Living body target detection method, device, equipment, medium and product

Abstract

The invention discloses a living body target detection method, device and equipment, a medium and a product. The living body target detection method comprises the following steps: acquiring channel state information of each subcarrier in a WIFI system in a detection area; preprocessing the channel state information to obtain intermediate data under the interested distance; feature extraction is conducted on the intermediate data based on the autocorrelation lagging number and the weighting function, and breathing features are determined; and determining whether a living body target exists in the detection area or not according to the breathing features and preset features. The method comprises the following steps: carrying out distance filtering on channel state information under an interested distance to obtain intermediate data, carrying out signal enhancement on the intermediate data based on an autocorrelation lag number and a weighting function, and carrying out feature extraction and target detection on the enhanced signal. The interference from targets beyond the interested distance or multipath signals is reduced, the detection robustness is further improved, and the detection accuracy of the living body target with weak amplitude change is improved.

Description

Technical Field

[0001] This invention relates to the field of wireless communication technology, and in particular to a method, apparatus, device, medium, and product for detecting live targets. Background Technology

[0002] One emerging application of Wi-Fi is motion sensing. The data embedded in Wi-Fi Channel State Information (CSI) contains rich information about the motion of objects (living or non-living) within the Wi-Fi signal coverage area. The motion of these objects causes the CSI signal to change over time. This characteristic has been well utilized in applications such as gesture recognition and living organism detection.

[0003] In automotive applications, Wi-Fi is currently widely used for connectivity and communication. Simultaneously, Wi-Fi based on CSI signals can also serve as an in-vehicle wireless motion sensor to detect living beings (people or pets) inside the vehicle. This is an important application scenario: detecting the presence of a person or pet when they are alone in the car with the windows closed and the engine off.

[0004] However, when the life form inside the cabin is an infant or child, their movements are relatively small. For example, when an infant is asleep, only their breathing causes chest rise and fall, which generates a small disturbance signal for CSI. This can lead to the problem that the amplitude of the CSI signal change cannot be detected, and in this case, the life form inside the cabin may not be detected. Summary of the Invention

[0005] This invention provides a method, apparatus, device, medium, and product for detecting live targets, enabling the detection of live targets with small movement amplitudes.

[0006] According to a first aspect of the present invention, a method for detecting a live target is provided, comprising:

[0007] Obtain the channel state information of each subcarrier in the WIFI system within the detection area;

[0008] The channel state information is preprocessed to obtain intermediate data at the distance of interest;

[0009] Feature extraction is performed on the intermediate data based on autocorrelation lags and weighting functions to determine respiratory characteristics;

[0010] Based on the respiratory characteristics and preset characteristics, determine whether there is a living target in the detection area.

[0011] According to a second aspect of the present invention, a live target detection device is provided, comprising:

[0012] The information acquisition module is used to acquire the channel status information of each subcarrier in the WIFI system under the detection area;

[0013] The data determination module is used to preprocess the channel state information to obtain intermediate data at the distance of interest;

[0014] The feature extraction module is used to extract features from the intermediate data based on the autocorrelation lag number and weighting function to determine respiratory features;

[0015] The target determination module is used to determine whether a living target exists in the detection area based on the respiratory characteristics and preset characteristics.

[0016] According to a third aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0017] At least one processor; and

[0018] A memory communicatively connected to the at least one processor; wherein,

[0019] The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the live target detection method according to any embodiment of the present invention.

[0020] According to a fourth aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions for causing a processor to execute and implement the live target detection method according to any embodiment of the present invention.

[0021] According to a fifth aspect of the present invention, embodiments of the present invention also provide a computer program product, the computer program product including a computer program, which, when executed by a processor, implements the live target detection method of any embodiment of the present invention.

[0022] The technical solution of this invention involves acquiring channel state information of each subcarrier in a WIFI system within a detection area; preprocessing the channel state information to obtain intermediate data at the range of interest; extracting features from the intermediate data based on autocorrelation hysteresis and a weighting function to determine breathing characteristics; and determining the presence of a live target in the detection area based on the breathing characteristics and preset features. By performing range filtering on the channel state information at the range of interest to obtain intermediate data, enhancing the signal of the intermediate data based on autocorrelation hysteresis and a weighting function, extracting breathing characteristics from the enhanced signal, and detecting the target, this method reduces interference from targets outside the range of interest or multipath signals, further improving the robustness of detection and increasing the accuracy of detecting live targets with weak amplitude variations.

[0023] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0024] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 This is a flowchart of a live target detection method provided in Embodiment 1 of the present invention;

[0026] Figure 2 This is a flowchart of a live target detection method provided in Embodiment 2 of the present invention;

[0027] Figure 3 This is an example diagram of channel impulse response data in a live target detection method according to Embodiment 2 of the present invention;

[0028] Figure 4 This is an example diagram of the channel state signal in a live target detection method according to Embodiment 2 of the present invention;

[0029] Figure 5 This is an example flowchart of a live target detection method provided in Embodiment 2 of the present invention;

[0030] Figure 6 This is a schematic diagram of the structure of a live target detection device according to Embodiment 3 of the present invention;

[0031] Figure 7 This is a schematic diagram of the structure of an electronic device that implements an embodiment of the present invention. Detailed Implementation

[0032] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0033] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0034] Example 1

[0035] Figure 1 This is a flowchart of a live target detection method provided in Embodiment 1 of the present invention. This embodiment is applicable to the detection of live targets. The method can be executed by a live target detection device, which can be implemented in hardware and / or software and can be configured in an electronic device. Figure 1 As shown, the method includes:

[0036] S110. Obtain the channel status information of each subcarrier in the WIFI system under the detection area.

[0037] In this embodiment, the detection area can be understood as the area where liveness detection is performed, such as an indoor space or the cabin of a vehicle or other transportation equipment. The WIFI system can be understood as a system that provides WIFI signals. Subcarriers can be understood as a fundamental component of Orthogonal Frequency Division Multiplexing (OFDM) technology, which distributes high-speed data streams across multiple subcarriers for transmission. Channel State Information (CSI) can be understood as key information describing the channel attributes of a communication link in the field of wireless communication; it describes the interaction between the WIFI signal and the environment, including the direct path and multipath of the signal from the transmitter to the receiver.

[0038] Specifically, the processor can acquire the channel state information of each subcarrier in the WIFI system within the detection area, where the channel state information can be a data cube. , where f is the subcarrier frequency, n is the data packet number, and m is the channel number, which contains the CSI data of F subcarriers of N consecutive data packets from all M available channels in the MIMO system.

[0039] For example, channel state information can be viewed as a channel transfer function, describing the delay, attenuation, and scattering experienced by the subcarrier signal as it propagates throughout the environment. It can be defined by a complex number:

[0040]

[0041] in, It is the subcarrier frequency. It is the data packet number. and These are the receiver and transmitter signals, respectively.

[0042] S120. Preprocess the channel state information to obtain intermediate data at the distance of interest.

[0043] In this embodiment, since the bandwidth of a Wi-Fi signal is typically tens of megabits, the distance resolution of the channel state information is poor. For example, a Wi-Fi signal can penetrate a vehicle and cover an area of ​​tens of meters around it. Usually, only liveness detection inside the vehicle cabin is needed, without considering other liveness around the vehicle. This characteristic of Wi-Fi signals makes it impossible to accurately identify whether the target is within the detection range or outside the range. The distance of interest (ROI) can be understood as the range for liveness detection, such as inside the aforementioned vehicle cabin, where the ROI can coincide with the detection area. Intermediate data can be understood as the channel state information at the ROI.

[0044] Specifically, the processor can preprocess the channel state information, remove the linear components of the phase response along the subcarrier dimension, and then perform range filtering on the removed signal so that only the response of the target (including multipath information) with a short-range path enters the feature extraction stage to obtain intermediate data at the distance of interest.

[0045] S130. Based on the autocorrelation lag number and weighting function, feature extraction is performed on the intermediate data to determine respiratory characteristics.

[0046] In this embodiment, the autocorrelation hysteresis number can be understood as the number of hysteresis subcarriers considered when performing autocorrelation calculations. The weighting function can be understood as a function used to mix the information on each subcarrier according to different weighting values. Respiratory features can be understood as characteristics used to characterize the respiration of a living organism, such as respiratory energy, respiratory frequency, and intensity.

[0047] Specifically, the processor can calculate the autocorrelation function of the intermediate data under each subcarrier using the autocorrelation hysteresis number to obtain the autocorrelation value of each subcarrier. Then, it can sum the autocorrelation values ​​of all subcarriers using a weighting function to obtain the autocorrelation function of all subcarriers. Finally, it can convert the autocorrelation function into a signal in the frequency domain using a set frequency domain conversion algorithm to extract features in the frequency domain and obtain breathing features.

[0048] S140. Based on respiratory characteristics and preset characteristics, determine whether there is a living target in the detection area.

[0049] In this embodiment, the preset features can be understood as the respiratory-related features of the target of interest, such as the set respiratory rate (e.g., 0.2-1Hz for infants) or respiratory energy (e.g., the energy contained in 60 breaths of respiratory rate).

[0050] Specifically, the processor can search for respiratory features based on preset features to determine whether there is a living target in the detection area. For example, if there is no living target in the detection area, the respiratory frequency in the respiratory features should be 0. However, when there is a living target, such as an infant, there will be a respiratory frequency of 0.2-1Hz. Therefore, if there is a respiratory frequency of 0.2-1Hz in the respiratory features, it can be determined that there is a target in the detection area.

[0051] The technical solution of this invention obtains intermediate data by performing range filtering on channel state information at the range of interest. Signal enhancement is then performed on the intermediate data based on autocorrelation hysteresis and a weighting function. Breathing features are extracted from the enhanced signal, and the target is detected. This reduces interference from targets outside the range of interest or from multipath signals, further improving detection robustness and accuracy for detecting live targets with weak amplitude variations.

[0052] Example 2

[0053] Figure 2 This is a flowchart of a live target detection method provided in Embodiment 2 of the present invention. This embodiment is a further refinement of the above embodiment. Figure 2 As shown, the method includes:

[0054] S201. Obtain the channel status information of each subcarrier in the WIFI system under the detection area.

[0055] S202. The channel state information is dephase linearized using the least squares method to obtain delinearized data.

[0056] It is known that the center frequencies of the receiver and transmitter are often not precisely synchronized, resulting in a carrier frequency offset (CFO). This causes the signal to carry the remaining CFO, leading to a time-varying CSI phase offset across subcarriers. In real-world wireless communication environments, signals propagate to the receiver via multiple paths, each with varying lengths and attenuations. When the receiver attempts to estimate the channel state, it is essentially estimating the superposition effect of multiple paths, which can cause nonlinear changes in the CSI phase.

[0057] In this embodiment, the least squares method can be understood as a way to find the best function match for the data by minimizing the sum of squared errors, which are typically the differences between the observed values ​​and the model's predicted values. Delinearized data can be understood as data after removing linear phase.

[0058] Specifically, in channel state information, linear phase is related to the linear slope and phase offset on the subcarrier. The unknown parameter matrix of the least squares method is the linear slope and phase offset. The processor can construct the least squares method by unwrapping the phase vector, frequency vector and unknown parameter matrix along the subcarrier, and determine the unknown parameter matrix through the pre-determined estimated solution. The determined unknown parameters are used to perform dephase linearization processing on the channel state information to obtain delinearized data.

[0059] Furthermore, based on the above embodiments, the step of performing phase delinearization processing on the channel state information using the least squares method to obtain delinearized data can be refined as follows:

[0060] Based on the subcarrier spacing between each subcarrier, determine the fixed values ​​of the parameter matrix; determine the unwrapped phase vector of each data packet along each subcarrier and the channel frequency vector of each channel in the WIFI system; construct a least squares problem based on the unwrapped phase vector, channel frequency vector, and unknown parameter matrix; determine the estimated solution of the problem based on the fixed values ​​to obtain the linear phase term; subtract the linear phase term from the channel state information to obtain the delinearized data.

[0061] In this embodiment, the subcarrier spacing can be understood as the interval between the center frequencies of two adjacent subcarriers in orthogonal frequency division multiplexing. The fixed values ​​of the parameter matrix can be understood as pre-calculated fixed values ​​that vary only with the channel frequencies used in the WiFi signal. Data packets can be understood as consecutive data packets sent by each available channel in the WiFi system, i.e., n in the data cube shown above. The unwrapped phase vector can be understood as an important concept used to describe and process phase information in CSI signals. The channel frequency vector can be understood as a vector representing the frequency of each available channel. The linear phase term can be understood as the portion of the linear phase in the channel state information.

[0062] Specifically, the processor can determine the fixed values ​​of the parameter matrix based on the subcarrier spacing between each subcarrier; determine the unwrapped phase vector of each data packet along each subcarrier and the channel frequency vector of each channel in the WIFI system; construct a least squares problem based on the unwrapped phase vector, the channel frequency vector and the unknown parameter matrix; determine the estimated solution of the problem based on the fixed values ​​to obtain the linear phase term; and subtract the linear phase term from the channel state information to obtain the delinearized data.

[0063] For example, the impulse response of a channel in the channel state information can be expressed as:

[0064]

[0065] in, It is the part that does not contain linear phase. In subcarrier linear slope on, It is on this subcarrier The phase shift on the time domain. Converting the above expression to the time domain means a time (or range) shift in the channel impulse response, primarily due to uncalibrated delays on the hardware path. IFT indicates performing an inverse Fourier transform:

[0066]

[0067] Therefore, the linear phase term can be characterized as The removal of the linear phase term is to estimate the parameter characteristics. The estimation of parameter characteristics can be accomplished by performing a least-squares fit along the subcarrier phase for each data packet. The least-squares problem can be written as:

[0068]

[0069] in, It is along the subcarrier The unwrapped phase vector, It is a frequency vector, which can be represented as:

[0070]

[0071] Where B is an unknown parameter. The unknown parameter matrix is ​​formed, and the estimated solution of this least squares estimation is... for:

[0072]

[0073] Among them, matrix It is a fixed value that can be calculated and stored in advance, and it only varies with the channel frequency used in the WIFI signal. The subcarrier spacing between each subcarrier in the OFDM WIFI signal can be utilized. To further simplify the problem and make it independent of the channel used, the linear phase term can be derived from the subcarrier index. Rewritten as follows,

[0074]

[0075] in, and The matrix of the frequency vector here can be rewritten as:

[0076]

[0077] and for Make the matrix It is a fixed value independent of the channel used, and can be obtained through... and Estimate and .

[0078] S203. Perform distance filtering on the delinearized data to obtain intermediate data at the distance of interest.

[0079] Specifically, the processor can set a filter based on the distance of interest, such as a low-pass filter, and use the filter to perform distance filtering on the delinearized data to obtain intermediate data at the distance of interest.

[0080] Furthermore, based on the above embodiments, the step of performing distance filtering on the delinearized data to obtain intermediate data at the distance of interest can be refined as follows:

[0081] The delinearized data is converted into channel impulse response data; the intermediate data at the range of interest is obtained by convolving the channel impulse response data with a range filter.

[0082] In this embodiment, the Channel Impulse Response (CIR) data can be understood as a response function to a unit pulse signal, reflecting the channel's attenuation and phase shift characteristics to different frequency components. The range filter can be understood as a filter pre-set according to the range of interest for filtering distances; for example, it could be a low-pass filter.

[0083] Specifically, the processor can perform an inverse Fourier transform on the delinearized data to obtain channel impulse response data. This data describes the multiple paths the propagating signal has traversed and provides useful information about the perceived locations of multiple objects in the sensing area. The time impulse response can also be converted to distance using free-space path delay. The processor can then convolve the channel impulse response data with a distance filter to obtain intermediate data at the distance of interest.

[0084] For example, from delinearized data Perform convolution operations along the subcarriers, as follows:

[0085]

[0086] in, For intermediate data, This is the distance filter in the desired frequency domain. In this operation, the delinearized data and the frequency response of the filter are compared. Perform convolution. This operation is similar to multiplying the CIR pulse shape by the range filter, i.e., converting it to CIR:

[0087]

[0088] Equivalent to:

[0089]

[0090] in It can be designed as a pulse centered at t=0, with a spectral shape of... .

[0091] For example, to illustrate the effect of range filtering more clearly, an inverse Fourier transform is performed on the channel state information, using an example of channel impulse response (CIR) data as a concrete demonstration. Figure 3 This invention provides an example diagram of channel impulse response data in a live target detection method according to Embodiment 2 of the present invention, as shown in the diagram. Figure 3 As shown, the horizontal axis represents the data packet number, and the vertical axis represents the distance, for example, Index=120, t=0. The color from white to black represents the subcarrier frequency from low to high. Figure 3 In the figure, 'a' represents the original channel impulse response data example, 'b' represents the delinearized channel impulse response data, and 'c' represents the channel impulse response data after range filtering. As can be seen from the figure, filtering the region of interest to between 100 and 150 effectively limits various interferences that may come from distant targets outside the region of interest or from multipath interference caused by targets within the region of interest.

[0092] For example, the effect can be demonstrated using CSI data within a single data packet. Figure 4 This is an example diagram of the channel state signal in a live target detection method provided in Embodiment 2 of the present invention, as shown below. Figure 4 The dashed line represents the original CIR data, the light line with dots represents the delinearized CIR data, and the black line represents the delinearized and range-filtered CSI data. It can be seen that the original data under the dashed line has a phase shift; after correction, it can be seen that the overall shift has returned to its original position to the left. The light line with dots indicates strong interference outside the region of interest. After range filtering and delinearization, the region of interest is divided between 100 and 150 degrees. The curve shows that the interference signal outside the region of interest is weakened, its fluctuations are reduced, and the signal within the region of interest is highlighted more clearly.

[0093] S204. Determine the autocorrelation matrix of each subcarrier in the intermediate data based on the autocorrelation hysteresis.

[0094] Understandably, the cross-correlation coefficients between CSI subcarriers describe the degree of variation between subcarriers. If there is no motion within the region of interest, the cross-correlation coefficient is usually close to 1, while if there is various activities within the region of interest, the cross-correlation coefficient is very small. The second characteristic is the variance of CSI over time, which is also very sensitive to changes between data packets. A large variance value usually indicates the presence of motion, while a low variance value indicates a high probability of no activity. The activity value is calculated based on autocorrelation.

[0095] In this embodiment, the autocorrelation matrix can be understood as a matrix that characterizes the correlation between each subcarrier and other subcarriers in matrix form.

[0096] Specifically, the processor can calculate the autocorrelation matrix of the CSI data from the intermediate data cube, which is a matrix of size [missing information]. , of the matrix, It is the number of subcarriers selected. It equals the lag number in the autocorrelation calculation. The purpose of calculating autocorrelation is to improve the signal-to-noise ratio of intermediate data and amplify any weak periodic motion contained in the data, so that any periodic variations in the intermediate data can be observed in their amplitude.

[0097] For example, the autocorrelation matrix of each subcarrier can be determined using the following formula. :

[0098]

[0099] Where l represents the autocorrelation lag number.

[0100] S205. Determine the autocorrelation function based on the respective correlation matrices and weighting functions.

[0101] Specifically, the processor can weight each correlation matrix with its corresponding weighting function and sum them over the whole to obtain the autocorrelation function corresponding to the intermediate data.

[0102] For example, the autocorrelation function can be determined using the following formula. :

[0103]

[0104] Wherein, the weighting function is It can be obtained in different ways, such as the variance function or autocorrelation function value on the fk-th subcarrier.

[0105] S206. Perform feature enhancement on the autocorrelation function to obtain the feature function to be extracted.

[0106] In this embodiment, the feature function to be extracted can be understood as the function after feature enhancement.

[0107] Specifically, the processor can perform a Hilbert transform on the autocorrelation function and set the frequency range of interest, such as less than 2Hz. The processor can then use a set filter to perform breathing frequency filtering on the autocorrelation function to obtain the feature function to be extracted.

[0108] S207. Perform feature extraction on the feature function to be extracted to obtain respiratory features.

[0109] Specifically, the processor can perform frequency domain transformation on the feature function to be extracted to obtain breathing features.

[0110] Furthermore, based on the above embodiments, the steps for extracting features from the feature function to be extracted and obtaining respiratory features can be refined as follows:

[0111] If the signal strength of the feature function to be extracted reaches the set strength, the spectrum is determined by the Fast Fourier Transform algorithm to obtain the respiratory spectrum energy and respiratory spectrum noise energy, which are used as respiratory features; otherwise, the feature function to be extracted is analyzed by the Relaxed RELAX algorithm to determine the respiratory frequency and respiratory amplitude, which are used as respiratory features.

[0112] In this embodiment, signal strength can be understood as the amplitude of the signal. Setting the strength can be understood as distinguishing the strength of the algorithm used, and as distinguishing whether the signal from which the feature function is to be extracted is a weak signal. The Fast Fourier Transform (FFT) algorithm can be understood as an algorithm that converts a time-domain signal into a frequency-domain signal. Breath spectrum energy can be understood as the frequency range of interest. Breath spectrum noise energy can be understood as the portion outside the frequency range of interest. The RELAX algorithm can be understood as a high-resolution spectrum analysis algorithm.

[0113] Specifically, the processor can compare the signal strength of the feature function to be extracted with a set strength. When the set strength is reached, a Fast Fourier Transform (FFT) can be performed on the feature signal to be extracted to obtain the respiratory spectrum energy and respiratory spectrum noise energy. The preset features are the frequency range of interest; if a living organism exists within this frequency range, then the features of the living organism are included. First, the spectrum is calculated using FFT, and approximately... Hertz (e.g., 0.2-1 Hz for infants) or per minute The energy contained within each breath is termed respiratory energy under the autocorrelation function. Here, we only use the respiratory signal as an example; it could also be calculated using heartbeats. However, because the chest displacement caused by breathing is significantly greater than cardiac activity (i.e., heartbeats), respiratory features are easier to extract than heartbeat features. Additionally, respiratory spectral noise energy, i.e., the total energy outside the frequency range of interest, is also a useful feature. Typically, respiratory spectral noise energy increases with the presence of any motion, primarily due to the Doppler effect in a multipath environment. Because of the multipath effect within the cabin, any motion will produce various Doppler shifts.

[0114] Specifically, when the signal strength does not reach the set strength, it can be determined that the signal is weak when the feature function to be extracted is being extracted. Since the breathing frequency cannot be determined by Fast Fourier Transform, a high-resolution spectral analysis algorithm, the RELAX algorithm, can be used for calculation. The RELAX algorithm is a simple, iterative, and computationally efficient algorithm used to estimate the number of sine waves contained in the signal and their corresponding frequencies. These sine waves and their corresponding frequencies cannot be detected by simple low-resolution peak search algorithms. The RELAX algorithm is used to analyze the feature function to be extracted, determining the breathing frequency and breathing amplitude, which are then used as breathing features. The preset feature at this point is either the breathing frequency or the breathing amplitude of interest.

[0115] S208. Based on respiratory characteristics and preset characteristics, determine whether there is a living target in the detection area.

[0116] The technical solution of this invention, by de-phase linearizing the channel state information, obtains delinearized data. After removing the linear phase, the CSI data better reflects the true environmental characteristics and target information, which helps improve the accuracy and reliability of live target detection. Range filtering is applied to the delinearized data to obtain intermediate data, reducing interference from targets or multipath signals outside the range of interest. The autocorrelation function is calculated on the intermediate data using autocorrelation hysteresis and a weighting function, introducing features related to breathing frequency to further improve detection robustness. A fast Fourier transform is performed on the autocorrelation function to add features other than breathing frequency to improve detection robustness. Using the RELAX algorithm on the autocorrelation function helps recover weak breathing signals from the spectrum, improving the accuracy of frequency estimation and increasing the detection accuracy for live targets with weak amplitude changes.

[0117] For example, to better understand this solution, a specific example will be used as a demonstration. Figure 5 Here is an example flowchart of a live target detection method provided in Embodiment 2 of the present invention, as shown below. Figure 5As shown, the steps may include: acquiring raw CSI data of the WIFI system in the detection area, constructing channel state information H(f,n,m) in the form of a data cube for each subcarrier; and performing phase delinearization processing on the channel state information using the least squares method to obtain delinearized data H. d (f,n,m); Perform distance filtering on the delinearized data to obtain intermediate data H at the distance of interest. dr (f,n,m); Outlier removal is performed on intermediate data, and the autocorrelation function R(l) is calculated using the autocorrelation lag number and weighting function; feature enhancement is performed on the autocorrelation function using Hilbert transform and respiratory frequency filtering to obtain the feature function to be extracted; feature extraction is performed on the feature function to be extracted using the fast Fourier transform algorithm to obtain respiratory spectrum energy and respiratory spectrum noise energy; based on the respiratory spectrum energy, respiratory spectrum noise energy, and preset features, it is determined whether there is a live target in the detection area; feature extraction is performed on the feature function to be extracted using the RELAX algorithm to determine the respiratory frequency and respiratory amplitude; based on the respiratory frequency, respiratory amplitude, and preset features, it is determined whether there is a live target in the detection area.

[0118] Example 3

[0119] Figure 6 This is a schematic diagram of a live target detection device provided in Embodiment 3 of the present invention. Figure 6 As shown, the device includes:

[0120] The information acquisition module 61 is used to acquire the channel status information of each subcarrier in the WIFI system under the detection area;

[0121] Data determination module 62 is used to preprocess the channel state information to obtain intermediate data at the distance of interest;

[0122] Feature extraction module 63 is used to extract features from the intermediate data based on autocorrelation lag number and weighting function to determine respiratory features;

[0123] The target determination module 64 is used to determine whether a living target exists in the detection area based on the breathing characteristics and preset characteristics.

[0124] The technical solution of this invention obtains intermediate data by performing range filtering on channel state information at the range of interest. Signal enhancement is then performed on the intermediate data based on autocorrelation hysteresis and a weighting function. Breathing features are extracted from the enhanced signal, and the target is detected. This reduces interference from targets outside the range of interest or from multipath signals, further improving detection robustness and accuracy for detecting live targets with weak amplitude variations.

[0125] Furthermore, the data determination module 62 includes:

[0126] The first determining submodule is used to perform phase delinearization processing on the channel state information using the least squares method to obtain delinearized data;

[0127] The second determining submodule is used to perform distance filtering on the delinearized data to obtain intermediate data at the distance of interest.

[0128] Specifically, the first determining submodule is used for:

[0129] The fixed values ​​of the parameter matrix are determined based on the subcarrier spacing between each of the subcarriers;

[0130] Determine the unwrapped phase vector of each data packet along each of the subcarriers and the channel frequency vector of each channel in the WIFI system;

[0131] Based on the unwrapped phase vector, the channel frequency vector, and the unknown parameter matrix, construct the least squares problem;

[0132] Based on the fixed value, an estimated solution to the problem is determined to obtain the linear phase term;

[0133] Subtracting the linear phase term from the channel state information yields delinearized data.

[0134] Specifically, the second determining submodule is used for:

[0135] The delinearized data is converted into channel impulse response data;

[0136] Intermediate data at the distance of interest is obtained by convolving the channel impulse response data with a distance filter.

[0137] Furthermore, the feature extraction module 63 includes:

[0138] The third determining submodule is used to determine the autocorrelation matrix of each subcarrier in the intermediate data based on the autocorrelation hysteresis number;

[0139] The fourth determining submodule is used to determine the autocorrelation function based on each of the autocorrelation matrices and weighting functions;

[0140] The fifth determining submodule is used to perform feature enhancement on the autocorrelation function to obtain the feature function to be extracted;

[0141] The sixth determining submodule is used to extract features from the feature function to be extracted, thereby obtaining respiratory features.

[0142] Specifically, the sixth determining submodule is used for:

[0143] If the signal strength of the feature function to be extracted reaches the set strength, the spectrum is determined by the Fast Fourier Transform algorithm to obtain the respiratory spectrum energy and respiratory spectrum noise energy as respiratory features.

[0144] Otherwise, the respiratory rate and respiratory amplitude are determined by analyzing the feature function to be extracted using the relaxed RELAX algorithm and used as the respiratory features.

[0145] The live target detection device provided in the embodiments of the present invention can execute the live target detection method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the method execution.

[0146] Example 4

[0147] Figure 7 A schematic diagram of an electronic device 70 that can be used to implement embodiments of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0148] like Figure 7 As shown, the electronic device 70 includes at least one processor 71 and a memory, such as a read-only memory (ROM) 72 and a random access memory (RAM) 73, communicatively connected to the at least one processor 71. The memory stores computer programs executable by the at least one processor. The processor 71 can perform various appropriate actions and processes based on the computer program stored in the ROM 72 or loaded into the RAM 73 from storage unit 78. The RAM 73 can also store various programs and data required for the operation of the electronic device 70. The processor 71, ROM 72, and RAM 73 are interconnected via a bus 74. An input / output (I / O) interface 75 is also connected to the bus 74.

[0149] Multiple components in electronic device 70 are connected to I / O interface 75, including: input unit 76, such as keyboard, mouse, etc.; output unit 77, such as various types of monitors, speakers, etc.; storage unit 78, such as disk, optical disk, etc.; and communication unit 79, such as network card, modem, wireless transceiver, etc. Communication unit 79 allows electronic device 70 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0150] Processor 71 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 71 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, digital signal processors (DSPs), and any suitable processor, controller, microcontroller, etc. Processor 71 performs the various methods and processes described above, such as live target detection methods.

[0151] In some embodiments, the live target detection method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 78. In some embodiments, part or all of the computer program may be loaded and / or mounted on electronic device 70 via ROM 72 and / or communication unit 79. When the computer program is loaded into RAM 73 and executed by processor 71, one or more steps of the live target detection method described above may be performed. Alternatively, in other embodiments, processor 71 may be configured to perform the live target detection method by any other suitable means (e.g., by means of firmware).

[0152] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0153] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0154] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0155] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0156] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0157] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0158] In one embodiment, the present invention further includes a computer program product, which includes a computer program that, when executed by a processor, implements the live target detection method of any embodiment of the present invention.

[0159] In implementing the computer program product, computer program code for performing the operations of this invention can be written in one or more programming languages ​​or a combination thereof. Programming languages ​​include object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0160] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0161] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for detecting live targets, characterized in that, include: Obtain the channel state information of each subcarrier in the WIFI system within the detection area; The channel state information is preprocessed to obtain intermediate data at the distance of interest; Feature extraction is performed on the intermediate data based on autocorrelation lags and weighting functions to determine respiratory characteristics; Based on the respiratory characteristics and preset characteristics, determine whether there is a living target in the detection area.

2. The method according to claim 1, characterized in that, The preprocessing of the channel state information to obtain intermediate data at the distance of interest includes: The channel state information is dephase linearized using the least squares method to obtain delinearized data. Distance filtering is performed on the delinearized data to obtain intermediate data at the distance of interest.

3. The method according to claim 2, characterized in that, The process of dephase linearizing the channel state information using the least squares method to obtain delinearized data includes: The fixed values ​​of the parameter matrix are determined based on the subcarrier spacing between each of the subcarriers; Determine the unwrapped phase vector of each data packet along each of the subcarriers and the channel frequency vector of each channel in the WIFI system; Based on the unwrapped phase vector, the channel frequency vector, and the unknown parameter matrix, construct the least squares problem; Based on the fixed value, an estimated solution to the problem is determined to obtain the linear phase term; Subtracting the linear phase term from the channel state information yields delinearized data.

4. The method according to claim 2, characterized in that, The step of performing distance filtering on the delinearized data to obtain intermediate data at the distance of interest includes: The delinearized data is converted into channel impulse response data; Intermediate data at the distance of interest is obtained by convolving the channel impulse response data with a distance filter.

5. The method according to claim 1, characterized in that, The step of extracting features from the intermediate data based on autocorrelation lags and weighting functions to determine respiratory features includes: Based on the autocorrelation hysteresis, determine the autocorrelation matrix of each subcarrier in the intermediate data; Based on the autocorrelation matrices and weighting functions described above, determine the autocorrelation function; The autocorrelation function is enhanced to obtain the feature function to be extracted; Feature extraction is performed on the feature function to be extracted to obtain respiratory features.

6. The method according to claim 5, characterized in that, The step of extracting features from the feature function to obtain respiratory features includes: If the signal strength of the feature function to be extracted reaches the set strength, the spectrum is determined by the Fast Fourier Transform algorithm to obtain the respiratory spectrum energy and respiratory spectrum noise energy as respiratory features. Otherwise, the respiratory rate and respiratory amplitude are determined by analyzing the feature function to be extracted using the relaxed RELAX algorithm and used as the respiratory features.

7. A live target detection device, characterized in that, include: The information acquisition module is used to acquire the channel status information of each subcarrier in the WIFI system under the detection area; The data determination module is used to preprocess the channel state information to obtain intermediate data at the distance of interest; The feature extraction module is used to extract features from the intermediate data based on the autocorrelation lag number and weighting function to determine respiratory features; The target determination module is used to determine whether a living target exists in the detection area based on the respiratory characteristics and preset characteristics.

8. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the live target detection method according to any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that cause a processor to execute the live target detection method according to any one of claims 1-6.

10. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the live target detection method according to any one of claims 1-6.

Images