Multi-person vital sign monitoring method and system based on space-time encoding metasurface

Abstract

The application discloses a kind of multi-person vital sign monitoring method and system based on space-time coding hypersurface, utilize the harmonic wave beam scanning area of attention generated by space-time coding hypersurface to carry out human detection, and allocate the beam of frequency orthogonal for each detected human target, to accurately estimate their respiratory and heartbeat rate.The system can monitor the vital sign of multiple people simultaneously, and accurately estimate the respiratory and heartbeat frequency of human target.The control ability of space-time coding hypersurface used in the application can reduce the noise reflected by non-related objects, thereby improving the signal-to-noise ratio of human chest echo.The application uses wireless signals, has unique advantages such as not easily disturbed by environmental factors and protecting visual privacy.Compared with traditional vital sign monitoring instruments, the non-contact characteristics of the application can reduce the stress and interference on the user, and can be used for long-term, remote health monitoring.

Description

TECHNICAL FIELD

[0001] The present application relates to the fields of new artificial electromagnetic materials, wireless communication technology and biomedical technology, and particularly relates to a multi-person vital sign monitoring method and system based on a space-time coding metasurface. BACKGROUND

[0002] Vital sign monitoring is of great significance for early detection of diseases, evaluation of treatment effectiveness and personal health management. However, traditional contact vital sign monitoring devices may cause inconvenience to the subjects, limiting their application in daily health monitoring and long-term tracking. Therefore, the concept of non-contact vital sign monitoring has attracted widespread attention. Through sensors and other devices, non-contact technology can accurately obtain physiological parameters of the human body. Since it does not require direct contact with the body of the subject, the monitoring process is more convenient and natural. However, existing non-contact vital sign monitoring technology still has problems such as low efficiency and susceptibility to environmental condition changes.

[0003] In recent years, the rapid development of wireless communication technology and the widespread deployment of mobile communication infrastructure have made non-contact vital sign monitoring methods based on wireless signals a hot topic. Vital sign monitoring using wireless signals is achieved by capturing the slight changes in reflected signals (such as radio absorption, scattering and polarization) caused by human chest movements. This method is less affected by environmental factors and has the unique advantage of protecting visual privacy. Although existing methods have achieved considerable results, they are mainly applicable to single-person scenarios. To realize the great application potential of wireless signal-based vital sign monitoring technology in multi-person living environments, many problems faced by traditional signal processing technology still need to be solved, such as insufficient signal resolution, multipath interference and signal space-time overlap. These problems currently greatly limit the efficiency and accuracy of multi-person vital sign monitoring.

[0004] In this context, space-time coding metasurfaces can dynamically control electromagnetic waves in the time domain, frequency domain and spatial domain, and can enhance the focusing ability of signals in a specific direction, improving the spatial resolution of signals. This characteristic provides new ideas and methods for solving the signal separation and identification problems in multi-person vital sign monitoring.

[0005] The present application proposes a multi-person vital sign monitoring method based on a space-time coding metasurface. The method first controls the beam to the possible location of the target human body through the space-time coding metasurface. After determining the presence of the target, the echo signal is processed using an improved variational mode decomposition algorithm, and finally the respiratory and heartbeat frequencies of multiple targets can be extracted. The present application can efficiently and accurately distinguish and identify the vital sign signals of multiple individuals in the monitoring area. SUMMARY

[0006] Technical problem

[0007] The purpose of the present application is to provide a non-contact multi-person vital sign monitoring method based on a spatiotemporal coding metasurface

[0008]

[0009] The ability to accurately control the electromagnetic beam and the variational modal decomposition algorithm can decompose complex signals, and a multi-person vital sign monitoring system based on a spatiotemporal coding metasurface is constructed, which can accurately and efficiently detect the breathing and heartbeat frequencies of multiple human targets.

[0010] Technical scheme

[0011] A multi-person vital sign monitoring method based on a spatiotemporal coding metasurface, comprising the following steps:

[0012] Step 1: Use the spatiotemporal coding metasurface to modulate the incident signal transmitted by the transmitter into a series of harmonic beams with different frequencies, and optimize the spatiotemporal coding to control the energy distribution of the harmonic beams in the spatial domain and the frequency domain, so that the energy of the beams of different harmonics is concentrated in different directions;

[0013] Step 2: Use the harmonic beams generated by the spatiotemporal coding metasurface to perform beam scanning in the area of interest, and in each scanning, decompose the echo signal received by the receiver in the frequency domain when the signal strength is greater than the threshold to obtain the breathing signal; according to the signal strength of the echo signal and the breathing signal, determine whether there is a human target in the current scanning area;

[0014] Step 3: If a human target is detected, assign a harmonic beam to the direction where the person is located, and then collect the echo signal of the person and extract the fine human physiological signal from the echo signal of the person.

[0015] Preferably, the spatiotemporal coding metasurface uses a 1-bit metasurface with a working frequency of 3.5GHz, which is composed of 32x32 units, and each digital unit includes two PIN diodes, each diode has the function of electronically and dynamically controlling the electromagnetic response of the reflection.

[0016] Preferably, by optimizing the spatiotemporal coding, the generated coding sequence is switched in a pre-designed time period, so as to control the power distribution and spatial propagation of the harmonic beams of the spatiotemporal coding metasurface, so as to modulate the harmonic components of different frequencies to different spatial positions.

[0017] Preferably, the step of determining whether there is a human body in the region of interest in step 2 comprises: presetting a signal strength threshold, if the difference between the echo signal strength in a certain direction and the strength of the echo signal is greater than the signal strength threshold, it is determined that there may be a human being in the direction; on this basis, the echo signal in the direction is decomposed to detect whether the human respiratory signal can be extracted from the echo signal; if the respiratory signal can be extracted from the echo signal, it is finally determined that there is a human body target in the direction.

[0018] Preferably, the improved VMD algorithm is used to extract the human vital sign signal in step 3, and the specific steps comprise:

[0019] Step 3.1: Construct an optimization problem, and the optimization objective function is:

[0020]

[0021]

[0022] Wherein, L(.) is the target function to be optimized, u i is the i-th signal component obtained by VMD algorithm decomposition, ω i is the frequency of u i , λ represents the Lagrange multiplier, α i is the penalty factor of u i , I is the total number of signal components obtained by VMD algorithm decomposition, is an impulse function, f l (t) represents a low-pass filter, f b (t) represents a band-pass filter, f i (t) is a filter acting on the i-th signal component, which can be f l (t) or f b (t), M represents the number of signal components belonging to the respiratory signal, s(t) represents the original signal input to the VMD algorithm, which is also the echo signal received by the receiving end, and ||.||2 2 2 represents the square of the modulus;

[0023] According to the Plancherel-Parseval Fourier isometry property under the L2 norm, ω→ω-ω i is changed, and formula (1) is rewritten as formula (2):

[0024]

[0025] Wherein is the frequency domain expression of u i , is the frequency domain expression of s(t), and so on.

[0026] Step 3.2: solve the optimization objective function by using the ADMM algorithm to calculate the i-th IMF in the (n+1)th iteration i.e. equation (3):

[0027]

[0028] Meanwhile, The center frequency of the i-th harmonic beam is According to equation (4):

[0029]

[0030] The application also provides a multi-person vital sign monitoring system based on a space-time coding metasurface, which comprises a transceiver module, a space-time coding metasurface module and a host computer. The transceiver module is connected with two antennas by using a universal software radio peripheral device and is used for signal transmission and reception. In the space-time coding metasurface module, a micro-programmed control unit is used to control the hardware of the metasurface. The space-time coding metasurface manipulates incident electromagnetic waves by time-varying switching of coding modes, so as to customize wireless sensing channels in the spatial and frequency domains. The host computer is responsible for communication with the universal software radio peripheral device and the micro-programmed control unit and performs echo signal processing. The host computer first judges whether a human target exists in a region of interest according to echo signals and feeds back to the micro-programmed control unit. The micro-programmed control unit controls the space-time coding metasurface to allocate a harmonic beam to the direction where the human target exists. The transceiver module collects the echo signals of the person and sends them to the host computer. The host computer extracts fine human vital sign signals from the echo signals.

[0031] Preferably, the host computer judges whether a human target exists according to the echo signal intensity, and then uses a series of bandpass filters to decompose the echo signals with a signal intensity greater than a threshold in the frequency domain to obtain a respiratory signal. Whether a human target exists is finally determined according to the respiratory signal.

[0032] Preferably, the micro-programmed control unit controls the power distribution and spatial propagation of the harmonic beam of the space-time coding metasurface by optimizing the space-time coding, and the generated coding sequence is cyclically switched in a pre-designed time period to realize modulation of different frequency components of the scattering pattern to different positions.

[0033] Preferably, the host computer extracts fine human vital sign signals by using the improved VMD algorithm.

[0034] Beneficial effects

[0035] (1) The space-time coding metasurface is used to realize multi-harmonic beam generation, and different harmonic beams are used for simultaneous multi-person vital sign detection and perception. Unlike existing human vital sign perception algorithms based on wireless signals, the application can realize simultaneous multi-person perception, and the perception functions of each target person are independent of each other.

[0036] (2) For the specific scene of monitoring human respiration and heartbeat, the application proposes an improved VMD algorithm. The improved algorithm uses a small amount of preset IMF, performs signal reconstruction based on a filter, and adaptively decomposes physiological signals from reflected signals, thereby improving the accuracy and reliability of vital sign signal extraction.

[0037] (3) The application utilizes the multi-harmonic beam generation and harmonic beam control capability of the space-time coded metasurface, which can detect personnel in the target scene, and simultaneously perform real-time sensing on the detected personnel, thereby realizing the integration of human detection and vital sign sensing based on wireless signals. BRIEF DESCRIPTION OF DRAWINGS

[0038] Figure 1 The figure is a schematic diagram of the multi-person vital sign sensing system based on the space-time coded metasurface proposed in the application.

[0039] Figure 2 The figure is a flow chart of the human echo signal processing based on the improved VMD algorithm proposed in the application.

[0040] Figure 3 The figure is a 1-bit metasurface used in the application. (a) Metasurface unit structure, (b) simulated metasurface unit amplitude and phase frequency characteristics, (c) experimentally obtained unit amplitude and phase frequency characteristics.

[0041] Figure 4 The figure is a space-time coding sequence used in the application. (a) 21 coding sequences optimized using the BPSO algorithm; (b) reference coordinate system; (c) near-field patterns of -3rd, -1st, +1st, +3rd harmonic beams, whose energies are concentrated at positions (-1.5m, +1m, 0m), (-0.5m, +1m, 0m), (+0.5m, +1m, 0m) and (+1.5m, +1m, 0m) respectively. LO represents the zeroth harmonic component. DETAILED DESCRIPTION

[0042] The multi-person vital sign monitoring method based on the space-time coded metasurface proposed in the application can detect multiple human targets in a non-contact manner and accurately and efficiently extract their respiration and heartbeat frequencies. This method can be realized by the system in the application, which includes three layers: a physical layer, a digital layer and a sensing layer. Through these three modules, the system can continuously detect the presence of human targets, while estimating the respiration and heartbeat frequencies of the identified individuals, as shown in Figure 1 The configuration and function of each layer are summarized as follows.

[0043] (1) Physical layer: space-time coded beam control

[0044] First, the physical layer consists of a transceiver module and a spatiotemporally coded metasurface module. The transceiver module utilizes a Universal Software Radio Peripheral (USRP) device connected to two antennas for signal transmission and reception. Simultaneously, a microprogrammable control unit (MCU) is employed within the metasurface module to control the metasurface hardware. Through spatiotemporal coding, the metasurface can manipulate incident electromagnetic waves and customize wireless sensing channels in the spatial and frequency domains. In this way, human detection and vital sign monitoring can be performed simultaneously using different harmonic beams. To achieve this functionality, a 1-bit metasurface operating at 3.5 GHz is used in the system. This metasurface consists of 32×32 units. Each digital unit includes two PIN diodes, each with electronic and dynamic control over the reflected electromagnetic response. The amplitude-frequency response and phase-frequency response of the designed units are as follows: Figure 3 As shown, at around 3.5 GHz, the reflected electromagnetic waves encoded as "0" and "1" have almost no amplitude difference, while the corresponding phase difference is approximately 180 degrees. Therefore, the unit cell can be effectively used to design phase-based digital metasurfaces. The near-field radiation pattern G of the metasurface is shown. i,j (f) can be expressed as:

[0045]

[0046] Among them, the single-tone signal emitted to the (m,n)th unit of the metasurface is used Indicate. Given amplitude A m,n Center frequency f and phase φ m,n w i,j,m,n Here are the transmission parameters. k = 2π / λ is the wave number. It is the reflection coefficient of the (m,n) element at the center frequency within the interval (l-1)τ≤t≤lτ. and β m,n Let B represent the modulation amplitude and phase of the (m,n) unit, respectively. For a 1-bit metasurface, the reflection amplitude B of each unit is... m,n It is uniform, while the phase β m,n Based on the digits "0" and "1" in the encoded sequence, it represents 0° or 180°. This is achieved through optimized spatiotemporal coding. It can simultaneously control the power distribution and spatial propagation of harmonic beams. The generated coded sequence cycles through a pre-designed time period to modulate different frequency components of the scattering pattern to different locations. The functions implemented by the encoding include... Figure 4 As shown. This allows for the differentiation of sensor signals targeting different individuals in both the spatial and frequency domains. Furthermore, comprehensive human detection and vital sign monitoring require system communication. Therefore, a host computer is integrated into the system to communicate with the USRP and MCU, and to process the echo signals.

[0047] (2) Digital layer: Scan-based multi-person detection

[0048] Secondly, the digital layer is mainly used for human target detection. Specifically, the harmonics generated by the spatiotemporal coded metasurface are used for beam scanning within the sensing area. In each scan, the echo signals of different harmonic frequencies received can be decomposed in the frequency domain and then used for human body detection. Then, two indicators, i.e., the intensity of the reflected signal and the breathing signal, are used to infer whether there is a human target in the current scanning area. If a human target is detected, a harmonic beam is assigned to this direction, and the echo signal of the person is collected for fine vital sign sensing. Meanwhile, other harmonic beams not used for vital sign sensing will continue to scan the sensing area to detect human targets. Specifically, the single-tone signal x T (t) = exp(2πf c t) is transmitted by the directional antenna Tx and directed to the spatiotemporal coded metasurface. With the directional antenna Tx, it can be assumed that there is no line-of-sight distance between Tx and the sensing area. That is, the transmitted signal x T (t) is first modulated by the metasurface and then reflected to the AoI for monitoring. The diagonal matrix H RIS ∈ C MN×MN is taken as the steering matrix of the metasurface, which contains M x N elements. The diagonal elements of H RIS form a vector Thus, the signal x RIS (t) reflected by the metasurface can be represented as:

[0049] x RIS (t) = H RIS (t)H T,RIS (t) x T (t) + z RIS,T (t) (2)

[0050] where H T,RIS (t) ∈ C MN×1 represents the channel between Tx and RIS, z RIS,T (t) represents the noise between Tx and the metasurface, and x T (t) represents the transmitted symbol at time t. According to the spatiotemporal coding method, the incident signal x T (t) is modulated into a series of harmonic components with different frequencies. Therefore, the signal x RIS (t) can be re-expressed as a combination of a series of harmonics:

[0051]

[0052] where f0= 1 / T0is the time-domain modulation frequency of the spatiotemporal coded metasurface, and A i(t) is the amplitude of the i-th harmonic. Thus, when the i-th harmonic beam illuminates the human target, the signal will be modulated by the motion of the human body. Subsequently, the omni-directional antenna Rx of the receiving terminal will receive the reflected echo of the target human body. Thus, the channel model H RIS,P,R (t) between the metasurface and Rx can be expressed as:

[0053]

[0054] where H P,R (t) denotes the channel between Rx and the human target illuminated by the i-th harmonic beam, H RIS,P (t) denotes the channel between the metasurface and the human target, denotes the complex reflection coefficient of the human at time t, f D denotes the Doppler shift caused by the human motion.

[0055] Thus, considering that there are two main propagation paths between the metasurface and Rx (i.e., the line-of-sight path and the “RIS-human-Rx” path), the received signal x R (t) consisting of a series of harmonic components can be expressed as:

[0056]

[0057] where z R (t) denotes the noise between Tx and Rx, H i (t) can be expressed as:

[0058]

[0059] Since the frequencies of these harmonics are orthogonal, it is relatively easy to separate the reflected signal in the frequency domain using a series of bandpass filters, which can then be used for human detection and vital sign monitoring. For the human detection task, two indicators are used to identify whether there is a human in the illumination direction. Specifically, in the absence of a human, a pre-scan is first performed on the entire AoI to obtain the intensity vector of the received echo signal from D directions Then, a signal intensity threshold μ is preset according to empirical knowledge. Thus, for direction d ∈ [1, D], the signal intensity difference ΔI d can be expressed as:

[0060]

[0061] where I d is from direction d. If ΔI dIf the value of the correlation coefficient is greater than the preset μ, the system will determine that there is a person in this direction. In addition, in order to further confirm whether the target is a person, the system also uses a breathing detection technology. Some preliminary experiments show that it is easier to extract the breathing signal than to obtain the heartbeat signal. Therefore, the breathing signal is used as another indicator for detecting human individuals. In other words, if the breathing signal is extracted, the target to be detected will eventually be identified as a human individual.

[0062] (3) Perception layer: vital sign perception based on improved VMD algorithm

[0063] The perception layer is used to finally obtain the breathing and heartbeat signals of multiple human body targets after determining that there is a human body target in the focused beam position. Since the detected human target is irradiated by different harmonic beams, a series of bandpass filters can be used to separate their echo signals in the frequency domain.

[0064] In addition, the beamforming technology supported by the spatiotemporal coded metasurface can improve the signal-to-noise ratio of the reflection signal of the target individual and reduce the interference from other people and environmental clutter. In order to efficiently process the echo signal to accurately obtain the breathing and heartbeat frequencies, the present application proposes an improved VMD algorithm as shown in Figure 2 The definition of IMF in VMD is:

[0065] u k (t)=A k (t)cos(φ k (t)) (8)

[0066] In the formula, u k (t) is a frequency-modulation-amplitude-modulation (AM-FM) signal, A k (t) is an amplitude envelope greater than zero, and φ k (t) represents the phase. This new modal definition method can make each signal component have a narrower bandwidth. To obtain the frequency bandwidth of each mode through the algorithm, first, the analytic signal of each mode is obtained by using the Hilbert transform, that is, the analytic signal of the signal is obtained by taking the inverse Fourier transform of the single-sided spectrum of the signal,

[0067]

[0068] Then, the analytic signal is multiplied by the complex exponential with the center frequency, and the translation of the single-sided spectrum to the respective baseband is realized in the frequency domain.

[0069]

[0070] The bandwidth estimation of the signal components utilizes the square of the L2 norm of the gradient of the analytic signal. To ensure strict enforcement of the constraints, a penalty factor and a Lagrange multiplier λ are introduced. This constraint problem can be expressed as:

[0071]

[0072] Where {u i}:={u1,u2,...,u I} and {w i}:={w1,w2,...,w I Let} represent the set of all IMFs and their center frequencies, respectively. To solve this constrained problem, iterative optimization using the Alternate Direction Multiplication (ADMM) method is employed. The signal component u can be obtained through iteration. k (t) and its center frequency ω k (t).

[0073] Based on this, in order to avoid respiratory signals s r (t) and heartbeat signal s h To address the spectral overlap between (t), this invention proposes a filter-based signal reconstruction mechanism and employs two filters for the IMF during the ADMM optimization process. Specifically, it is assumed that the first M IMFs belong to s r (t), while other IMFs belong to s h (t). Then, in each iteration of ADMM optimization, the high-frequency components in the first M IMFs are processed using a low-pass filter, and noise in the remaining IMFs is removed using a band-pass filter. In this case, the original echo signal s(t) can be reconstructed as:

[0074]

[0075] Among them, u m (t) belongs to s r The m-th possible IMF of (t), u n (t) belongs to s h The nth possible IMF of (t), f l (t) is a low-pass filter, f b (t) is a bandpass filter. Furthermore, the improved VMD algorithm proposes a frequency-dependent adaptive penalty factor α to extract IMFs with narrower bandwidths. Specifically, a penalty factor based on a standard frequency is proposed for each IMF. For the i-th IMF (i≤i), its penalty factor can be expressed as:

[0076]

[0077] Where, α intdenotes the initial penalty factor, ζ is the scaling factor. ω r denotes the reference frequency, which is preset according to the observation that the respiratory frequency and the heartbeat frequency are generally around 0.25 Hz and 1.35 Hz, respectively. With the adaptive α, when the center frequency of the ith IMF is close to its reference frequency ω r , the IMF will be given a larger weight in the ADMM optimization process, so as to obtain a narrower frequency band signal and filter out the out-of-band noise. In general, the quadratic optimization problem in formula (11) can be rewritten as:

[0078]

[0079] where f i (t) denotes the filter performed on the ith IMF, which can be a low-pass or band-pass filter. Then, according to the Plancherel-Parseval Fourier isometry property under the L2 norm, ω→ω-ω i , formula (14) can be rewritten as formula (15):

[0080]

[0081]

[0082] Finally, by using the ADMM algorithm, the ith IMF in the (n+1)th iteration can be calculated as: i.e. formula (16):

[0083]

[0084] At the same time, the center frequency of the ith IMF can be calculated according to formula (16):

[0085]

[0086] By using the proposed filter-based signal reconstruction mechanism and adaptive penalty factor, the VMD algorithm can accurately extract human physiological signals.

Claims

1. A method for monitoring the vital signs of multiple individuals based on a spatiotemporally encoded metasurface, characterized in that, Includes the following steps: Step 1: The incident signal emitted by the transmitter is modulated into a series of harmonic beams with different frequencies using a spatiotemporally coded metasurface. By optimizing the spatiotemporal coding, the energy distribution of the harmonic beams in the spatial and frequency domains is controlled, so that the energy of the beams of different harmonics is concentrated in different directions. Step 2: Use the harmonic beam generated by the spatiotemporally coded metasurface to perform beam scanning in the region of interest. In each scan, decompose the echo signal with a signal strength greater than the threshold received by the receiver in the frequency domain to obtain the breathing signal. Determine whether there is a human target in the current scanning area based on the signal strength of the echo signal and the breathing signal. Step 3: If a human target is detected, a harmonic beam is allocated in the direction of the person to collect their echo signal. An improved VMD algorithm is then used to extract fine human physiological signals from this echo signal. The improved VMD algorithm is implemented by introducing a filter-based signal reconstruction mechanism and a frequency-dependent adaptive penalty factor, and specifically includes the following sub-steps: Step 3.1: Construct the optimization problem, with the objective function as follows: ； (1)； in, Let be the objective function to be optimized. Let i be the i-th signal component obtained by the VMD algorithm. for frequency, Represents the Lagrange multiplier. yes The adaptive penalty factor is given, where I is the total number of signal components obtained by the VMD algorithm. Represents a low-pass filter. Represents a bandpass filter. For a filter acting on the i-th IMF, it may be: or M represents the number of signal components belonging to the respiratory signal. This represents the original signal input to the VMD algorithm, which is also the echo signal received by the receiver. Represents the square of the modulus; It is a pulse function; Based on the Plancherel-Parseval Fourier isometry property under the L2 criterion, change Formula (1) can be rewritten as Formula (2): ， (2) in yes The frequency domain representation, yes The frequency domain representation, , , , They are , , The frequency domain representation; Step 3.2: Using the ADMM algorithm, solve the objective function and calculate the i-th IMF in the (n+1)-th iteration. That is, formula (3): (3)； at the same time, center frequency Calculate according to formula (4): (4)。 2. The method for monitoring the vital signs of multiple individuals based on a spatiotemporally encoded metasurface according to claim 1, characterized in that, The spatiotemporal coding metasurface is a 1-bit metasurface with an operating frequency of 3.5 GHz. The metasurface consists of 32×32 units, each of which includes two PIN diodes. Each diode has the function of electronic and dynamic control of reflected electromagnetic response.

3. The method for monitoring the vital signs of multiple individuals based on a spatiotemporally encoded metasurface according to claim 1, characterized in that, By optimizing the spatiotemporal coding, the generated coding sequence is cyclically switched within a pre-designed time period, thereby controlling the power distribution and spatial propagation of the spatiotemporal coded metasurface harmonic beam, so as to modulate harmonic components of different frequencies to different spatial locations.

4. The method for monitoring the vital signs of multiple individuals based on a spatiotemporally encoded metasurface according to claim 1, characterized in that, Step 2, which determines whether there is a human body in the area of ​​interest, includes: setting a preset signal strength threshold; if the difference between the intensity of the echo signal and the intensity of the echo signal in a certain direction is greater than the signal strength threshold, it is determined that there may be a human body in that direction; based on this, the echo signal in that direction is decomposed to detect whether a human breathing signal can be extracted from the echo signal; if a breathing signal can be extracted from the echo signal, it is finally determined that there is a human target in that direction.

5. A multi-person vital signs monitoring system based on spatiotemporally coded metasurfaces, characterized in that, The system includes a transceiver module, a spatiotemporally coded metasurface module, and a host computer. The transceiver module is connected to two antennas using general-purpose software-defined radio peripherals for signal transmission and reception. In the spatiotemporally coded metasurface module, a microprogrammable control unit is used to control the metasurface hardware. The spatiotemporally coded metasurface manipulates incident electromagnetic waves by switching coding modes in a time-varying manner, thereby customizing wireless sensing channels in the spatial and frequency domains. The host computer is responsible for communicating with the general-purpose software-defined radio peripherals and the microprogrammable control unit, and for processing echo signals. The host first determines whether there is a human target in the area of ​​interest based on the echo signal and feeds it back to the microprogramming control unit. The microprogramming control unit controls the spatiotemporal coding metasurface to allocate a harmonic beam in the direction where a human target exists. The transceiver module collects the echo signal of the person and sends it to the host. The host then extracts detailed human vital signs signals from the signal. The host computer uses an improved VMD algorithm to extract detailed human vital signs signals. The specific steps include: Step 3.1: Construct the optimization problem, with the objective function as follows: ， (1)； in, Let be the objective function to be optimized. Let i be the i-th signal component obtained by the VMD algorithm. for frequency, Represents the Lagrange multiplier. yes The adaptive penalty factor is given, where I is the total number of signal components obtained by the VMD algorithm. For impulse functions, Represents a low-pass filter. Represents a bandpass filter. For a filter acting on the i-th IMF, it may be: or M represents the number of signal components belonging to the respiratory signal. This represents the original signal input to the VMD algorithm, which is also the echo signal received by the receiver. Represents the square of the modulus; Based on the Plancherel-Parseval Fourier isometry property under the L2 criterion, change Formula (1) can be rewritten as Formula (2): ， (2)； in yes The frequency domain representation, yes The frequency domain representation, , , , Similarly; Step 3.2: Using the ADMM algorithm, solve the objective function and calculate the i-th IMF in the (n+1)-th iteration. That is, formula (3): (3)； at the same time, center frequency Calculate according to formula (4): (4)。 6. A multi-person vital sign monitoring system based on a spatiotemporally encoded metasurface according to claim 5, characterized in that, The host determines whether a human target may exist based on the intensity of the echo signal. Then, it uses a series of bandpass filters to decompose the echo signal with a signal intensity greater than a threshold in the frequency domain to obtain the breathing signal. The presence or absence of the breathing signal is then used to ultimately determine whether a human target exists.

7. A multi-person vital signs monitoring system based on a spatiotemporally encoded metasurface according to claim 5, characterized in that, The microprogrammable control unit optimizes the spatiotemporal coding and simultaneously controls the power distribution and spatial propagation of the spatiotemporal coded metasurface harmonic beam. The generated coding sequence is cyclically switched within a pre-designed time period to modulate different frequency components of the scattering pattern to different positions.

Images