Vital sign signal detection method of radar communication integrated signal based on OFDM (Orthogonal Frequency Division Multiplexing)

Abstract

The invention discloses an OFDM (Orthogonal Frequency Division Multiplexing)-based vital sign signal detection method for radar communication integrated signals. Belongs to the technical field of radar communication integration. According to the method, the micro-motion feature information of the chest of the human body carried in the echoes is utilized more comprehensively, sensitive subcarriers are screened by calculating the slow time variance of the subcarriers in the aspect of accurate signal extraction, signals corresponding to weak motion of the chest are accurately captured, the separation precision of breathing and heartbeat signals is greatly improved, and the accuracy of accurate signal extraction is improved. Wavelet transform soft threshold denoising is combined to suppress high-frequency noise and reserve effective low frequency, and VMD decomposes a non-stationary signal into a modal function, so that the signal quality is optimized in an omnibearing manner, and vital sign monitoring is more accurate. The cross-field application advantage is remarkable, the safety of personnel in the vehicle is guaranteed in unmanned driving, the physical abnormity of dangerous personnel is remotely informed in intelligent security and protection, remote accurate diagnosis of medical care is assisted during remote medical treatment, and the multi-field requirements are practically met.

Description

Technical Field

[0001] The present invention belongs to the technical field of integrated radar and communication, and particularly relates to a method for detecting vital sign signals of integrated radar and communication signals based on OFDM. Background Art

[0002] With the development of modern medical health monitoring technology, the real-time monitoring of vital sign signals has become an important means for health management and disease diagnosis. Existing technologies such as electrocardiogram (ECG), pulse wave and other vital sign monitoring methods, although with high precision, usually rely on dedicated sensors and attachment devices for detecting vital sign signals, which bring many limitations.

[0003] In recent years, the application of integrated radar and communication technology in vital sign monitoring has gradually attracted attention. By detecting the minute movement changes of the human chest through radar signals, non-contact vital sign monitoring can be achieved. Although the commonly used LFM signals and stepped frequency (SF) signals in current research have broadband advantages, they have obvious defects: LFM signals have serious time-frequency coupling, which brings difficulties to the analysis and extraction of micro-Doppler and is prone to range-Doppler ambiguity; when SF signals use pulse synthesis bandwidth, it will cause range image shift and even waveform distortion of the echo signal under target movement, and the construction of its micro-motion modulation model is more complex.

[0004] In contrast, OFDM signals exhibit significant technical advantages. As an efficient multi-carrier transmission method, OFDM signals are large-bandwidth signals synthesized by multiple sub-band signals of different frequencies. Their thumbtack-shaped ambiguity function effectively avoids the time-frequency coupling problem, has both broadband high-resolution ability and narrowband micro-Doppler characteristics, is more conducive to the detection and extraction of micro-moving targets, and the sensitivity of each sub-carrier to phase changes can accurately capture millimeter-level chest fluctuations. OFDM signals can provide higher resolution and better anti-interference ability in radar applications, can effectively resist multi-path interference, and have advantages for accurately detecting minute vital sign changes of the human body. However, in a complex environment, there are a large number of interference clutter, especially stationary clutter, environmental noise, and noise caused by human movement. The vital sign signals are very weak and are often submerged in the noise. These noises will affect the accuracy of radar signals, and even cause the loss or misjudgment of vital sign signals, making target positioning and the extraction of vital signs more complex and difficult. Summary of the Invention

[0005] The purpose of the present invention is to overcome the defects of the above-mentioned existing technologies, and provide a method for detecting vital sign signals of integrated radar and communication signals based on OFDM, which can accurately estimate the time delay and Doppler frequency shift of the target in a noise environment, especially in the case of low signal-to-noise ratio, can effectively remove noise interference, and improve the estimation accuracy of respiratory and heartbeat parameters.

[0006] The technical problem proposed by the present invention is solved as follows:

[0007] A method for detecting vital sign signals of an integrated radar and communication signal based on OFDM includes the following steps:

[0008] Step 1: Mix the transmitted signal carrying communication information with the transmitted carrier frequency to obtain the up-converted integrated radar and communication transmitted signal S TX (t), where t is time. After detecting the human chest part, the integrated radar and communication echo signal S RX (t) carrying vital sign information is obtained after down-conversion processing;

[0009] Step 2: Convert the integrated radar and communication echo signal S RX (t) into N parallel data streams, then perform the operation of removing the cyclic prefix, and then perform the N-point FFT transform to obtain the frequency-domain information Y RX (k) of the N sub-carrier echo signals. N is a positive integer; perform the N-point FFT transform on the transmitted signal S TX (t) to obtain the frequency-domain information Y TX (k) of the N sub-carrier transmitted signals; Divide Y RX (k) by Y TX (k) to obtain the frequency-domain target correlation signal H(k) containing vital sign signals, where k is the frequency;

[0010] Step 3: Perform the N-point IFFT transform on the frequency-domain target correlation signal H(k) to obtain the time-domain target correlation signal h(t). Extract the phase information of the time-domain target correlation signal h(t) on each sub-carrier, and then calculate its average phasor in the slow-time dimension as the background noise phase component; For each sub-carrier, subtract the average phasor in the slow-time dimension from the phase information of the time-domain target correlation signal h(t) to obtain the phase information of the time-domain target correlation signal h(t) on each sub-carrier after removing the background noise; Based on the phase information after removing the background noise, use complex exponential operations for signal reconstruction to generate the reconstructed signal β(t);

[0011] Step 4: Extract the phase φ(t) of the reconstructed signal ββ(t), and calculate the phase fluctuation variance of φ(t) in the slow-time dimension of each sub-carrier; Screen the sub-carriers with phase fluctuation variance greater than the set threshold, and extract the signal χ(t) of the reconstructed signal ββ(t) on the screened sub-carriers as the sensitive signal;

[0012] Step 5: Decompose the sensitive signal χ(t) using wavelet basis functions to obtain a wavelet decomposition coefficient vector; adopt the soft threshold method to remove the noise components in the wavelet decomposition coefficient vector to obtain a denoised wavelet decomposition coefficient vector; perform iterative reconstruction based on the denoised wavelet decomposition coefficient vector and the wavelet basis functions to obtain a reconstructed sensitive signal f denoised (t);

[0013] Step 6: Extract the phase of the reconstructed sensitive signal f denoised (t) and perform phase unwrapping processing to obtain a signal g(t) with unwrapped phase. Perform adaptive decomposition on the signal g(t) to obtain z narrowband modes u z (t) with different frequencies, where Z is a positive integer;

[0014] Step 7: Perform FFT transformation on the decomposed multiple narrowband modes to obtain spectra, extract the peaks from the spectra, and screen the respiratory signal and the heartbeat signal based on the frequency points corresponding to the spectral peaks.

[0015] Furthermore, in Step 1, the radar communication integrated echo signal S RX (t) is expressed as:

[0016]

[0017] where 0 ≤ n ≤ N - 1, N is the number of subcarriers of the transmit carrier frequency, 0 ≤ m ≤ M - 1, M is the total number of OFDM signal symbols, A(n, m) represents the communication information modulated on the nth subcarrier in the mth symbol, j is the imaginary part symbol, Δf is the subcarrier spacing, τ is the time delay, f c is the carrier frequency, and noise is the noise;

[0018] The time delay τ is expressed as:

[0019] τ = 2R1 / c0

[0020] where c0 is the speed of light, R is the distance between the human chest cavity and the radar, V0 is the human radial motion speed, B b is the chest motion amplitude generated by breathing, ω b is the breathing frequency, H h is the chest motion amplitude generated by the heartbeat, ω h is the heartbeat frequency, is the phase difference generated by the heartbeat and breathing.

[0021] Furthermore, in Step 3, for each subcarrier, the phase information of the time-domain target correlation signal h(t) minus the average phasor in the slow-time dimension is expressed as:

[0022] Y pvmc，n (t) = Yn (t) - Y mean，n

[0023] Wherein, Y pvmc，n (t) represents the phase information of eliminating background noise of the time-domain target-related signal on the nth subcarrier, and Y n (t) represents the phase information of the time-domain target-related signal on the nth subcarrier, and Y mean，n represents the background noise phase component corresponding to the nth subcarrier.

[0024] Furthermore, in step 4, extract the phase φ(t) of the reconstructed signal β(t), and calculate the phase fluctuation variance of φ(t) in the slow time dimension of each subcarrier:

[0025]

[0026] Wherein, is the phase fluctuation variance of the nth subcarrier, and φ m，n (m) is the instantaneous phase of the nth subcarrier at the mth symbol, is the mean value of the phase of the nth subcarrier.

[0027] Furthermore, the specific process of step 5 is as follows:

[0028] Use the wavelet basis function to perform L-level decomposition on the sensitive signal χ(t), which is expressed as:

[0029]

[0030] Wherein, l is the scale parameter, 1 ≤ l ≤ L, and L is the decomposition level; d l is the detail coefficient of the lth layer, ψ l (t) is the wavelet basis function of the lth layer, a L is the approximation coefficient, and κ L (t) is the scaling function;

[0031] Extract the approximation coefficients and detail coefficients of all layers to construct the wavelet decomposition coefficient vector ζ = [a L , d L , d L-1 , d L-2 , d L-3 , …, d1]; adopt the soft threshold method to remove the noise components in the wavelet decomposition coefficient vector, which is expressed as:

[0032] ζ denoised = sign(v)·max(|ζ| - λ, 0)

[0033] Wherein, ζ denoisedDenote the wavelet decomposition coefficient vector after denoising as ζ, λ as the set threshold, max as taking the maximum value, | as taking the absolute value of the element, and sign(·) as the sign function, which is defined as:

[0034]

[0035] The wavelet decomposition coefficient vector ζ after denoising denoised is expressed as:

[0036] ζ denoised = [a denoised,L , d denoised，k , d denoised,L-1 , d denoised,L-2 , d denoised,L-3 , …, d denoised,1

[0037] According to the wavelet decomposition coefficient vector ζ after denoising denoised and the wavelet basis function, perform L-layer iterative reconstruction to obtain the sensitive signal f denoised (t) after removing noise:

[0038]

[0039] Furthermore, in step 6, the process of adaptively decomposing the signal g(t) to obtain z narrowband modes u z (t) with different frequencies is expressed as:

[0040]

[0041] where 1 ≤ z ≤ Z, Z is the number of narrowband modes, u z (t) is the z-th narrowband mode, ω z represents the center frequency of the z-th mode, δ(t) is the Dirac function, represents the partial derivative with respect to time t, and || ||2 represents calculating the two-norm.

[0042] Furthermore, in step 7, filter 0.1 Hz to 0.5 Hz as the frequency estimate of the respiratory signal, and 0.8 Hz to 2 Hz as the frequency estimate of the heartbeat signal.

[0043] The beneficial effects of the present invention are:

[0044] ​The method of the present invention makes more comprehensive use of the micro-motion feature information of the human chest carried in the echo. In terms of precise signal extraction, sensitive subcarriers are screened by calculating the slow-time variance of the subcarriers to accurately capture the signals corresponding to the weak chest movements, greatly improving the separation accuracy of the respiration and heartbeat signals. Combining wavelet transform soft threshold denoising to suppress high-frequency noise and retain effective low-frequency components, and VMD to decompose non-stationary signals into mode functions, comprehensively optimizes the signal quality and makes the vital sign monitoring more accurate. Its cross-domain application advantages are significant, ensuring the safety of vehicle occupants in driverless driving, remotely detecting the physical abnormalities of dangerous personnel in intelligent security, and assisting medical staff in remote and accurate diagnosis in telemedicine, effectively meeting the needs of multiple fields. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 is a schematic flowchart of the method of the present invention;

[0046] Figure 2 is a flowchart of the subcarrier selection method based on slow-time variance in the method of the present invention;

[0047] Figure 3 is a graph showing the change of the phase of the subcarriers selected in the method of the embodiment over time;

[0048] Figure 4 is a phase diagram of the reconstructed echo signal after selecting subcarriers in the method of the embodiment;

[0049] Figure 5 is a flowchart of the wavelet decomposition method based on soft threshold processing in the method of the present invention;

[0050] Figure 6 is a comparison diagram before and after denoising the received signal in the method of the embodiment;

[0051] Figure 7 is a decomposition result diagram of the denoised signal in the method of the embodiment;

[0052] Figure 8 is a parameter estimation result diagram of the vital sign signals in the method of the embodiment, where (a) is the extracted respiration signal and its spectrum, and (b) is the extracted heartbeat signal and its spectrum. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0053] The present invention will be further described below with reference to the drawings and embodiments.

[0054] This embodiment provides a method for detecting vital sign signals of an integrated radar communication signal based on OFDM. During the process of detecting human vital sign signals using an OFDM signal radar, the echo at the radar receiving end is processed to extract the phase information of the received signal, and then the mean value in the slow-time dimension is calculated, representing the phase component of the background noise. Next, the phase at each moment is subtracted from the mean phase to eliminate the influence of the static background and retain the dynamic vital sign information. Then, to improve the signal quality, the slow-time variance of each subcarrier is calculated, and the variance is used as a measure of the subcarrier sensitivity. The slow-time statistical characteristics are introduced into the radar subcarrier selection. By calculating the slow-time phase fluctuation variance of each subcarrier, its sensitivity to the weak movement of the chest cavity is quantified, and the subcarriers with high sensitivity are selected by sorting to screen out the subcarriers most sensitive to the target movement. The selected subcarriers are used for further signal analysis to more accurately extract vital sign information. In addition, wavelet transform is used to perform multi-level denoising on the signal. The high-frequency noise is suppressed by the soft threshold method, and the low-frequency components are retained. At the same time, combined with the variational mode decomposition technique, after the denoised signal is unwrapped, the unwrapped phase is decomposed by VMD, and the signal is decomposed into multiple intrinsic mode functions. The non-stationary signal is effectively processed by the non-linear optimization decomposition method, and the modes are screened to obtain the respiration and heartbeat signals, and the key signal components of respiration and heartbeat are accurately extracted.

[0055] The method for detecting vital sign signals of an integrated radar communication signal based on OFDM is a method that can effectively denoise in a complex environment, especially at low signal-to-noise ratios, accurately extract phase signals, improve the separation accuracy of respiration and heartbeat signals and the stability of vital sign monitoring accuracy, and improve the estimation accuracy of respiration and heartbeat parameters, providing support for real-time health monitoring and efficient and accurate solutions for fields such as unmanned driving, intelligent security, and remote medical treatment.

[0056] The flow schematic diagram of the method in this embodiment is as Figure 1 shown, and specifically includes the following steps:

[0057] Step 1: Mix the transmitted signal s(t) carrying communication information with the transmitted carrier frequency exp{j2πf c t} to obtain the up-converted integrated radar communication transmitted signal S Tx (t). After detecting the chest part of the human body, the integrated radar communication echo signal S RX (t) carrying vital sign information is obtained after down-conversion processing.

[0058] The integrated radar communication echo signal S RX (t) is expressed as:

[0059]

[0060] where t is time, 0 ≤ n ≤ N - 1, N is the number of sub - carriers of the transmitted carrier frequency, 0 ≤ m ≤ M - 1, M is the total number of OFDM signal symbols, A(n, m) represents the communication information modulated on the n - th sub - carrier in the m - th symbol, j is the imaginary part symbol, Δf is the sub - carrier spacing, τ is the time delay, f c is the carrier frequency, and noise is the noise;

[0061] The time delay τ is expressed as:

[0062] τ = 2R1 / c0

[0063] where c0 is the speed of light, R is the distance between the human chest and the radar, V0 is the radial motion speed of the human body, B b is the amplitude of chest motion generated by breathing, ω b is the breathing frequency, H h is the amplitude of chest motion generated by heartbeat, ω h is the heartbeat frequency, is the phase difference generated by heartbeat and breathing. Here, it is assumed that the breathing and heartbeat signals perform periodic motion, and the small changes in vital signs are mainly reflected in the frequency shift, representing activities such as human motion, heartbeat, or breathing.

[0064] Step 2: After converting the integrated radar communication echo signal S RX (t) into N - channel parallel data streams, perform the operation of removing the cyclic prefix, and then perform the N - point FFT transform to obtain the frequency - domain information Y RX (k) of the N sub - carrier echo signals; perform the N - point FFT transform on the transmitted signal S Tx (t) to obtain the frequency - domain information Y TX (k) of the N sub - carrier transmitted signals; because the existence of communication information will affect the detection effect of vital sign signals, the influence of communication information needs to be eliminated, which can be obtained by dividing the received signal by the transmitted signal, and the frequency - domain target - related signal H(k) containing vital sign signals is obtained:

[0065]

[0066] where k is the frequency.

[0067] Step 3: The frequency - domain target - related signal H(k) extracted in Step 2 has eliminated the influence of communication modulation, but still contains the echo interference of other stationary objects in the complex environment and other noise interferences; perform the N - point IFFT transform on the signal H(k) to obtain the time - domain target - related signal h(t), extract the phase information of the signal h(t) on each sub - carrier, and then calculate its average phasor in the slow - time dimension as the background noise phase component.

[0068] For each sub - carrier, the phase information of the time - domain target - related signal h(t) is subtracted by the average phasor in the slow - time dimension to eliminate the influence of the static background and retain the dynamic vital - sign information, which is expressed as:

[0069] Y pvmc，n (t)=Y n (t)-Y mean，n

[0070] where, Y pvmc，n (t) represents the phase information of the time - domain target - related signal after removing the background noise on the nth sub - carrier, Y n (t) represents the phase information of the time - domain target - related signal on the nth sub - carrier, and Y mean，n represents the background - noise phase component corresponding to the nth sub - carrier.

[0071] Use complex - exponential operation to reconstruct the above - obtained dynamic vital - sign signal Y pvmc，n (t) to generate a reconstructed signal β(t), which is the target signal after removing the static interference and highlighting the dynamic target signal after removing the background noise, facilitating subsequent vital - sign analysis and extraction.

[0072] Step 4: Based on the sensitivity of the vital - sign signal to each sub - carrier, screen out the sub - carriers that are most sensitive to small movements, reducing the computational complexity and improving the detection accuracy.

[0073] Extract the phase φ(t) of the reconstructed signal β(t), and calculate the phase - fluctuation variance of φ(t) in the slow - time dimension of each sub - carrier:

[0074]

[0075] where, is the phase - fluctuation variance of the nth sub - carrier, φ m，n (m) is the instantaneous phase of the nth sub - carrier at the mth symbol, is the average value of the phase of the nth sub - carrier.

[0076] The magnitude of the variance reflects the fluctuation of the sub - carrier phase in the slow - time. Sub - carriers with a large variance are more sensitive to small phase changes caused by vital signs. Then, by setting a threshold or sorting, select sub - carriers with a larger variance, as Figure 2 shown.

[0077] Denote the set of sub - carrier serial numbers κ selected that are most sensitive to small movements as:

[0078]

[0079] Extract the signal χ(t) of the reconstructed signal β(t) on the selected subcarriers as the signal most sensitive to minute motion.

[0080] The phase change diagram of the selected subcarriers over time is as Figure 3 shown. It can be seen that the phase of the selected subcarriers is relatively sensitive to the weak thoracic signals, with obvious trajectory changes. However, it can still be seen that there is some noise interference. The phase diagram of the echo signal reconstructed after selecting the subcarriers is as Figure 4 shown.

[0081] Step 5: For the sensitive signal χ(t) selected in Step 4, perform L-level decomposition on the signal using the wavelet basis function sym2, which is expressed as:

[0082]

[0083] where l is the scale parameter, 1 ≤ l ≤ L, and L is the decomposition level; d l is the detail coefficient of the l-th layer, ψ l (t) is the wavelet basis function of the l-th layer, a L is the approximation coefficient, and κ L (t) is the scaling function.

[0084] Solving the above formula gives the wavelet decomposition coefficient vector ζ = [a L , d L , d L-1 , d L-2 , d L-3 , …, d1];

[0085] Then, remove the noise components in the wavelet decomposition coefficient vector ζ through the soft thresholding method. The soft thresholding method means that the wavelet coefficients with amplitudes less than the set threshold λ are directly set to 0, while the wavelet coefficients with amplitudes greater than the set threshold are shrunk towards 0 by a certain value (the shrinkage amount is the threshold λ); after such processing, most of the noise interference can be removed, and the wavelet coefficient part more likely corresponding to the original signal is retained, thus achieving the preliminary denoising effect of the signal and obtaining the denoised coefficient vector ζ denoised . The specific formula is as follows:

[0086] ζ denoised = sign(ζ)·max(|ζ| - λ, 0)

[0087] where ζ denoised represents the denoised coefficient vector, λ is the set threshold, max represents taking the maximum value, || represents taking the absolute value of the elements, and sign(·) is the sign function, which is defined as:

[0088]

[0089] The coefficient vector ζ after denoising denoised is expressed as:

[0090] ζ denoised = [a denoised,L , d denoised,L , d denoised,L-1 , d denoised，L-2 , d denoised,L-3 , …, d denoised,1

[0091] Based on the coefficient vector ζ after denoising denoised and the wavelet basis function sym2, perform L-layer iterative reconstruction:

[0092]

[0093] where f denoised (t) represents the sensitive signal after removing noise.

[0094] Through the above steps, the reconstructed sensitive signal f denoised (t) after denoising is obtained, and the reconstructed sensitive signal still retains the vital sign frequency band.

[0095] The flow chart of the wavelet decomposition method based on soft threshold processing is as shown in Figure 5 . The comparison chart before and after denoising the received signal is as shown in Figure 6 . It can be seen that the noise is significantly suppressed.

[0096] Step 6: Extract the phase of the reconstructed sensitive signal f denoised (t) and perform phase unwrapping processing to obtain the signal g(t) after phase unwrapping. Perform adaptive decomposition on the signal g(t) to obtain z narrowband modes u z (t) with different frequencies, and then adaptively separate the breathing and heartbeat components from the denoised signal:

[0097]

[0098] where 1 ≤ z ≤ Z, Z is the number of narrowband modes, and u z (t) is the z-th narrowband mode, ω z represents the center frequency of the z-th mode, δ(t) is the Dirac function, represents the partial derivative with respect to time t, and || ||2 represents the calculation of the second norm. The decomposition result diagram of the denoised signal is as shown in Figure 7 . It can be seen the variation of each mode.

[0099] Step 7: Perform FFT transformation on the decomposed multiple narrowband modes to obtain the spectrum, extract the peaks of the spectrum, and select the breathing signal and heartbeat signal according to the frequency points corresponding to the spectrum peaks. ​

[0100] Specifically, the frequency estimation of the respiration signal is screened in the range of 0.1 Hz to 0.5 Hz, and the frequency estimation of the heartbeat signal is screened in the range of 0.8 Hz to 2 Hz. The parameter estimation result diagram of the vital sign signal is as Figure 8 shown, where (a) is the extracted respiration signal and its spectrum, and (b) is the extracted heartbeat signal and its spectrum.

Claims

1. A vital sign signal detection method based on OFDM radar communication integrated signal, characterized in that: The following steps are involved: Step 1: Mix the transmission signal carrying communication information with the transmission carrier frequency to obtain the radar communication integrated transmission signal S after up-conversion processing TX (t), t is the time. After detecting the chest part of the human body, the radar communication integrated echo signal S carrying vital signs information is obtained after down-conversion processing. RX (t); Step 2: Integrate the radar communication echo signal S RX (t) After being converted into N parallel data streams, the cyclic prefix is ​​removed, and then an N-point FFT transformation is performed to obtain the frequency domain information Y of the N subcarrier echo signals. RX (k), N is a positive integer; for the transmitted signal S TX (t) Perform N-point FFT transformation to obtain the frequency domain information Y of the N subcarrier transmission signals TX (k); Y RX (k) and Y TX (k) is divided to obtain the frequency domain target related signal H(k) containing the vital sign signal, where k is the frequency; Step 3: Perform an N-point IFFT transform on the frequency domain target correlation signal H(k) to obtain the time domain target correlation signal h(t), extract the phase information of the time domain target correlation signal h(t) on each subcarrier, and then calculate its average phasor in the slow time dimension as the background noise phase component; for each subcarrier, subtract the average phasor in the slow time dimension from the phase information of the time domain target correlation signal h(t) to obtain the phase information of the time domain target correlation signal h(t) with the background noise eliminated on each subcarrier; based on the phase information with the background noise eliminated, use the complex exponential operation to reconstruct the signal to generate the reconstructed signal β(t); Step 4: Extract the phase φ(t) of the reconstructed signal β(t), and calculate the phase fluctuation variance of φ(t) in the slow time dimension of each subcarrier; select the subcarriers whose phase fluctuation variance is greater than the set threshold, and extract the signal χ(t) of the reconstructed signal β(t) on the selected subcarrier as a sensitive signal; Step 5: Use the wavelet basis function to decompose the sensitive signal χ(t) to obtain the wavelet decomposition coefficient vector; use the soft threshold method to remove the noise component in the wavelet decomposition coefficient vector to obtain the denoised wavelet decomposition coefficient vector; iteratively reconstruct based on the denoised wavelet decomposition coefficient vector and the wavelet basis function to obtain the denoised reconstructed sensitive signal f denoised (t); Step 6: Extract and reconstruct sensitive signal f denoised The phase of (t) is unwrapped to obtain the phase unwrapped signal g(t), and the signal g(t) is adaptively decomposed to obtain Z narrowband modes u with different frequencies. z (t), Z is a positive integer; Step 7: Perform FFT transformation on the decomposed multiple narrow-band modes to obtain the spectrum, extract the peak value of the spectrum, and filter the respiratory signal and heartbeat signal according to the frequency point corresponding to the spectrum peak value.

2. The method for detecting vital signs based on OFDM integrated radar communication signal according to claim 1, characterized in that: In step 1, the radar communication integrated echo signal S RX (t) is expressed as: Where, 0≤n≤N-1, N is the number of subcarriers of the transmission carrier frequency, 0≤m≤M-1, M is the total number of OFDM signal symbols, A(n,m) represents the communication information modulated on the nth subcarrier in the mth symbol, j is the imaginary symbol, Δf is the subcarrier spacing, τ is the delay, f c is the carrier frequency, noise is the noise; The time delay τ is expressed as: τ=2R1 / c0 Where c0 is the speed of light, R is the distance between the human chest and the radar, V0 is the radial motion speed of the human body, B b is the amplitude of chest movement caused by breathing, ω b is the respiratory rate, H h is the amplitude of chest movement caused by heartbeat, ω h is the heart rate, is the phase difference caused by heartbeat and breathing.

3. The method for detecting vital signs based on OFDM integrated radar communication signal according to claim 1, characterized in that: In step 3, for each subcarrier, the phase information of the time domain target correlation signal h(t) minus the average phasor in the slow time dimension is expressed as: AND pvmc，n (t)=Y n (t)-Y mean，n Among them, Y pvmc，n (t) represents the phase information of the time domain target correlation signal after background noise is eliminated on the nth subcarrier, Y n (t) represents the phase information of the time domain target correlation signal on the nth subcarrier, Y mean，n Represents the background noise phase component corresponding to the nth subcarrier.

4. The method for detecting vital signs based on OFDM integrated radar communication signal according to claim 1, characterized in that: In step 4, the phase φ(t) of the reconstructed signal ββ(t) is extracted, and the phase fluctuation variance of φ(t) is calculated in the slow time dimension of each subcarrier: in, is the phase fluctuation variance of the nth subcarrier, φ m，n (m) is the instantaneous phase of the nth subcarrier in the mth symbol, is the mean phase value of the nth subcarrier.

5. The method for detecting vital signs based on OFDM integrated radar communication signal according to claim 1, characterized in that: The specific process of step 5 is: The sensitive signal X(t) is decomposed into L levels using wavelet basis functions, which can be expressed as: Where l is the scale parameter, 1≤l≤L, L is the number of decomposition levels; d l is the detail coefficient of the lth layer, ψ l (t) is the wavelet basis function of the first layer, a L is the approximation coefficient, κ L (t) is the scale function; Extract the approximate coefficients and detail coefficients of all layers to construct the wavelet decomposition coefficient vector ζ = [a L , d L , d L-1 , d L-2 , d L-3 ,…,d1]; the soft threshold method is used to remove the noise component in the wavelet decomposition coefficient vector, which is expressed as: g denoised =sign(ζ)·max(|ζ|-λ,0) Among them, denoised represents the wavelet decomposition coefficient vector after denoising, λ is the set threshold, max represents the maximum value, | represents the absolute value of the element, and sign(·) is the sign function, which is defined as: The wavelet decomposition coefficient vector ζ after denoising denoised It is expressed as: ζ denoised =[a denoised,L ,d denoised，L ,d denoised,L-1 ,d denoised,L-2 ,d denoised,L-3 ,…,d denoised,1 ] According to the denoised wavelet decomposition coefficient vector ζ denoised The L-layer iterative reconstruction is performed with the wavelet basis function to obtain the sensitive signal f after noise removal. denoised (t):

6. The method for detecting vital signs based on OFDM radar communication integrated signal according to claim 1, characterized in that: In step 6, the signal g(t) is adaptively decomposed to obtain z narrowband modes u with different frequencies. z The process of (t) is expressed as: Where 1≤z≤Z, Z is the number of narrowband modes, u z (t) is the zth narrowband mode, ω z represents the center frequency of the zth mode, δ(t) is the Dirac function, represents the partial derivative with respect to time t, and || ||2 represents the second norm.

7. The method for detecting vital signs based on OFDM integrated radar communication signal according to claim 1, characterized in that: In step 7, 0.1 Hz to 0.5 Hz is selected as the frequency estimation of the breathing signal, and 0.8 Hz to 2 Hz is selected as the frequency estimation of the heartbeat signal.

Images