分复用传输的无线信号 进行处理，也适用于对利用时分复用传输的无线信号进行处理，还适用于其他利用调制解调技术传输的无线信号进行处理，这里不作赘述。
本申请实施例描述的设备以及应用场景是为了更加清楚的说明本申请实施例的技术方案，并不构成对于本申请实施例提供的技术方案的限定，本领域技术人员可知，随着新应用场景的出现，本申请实施例提供的技术方案对于类似的技术问题，同样适用。
本领域技术人员可以理解的是，图1中示出的无线网络WiFi感知系统并不构成对本申请实施例的限定，可以包括比图示更多或更少的模块，或者组合某些部件，或者不同的部件布置。
根据上述无线网络WiFi感知系统，下面对本申请的无线网络WiFi感知方法的各个实施例进行说明。
如图2所示，图2示出了本申请一个实施例提供的无线网络WiFi感知方法的流程示意图，该无线网络WiFi感知方法应用于无线网络WiFi感知系统。该无线网络WiFi感知方法包括但不限于有步骤S100、步骤S200和步骤S300。
步骤S100，获取路径传输的WiFi信号。
在一实施例中，可以为多路径传输WiFi信号，也可以为单路径传输WiFi信号，在有多个可用天线的情况下，多个天线进行多路径WiFi信号传输，根据频分复用或者时分复用技术，WiFi信号能够在各个路径传输中互不干扰，提高信道的利用率。通过硬件的前端采集路径传输的WiFi信号或者通过射频设备前端采集路径传输的WiFi雷达信号作为WiFi信号，从而获取路径传输的WiFi信号，有利于根据获取的路径传输的WiFi信号进行后续信号处理。
步骤S200，实时检测WiFi信号覆盖区域的场景状态。
在一实施例中，信号在实际传播环境下，无线信号会受到多个障碍物的影响，会沿着包括折射、发射在内的多条路径到达接收机，其中不同路径上的信号会经历不同程度的衰减和时延等失真，因此，WiFi信号具有距离限制和精度限制，通过实时检测WiFi信号覆盖区域的场景状态，能够得到不同时刻区域内的动态变化情况，即场景状态，有利于后续根据场景状态进行信号处理，以缓解距离限制和精度限制。
步骤S300，根据场景状态对WiFi信号进行信号处理，生成频谱图。
在一实施例中，场景状态可以为对活动识别、手势识别或者跌倒检测场景，可以为物体和天线相对距离和速度的变化，根据上述不同场景状态对WiFi信号进行信号处理，生成频谱图，实现了WiFi感知的多维度和高精度，频谱图能够反映空间变化与信号变化之间的映射关系，根据频谱图识别WiFi信号，从而能够用于实现粗粒度和精细粒度的感知任务。
在一实施例中，CSI是WiFi通信的延伸，CSI用于估计收发之间的通信信道，同时提供幅度和相位的信息，通过适当的无线硬件可以提高相位的角度和幅度的差值。PWR信息是基于雷达的原理，PWR信息将来自接入点的发送信号与来自监控区域的发射信号相关联，而且WiFi雷达定位相对粗糙。因此，CSI在视线配置中具有更好的性能，能够离线处理较大的数据量；而PWR信息在WiFi接入点和雷达接收器在空间配置中具有更好的性能，具有实时性。通过将两者结合起来，根据不同检测场景选择适当的CSI和PWR信息中的一种进行信号处理，能够提升WiFi感知的实时性和精确度。
在一实施例中，WiFi信号为多路径的WiFi信号，如图3所示，根据场景状态对WiFi信号进行信号处理，生成频谱图，包括但不限于有步骤S310和步骤S320。
步骤S310，在场景状态为视线配置的情况下，根据信道状态信息对WiFi信号进行信号处理，生成频谱图。
在一实施例中，视线配置包括场景中出现的人体活动识别、手势识别或者跌倒检测等，由于收发之间的物理隔离，存在传感范围和传感精度的限制。因此，采用信道状态信息对WiFi信号进行信号处理，能够实现较广泛的检测范围和更高的检测精度，还能够对大量数据进行处理。
在一实施例中，信道状态信息能够计算频分复用每个子载波的信道信息，也可以计算时分复用每个子载波的信道信息，下面以正交频分复用为例，对采用信道状态信息对WiFi信号进行信号处理，生成频谱图的处理过程进行说明。
如图4所示，在场景状态为视线配置的情况下，根据信道状态信息对WiFi信号进行信号处理，生成频谱图，包括但不限于有以下步骤：
步骤S311，对各个路径的WiFi信号进行降噪处理，得到多个降噪信号。
在一实施例中，通过硬件前端采集的数据信号通常携带有噪音，先对全部56个正交频分复用子载波上的峰值均匀采样，然后对采样后的各个路径的WiFi信号进行降噪处理，得到多个降噪信号，能够为检测提供高信噪比基础信号。
在一实施例中，在多天线的情况下，如图5所示，对各个路径的WiFi信号进行降噪处理，得到多个降噪信号，包括但不限于有以下步骤：
步骤S3111，根据各个路径的所述WiFi信号，得到信道状态信息信号。
在一实施例中，正交频分复用(Orthogonal Frequency Division Multiplexing，OFDM)广泛用于多种WiFi标准中，OFDM系统中的带宽在多个重叠的正交子载波之间共享。将OFDM信号按照以下公式进行定义：
其中，x(t)为OFDM信号，N为子载波数，an矩形符号序列中的第n个符号，Ts为OFDM的符号周期，t为时间，j为系数。
在一实施例中，矩阵符号序列可以为正交相移键控(Quadrature Phase Shift Keying，QPSK)，也可以为正交振幅调制(Quadrature Amplitude Modulation，QAM)。j为可以根据实际情况可调节的系数。
在一实施例中，接收信号包括多路反射和直接信号，反射信号中包括延迟和相位偏移的信号来自移动的人和静止的物体，将接收信号按照以下公式进行定义：
其中，y(t)为接收信号，Ap为第p路的衰减因子，τ为延时，fd为多普勒位移。n(t)为加斯白噪声，j为可调节的系数。
在一实施例中，接收信号包括多个路径的WiFi信号，根据对各个路径的WiFi信号的叠加得到信道状态信息信号，即CSI信号，有利于后续进行信号的降噪处理。
步骤S3112，计算两个相邻天线的信道状态信息信号的比值，得到多个降噪信号。
在一实施例中，由于WiFi无线接入点的功率和数据速率变化，根据步骤S3111得到的信道状态信号是高噪声的。根据两个复数的除法规律，CSI信号比值仍然是复值，其结果振幅 是其CSI信号振幅的商，相位是相邻两个CSI信号相位之间的差。对于幅度而言，脉冲噪声是一个缩放噪声，它放大同一接收器上同一电平处每个天线的功率，即虽然功率缩放随时间而变化，但它在同一接收器上的不同天线之间是一致的，可以通过计算两个天线的振幅系数来消除噪声。对于相位而言，由于接收器上的不同天线共享相同的时钟，因此相位失调(如载波)为频偏和采样频偏是相同的，相位频偏是随机的，而且是时变的，但是它们是相同的两根天线，能够通过计算两根天线的相位差来有效抵消。因此，通过计算对两个相邻天线的信道状态信息信号的比值，得到多个降噪信号，能够消除CSI信号幅度和时变相位失调中的大部分噪声，为检测提供高信噪比基础信号；并且，根据多输入多输出(Multi Input Multi Ouput，MIMO)技术来获得多个CSI信号，能够实现更长的检测范围和更高的检测精度。
在一实施例中，载波频率fc处的CSI信号表示为H(fc,t)，计算同一接收机两个相邻天线的信道状态信息的比值，得到降噪信号通过以下公式表示：
(f,t)＝H1(fc,t)/H2(fc,t)
其中，(f,t)表示降噪信号，H1(fc,t)表示一个天线CSI信号，H2(fc,t)表示另一个天线的CSI信号。
将上述公式展开表示为：
其中，(f,t)表示为降噪信号，Hs,1为第一个天线的CSI信号，Hs,2为第二个天线的CSI信号，S(t)表示为幅度脉冲噪声，表示时变相位偏移量，d1(t)表示第一个天线的目标反射路径长度，d2(t)表示第二个天线的目标反射路径长度，A1表示第一个天线的衰减因子，A2表示第二个天线的衰减因子，j和λ均为可调节的系数。
对上述公式进行优化，得到以下公式：
其中，(f,t)表示为降噪信号，d1(t)表示第一个天线的目标反射路径长度，d2(t)表示第二个天线的目标反射路径长度，A1表示第一个天线的衰减因子，A2表示第二个天线的衰减因子，j和λ均为可调节的系数。
步骤S312，对各个降噪信号进行时变相关性分析，得到多个主成分信号。
在一实施例中，由于从硬件设备前端采集的WiFi信号数据量较大，因此得到的信道状态信息信号的数据量也很大。示例性地，同一个收发机由1个发射天线和3个接收天线组成，成功率为1kHz，每秒1×3×1k＝3k个复数CSI信号。还可以包含更多的收发机，而每个收发机可以具有较多天线，因此，得到的降噪信号数据量较大，需要降维来降低整体计算复杂度。采用主成分分析(Principal Components Analysis，PCA)算法识别各个降噪信号的CSI信号流之间的时变相关性，然后结合这些时变相关性来提取表示CSI信号测量变化的成分；也可以采用其他降维算法，能够从较大的数据量中提取主要成分即可，这里不作赘述。选择主成分的数量在分类性能和计算复杂度之间进行权衡，示例性地，从两个或三个主分量捕获信号方差的70％，也可以为60％，根据该方差比例，提取CSI信号的六个主要部分，如果第一个主要部分包含了由于来自静止物体的反射而产生的噪声，将第一个主要部分丢弃，只使用五个 主要部分。上述算法能够从大量的数据中得到CSI信号的主成分信号，既能够保证分类性能，还能够减少计算复杂度，有利于根据主成分信号进行后续处理。
步骤S313，根据多个主成分信号，生成频谱图。
如图6所示，根据多个主成分信号，生成频谱图，包括但不限于有以下步骤：
步骤S3131，将各个主成分信号划分为多个相同长度的信号片段，并对各个信号片段进行傅里叶变换，得到各个主成分信号对应的光谱图。
在一实施例中，CSI信号对周围环境高度敏感，由于人体的射频反射在执行不同活动时表现出不同的频率，通常需要一个扫描背景的校准过程，本申请实施例利用短时傅里叶变换(short-term Fourier transform，STFT)对各个主成分信号进行光谱图转换。在一实施例中，STFT应用滑动窗口将主成分信号划分为多个相同长度的信号片段，然后对每个片段中的样本执行快速傅立叶变换(fast Fourier transform，FFT)，得到各个主成分信号对应的光谱图。还可以利用小波变换算法进行光谱图转换，只要能进行光谱图转换即可，这里不作赘述。光谱图包含了FFT的时间、频率和振幅三个维度，通过得到光谱图，有利于后续根据光谱图计算得到频谱图。
STFT算法通过以下公式表示：
其中，X(t,k)表示傅里叶变换结果，k表示频率指数，x[n]表示时域输入信号，w[n-t]表示窗口函数，j表示可调节的系数。
步骤S3132，对各个光谱图进行算术计算处理，生成频谱图。
示例性地，根据步骤S312得到五个主要部分，光谱图是从五个主要部分中生成的，然后对光谱图对应的五个主要部分进行求平均算术计算，生成频谱图。频谱图能够反映空间变化与信号变化之间的映射关系，从而对WiFi信号进行感知和识别。其中，算术计算还可以为求最大值或者最小值计算，能够提高感知精度即可。
步骤S320，在场景状态为空间配置的情况下，根据被动WiFi雷达信息对WiFi信号进行信号处理，生成频谱图。
在一实施例中，信道状态信息能够计算频分复用每个子载波的信道信息，也可以计算时分复用每个子载波的信道信息，下面以正交频分复用为例，对采用被动WiFi雷达信息对WiFi信号进行信号处理，生成频谱图的处理过程进行说明。
在一实施例中，PWR信息通常捕获比CSI持续时间长得多的信号，以确保能够捕获足够数量的WiFi信号，因此，PWR信息使用整个WiFi信号，即PWR信息不处理每个子载波的信号，而是将整个OFDM视为一个信号，PWR信息不能访问每个子载波内的信息。空间配置包括收发机与区域内物体之间的距离，或者粗略计算物体的动作，因此，采用PWR信息对整个WiFi信号进行实时信号处理，不关注每个子载波内的信息，生成频谱图。频谱图能够反映空间变化与整体信号变化之间的映射关系，有利于对WiFi信号进行感知和识别。
在一实施例中，WiFi信号为多路径的WiFi雷达信号，如图7所示，在场景状态为空间配置的情况下，根据被动WiFi雷达信息对WiFi信号进行信号处理，生成频谱图，包括但不限于有以下步骤：
步骤S321，对各个路径的WiFi雷达信号进行模糊性度量处理，得到各个度量信号。
在一实施例中，利用互模糊函数(Cross-Ambiguity Function，CAF)对各个路径的WiFi信号进行模糊性度量处理，得到各个度量信号，能够有效提取通过雷达采集到的信号，有利于后续对度量信号快速处理。互模糊函数可以为低复杂度的互模糊函数，也可以利用其他版本的互模糊函数，能够有效提取距离和多普勒信息即可，这里不作赘述。其中，CAF是一种用于被动式雷达领域的提取目标距离和多普勒信息的有效工具，它需要两个信道，监控信道从监控区域收集信号，而参考信道直接从发射机测量信号。
步骤S322，对各个度量信号进行干扰信号消除处理，得到各个去干扰信号。
在一实施例中，PWR信息的主要干扰源是来自直接进入监控信道的其他WiFi热点的信号，即直接干扰信息，该直接干扰信息具有比从运动目标反射的信号更高的能量，并且可以屏蔽CAF表面的多普勒脉冲。利用改进的CLEAN算法对各个度量信号进行直接干扰信号消除处理，得到去干扰信号，能够实现提高目标信号的信噪比。其中，改进的CLEAN算法可以为CLEAN-PSF算法，也可以为CLEAN-SC算法，改进的CLEAN算法共享一个相似的结构体给CAF过程，仅从参考信道产生自模糊表面，通过以下公式进行表示：
其中，表示去干扰信号，CAFself表示自模糊面，CAFk(τ,fd)表示k个度量信号，ak表示第k个CAF面的最大振幅，Tk表示第k个CAF面的最大相移，τ表示时延。
步骤S323，对各个去干扰信号进行噪音消除处理，得到各个多普勒脉冲。
在一实施例中，由于时间间隙内的相关性，在CLEAN算法之后，会存在残留噪声，通过步骤S322对信号去除干扰信号之后，进一步降低CAF表面的噪声。利用虚假报警率(Constant False-Alarm Rate，CFAR)估计去干扰信号的背景噪声分布，并将其应用于CAF表面，得到各个多普勒脉冲。通过得到多普勒脉冲，有利于后续根据多普勒脉冲生成频谱图。
将虚假报警率应用于CAF表面通过以下公式表示：
其中，Λ表示CAF表面的阈值，i表示距离指数，j表示多普勒指数，Nτ表示多普勒指数的处理时长，表示距离，Rτ表示处理时长上限，表示距离上限，表示各个去干扰信号。
在一实施例中，高于阈值的强脉冲代表人类活动，否则推断没有发生运动，因此，能够对人类的活动进行测量。
步骤S324，选择最大的多普勒脉冲生成频谱图。
在一实施例中，根据步骤S323得到多个多普勒脉冲，通过从CAF表面内的每个多普勒中选择最大多普勒脉冲，并结合一系列测量结果，生成频谱图。其中，频谱图为PWR信息的多普勒谱图。多普勒谱图能够反映空间变化与整体信号变化之间的映射关系，从而对WiFi信号进行感知和识别。
如图8所示，整体流程为首先获取路径传输的WiFi信号，可以为多路径的WiFi信号，也可以为单路径的WiFi信号，以多路径WiFi信号为例，WiFi信号可以通过硬件的前端获取多路径的WiFi信号或者通过射频设备前端获取多路径的WiFi雷达信号，根据检测的场景状 态，在场景状态为视线配置的情况下，示例性地，进行手势识别和跌倒识别等，利用信道状态信息对通过硬件的前端获取多路径的WiFi信号进行处理，首先计算同一收发机的各个天线的CSI信号，再根据得到的CSI信号计算同一接收机相邻两个天线的CSI信号的商，得到降噪信号，然后利用PCA算法对降噪信号进行相关性分析，提取主要成分，然后根据主成分信号的相位和幅度得到各个主成分对应的频谱图，随后对各个频谱图求均值，生成最终的频谱图，以观察WiFi信号的变化，对WiFi信号进行识别，从而实现高精度的WiFi感知任务。在场景状态为空间配置的情况下，示例性地，进行收发机与区域内物体之间的距离计算，以及粗略计算物体的动作等，利用被动WiFi雷达信息对通过射频设备前端获取多路径的WiFi雷达信号进行处理，首先通过模糊函数得到各个度量信号，对度量信号表面的干扰信号进行消除，得到去干扰信号，紧接着再对去干扰信号进行噪音消除处理，得到多个多普勒脉冲，从多普勒脉冲中选择最大值生成频谱图，以观察WiFi信号的变化，对WiFi信号进行识别，结合信道状态信息能够实现多维度的WiFi感知任务。上述实时检测场景状态，根据不同时刻下的场景状态，相应地选择信道状态信息和被动WiFi雷达信息的一种对WiFi信号进行处理，生成频谱图反映信号的变化状态，通过不同场景下交叉运用信道状态信息和被动WiFi雷达信息对信号进行处理，能够实现多维度和高精度的WiFi感知。
在一实施例中，上述无线网络WiFi感知方法能够应用于跌倒检测任务、手势识别和动作检测等精细粒度的感知任务，还能够应用于距离检测等粗粒度任务，具有广泛的应用范围。
参照图9，图9示出了本申请实施例提供的计算机设备900。该计算机设备900可以是服务器或者终端，该计算机设备900的内部结构包括但不限于：
存储器910，设置为存储程序；
处理器920，设置为执行存储器910存储的程序，当处理器920执行存储器910存储的程序时，处理器920设置为执行上述的无线网络WiFi感知方法。
处理器920和存储器910可以通过总线或者其他方式连接。
存储器910作为一种非暂态计算机可读存储介质，可设置为存储非暂态软件程序以及非暂态性计算机可执行程序，如本申请任意实施例描述的无线网络WiFi感知方法。处理器920通过运行存储在存储器910中的非暂态软件程序以及指令，从而实现上述的无线网络WiFi感知方法。
存储器910可以包括存储程序区和存储数据区，其中，存储程序区可存储操作系统、至少一个功能所需要的应用程序；存储数据区可存储执行上述的无线网络WiFi感知方法。此外，存储器910可以包括高速随机存取存储器，还可以包括非暂态存储器，比如至少一个磁盘存储器件、闪存器件、或其他非暂态固态存储器件。在一些实施方式中，存储器910可包括相对于处理器920远程设置的存储器，这些远程存储器可以通过网络连接至该处理器920。上述网络的实例包括但不限于互联网、企业内部网、局域网、移动通信网及其组合。
实现上述的无线网络WiFi感知方法所需的非暂态软件程序以及指令存储在存储器910中，当被一个或者多个处理器920执行时，执行本申请任意实施例提供的无线网络WiFi感知方法。
本申请实施例还提供了一种计算机可读存储介质，存储有计算机可执行指令，计算机可执行指令用于执行上述的无线网络WiFi感知方法。
在一实施例中，该存储介质存储有计算机可执行指令，该计算机可执行指令被一个或多个控制处理器920执行，比如，被上述计算机设备900中的一个处理器920执行，可使得上 述一个或多个处理器920执行本申请任意实施例提供的无线网络WiFi感知方法。
本申请实施例包括：首先获取路径传输的WiFi信号，有利于后续对路径传输的WiFi信号进行处理；通过实时检测WiFi信号覆盖区域的场景状态，能够得到不同时刻区域内的动态变化；根据场景状态对WiFi信号进行信号处理，生成频谱图，实现了WiFi感知的多维度和高精度，频谱图能够反映空间变化与信号变化之间的映射关系，根据频谱图识别WiFi信号，从而能够用于实现粗粒度和精细粒度的感知任务。即是说，本申请实施例的方案通过实时检测的场景状态对多路径的WiFi信号进行处理，得到频谱图，实现了WiFi感知的多维度和高精度，从而能够用于实现粗粒度和精细粒度的感知任务。与相关技术相比，能够缓解多径效应造成的WiFi感知能力下降问题，实现了多维度和高精度的WiFi感知，不仅能够用于实现粗粒度的感知任务，还能够用于实现精细粒度的感知任务。
以上所描述的实施例仅仅是示意性的，其中作为分离部件说明的单元可以是或者也可以不是物理上分开的，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部模块来实现本实施例方案的目的。
本领域普通技术人员可以理解，上文中所公开方法中的全部或某些步骤、系统可以被实施为软件、固件、硬件及其适当的组合。某些物理组件或所有物理组件可以被实施为由处理器，如中央处理器、数字信号处理器或微处理器执行的软件，或者被实施为硬件，或者被实施为集成电路，如专用集成电路。这样的软件可以分布在计算机可读介质上，计算机可读介质可以包括计算机存储介质(或非暂时性介质)和通信介质(或暂时性介质)。如本领域普通技术人员公知的，术语计算机存储介质包括在设置为存储信息(诸如计算机可读指令、数据结构、程序模块或其他数据)的任何方法或技术中实施的易失性和非易失性、可移除和不可移除介质。计算机存储介质包括但不限于RAM、ROM、EEPROM、闪存或其他存储器技术、CD-ROM、数字多功能盘(DVD)或其他光盘存储、磁盒、磁带、磁盘存储或其他磁存储装置、或者可以设置为存储期望的信息并且可以被计算机访问的任何其他的介质。此外，本领域普通技术人员公知的是，通信介质通常包括计算机可读指令、数据结构、程序模块或者诸如或其他传输机制之类的调制数据信号中的其他数据，并且可包括任何信息递送介质。

Claims (10)

  1. 一种无线网络WiFi感知方法，所述方法包括：
    获取路径传输的WiFi信号；
    实时检测所述WiFi信号覆盖区域的场景状态；
    根据所述场景状态对所述WiFi信号进行信号处理，生成频谱图。
  2. 根据权利要求1所述的方法，其中，所述根据所述场景状态对所述WiFi信号进行信号处理，生成频谱图，包括：
    在所述场景状态为视线配置的情况下，根据信道状态信息对所述WiFi信号进行信号处理，生成所述频谱图。
  3. 根据权利要求2所述的方法，其中，所述WiFi信号为多路径的WiFi信号；
    所述在所述场景状态为视线配置的情况下，根据信道状态信息对所述WiFi信号进行信号处理，生成所述频谱图，包括：
    对各个路径的所述WiFi信号进行降噪处理，得到多个降噪信号；
    对各个所述降噪信号进行时变相关性分析，得到多个主成分信号；
    根据多个所述主成分信号，生成所述频谱图。
  4. 根据权利要求3所述的方法，其中，在多天线的情况下，所述对各个路径的所述WiFi信号进行降噪处理，得到多个降噪信号，包括：
    根据各个路径的所述WiFi信号，得到信道状态信息信号；
    计算两个相邻天线的所述信道状态信息信号的比值，得到多个所述降噪信号。
  5. 根据权利要求3所述的方法，其中，所述根据多个所述主成分信号，生成所述频谱图，包括：
    将各个所述主成分信号划分为多个相同长度的信号片段，并对各个所述信号片段进行傅里叶变换，得到各个所述主成分信号对应的光谱图；
    对各个所述光谱图进行算术计算处理，生成所述频谱图。
  6. 根据权利要求1所述的方法，其中，所述根据所述场景状态对所述WiFi信号进行信号处理，生成频谱图，包括：
    在所述场景状态为空间配置的情况下，根据被动WiFi雷达信息对所述WiFi信号进行信号处理，生成所述频谱图。
  7. 根据权利要求6所述的方法，其中，所述WiFi信号为多路径的WiFi雷达信号；
    所述在所述场景状态为空间配置的情况下，根据被动WiFi雷达信息对所述WiFi信号进行信号处理，生成所述频谱图，包括：
    对各个路径的所述WiFi雷达信号进行模糊性度量处理，得到各个度量信号；
    对各个所述度量信号进行干扰信号消除处理，得到各个去干扰信号；
    对各个所述去干扰信号进行噪音消除处理，得到各个多普勒脉冲；
    选择最大的所述多普勒脉冲生成所述频谱图。
  8. 一种无线网络WiFi感知系统，包括：
    信号获取模块，设置为获取路径传输的WiFi信号；
    检测模块，设置为实时检测所述WiFi信号覆盖区域的场景状态；
    信号处理模块，设置为根据所述场景状态对所述WiFi信号进行信号处理，生成频谱图。
  9. 一种计算机设备，所述计算机设备包括存储器和处理器，所述存储器中存储有计算机可读指令，所述计算机可读指令被一个或多个所述处理器执行时，使得一个或多个所述处理器执行如权利要求1至7中任一项所述方法的步骤。
  10. 一种计算机可读存储介质，所述存储介质可被处理器读写，所述存储介质存储有计算机指令，所述计算机可读指令被一个或多个处理器执行时，使得一个或多个处理器执行如权利要求1至7中任一项所述方法的步骤。
PCT/CN2023/096573 2022-07-25 2023-05-26 无线网络WiFi感知方法、系统及计算机设备 WO2024021824A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202210875975.3 2022-07-25
CN202210875975.3A CN117498967A (zh) 2022-07-25 2022-07-25 无线网络WiFi感知方法、系统及计算机设备

Publications (1)

Publication Number Publication Date
WO2024021824A1 true WO2024021824A1 (zh) 2024-02-01

Family

ID=89678743

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2023/096573 WO2024021824A1 (zh) 2022-07-25 2023-05-26 无线网络WiFi感知方法、系统及计算机设备

Country Status (2)

Country Link
CN (1) CN117498967A (zh)
WO (1) WO2024021824A1 (zh)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110730473A (zh) * 2019-09-03 2020-01-24 中国人民解放军陆军工程大学 面向WiFi活动识别的信号特征提取方法
CN111082879A (zh) * 2019-12-27 2020-04-28 南京邮电大学 一种基于深度时空模型的wifi感知方法
EP4030806A1 (en) * 2021-01-19 2022-07-20 Vestel Elektronik Sanayi ve Ticaret A.S. Methods and apparatuses for sensing application identification and prediction
CN114782935A (zh) * 2022-05-17 2022-07-22 山东建筑大学 基于视觉-WiFi信号联合检测的驾驶接管险态感知方法

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110730473A (zh) * 2019-09-03 2020-01-24 中国人民解放军陆军工程大学 面向WiFi活动识别的信号特征提取方法
CN111082879A (zh) * 2019-12-27 2020-04-28 南京邮电大学 一种基于深度时空模型的wifi感知方法
EP4030806A1 (en) * 2021-01-19 2022-07-20 Vestel Elektronik Sanayi ve Ticaret A.S. Methods and apparatuses for sensing application identification and prediction
CN114782935A (zh) * 2022-05-17 2022-07-22 山东建筑大学 基于视觉-WiFi信号联合检测的驾驶接管险态感知方法

Also Published As

Publication number Publication date
CN117498967A (zh) 2024-02-02

Similar Documents

Publication Publication Date Title
Zheng et al. Super-resolution delay-Doppler estimation for OFDM passive radar
US9379924B2 (en) Cognitive radio spectrum sensing with improved edge detection of frequency bands
CN107171749B (zh) 确定运动物体直接反射的无线信号的多普勒频移的方法
US9277413B2 (en) Cooperative cognitive radio spectrum sensing using a hybrid data-decision method
CN108957399B (zh) 一种利用lte信号信道状态信息实现运动检测的方法
CN109490826B (zh) 一种基于无线电波场强rssi的测距与位置定位方法
CN112187316A (zh) 信号处理方法、装置、接收机及存储介质
US10972141B2 (en) Method for estimating arrival time based on noise cancellation
Li et al. A taxonomy of WiFi sensing: CSI vs passive WiFi radar
KR20200047086A (ko) 클라우드 데이터 처리 gnss 재밍 감시 방법 및 시스템
US8837558B1 (en) Systems, methods, and devices for improving signal detection in communication systems
Bocus et al. UWB and WiFi systems as passive opportunistic activity sensing radars
CN114095102B (zh) 一种基于时频分析的无人机遥控信号识别方法
CN111948618A (zh) 一种基于卫星外辐射源的前向散射目标探测方法及系统
WO2024021824A1 (zh) 无线网络WiFi感知方法、系统及计算机设备
CN107728166B (zh) 一种基于时域分组处理的卫星导航接收机多干扰抑制方法
CN108718223B (zh) 一种非合作信号的盲频谱感知方法
US7783110B2 (en) Semicoherent channel estimator
Rock et al. CNNs for interference mitigation and denoising in automotive radar using real-world data
Soto et al. Wi-fi csi-based human presence detection using dtw features and machine learning
CN116106888A (zh) 无接触式运动目标速度测量方法、系统、设备及存储介质
Sgammini et al. SVD-based RF interference detection and mitigation for GNSS
Boudlal et al. Design and Deployment of a Practical Wireless Sensing System for HAR with WiFi CSI in the 5GHz Band
CN113271161B (zh) 一种电磁信号的筛选方法及装置
WO2023160546A1 (zh) 感知方法、装置及通信设备

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: 23845038

Country of ref document: EP

Kind code of ref document: A1