A life detection method and system based on radar signal

Abstract

The present application proposes a life detection method and system based on radar signals, the method comprising: using an ultra-wideband radar to detect the area to be detected, obtaining a radar echo matrix, and filtering out static clutter in the radar echo matrix; extracting the body surface vibration signal of the human target in the area to be detected according to the radar echo matrix after filtering out the clutter; using the modal decomposition method to perform modal decomposition on the body surface vibration signal; using the wavelet analysis method to determine the main frequency of each intrinsic mode function, and screening out the intrinsic mode function whose main frequency is within the preset breathing and heartbeat frequency band; reconstructing the breathing and heartbeat signals based on the screened intrinsic mode functions; and determining the vital signs information of the detected human target based on the stochastic resonance system and the reconstructed breathing and heartbeat signals. The technical solution provided by the present invention has strong anti-interference ability and can accurately and quickly detect the breathing frequency and heartbeat frequency of the target to be detected.

Description

Technical Field

[0001] The present application relates to the field of life detection technology, and in particular to a life detection method and system based on radar signals. Background Art

[0002] Life detection radar is a high-tech life-saving device developed by integrating micro-power ultra-wideband radar technology and biomedical engineering technology. It is professionally used for searching and locating survivors in emergency rescue missions such as earthquake disasters and landslides. It uses the characteristics of nanosecond electromagnetic wave pulses, such as wide spectrum, strong penetration, high resolution, good anti-interference, and low power consumption, to penetrate non-metallic media and emit nanosecond pulse electromagnetic waves to long distances. Based on the time domain Doppler effect generated by human movement on the radar echo, it analyzes whether there is a human target in the detection range and obtains the specific location and vital signs information of the human target.

[0003] Existing life detection algorithms have poor anti-interference capabilities. When there is strong electromagnetic interference or high-intensity noise in the detection environment, it is difficult to separate and extract weak vital signs signals from radar echo signals. At the same time, it is difficult for existing algorithms to further separate heartbeat signals. Even if the heartbeat frequency can be extracted, the signal-to-noise ratio is very high and the accuracy is poor, which in turn affects the timeliness and accuracy of rescue. Summary of the invention

[0004] The present application provides a life detection method and system based on radar signals, so as to at least solve the technical problems in the related art of poor anti-interference ability and inability to accurately extract heartbeat frequency.

[0005] The first embodiment of the present application provides a life detection method based on radar signals, the method comprising:

[0006] Using an ultra-wideband radar to detect the area to be detected, obtaining a radar echo matrix corresponding to the area to be detected, and filtering out static clutter in the radar echo matrix;

[0007] Extract the body surface vibration signal of the human target in the area to be detected according to the radar echo matrix after filtering out the clutter;

[0008] Performing modal decomposition on the body surface vibration signal by using a modal decomposition method to obtain each eigenmode function corresponding to the filter body surface vibration signal;

[0009] Determine the main frequency of each eigenmode function by wavelet analysis, and select the eigenmode function whose main frequency is within the preset breathing and heartbeat frequency band;

[0010] Reconstruct the breathing and heartbeat signals based on the screened intrinsic mode functions;

[0011] The life information of the detected human target is determined based on the stochastic resonance system and the reconstructed breathing and heartbeat signals.

[0012] A second aspect of the present application provides a life detection system based on radar signals, the life detection system comprising:

[0013] A first filtering module is used to detect the area to be detected by using an ultra-wideband radar, obtain a radar echo matrix corresponding to the area to be detected, and filter out static clutter in the radar echo matrix;

[0014] An extraction module is used to extract the body surface vibration signal of the human target in the detection area according to the radar echo matrix after filtering out the clutter;

[0015] A modal decomposition module, used for performing modal decomposition on the body surface vibration signal by using a modal decomposition method to obtain each intrinsic mode function corresponding to the body surface vibration signal;

[0016] A screening module, used for determining the main frequency of each eigenmode function by wavelet analysis, and screening out the eigenmode functions whose main frequency is within the preset breathing and heartbeat frequency band;

[0017] A reconstruction module, used for reconstructing the breathing and heartbeat signals based on the screened intrinsic mode functions;

[0018] The determination module is used to determine the life information of the detected human target based on the stochastic resonance system and the reconstructed breathing and heartbeat signals.

[0019] The technical solution provided by the embodiments of the present application brings at least the following beneficial effects:

[0020] The present application proposes a life detection method and system based on radar signals, the method includes: using ultra-wideband radar to detect the area to be detected, obtaining the radar echo matrix corresponding to the area to be detected, and filtering out the static clutter in the radar echo matrix; extracting the body surface vibration signal of the human target in the area to be detected according to the radar echo matrix after filtering out the clutter; using the modal decomposition method to perform modal decomposition on the body surface vibration signal to obtain each intrinsic mode function corresponding to the body surface vibration signal; using the wavelet analysis method to determine the main frequency of each intrinsic mode function, and screening out the intrinsic mode function whose main frequency is within the preset breathing and heartbeat frequency band; reconstructing the breathing and heartbeat signals based on the screened intrinsic mode functions; determining the life information of the detected human target based on the stochastic resonance system and the reconstructed breathing and heartbeat signals. The technical method provided by the present invention has strong anti-interference ability and can accurately separate and extract the vital sign signal from the radar echo signal, and can also accurately extract the heartbeat frequency.

[0021] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0023] Figure 1 A schematic diagram of a flow chart of a life detection method based on radar signals according to an embodiment of the present application;

[0024] Figure 2 The figure is a structural diagram of a life detection system based on radar signals according to an embodiment of the present application. DETAILED DESCRIPTION

[0025] Embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application.

[0026] A life detection method and system based on radar signals according to an embodiment of the present application will be described below with reference to the accompanying drawings.

[0027] Example 1

[0028] Figure 1 FIG. 1 is a flow chart of a life detection method based on radar signals according to an embodiment of the present application. Figure 1 As shown, it may include:

[0029] Step 1: Use an ultra-wideband radar to detect the area to be detected, obtain a radar echo matrix corresponding to the area to be detected, and filter out static clutter in the radar echo matrix;

[0030] It should be noted that, assuming that the pulse signal transmitted by the ultra-wideband pulse radar is s(τ), the echo signal received by the antenna is: Where n(τ,t) is the random noise signal of the receiver. Then, the echo signal is discretized to obtain: Where R(m,n) is the radar echo matrix, δ s and T s They represent the sampling interval of fast time and slow time respectively, M and N represent the total number of sampling points of fast time and slow time respectively, τ v (t) is the time delay of the human body micro-motion echo, t and τ are slow time and fast time respectively, a v is the amplitude of the human body micro-motion echo, ai is the amplitude of the echo from a stationary object in the environment, τ i (t) is the time delay of the echo from a stationary object in the environment.

[0031] It should be noted that filtering out static clutter in the radar echo matrix includes:

[0032] The static clutter in the radar echo matrix is ​​filtered out by using the subtraction average method.

[0033] The specific implementation process of the subtraction average method is as follows: first, the stationary object echo signal of a certain sampling period is found within the period, and then the target echo signal is obtained by subtracting the reference received pulse from each received pulse. The expression of the reference received pulse is: In the formula, M is the number of sampling points in each sampling period, i is the number of current sampling periods, and N is the total number of reference pulse sampling periods. Therefore, the processed signal is: R[m,n]=R[m,n]-C[m].

[0034] Step 2: extracting the body surface vibration signal of the human target in the area to be detected according to the radar echo matrix after filtering out the clutter;

[0035] In the disclosed embodiment, extracting the body surface vibration signal of the human target in the area to be detected according to the radar echo matrix after filtering out the clutter includes:

[0036] Calculating the state parameters of the slow time series corresponding to each range gate in the radar echo matrix after filtering out clutter;

[0037] It should be noted that the state parameter includes: permutation entropy value or approximate entropy value or sample entropy value or fuzzy entropy value or variance or standard deviation or kurtosis or energy.

[0038] Determine the location of the human target in the area to be detected by using the range gate corresponding to the state parameter that meets the preset conditions;

[0039] According to the location of the human target, a slow time series corresponding to the range gate of the location of the human target is extracted from the radar echo matrix after clutter filtering, and the slow time series is used as the body surface vibration signal of each human target.

[0040] It should be noted that the method of determining the position of the human target in the to-be-detected area by using the range gate corresponding to the state parameter that meets the preset conditions includes:

[0041] The range gate corresponding to the minimum value of the permutation entropy value is used as the range gate where the human target in the area to be detected is located, and then the position of the human target is determined based on the range gate;

[0042] For example, let the fast time point m (i.e., range gate m) and the slow time sequence length be N. Then the slow time sequence can be expressed as X[m,i]={y(1),y(2),...,y(i)},i=1,2,3...,N; Y∈R M×N , reconstruct the phase space and obtain the reconstruction matrix Y:

[0043] j=1,2,3,...,K, where τ d is the delay time, q is the embedding dimension, K = N-(q-1)τ d In this embodiment, the parameters of permutation entropy are set as follows: q is 7, τ d is 1. Each row of the phase space matrix is ​​called a reconstruction component. Arrange the elements in the jth reconstruction component in ascending order to obtain a new reconstruction component: X j ={x(j+(j1-1)τ d )≤x(j+(j2-1)τ d )≤...≤x(j+(j q -1)τ d )}, where j1,j2,...,j q Represents X j The position of each element in the dimension. If the elements are equal, they are sorted by the value of j. For any reconstructed component of the phase space matrix, a set of sequences can be obtained: S(l) = {j1, j2, ..., j q}, where l = {1, 2, ..., K}, and K ≤ q! . The q-dimensional sequence generates q! S(l) sortings. Calculate the probability of each S(l) sequence in different sortings, and the sum of all the probabilities of the sortings is 1. Then the permutation entropy of X[m,i] can be defined as: H p The size of H represents the complexity of the slow time series X[m,i], p The smaller the value, the more regular the time series changes. Conversely, the more random the time series changes. M×N In , the slow-time directional data of the human target position range changes relatively regularly, so the permutation entropy value corresponding to the human target area is low.

[0044] Alternatively, the range gate corresponding to the minimum value of the approximate entropy value is used as the range gate where the human target is located in the area to be detected, and then the position of the human target is determined based on the range gate;

[0045] For example, let the fast time point m (i.e., range gate m) and the slow time sequence length be N. Then the slow time sequence can be expressed as X[m,i]={y(1),y(2),...,y(i)},i=1,2,3...,N; Y∈R M×N, reconstruct a b-dimensional vector for the slow time series; wherein the b-dimensional vector can be expressed as Y k ={y(k),y(k+1),...,y(k+b-1)}, k=1,2,3...,N-b+1, parameter b is usually set to 2 or 3;

[0046] Calculate the distance between any vector in the reconstructed b-dimensional vector and other vectors in the reconstructed b-dimensional vector, wherein the distance is the maximum value of the absolute value of the difference between the corresponding elements of the two vectors;

[0047] The number E of vectors whose distances between each vector in the reconstructed b-dimensional vector and other vectors in the reconstructed b-dimensional vector are less than a given threshold r is counted. r , and calculate E r Ratio of the total number (N-b+1) And let the standard deviation of the slow time series X[m1,i] be ζ, and the given threshold r is usually 0.2ζ;

[0048] For each vector generated Take the logarithm respectively, and then log (N-b+1) logarithmized Find the average value and record it as

[0049] Increase b by 1 and repeat the above steps to get φ b+1 (r), thus the approximate entropy A of the range gate m is obtained as A = φ b+1 (r)-φ b (r).

[0050] Alternatively, the range gate corresponding to the minimum value of the sample entropy value is used as the range gate where the human target is located in the area to be detected, and then the position of the human target is determined based on the range gate;

[0051] Alternatively, the range gate corresponding to the minimum fuzzy entropy value is used as the range gate where the human target is located in the area to be detected, and then the position of the human target is determined based on the range gate;

[0052] Alternatively, the range gate corresponding to the maximum variance value is used as the range gate where the human target is located in the area to be detected, and then the position of the human target is determined based on the range gate;

[0053] For example, the variance of each range gate is calculated, and then the range gate with the largest variance is selected as the range gate where the human target is located. The calculation formula of the variance of the range gate m is:

[0054]

[0055]

[0056] Where m is the range gate index, N is the slow time sampling point, and Rm,i represents the amplitude value of the range gate m at the i-th slow time sampling point.

[0057] Alternatively, the range gate corresponding to the maximum value of the standard deviation is used as the range gate where the human target is located in the area to be detected, and then the position of the human target is determined based on the range gate,

[0058] For example, the standard deviation of each range gate is calculated, and then the range gate with the maximum standard deviation is selected as the range gate where the human target is located. The standard deviation calculation formula of the range gate m is:

[0059]

[0060] Alternatively, the range gate corresponding to the maximum kurtosis value is used as the range gate where the human target is located in the area to be detected, and then the position of the human target is determined based on the range gate.

[0061] For example, the kurtosis of each range gate is calculated, and then the range gate with the maximum kurtosis is selected as the range gate where the human target is located. The kurtosis calculation formula of the range gate m is:

[0062]

[0063] Alternatively, the range gate corresponding to the maximum energy value is used as the range gate where the human target is located in the area to be detected, and then the position of the human target is determined based on the range gate.

[0064] For example, the energy of each range gate is first calculated, and then the maximum energy range gate is selected as the range gate where the human target is located. The energy calculation formula of the range gate m is:

[0065]

[0066] It should be noted that, then determining the position of the human target based on the range gate includes:

[0067] According to the distance gate P where the human target is located pos , using the formula The distance of the human target is estimated, where Range is the distance between the human target and the radar antenna. v is the electromagnetic wave propagation speed, which is equal to 3×10 8 m / s, δ s is the fast time sampling interval.

[0068] In the embodiment of the present disclosure, before performing modal decomposition on the body surface vibration signal using the modal decomposition method, the method further includes: performing Butterworth filtering on the body surface vibration signal;

[0069] Wherein, performing Butterworth filtering on the body surface vibration signal includes:

[0070] A Butterworth filter is used to filter out high-frequency signals greater than a preset frequency in the body surface vibration signal.

[0071] It should be noted that the preset frequency may be 3 Hz, and the relationship between the amplitude and frequency of the Butterworth filter is as follows:

[0072]

[0073] Where: H(jη) is the output amplitude of the Butterworth filter; ηc is the cutoff frequency; Nf is the filter order. The transition band of the filter is related to its size. The larger the order, the steeper the transition band.

[0074] Step 3: performing modal decomposition on the filtered body surface vibration signal using a modal decomposition method to obtain each intrinsic mode function corresponding to the filtered body surface vibration signal;

[0075] It should be noted that the modal decomposition method may include: an integrated empirical mode decomposition method, an empirical mode decomposition method, an empirical mode decomposition method based on a time-varying filter, and an empirical mode decomposition method based on a window average.

[0076] For example, the ensemble empirical mode decomposition (EEMD) adds Gaussian white noise to the original signal to make the original signal continuous at different scales, thereby promoting mode decomposition. The decomposition process is:

[0077] (1) Add white noise to the filtered body surface vibration signal x(t) to obtain the noisy x'(t);

[0078] (2) Perform EMD empirical mode decomposition on x'(t) to obtain the corresponding intrinsic mode function IMF;

[0079] (3) Repeat (1) and (2) M times, requiring that the white noise added to the original signal is different each time, but its standard deviation is the same;

[0080] (4) The intrinsic mode functions (IMFs) of the M-th corresponding layers are superimposed and averaged to obtain the final intrinsic mode function (IMF) of the corresponding layer.

[0081] For example, performing EMD empirical mode decomposition on x'(t) includes:

[0082] (1) Determine all the maximum and minimum points on the signal x'(t), connect all the maximum and minimum points with cubic spline curves, and use these two curves as the upper and lower envelopes of x'(t). Calculate their average value curve q1(t), subtract q1(t) from x'(t), and get h1(t) = x'(t) - q1(t);

[0083] (2) Determine whether h1(t) satisfies the IMF condition. If h1(t) does not satisfy the IMF condition, use h1(t) as the original signal and repeat the above step (1) to obtain h2(t); do this k times in sequence until h1(t) satisfies the IMF condition. k (t) becomes an IMF. In this way, the first IMF is decomposed from the original signal, which is called the first-order IMF, that is, c1(t) = h k (t). Subtract c1(t) from the original signal to obtain the first-order residual signal r1(t), that is, r1(t) = x'(t) - c1(t).

[0084] (3) Take r1(t) as the new original signal and repeat steps (1) and (2) to obtain r2(t). g (t) is also screened in the same way, so that the second-order IMF, ..., the n-th-order IMF and the second-order residual signal, ..., the n-th-order residual signal are obtained in turn. When the n-th-order IMF component c n (t) or the remainder r n (t) is less than the preset value or r n When (t) becomes a monotonic function, the screening ends.

[0085] (4) Finally Among them, c g (t) represents the IMF component obtained by the g-th screening, n is the number of screenings, r n (t) is the remaining component that cannot be further decomposed.

[0086] Step 4: using wavelet analysis to determine the main frequencies of the eigenmode functions, and screening out the eigenmode functions whose main frequencies are within the preset breathing and heartbeat frequency bands;

[0087] It should be noted that determining the main frequency of each eigenmode function includes:

[0088] Analyzing the time-frequency characteristics of each of the intrinsic mode functions using a wavelet analysis method to determine the wavelet variance corresponding to each frequency of each of the intrinsic mode functions;

[0089] It should be noted that in this embodiment, complex morlet wavelet is selected as the basic wavelet function, wavelet analysis is performed on each intrinsic mode function to obtain the wavelet transform coefficient matrix corresponding to each intrinsic mode function, and the wavelet variance corresponding to each frequency in each wavelet transform coefficient matrix is ​​calculated;

[0090] The frequency corresponding to the maximum value of the wavelet variance in each eigenmode function is taken as the main frequency corresponding to each eigenmode function.

[0091] Step 5: Reconstruct the breathing and heartbeat signals based on the selected intrinsic mode functions;

[0092] In the disclosed embodiment, the reconstructing the breathing and heartbeat signals based on the screened intrinsic mode functions includes:

[0093] Adding the eigenmode functions of the filtered respiratory signals whose main frequencies are within the preset respiratory frequency band to obtain a reconstructed respiratory signal;

[0094] The reconstructed heartbeat signal is obtained by adding the eigenmode functions of the screened heartbeat signals whose main frequencies are within the preset heartbeat frequency band.

[0095] The preset respiratory frequency band range may be 0.1 Hz to 0.5 Hz, and the preset heart rate frequency band range may be 0.8 Hz to 3.0 Hz.

[0096] Step 6: Determine the life information of the detected human target based on the stochastic resonance system and the reconstructed breathing and heartbeat signals.

[0097] In the disclosed embodiment, the method of determining the life information of the detected human target based on the stochastic resonance system and the reconstructed breathing and heartbeat signals includes:

[0098] Calculating the fourth-order cumulant of the respiratory signal, and performing a fast Fourier transform on the fourth-order cumulant of the respiratory signal to extract the respiratory frequency of the human target;

[0099] The heartbeat signal is input into the stochastic resonance system, and the parameters of the stochastic resonance system are adaptively optimized using an optimization algorithm to obtain an optimized stochastic resonance result, and then the optimized stochastic resonance result is subjected to a fast Fourier transform to extract the heartbeat frequency of the human target.

[0100] It should be noted that the optimization algorithms include: particle swarm algorithm, artificial fish swarm algorithm and ant colony algorithm.

[0101] For example, the established stochastic resonance system can be a nonlinear bistable system, and its model is: Where a and b are structural parameters of the stochastic resonance system, both greater than 0, s(t) is the weak periodic signal to be measured, and η(t) is the noise signal.

[0102] The fourth-order Runge-Kutta algorithm is used to numerically solve the model, as shown in the following formula:

[0103] k1=h*(a*x(i)-b*x(i).^3+x1(i));

[0104] k2=h*(a*(x(i)+k1 / 2)-b*(x(i)+k1 / 2).^3+x1(i));

[0105] k3=h*(a*(x(i)+k2 / 2)-b*(x(i)+k2 / 2).^3+x1(i+1));

[0106] k4=h*(a*(x(i)+k3)-b*(x(i)+k3).^3+x1(i+1));

[0107] x(i+1)=x(i)+(1 / 6)*(k1+2*k2+2*k3+k4);

[0108] Where x(i) is the system output, x1(i) is the system input, and h is the numerical calculation step.

[0109] In the disclosed embodiment, the reconstructed heartbeat signal is used as the system input, the signal-to-noise ratio of the stochastic resonance system output signal is used as the fitness function, and the particle swarm algorithm is used to collaboratively optimize the three parameters of the stochastic resonance system, namely, the structural parameters a, b and the calculation step length h.

[0110] Among them, the calculation formula of the signal-to-noise ratio can be S=|x(k0)| 2 , x(k0)=maxx(i), B is optional.

[0111] Wherein, the particle swarm algorithm comprises:

[0112] Assume that in D-dimensional space, the particle group X = (x1, ..., xe, ..., xE) consists of E particles, where the position of the e-th particle is represented by a D-dimensional vector X e =(X e1 , X e2 , ..., X eD ) T , the particle velocity is represented by V e =(V e1 , V e2 , ..., V eD )T , the individual extreme value is represented by Pbest e =(P e1 , P e2 , ..., P eD ) T , the group extreme value is expressed as Nbest = (N1, N2, ..., N D ) T In this embodiment, D is 3.

[0113] Step 1: Set the number of particles E, the search range of parameters a, b and h, the maximum number of iterations Tmax, and the maximum search speed to 10% to 20% of the maximum adjustment step. The maximum adjustment step here refers to the difference between the upper limit of the set particle position range and the lower limit of the particle position range. Randomly initialize the position and speed of each particle, set the individual extreme value coordinates of each particle to its current position, and set the group extreme value to the current position of the particle with the largest individual extreme value.

[0114] Step 2: Calculate the fitness function value of each particle according to the above formula, and compare it with the current individual extreme value of the particle. If it is greater than the latter, set Pbest e is the position of the particle, and the individual extreme value is updated. If the maximum individual extreme value of all particles in the neighborhood of the particle is greater than the current Nbest, Nbest is set to the position of the particle, and the group extreme value is updated.

[0115] Step 3: Update the positions and velocities of all particles according to the formula. The updated calculation formula is: Vi(t+1)=wVi(t)+c1r1(t)[Pbest e (t)-Ve(t)]+c2r2(t)[Nbest(t)-Ve(t)], Xe(t+1)=Xe(t)+Ve(t+1), where j=1, 2, …, D; t is the current iteration number; r1, r2 are random numbers uniformly distributed between [0, 1]; c1, c2 are learning factors, usually c1=c2=2; w is the inertia weight, that is, the coefficient for maintaining the original speed, and its calculation formula is In the formula, w max is the upper limit of inertia weight, w min is the lower limit of inertia weight, T max is the maximum number of iterations.

[0116] Step 4: Determine whether the current number of iterations reaches the maximum number of iterations Tmax. If the condition is met, stop the iteration and output the optimal parameters a, b and h found by the particle swarm algorithm. Otherwise, go to step 2.

[0117] It should be noted that after obtaining the system output result when the stochastic resonance system parameters are optimal, it is also necessary to perform fast Fourier transform on the system output result when the stochastic resonance system parameters are optimal to extract the peak frequency, and then obtain the heart rate of the human target after scale transformation.

[0118] Specifically, the specific process of the scale transformation is: fh = fmax × h × fs

[0119] Where fh is the heartbeat frequency, fmax is the peak frequency, and fs is the slow time sampling frequency of the radar.

[0120] In summary, the present application proposes a life detection method based on radar signals, which uses an ultra-wideband radar to detect the area to be detected, obtains the radar echo matrix corresponding to the area to be detected, and filters out the static clutter in the radar echo matrix; extracts the body surface vibration signal of the human target in the area to be detected according to the radar echo matrix after filtering out the clutter; uses the modal decomposition method to perform modal decomposition on the body surface vibration signal to obtain each intrinsic mode function corresponding to the body surface vibration signal; uses the wavelet analysis method to determine the main frequency of each intrinsic mode function, and selects the intrinsic mode function whose main frequency is within the preset breathing and heartbeat frequency band; reconstructs the breathing and heartbeat signals based on the selected intrinsic mode functions; and determines the life information of the detected human target based on the stochastic resonance system and the reconstructed breathing and heartbeat signals. The technical method provided by the present invention has strong anti-interference ability and can accurately and quickly detect the breathing frequency and heartbeat frequency of the human target in the environment.

[0121] Example 2

[0122] Further, Figure 2 A life detection system based on radar signals is provided according to an embodiment of the present application, such as Figure 2 As shown, the life detection system comprises:

[0123] The first filtering module 100 is used to detect the area to be detected by using an ultra-wideband radar, obtain a radar echo matrix corresponding to the area to be detected, and filter out static clutter in the radar echo matrix;

[0124] An extraction module 200 is used to extract the body surface vibration signal of the human target in the area to be detected according to the radar echo matrix after filtering out the clutter;

[0125] A modal decomposition module 300 is used to perform modal decomposition on the body surface vibration signal by using a modal decomposition method to obtain each intrinsic mode function corresponding to the body surface vibration signal;

[0126] A screening module 400 is used to determine the main frequency of each eigenmode function by wavelet analysis, and screen out the eigenmode functions whose main frequency is within a preset breathing and heartbeat frequency band;

[0127] A reconstruction module 500, for reconstructing the breathing and heartbeat signals based on the screened intrinsic mode functions;

[0128] The determination module 600 is used to determine the life information of the detected human target based on the stochastic resonance system and the reconstructed breathing and heartbeat signals.

[0129] In the disclosed embodiment, extracting the body surface vibration signal of the human target in the area to be detected according to the radar echo matrix after filtering out the clutter includes:

[0130] Calculating the state parameters of the slow time series corresponding to each range gate in the radar echo matrix after filtering out clutter;

[0131] The range gate corresponding to the state parameter that meets the preset conditions is used as the position of the human target in the area to be detected;

[0132] According to the location of the human target, a slow time series corresponding to the range gate of the location of the human target is extracted from the radar echo matrix after clutter filtering, and the slow time series is used as the body surface vibration signal of each human target.

[0133] It should be noted that the state parameters include: permutation entropy value or approximate entropy value or sample entropy value or fuzzy entropy value or variance or standard deviation or kurtosis or energy;

[0134] It should be noted that the method of using the range gate corresponding to the state parameter that meets the preset condition as the position of the human target in the area to be detected includes:

[0135] The distance gate corresponding to the minimum value of the permutation entropy value is used as the position of the human target in the area to be detected, or,

[0136] The range gate corresponding to the minimum approximate entropy value is used as the range gate where the human target in the to-be-detected area is located, and then the position of the human target is determined based on the range gate, or,

[0137] The range gate corresponding to the minimum sample entropy value is used as the range gate where the human target is located in the area to be detected, and then the position of the human target is determined based on the range gate, or,

[0138] The range gate corresponding to the minimum fuzzy entropy value is used as the range gate where the human target in the to-be-detected area is located, and then the position of the human target is determined based on the range gate, or,

[0139] The range gate corresponding to the maximum variance is used as the position of the human target in the area to be detected, or,

[0140] The range gate corresponding to the maximum value of the standard deviation is used as the position of the human target in the area to be detected, or,

[0141] The range gate corresponding to the maximum kurtosis value is used as the position of the human target in the area to be detected, or,

[0142] The range gate corresponding to the maximum energy value is used as the position of the human target in the area to be detected.

[0143] In the embodiment of the present disclosure, before performing modal decomposition on the body surface vibration signal using the modal decomposition method, the method further includes: performing Butterworth filtering on the body surface vibration signal;

[0144] Wherein, performing Butterworth filtering on the body surface vibration signal includes:

[0145] A Butterworth filter is used to filter out high-frequency signals greater than a preset frequency in the body surface vibration signal.

[0146] It should be noted that the modal decomposition method includes:

[0147] Integrated empirical mode decomposition method, empirical mode decomposition method, empirical mode decomposition method based on time-varying filter and window average empirical mode decomposition method.

[0148] Specifically, the main frequencies of the eigenmode functions are determined by wavelet analysis, including:

[0149] Analyzing the time-frequency characteristics of each of the intrinsic mode functions using a wavelet analysis method to determine the wavelet variance corresponding to each frequency of each of the intrinsic mode functions;

[0150] The frequency corresponding to the maximum value of the wavelet variance in each eigenmode function is taken as the main frequency corresponding to each eigenmode function.

[0151] In the embodiment of the present disclosure, the reconstruction module 500 is specifically used for:

[0152] Adding the eigenmode functions of the filtered respiratory signals whose main frequencies are within the preset respiratory frequency band to obtain a reconstructed respiratory signal;

[0153] The reconstructed heartbeat signal is obtained by adding the eigenmode functions of the screened heartbeat signals whose main frequencies are within the preset heartbeat frequency band.

[0154] In the embodiment of the present disclosure, the determination module 600 is specifically used to:

[0155] Calculating the fourth-order cumulant of the respiratory signal, and performing a fast Fourier transform on the fourth-order cumulant of the respiratory signal to extract the respiratory frequency of the human target;

[0156] The heartbeat signal is input into the stochastic resonance system, and the parameters of the stochastic resonance system are adaptively optimized using an optimization algorithm to obtain an optimized stochastic resonance result, and then the optimized stochastic resonance result is subjected to a fast Fourier transform to extract the heartbeat frequency of the human target.

[0157] Furthermore, the optimization algorithm includes: particle swarm algorithm, artificial fish swarm algorithm and ant colony algorithm.

[0158] In summary, the life detection system based on radar signals proposed in this application has strong anti-interference ability and can accurately and quickly detect the breathing rate and heart rate of human targets in the environment.

[0159] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0160] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, fragment or portion of code comprising one or more executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present application belong.

[0161] Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limitations on the present application. Ordinary technicians in this field can change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. A life detection method based on radar signals, characterized in that: The method comprises: Using an ultra-wideband radar to detect the area to be detected, obtaining a radar echo matrix corresponding to the area to be detected, and filtering out static clutter in the radar echo matrix; Extract the body surface vibration signal of the human target in the area to be detected according to the radar echo matrix after filtering out the clutter; Performing modal decomposition on the body surface vibration signal by using a modal decomposition method to obtain each eigenmode function corresponding to the body surface vibration signal; Determine the main frequency of each eigenmode function by wavelet analysis, and select the eigenmode function whose main frequency is within the preset breathing and heartbeat frequency band; Reconstruct the breathing and heartbeat signals based on the screened intrinsic mode functions; Determining the life information of the detected human target based on the stochastic resonance system and the reconstructed breathing and heartbeat signals; The method of determining the life information of the detected human target based on the stochastic resonance system and the reconstructed breathing and heartbeat signals includes: Calculating the fourth-order cumulant of the respiratory signal, and performing a fast Fourier transform on the fourth-order cumulant of the respiratory signal to extract the respiratory frequency of the human target; The heartbeat signal is input into the stochastic resonance system, and the parameters of the stochastic resonance system are adaptively optimized using an optimization algorithm to obtain an optimized stochastic resonance result, and then the optimized stochastic resonance result is subjected to a fast Fourier transform to extract the heartbeat frequency of the human target.

2. The life detection method according to claim 1, characterized in that: The method of extracting the body surface vibration signal of the human target in the area to be detected according to the radar echo matrix after filtering out the clutter includes: Calculating the state parameters of the slow time series corresponding to each range gate in the radar echo matrix after filtering out clutter; Determine the location of the human target in the area to be detected by using the range gate corresponding to the state parameter that meets the preset conditions; According to the location of the human target, a slow time series corresponding to the range gate of the location of the human target is extracted from the radar echo matrix after clutter filtering, and the slow time series is used as the body surface vibration signal of each human target.

3. The life detection method according to claim 2, characterized in that: The state parameters include: permutation entropy value or approximate entropy value or sample entropy value or fuzzy entropy value or variance or standard deviation or kurtosis or energy; The method of determining the position of the human target in the area to be detected by using a range gate corresponding to a state parameter that meets a preset condition includes: The range gate corresponding to the minimum value of the permutation entropy value is used as the range gate where the human target is located in the area to be detected, and then the position of the human target is determined based on the range gate, or, The range gate corresponding to the minimum approximate entropy value is used as the range gate where the human target is located in the area to be detected, and then the position of the human target is determined based on the range gate, or, The range gate corresponding to the minimum sample entropy value is used as the range gate where the human target is located in the area to be detected, and then the position of the human target is determined based on the range gate, or, The range gate corresponding to the minimum fuzzy entropy value is used as the range gate where the human target in the to-be-detected area is located, and then the position of the human target is determined based on the range gate, or, The range gate corresponding to the maximum variance value is used as the range gate where the human target is located in the area to be detected, and then the position of the human target is determined based on the range gate, or, The range gate corresponding to the maximum value of the standard deviation is used as the range gate where the human target is located in the area to be detected, and then the position of the human target is determined based on the range gate, or The range gate corresponding to the maximum kurtosis value is used as the range gate where the human target is located in the area to be detected, and then the position of the human target is determined based on the range gate, or, The range gate corresponding to the maximum energy value is used as the range gate where the human target is located in the area to be detected, and then the position of the human target is determined based on the range gate.

4. The life detection method according to claim 1, characterized in that: Before performing modal decomposition on the body surface vibration signal using the modal decomposition method, the method further includes: performing Butterworth filtering on the body surface vibration signal; Wherein, performing Butterworth filtering on the body surface vibration signal comprises: A Butterworth filter is used to filter out high-frequency signals greater than a preset frequency in the body surface vibration signal.

5. The life detection method according to claim 1, characterized in that: The modal decomposition method comprises: Integrated empirical mode decomposition method, empirical mode decomposition method, empirical mode decomposition method based on time-varying filter and window average empirical mode decomposition method.

6. The life detection method according to claim 5, characterized in that: Determining the main frequency of each eigenmode function includes: Analyzing the time-frequency characteristics of each of the intrinsic mode functions using a wavelet analysis method to determine the wavelet variance corresponding to each frequency of each of the intrinsic mode functions; The frequency corresponding to the maximum value of the wavelet variance in each eigenmode function is taken as the main frequency corresponding to each eigenmode function.

7. The life detection method according to claim 6, characterized in that: The reconstructing the breathing and heartbeat signals based on the screened intrinsic mode functions comprises: Adding the eigenmode functions of the filtered respiratory signals whose main frequencies are within the preset respiratory frequency band to obtain a reconstructed respiratory signal; The reconstructed heartbeat signal is obtained by adding the eigenmode functions of the screened heartbeat signals whose main frequencies are within the preset heartbeat frequency band.

8. The life detection method according to claim 1, characterized in that: The optimization algorithms include: particle swarm algorithm, artificial fish swarm algorithm and ant colony algorithm.

9. A life detection system based on radar signals, characterized in that: The life detection system comprises: A first filtering module is used to detect the area to be detected by using an ultra-wideband radar, obtain a radar echo matrix corresponding to the area to be detected, and filter out static clutter in the radar echo matrix; An extraction module is used to extract the body surface vibration signal of the human target in the detection area according to the radar echo matrix after filtering out the clutter; A modal decomposition module, used for performing modal decomposition on the body surface vibration signal by using a modal decomposition method to obtain each intrinsic mode function corresponding to the body surface vibration signal; A screening module, used for determining the main frequency of each eigenmode function by wavelet analysis, and screening out the eigenmode functions whose main frequency is within the preset breathing and heartbeat frequency band; A reconstruction module, used for reconstructing the breathing and heartbeat signals based on the screened intrinsic mode functions; A determination module, used for determining the life information of the detected human target based on the stochastic resonance system and the reconstructed breathing and heartbeat signals; The determination module is further used to calculate the fourth-order cumulant of the respiratory signal and perform a fast Fourier transform on the fourth-order cumulant of the respiratory signal to extract the respiratory frequency of the human target; The heartbeat signal is input into the stochastic resonance system, and the parameters of the stochastic resonance system are adaptively optimized using an optimization algorithm to obtain an optimized stochastic resonance result, and then the optimized stochastic resonance result is subjected to a fast Fourier transform to extract the heartbeat frequency of the human target.

Images