Non-contact vital sign monitoring method, electronic equipment and storage medium

Abstract

The invention provides a non-contact vital sign monitoring method which can be applied to the technical fields of wireless signal processing, physiological parameter examination and health monitoring. The method comprises the following steps: based on authorization of a target user, collecting channel state information of the target user in real time through a non-contact detection technology, and performing feature extraction on the channel state information in a time-frequency domain to obtain wireless time-frequency features; constructing a multi-signal sparse coefficient representation model based on a second-order regularization normal form by utilizing the pre-trained multi-dimensional vital sign self-adaptive dictionary, the wireless time-frequency characteristics and a preset threshold value; and performing distributed iteration processing on the multi-signal sparse coefficient representation model by using an alternating direction multiplier algorithm to obtain a multi-signal sparse coefficient matrix representing vital signs, and processing the multi-signal sparse coefficient matrix representing the vital signs to obtain multi-dimensional vital sign information of the target user. The invention further provides electronic equipment for non-contact vital sign monitoring and a storage medium.

Description

Technical Field

[0001] The present invention relates to the technical fields of wireless signal processing, physiological parameter detection, and health monitoring, and particularly to a non-contact vital sign monitoring method, an electronic device, and a storage medium. Background Art

[0002] With the in-depth development of the integration of wireless communication technology and health care, the use of passive sensing means such as Wi-Fi and millimeter-wave radar to achieve non-contact vital sign detection has become a research hotspot. Some vital signs, such as respiratory signals and heartbeat signals, have characteristics such as low frequency, weakness, and high noise. Traditional spectrum analysis methods have performance degradation under low signal-to-noise ratio conditions and are easily affected by environmental interference.

[0003] At present, non-contact vital sign detection technology has made preliminary progress. Especially in the detection of respiration and heartbeat using CSI (Channel State Information), there have been multiple research results. However, the existing non-contact vital sign detection technology faces multiple technical challenges in practical applications. First, in a low signal-to-noise ratio environment, traditional spectrum analysis methods are difficult to effectively distinguish target signals from noise, resulting in unstable estimation results. Second, the fundamental wave and its high-order harmonics of some vital signs, such as respiratory signals, are prone to overlap with the frequency of heartbeat signals, causing interference between the two and affecting the accurate detection of heartbeat frequency. Third, existing methods mostly rely on fixed filters and manual parameters, lacking generality and adaptive capabilities and being unable to adapt to different human positions, orientations, and environmental conditions. Finally, due to ignoring the non-stationary characteristics and time-domain structure of signals, most methods are difficult to track the dynamic changes of vital signs in real time. The above technical problems mainly stem from the lack of modeling and utilization of the sparse characteristics of signals in traditional technical solutions, the lack of a robust signal separation mechanism, and a time-frequency joint analysis framework. Summary of the Invention

[0004] In view of the above problems, the present invention provides a non-contact vital sign monitoring method, an electronic device, and a storage medium for at least solving one of the above technical problems.

[0005] According to a first aspect of the present invention, there is provided a non-contact vital sign monitoring method, including:

[0006] Based on the authorization of the target user, the channel state information of the target user is collected in real time through non-contact detection technology, and feature extraction is performed on the channel state information in the time-frequency domain to obtain wireless time-frequency features;

[0007] Construct a multi-signal sparse coefficient representation model based on the second-order regularization paradigm by using a pre-trained multi-dimensional vital sign adaptive dictionary, wireless time-frequency features, and a preset threshold. Among them, the pre-trained multi-dimensional vital sign adaptive dictionary is obtained through iterative convex optimization operations. The multi-dimensional vital sign adaptive dictionary is a learnable template library with multi-dimensional physiological feature information.

[0008] Use the alternating direction multiplier algorithm to perform distributed iterative processing on the multi-signal sparse coefficient representation model to obtain a multi-signal sparse coefficient matrix representing vital signs, and process the multi-signal sparse coefficient matrix representing vital signs to obtain multi-dimensional vital sign information of the target user.

[0009] According to an embodiment of the present invention, the above multi-dimensional vital sign information includes respiratory rate and heart rate.

[0010] Among them, the pre-trained multi-dimensional vital sign adaptive dictionary includes a pre-trained respiratory rate adaptive dictionary and a pre-trained heart rate adaptive dictionary.

[0011] Among them, the multi-signal sparse coefficient matrix includes a respiratory rate sparse coefficient matrix and a heart rate sparse coefficient matrix.

[0012] According to an embodiment of the present invention, the above pre-trained multi-dimensional vital sign adaptive dictionary obtained through iterative convex optimization operations includes:

[0013] By introducing a sine function based on physiological prior information and performing non-linear constraints on the joint optimization objective function for the multi-dimensional vital sign adaptive dictionary, a joint optimization objective function with prior constraints is obtained. Among them, the physiological prior information includes the periodic physiological signals of organisms.

[0014] Randomly select multiple wireless time-frequency samples from the wireless time-frequency feature samples related to vital signs to initialize the multi-dimensional vital sign adaptive dictionary, and obtain the initialized multi-dimensional vital sign adaptive dictionary.

[0015] Fix the parameters of the initialized multi-dimensional vital sign self-adaptive dictionary, and use the joint optimization objective function to perform convex optimization operations on the training sparse coefficient matrix corresponding to the multi-dimensional vital sign self-adaptive dictionary by introducing a weighted quadratic diagonal matrix.

[0016] Numerically fix the convex-optimized training sparse coefficient matrix, and use the joint optimization objective function with prior constraints to perform gradient descent processing on the initialized multi-dimensional vital sign self-adaptive dictionary based on the Lagrange multiplier to obtain the processed multi-dimensional vital sign adaptive dictionary.

[0017] Iteratively perform convex optimization operations and gradient descent processing operations until the preset training conditions are met, and obtain the pre-trained multi-dimensional vital sign adaptive dictionary.

[0018] According to an embodiment of the present invention, the above convex optimization operation on the training sparse coefficient matrix corresponding to the multi-dimensional vital sign self-adaptive dictionary by introducing a weighted quadratic diagonal matrix includes:

[0019] Calculate the Euclidean norm of each column vector in the training sparse coefficient matrix;

[0020] Obtain the second-order norm of the training sparse coefficient matrix based on the Euclidean norm of each column vector;

[0021] Perform a twice-differentiable simulation on the second-order norm of the training sparse coefficient matrix by introducing a weighted quadratic diagonal matrix;

[0022] Use the result of the twice-differentiable simulation to convert the joint optimization objective function into a convex optimization objective function;

[0023] Perform function expansion and derivative processing on the convex optimization objective function to obtain the convex-optimized training sparse coefficient matrix.

[0024] According to an embodiment of the present invention, the above processing of the initialized multi-dimensional vital sign self-adaptive dictionary by using the joint optimization objective function with prior constraints through gradient descent processing based on Lagrange multipliers to obtain the processed multi-dimensional vital sign adaptive dictionary includes:

[0025] Introduce Lagrange multipliers for each prior constraint term in the joint optimization objective function with prior constraints, and then expand the joint optimization objective function with prior constraints into a Lagrangian function;

[0026] Fix the Lagrange multipliers, calculate the gradient information corresponding to the initialized multi-dimensional vital sign self-adaptive dictionary by using the Lagrangian function, and update the parameters of the initialized multi-dimensional vital sign self-adaptive dictionary by using the gradient information;

[0027] Fix the parameters of the updated multi-dimensional vital sign self-adaptive dictionary and update the Lagrange multipliers;

[0028] Iteratively perform gradient calculation, adaptive dictionary parameter update, and Lagrange multiplier update until a preset convergence condition is met to obtain the processed multi-dimensional vital sign adaptive dictionary.

[0029] According to an embodiment of the present invention, the above preset convergence condition is that the norm between the multi-dimensional vital sign adaptive dictionary updated in the current stage and the multi-dimensional vital sign adaptive dictionary updated in the previous stage is less than a preset value.

[0030] According to an embodiment of the present invention, the above distributed iterative processing of the multi-signal sparse coefficient representation model by using the alternating direction multiplier algorithm to obtain the multi-signal sparse coefficient matrix representing vital signs includes:

[0031] By introducing an auxiliary processing signal, a penalty parameter of the alternating direction multiplier algorithm, and a Lagrange multiplier matrix, the multi-dimensional optimization objective function in the multi-signal sparse coefficient representation model is converted into an augmented Lagrangian function;

[0032] Using a second-order regularization term optimization algorithm, the augmented Lagrangian function is used to perform distributed alternating iterative updates on the auxiliary processing signal, the penalty parameter, and the Lagrange multiplier matrix until a preset iteration termination condition is satisfied, and a multi-signal sparse coefficient matrix representing vital signs is obtained.

[0033] According to an embodiment of the present invention, the above-mentioned processing of the multi-signal sparse coefficient matrix representing vital signs to obtain multi-dimensional vital sign information of the target user includes:

[0034] Traverse the non-zero elements in the multi-signal sparse coefficient matrix representing vital signs to obtain the position information of the non-zero elements;

[0035] Based on the position information of the non-zero elements, the frequency information corresponding to the non-zero elements is retrieved from the pre-trained multi-dimensional vital sign adaptive dictionary;

[0036] By screening the frequency information corresponding to the non-zero elements, multi-dimensional vital sign information of the target user is obtained.

[0037] A second aspect of the present invention provides an electronic device, including: one or more processors; a memory for storing one or more computer programs, wherein the above-mentioned one or more processors execute the above-mentioned one or more computer programs to implement the steps of the above-mentioned method.

[0038] A third aspect of the present invention further provides a computer-readable storage medium, on which a computer program or instruction is stored, and when the computer program or instruction is executed by a processor, the steps of the above-mentioned method are implemented.

[0039] The non-contact vital sign monitoring method provided by the present invention overcomes the problems of low accuracy and poor adaptability of the traditional fixed dictionary sparse representation method in complex application scenarios by introducing a pre-trained multi-dimensional vital sign adaptive dictionary. The non-contact vital sign monitoring method provided by the present invention enhances the environmental adaptability through the pre-trained multi-dimensional vital sign adaptive dictionary, improves the robustness of the monitoring of vital sign information, and improves the estimation accuracy of vital sign information through the multi-signal sparse coefficient representation model; by fusing physiological prior information into the construction process of the pre-trained multi-dimensional vital sign adaptive dictionary, the accuracy and stability of vital sign information recognition are improved. Description of the Drawings

[0040] Through the following description of the embodiments of the present invention with reference to the accompanying drawings, the above content and other objects, features and advantages of the present invention will become clearer. In the drawings:

[0041] Figure 1 is an application scenario diagram of a non-contact vital sign monitoring method according to an embodiment of the present invention;

[0042] Figure 2 is a flowchart of a non-contact vital sign monitoring method according to an embodiment of the present invention;

[0043] Figure 3 is a process diagram of a vital sign detection method based on an adaptive dictionary according to an embodiment of the present invention;

[0044] Figure 4 is a schematic structural diagram of a non-contact vital sign monitoring device according to an embodiment of the present invention;

[0045] Figure 5 is a block diagram of an electronic device suitable for implementing a non-contact vital sign monitoring method according to an embodiment of the present invention. Detailed Embodiments

[0046] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. However, it should be understood that these descriptions are merely exemplary and are not intended to limit the scope of the present invention. In the following detailed description, for the sake of explanation, many specific details are set forth to provide a comprehensive understanding of the embodiments of the present invention. However, obviously, one or more embodiments can also be implemented without these specific details. In addition, in the following description, descriptions of well-known structures and technologies are omitted to avoid unnecessarily confusing the concepts of the present invention.

[0047] The terms used herein are merely for describing specific embodiments and are not intended to limit the present invention. The terms "including", "comprising", etc. used herein indicate the presence of the described features, steps, operations, and / or components, but do not exclude the presence or addition of one or more other features, steps, operations, or components.

[0048] All terms used herein (including technical and scientific terms) have the meanings commonly understood by those skilled in the art, unless otherwise defined. It should be noted that the terms used herein should be interpreted as having a meaning consistent with the context of this specification and should not be interpreted in an idealized or overly rigid manner.

[0049] In the case of using expressions such as "at least one of A, B, and C, etc.", generally, it should be interpreted according to the meaning that those skilled in the art usually understand this expression (for example, "a system having at least one of A, B, and C" should include, but not be limited to, a system having only A, only B, only C, having A and B, having A and C, having B and C, and / or having A, B, and C, etc.).

[0050] Existing non-contact vital sign monitoring methods mainly include: using Wi-Fi or radar devices to collect continuous environmental signals; extracting the amplitude or phase change of the signals as a reflection of human micro-movement; estimating the dominant frequency by means of band-pass filtering, FFT, wavelet transform, etc.; and estimating the respiration and heart rate frequencies by observing the spectral peaks. However, existing non-contact vital sign monitoring methods have technical problems such as poor environmental adaptability, inaccurate detected vital sign information, and relatively poor robustness of the analysis framework or model adopted.

[0051] To solve one of the problems in the prior art, the present invention provides a non-contact vital sign monitoring method. By collecting wireless signals around the human body (such as the channel state information of Wi-Fi), constructing a time-frequency feature representation of the signals, and designing an adaptive dictionary for sparse representation, the dominant frequencies of respiration and heartbeats are extracted. Among them, the adaptive dictionary is dynamically generated according to the actually collected data, integrating the periodic characteristics of physiological signals and prior knowledge, which improves the accuracy and robustness of frequency estimation. Compared with the traditional sparse representation method using a fixed dictionary, the present invention can stably extract the vital sign frequencies under different environments, postures, and noise interferences, with strong adaptability and small errors, and is especially suitable for complex scenarios such as non-contact and continuous monitoring.

[0052] Figure 1 It is an application scenario diagram of the non-contact vital sign monitoring method according to an embodiment of the present invention.

[0053] As Figure 1 shown, the application scenario 100 according to this embodiment may include scenarios such as wireless signal processing, physiological parameter detection, and health monitoring. The network 104 is used to provide a medium for communication links between the first terminal device 101, the second terminal device 102, the third terminal device 103, and the server 105. The network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables, etc.

[0054] Users can use the first terminal device 101, the second terminal device 102, and the third terminal device 103 to interact with the server 105 via the network 104 to receive or send messages, etc. Various communication client applications can be installed on the first terminal device 101, the second terminal device 102, and the third terminal device 103, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social platform software, etc. (for example only).

[0055] The first terminal device 101, the second terminal device 102, and the third terminal device 103 can be various electronic devices with a display screen and supporting web browsing, including but not limited to smartphones, tablets, laptop computers, desktop computers, and so on.

[0056] The server 105 can be a server that provides various services, such as a background management server that supports the websites browsed by users using the first terminal device 101, the second terminal device 102, and the third terminal device 103 (for example only). The background management server can analyze and process data such as user requests received, and feedback the processing results (such as web pages, information, or data obtained or generated according to user requests) to the terminal device.

[0057] It should be noted that the non-contact vital sign monitoring method provided by the embodiments of the present invention can generally be executed by the server 105. Correspondingly, the non-contact vital sign monitoring device provided by the embodiments of the present invention can generally be set in the server 105. The non-contact vital sign monitoring method provided by the embodiments of the present invention can also be executed by a server or a server cluster different from the server 105 and capable of communicating with the first terminal device 101, the second terminal device 102, the third terminal device 103, and / or the server 105. Correspondingly, the non-contact vital sign monitoring device provided by the embodiments of the present invention can also be set in a server or a server cluster different from the server 105 and capable of communicating with the first terminal device 101, the second terminal device 102, the third terminal device 103, and / or the server 105.

[0058] It should be understood that Figure 1 the numbers of terminal devices, networks, and servers in

[0059] are merely illustrative. According to the implementation requirements, there can be any number of terminal devices, networks, and servers.

[0059] Based on the Figure 1 scenario described below, the non-contact vital sign monitoring method of the disclosed embodiments will be described in detail through Figures 2 - 3 .

[0060] It should be noted that in the embodiments of this application, some existing industry solutions such as certain software, components, models, etc. may be mentioned. They should be regarded as exemplary. The purpose is only to illustrate the feasibility in the implementation of the technical solution of this application, but it does not mean that the applicant has already or necessarily used this solution.

[0061] At the same time, in the technical solution of the present invention, the user information involved (including but not limited to user personal information, user image information, user device information, such as location information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) are all information and data that have been authorized by the user or fully authorized by all parties. And the processing of relevant data such as collection, storage, use, processing, transmission, provision, disclosure, and application complies with relevant laws, regulations, and standards, takes necessary confidentiality measures, does not violate public order and good customs, and provides corresponding operation entrances for users to choose to authorize or refuse.

[0062] In addition, in the scenario of making automated decisions using personal information, the methods, devices, and systems provided in the embodiments of the present invention all provide corresponding operation entrances for users to choose to agree or refuse the results of automated decisions; if the user chooses to refuse, the expert decision-making process will be entered. The expression "automated decision" here refers to the activity of automatically analyzing and evaluating an individual's behavior habits, interests and hobbies, or economic, health, credit status, etc. through a computer program and making a decision. The expression "expert decision" here refers to the activity of making a decision by a person who specializes in a certain field, has specialized experience, knowledge, and skills, and has reached a certain professional level.

[0063] Figure 2 It is a flowchart of a non-contact vital sign monitoring method according to an embodiment of the present invention.

[0064] As Figure 2 shown, the non-contact vital sign monitoring method of this embodiment includes operations S210 to S230.

[0065] In operation S210, based on the authorization of the target user, the channel state information of the target user is collected in real time through non-contact detection technology, and feature extraction is performed on the channel state information in the time-frequency domain to obtain wireless time-frequency features.

[0066] After obtaining the authorization of the target user, the channel state information around the target user is acquired. Among them, the channel state information includes WiFi signals, millimeter-wave radar signals, etc. Feature extraction is performed on the acquired channel state information in the frequency domain and time domain to obtain wireless time-frequency features.

[0067] In operation S220, a multi-signal sparse coefficient representation model based on a second-order regularization paradigm is constructed using a pre-trained multi-dimensional vital sign adaptive dictionary, wireless time-frequency features, and a preset threshold. The pre-trained multi-dimensional vital sign adaptive dictionary is obtained through iterative convex optimization operations. The multi-dimensional vital sign adaptive dictionary is a learnable template library with multi-dimensional physiological feature information.

[0068] The pre-trained multi-dimensional vital sign adaptive dictionary is a dictionary that integrates the prior information of the organism's periodic physiology.

[0069] In this application, the dictionary is a set composed of basis vectors (or feature vectors).

[0070] In this application, the input signal is represented as a sparse linear combination of the dictionary as shown in formula (1):

[0071] (1),

[0072] where represents the sparse coefficient vector, most of whose elements are 0, represents a vector space with dimensions. If the dictionary is fixed, it is called a static dictionary; if the dictionary is automatically updated or trained and generated according to the input data, it is called an adaptive dictionary. The adaptive dictionary generates a dictionary that better suits a specific task by learning the statistical features in the data, thereby enhancing the expression ability. The core idea of the adaptive dictionary is: given the training data set , simultaneously solve the optimal dictionary and the sparse coefficient such that the reconstruction error is minimized and the sparsity is maximized, which can be expressed by formula (2):

[0073] (2),

[0074] where represents each column vector in matrix D, is the regularization parameter that controls the degree of sparsity, represents the square of the L2 norm (the same below), represents the L2 norm (the same below).

[0075] In the monitoring of respiration and heartbeat, the use of an adaptive dictionary is like a "learning template library". If a set of "templates" is used to represent or identify a person's current respiration and heartbeat states, a static dictionary is a fixed template that uses a set of unchanging reference patterns to fit, regardless of whether the respiration and heartbeat are fast or slow; while an adaptive dictionary will automatically update the templates according to the recent changes in the subject's respiration and heartbeat. For example, after the human body exercises, the respiration and heartbeat become faster, and the dictionary will add these "fast-paced" templates; after the human body falls asleep, the respiration becomes shallow and slow, and the dictionary will also capture this change. The present invention adopts an adaptive dictionary to effectively cope with complex and changeable environments and improve the detection accuracy of human physiological parameters.

[0076] In operation S230, the alternating direction multiplier algorithm is used to perform distributed iterative processing on the multi-signal sparse coefficient representation model to obtain a multi-signal sparse coefficient matrix representing vital signs, and the multi-signal sparse coefficient matrix representing vital signs is processed to obtain multi-dimensional vital sign information of the target user.

[0077] According to an embodiment of the present invention, the above multi-dimensional vital sign information includes respiration rate and heartbeat rate; among them, the pre-trained multi-dimensional vital sign adaptive dictionary includes a pre-trained respiration rate adaptive dictionary and a pre-trained heartbeat rate adaptive dictionary; among them, the multi-signal sparse coefficient matrix includes a respiration rate sparse coefficient matrix and a heartbeat rate sparse coefficient matrix.

[0078] It should be particularly noted that the above non-contact vital sign monitoring method provided by the present invention is used to improve the detection accuracy of the vital sign information of the target user under complex scenarios or multi-scenario conditions. The detected vital sign information generally refers to respiration information and heartbeat information, which belongs to intermediate information in the field of medical health. The purpose of the present invention is not to directly diagnose the health status of the target user; in addition, the above operations S210~S230 and other operations of the embodiments of the present invention are all operations for processing intermediate information implemented by devices such as computers.

[0079] According to an embodiment of the present invention, the above-mentioned pre-trained multi-dimensional vital sign adaptive dictionary is obtained through iterative convex optimization operations, including: by introducing a sine function based on physiological prior information and performing non-linear constraints on the joint optimization objective function for the multi-dimensional vital sign adaptive dictionary, a joint optimization objective function with prior constraints is obtained, where the physiological prior information includes the periodic physiological signals of an organism; randomly selecting multiple wireless time-frequency samples from the wireless time-frequency feature samples related to vital signs to initialize the multi-dimensional vital sign adaptive dictionary, obtaining the initialized multi-dimensional vital sign adaptive dictionary; solidifying the parameters of the initialized multi-dimensional vital sign adaptive dictionary, using the joint optimization objective function, and performing a convex optimization operation on the training sparse coefficient matrix corresponding to the multi-dimensional vital sign adaptive dictionary by introducing a weighted quadratic diagonal matrix; numerically solidifying the convex-optimized training sparse coefficient matrix, and performing gradient descent processing based on Lagrange multipliers on the initialized multi-dimensional vital sign adaptive dictionary using the joint optimization objective function with prior constraints, obtaining the processed multi-dimensional vital sign adaptive dictionary; iteratively performing convex optimization operations and gradient descent processing operations until a preset training condition is met, obtaining the pre-trained multi-dimensional vital sign adaptive dictionary.

[0080] Taking respiration and heartbeat as vital sign information below, the pre-training process of the above-mentioned pre-trained multi-dimensional vital sign adaptive dictionary will be further described in detail.

[0081] In the vital sign perception method based on sparse representation, the existing technical solutions use fixed bases to construct dictionaries. The advantages of this type of method are simple implementation and relatively small computational overhead, and usually do not need to rely on actual data for dictionary design. However, in the face of different usage scenarios or individual differences, fixed dictionaries are difficult to effectively capture the change characteristics of signals, resulting in weak adaptability to environmental changes, insufficient robustness, and affecting detection accuracy and stability. Therefore, the present invention adopts a data-driven adaptive dictionary construction method.

[0082] Constructing an adaptive dictionary model for respiration and heartbeat is shown in formula (3):

[0083] (3),

[0084] where X and Y represent the samples of respiration and heartbeat signals in the frequency domain, represents the adaptive dictionary of the respiration signal to be designed, represents the adaptive dictionary of the heartbeat signal to be designed, represents the transpose matrix of (the same below), represents The transposed matrix (the same below), R and H represent the sparse coefficient matrices corresponding to the adaptive dictionary, and α, β, λ1, λ2, γ are different regularization factors. denotes L 2,1 norm (the same below). denotes L F the square of the norm (the same below). The reason for choosing the L 2,1 norm is that within a short period of time, the breathing and heart rate of the human body usually remain stable, and the perception results of multiple measurements are basically the same. Therefore, this norm can be used to construct a sparse representation under multiple measurements, so as to more accurately characterize the signal features.

[0085] During the process of updating the adaptive dictionary, based on the periodic waveform properties of the breathing and heartbeat signals, this paper uses the sine function waveform to approximate the breathing and heartbeat signal waveforms and uses it as a constraint for updating the adaptive dictionary, which helps to prevent overfitting, reduce redundant information, and improve the signal reconstruction quality. In addition, adding physiological prior constraints can ensure the rationality of the signal, improve the computational stability, accelerate the convergence speed, and make the sparse representation model more robust and generalization-capable in complex environments. The periodic constraints on the adaptive dictionary are shown in formulas (4) and (5):

[0086] (4),

[0087] (5),

[0088] where, denotes a certain element in denotes a certain element in and are the L 2,1 norms of R and H. Although the L 2,1 norm is convex, due to its non-smoothness, non-differentiability at zero points, and the complexity of the optimization equation, the closed-form solution cannot be obtained by traditional convex optimization methods.

[0089] According to the embodiments of the present invention, the above-mentioned convex optimization operation on the training sparse coefficient matrix corresponding to the multi-dimensional vital sign self-adaptive dictionary by introducing a weighted quadratic diagonal matrix includes: calculating the Euclidean norm of each column vector in the training sparse coefficient matrix; obtaining the second-order norm of the training sparse coefficient matrix based on the Euclidean norm of each column vector; performing a twice-differentiable simulation on the second-order norm of the training sparse coefficient matrix by introducing a weighted quadratic diagonal matrix; converting the joint optimization objective function into a convex optimization objective function by using the result of the twice-differentiable simulation; and performing function expansion and derivative processing on the convex optimization objective function to obtain the convex-optimized training sparse coefficient matrix.

[0090] According to an embodiment of the present invention, the above-mentioned gradient descent processing based on Lagrange multipliers for the initialized multi-dimensional vital sign adaptive dictionary using the joint optimization objective function with prior constraints to obtain the processed multi-dimensional vital sign adaptive dictionary includes: expanding the joint optimization objective function with prior constraints into a Lagrangian function after introducing Lagrange multipliers for each prior constraint term in the joint optimization objective function with prior constraints; fixing the Lagrange multipliers, calculating the gradient information corresponding to the initialized multi-dimensional vital sign adaptive dictionary using the Lagrangian function, and updating the parameters of the initialized multi-dimensional vital sign adaptive dictionary using the gradient information; fixing the parameters of the updated multi-dimensional vital sign adaptive dictionary and updating the Lagrange multipliers; iteratively performing gradient calculation, adaptive dictionary parameter update, and Lagrange multiplier update until a preset convergence condition is satisfied to obtain the processed multi-dimensional vital sign adaptive dictionary.

[0091] According to an embodiment of the present invention, the above-mentioned preset convergence condition is that the norm between the multi-dimensional vital sign adaptive dictionary updated in the current stage and the multi-dimensional vital sign adaptive dictionary updated in the previous stage is less than a preset value.

[0092] Next, taking respiration and heartbeat as vital sign signals, the alternating iterative update process of the multi-dimensional vital sign adaptive dictionary involved in the present invention will be further described in detail through specific embodiments.

[0093] First, initialize the dictionary and , and randomly select n sample data of respiration signal and heartbeat signal data to initialize the dictionaries and .

[0094] Secondly, update the sparse coefficient matrix, fix the dictionaries and , and the update problems of the sparse coefficient matrices R and H are shown in formula (6):

[0095] (6).

[0096] The norm of matrix R is defined as shown in formula (7):

[0097] (7),

[0098] where is the i-th column of matrix R, as shown in formula (8):

[0099] (8).

[0100] So its Euclidean norm is as shown in formula (9):

[0101] (9).

[0102] Therefore, the norm of matrix R is shown in Equation (10):

[0103] (10).

[0104] Since the square root function is not twice differentiable, it is usually approximated by a weighted quadratic form. The present invention introduces a diagonal matrix, as shown in Equation (11):

[0105] (11),

[0106] Then, can be approximated as shown in Equation (12):

[0107] (12),

[0108] This can be written in matrix form . Where , represents the i-th row of R, and represents the i-th element of the diagonal matrix. The same applies to the heartbeat signal matrix H.

[0109] Therefore, the above problems can be regarded as convex problems with respect to matrices R and H respectively. By taking the derivative and setting it to 0, the closed-form solutions of matrices R and H can be obtained. Expanding the objective function gives Equation (13):

[0110] (13).

[0111] Taking the derivative of matrix R and setting it to 0 gives Equation (14):

[0112] (14).

[0113] From this, the sparse coefficient matrix corresponding to respiration in the pre-training stage, as shown in Equation (15), can be obtained:

[0114] (15).

[0115] Similarly, the sparse coefficient matrix corresponding to heartbeat in the pre-training stage, as shown in Equation (16), can be obtained:

[0116] (16).

[0117] Finally, update the dictionary, fix the sparse coefficient matrices R and H, and the dictionary and Update, as shown in Formulas (17) and (18):

[0118] (17),

[0119] (18),

[0120] where X represents the respiratory data sample matrix, represents the heartbeat data sample matrix, represents the identity matrix.

[0121] Since the original objective function is an optimization problem with respect to matrices and and the additional constraints are non - linear constraints imposed on the individual elements of the matrices and the Lagrange multiplier method can be used to solve it. The present invention introduces a set of Lagrange multipliers , corresponding to each constraint term . Then the overall Lagrangian function is as shown in Formula (19):

[0122] (19).

[0123] The gradient of the Lagrangian function is as shown in Formula (20):

[0124] (20),

[0125] where is a matrix with the same dimension as , and its -th element is .

[0126] First, fix the multiplier , and update . Use gradient descent to update , as shown in Formula (21):

[0127] (21).

[0128] Second, fix and update the multiplier , as shown in Formula (22):

[0129] (22).

[0130] Judge the convergence condition, as shown in Formula (23):

[0131] (23).

[0132] For a matrix Similarly

[0133] Repeat the alternating iterative update of the sparse coefficient matrix and the corresponding adaptive dictionary until the objective function value in the adaptive dictionary model of breathing and heartbeat is less than the set threshold , and obtain the pre-trained adaptive dictionary of breathing information and the adaptive dictionary of heartbeat information

[0134] According to an embodiment of the present invention, the above-mentioned distributed iterative processing of the multi-signal sparse coefficient representation model by using the alternating direction multiplier algorithm to obtain the multi-signal sparse coefficient matrix representing vital signs includes: converting the multi-dimensional optimization objective function in the multi-signal sparse coefficient representation model into an augmented Lagrangian function by introducing an auxiliary processing signal, a penalty parameter of the alternating direction multiplier algorithm, and a Lagrangian multiplier matrix; using the second-order regularization term optimization algorithm to perform distributed alternating iterative updates on the auxiliary processing signal, the penalty parameter, and the Lagrangian multiplier matrix through the augmented Lagrangian function until the preset iteration termination condition is satisfied, and obtaining the multi-signal sparse coefficient matrix representing vital signs

[0135] According to an embodiment of the present invention, the above-mentioned processing of the multi-signal sparse coefficient matrix representing vital signs to obtain the multi-dimensional vital sign information of the target user includes: traversing the non-zero value elements in the multi-signal sparse coefficient matrix representing vital signs to obtain the position information of the non-zero value elements; based on the position information of the non-zero value elements, obtaining the frequency information corresponding to the non-zero value elements in advance from the pre-trained multi-dimensional vital sign adaptive dictionary; and obtaining the multi-dimensional vital sign information of the target user by screening the frequency information corresponding to the non-zero value elements

[0136] Next, taking breathing information and heartbeat information as examples, the above-mentioned process of obtaining the multi-dimensional vital sign information of the target user involved in the present invention will be further described in detail

[0137] Based on the pre-trained multi-dimensional vital sign (breathing information and heartbeat information) adaptive dictionary, obtain an effective sparse representation of the target breathing and heartbeat signals, as shown in formula (24):

[0138] (24),

[0139] Among them, and are sparse coefficient matrices. The matrices and obtained by solving this problemThe non-zero elements are the estimated respiratory and heart rate values. If X and Y are matrices, then the entire sparse representation problem can be generalized to the Multiple Measurement Vectors (MMV) problem. In this case, it is necessary to solve the sparse representation of multiple observed signals. The present invention uses the ADMM method to solve this sparse representation problem and transforms the original problem into the form shown in formula (25):

[0140] (25).

[0141] Introduce an auxiliary variable such that , as shown in formula (26):

[0142] (26).

[0143] Introduce the Lagrange multiplier matrix U and construct the augmented Lagrangian function, as shown in formula (27):

[0144] (27),

[0145] where ρ is the penalty parameter of ADMM.

[0146] The ADMM alternating optimization needs to update in turn. First, it is necessary to update , as shown in formula (28):

[0147] (28).

[0148] Take the derivative of and set the derivative to 0, as shown in formula (29):

[0149] (29),

[0150] Obtain formula (30), (30).

[0151] Update Z, as shown in formula (31):

[0152] (31).

[0153] Formula (31) is an optimization problem with a regularization term, and its solution can be obtained by the row-level soft threshold function, as shown in formula (32):

[0154] (32),

[0155] where .

[0156] Update U

[0157] The condition for stopping the ADMM iteration can be set as shown in formula (33):

[0158] (33)

[0159] For the solution is the same.

[0160] Next, in combination with Figure 3 the above non-contact vital sign monitoring method provided by the present invention will be further described in detail.

[0161] Figure 3 is a process diagram of a vital sign detection method based on an adaptive dictionary according to an embodiment of the present invention.

[0162] As Figure 3 shown, first, based on user authorization, a CSI data set around the user is collected, and the CSI data set is preprocessed to obtain CSI training data samples; an adaptive dictionary model is constructed, such as a respiration information adaptive dictionary model and a heartbeat information adaptive dictionary model, and the constructed adaptive dictionary model is initialized through the CSI training data samples. During the iterative update process of the adaptive dictionary model, first, the parameters of the adaptive dictionary model are fixed, and the sparse coefficient matrix corresponding to the adaptive dictionary model is updated; then, the value of the updated sparse coefficient matrix is fixed, and the initialized adaptive dictionary model is updated until the objective function value is less than a set threshold, and a pre-trained multi-dimensional vital sign adaptive dictionary (or dictionary model) is obtained. Then, the pre-trained multi-dimensional vital sign adaptive dictionary is used to solve the multi-dimensional sparse coefficient matrix, and the vital sign information, such as respiration rate and heartbeat rate, is output using the multi-dimensional sparse coefficient matrix.

[0163] Through the above embodiments or specific implementation manners, the non-contact vital sign monitoring method provided by the present invention has been described in detail. Through the above embodiments or specific implementation manners, it can be seen that compared with the existing sparse representation method using a fixed basis to construct a dictionary, the sparse representation vital sign frequency estimation method based on a pre-trained adaptive dictionary provided by the present invention has significantly improved adaptability and robustness. The core technical feature of the method provided by the present invention is that it does not rely on a preset fixed basis, but dynamically constructs a dictionary in a data-driven manner according to the actually collected CSI or other wireless sensing signals, and captures the atomic elements in the signal that are highly correlated with the respiration and heartbeat frequency characteristics. The advantages and technical effects of this technology are mainly reflected in the following aspects:

[0164] (1) Enhance environmental adaptability: Since the dictionary is adaptively constructed based on the current scenario and target state, it can better reflect the change characteristics of signals in different environments, different individuals, or different postures, and is applicable to complex and variable vital sign monitoring scenarios.

[0165] (2) Improve the accuracy of frequency estimation: The adaptive dictionary can effectively suppress the interference components unrelated to the target frequency, enhance the expression ability of the target frequency components during the sparse solution process, and thus improve the estimation accuracy of respiratory and heart rates.

[0166] (3) Enhance robustness: Compared with the fixed dictionary method whose performance drops sharply in scenarios with low signal-to-noise ratio or large channel perturbations, the present invention significantly enhances the robustness to noise and multipath effects by adjusting the dictionary structure in real time to maintain a stable representation ability.

[0167] (4) Flexibly integrate various prior knowledge: This method supports embedding physiological models (such as periodic characteristics, sine wave constraints, etc.) as constraints into the dictionary construction process to further improve the stability and accuracy of frequency recognition.

[0168] It can be seen that by introducing the adaptive dictionary mechanism, the present invention fundamentally overcomes the problems of low accuracy and poor adaptability of traditional fixed dictionary sparse representation methods in complex application scenarios, and has stronger practicality and promotion value.

[0169] Based on the above non-contact vital sign monitoring method, the present invention also provides a non-contact vital sign monitoring device. The following will be combined with Figure 4 to describe this device in detail.

[0170] Figure 4 is a schematic structural diagram of a non-contact vital sign monitoring device according to an embodiment of the present invention.

[0171] As Figure 4 shown, the above non-contact vital sign monitoring device 400 includes a CSI data acquisition and preprocessing module 410, a multi-signal sparse coefficient representation model construction module 420, and a multi-dimensional vital sign information acquisition module 430.

[0172] The CSI data acquisition and preprocessing module 410 is used to, based on the authorization of the target user, collect the channel state information of the target user in real time through non-contact detection technology, and perform feature extraction on the channel state information in the time-frequency domain to obtain wireless time-frequency features; in one embodiment, the CSI data acquisition and preprocessing module 410 can be used to perform the operation S210 described above, which will not be elaborated here.

[0173] The multi-signal sparse coefficient representation model construction module 420 is configured to construct a multi-signal sparse coefficient representation model based on a second-order regularization paradigm by using a pre-trained multi-dimensional vital sign adaptive dictionary, wireless time-frequency features, and a preset threshold. The pre-trained multi-dimensional vital sign adaptive dictionary is obtained through iterative convex optimization operations. The multi-dimensional vital sign adaptive dictionary is a learnable template library with multi-dimensional physiological feature information. In one embodiment, the multi-signal sparse coefficient representation model construction module 420 can be used to perform the operation S220 described above, which will not be elaborated here.

[0174] The multi-dimensional vital sign information acquisition module 430 is configured to perform distributed iterative processing on the multi-signal sparse coefficient representation model by using the alternating direction multiplier algorithm to obtain a multi-signal sparse coefficient matrix representing vital signs, and process the multi-signal sparse coefficient matrix representing vital signs to obtain the multi-dimensional vital sign information of the target user. In one embodiment, the multi-dimensional vital sign information acquisition module 430 can be used to perform the operation S230 described above, which will not be elaborated here.

[0175] According to an embodiment of the present invention, any multiple of the CSI data acquisition and preprocessing module 410, the multi-signal sparse coefficient representation model construction module 420, and the multi-dimensional vital sign information acquisition module 430 can be combined and implemented in one module, or any one of them can be split into multiple modules. Or, at least part of the functions of one or more of these modules can be combined with at least part of the functions of other modules and implemented in one module. According to an embodiment of the present invention, at least one of the CSI data acquisition and preprocessing module 410, the multi-signal sparse coefficient representation model construction module 420, and the multi-dimensional vital sign information acquisition module 430 can be at least partially implemented as a hardware circuit, such as a field programmable gate array (FPGA), a programmable logic array (PLA), a system on chip, a system on substrate, a system on package, an application specific integrated circuit (ASIC), or can be implemented by any other reasonable way of integrating or packaging circuits and other hardware or firmware, or can be implemented in any one of the three implementation manners of software, hardware, and firmware or in any appropriate combination of several of them. Or, at least one of the CSI data acquisition and preprocessing module 410, the multi-signal sparse coefficient representation model construction module 420, and the multi-dimensional vital sign information acquisition module 430 can be at least partially implemented as a computer program module, which can perform corresponding functions when the computer program module is run.

[0176] Figure 5 It is a block diagram of an electronic device suitable for implementing the non-contact vital sign monitoring method according to an embodiment of the present invention.

[0177] As Figure 5As shown, the electronic device 500 according to an embodiment of the present invention includes a processor 501, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 502 or a program loaded from a storage section 508 into a random access memory (RAM) 503. The processor 501 may include, for example, a general-purpose microprocessor (e.g., CPU), an instruction set processor, and / or a related chipset, and / or a dedicated microprocessor (e.g., an application specific integrated circuit (ASIC)), etc. The processor 501 may also include on-board memory for caching purposes. The processor 501 may include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of the present invention.

[0178] In the RAM 503, various programs and data required for the operation of the electronic device 500 are stored. The processor 501, the ROM 502, and the RAM 503 are connected to each other via a bus 504. The processor 501 performs various operations of the method flow according to an embodiment of the present invention by executing the program in the ROM 502 and / or the RAM 503. It should be noted that the program may also be stored in one or more memories other than the ROM 502 and the RAM 503. The processor 501 may also perform various operations of the method flow according to an embodiment of the present invention by executing the program stored in the one or more memories.

[0179] According to an embodiment of the present invention, the electronic device 500 may further include an input / output (I / O) interface 505, and the input / output (I / O) interface 505 is also connected to the bus 504. The electronic device 500 may further include one or more of the following components connected to the input / output (I / O) interface 505: an input section 506 including a keyboard, a mouse, etc.; an output section 507 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc. and a speaker, etc.; a storage section 508 including a hard disk, etc.; and a communication section 509 including a network interface card such as a LAN card, a modem, etc. The communication section 509 performs communication processing via a network such as the Internet. A drive 510 is also connected to the input / output (I / O) interface 505 as needed. A removable medium 511, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 510 as needed so that a computer program read therefrom can be installed into the storage section 508 as needed.

[0180] The present invention also provides a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiments; or may exist independently without being assembled into the device / apparatus / system. The above computer-readable storage medium carries one or more programs, and when the above one or more programs are executed, the method according to the embodiments of the present invention is implemented.

[0181] According to an embodiment of the present invention, the computer-readable storage medium may be a non-volatile computer-readable storage medium, for example, it may include but is not limited to: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the above. In the present invention, the computer-readable storage medium may be any tangible medium that contains or stores a program, and this program can be used by or in combination with an instruction execution system, apparatus, or device. For example, according to an embodiment of the present invention, the computer-readable storage medium may include the above-described ROM 502 and / or RAM 503 and / or one or more memories other than ROM 502 and RAM 503.

[0182] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagram may represent a module, a program segment, or a part of code, and the above module, program segment, or part of code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than marked in the accompanying drawings. For example, two consecutive blocks shown may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram or flowchart, and the combination of blocks in the block diagram or flowchart, can be implemented by a dedicated hardware-based system for performing the specified functions or operations, or can be implemented by a combination of dedicated hardware and computer instructions.

[0183] Those skilled in the art can understand that the features described in the various embodiments of the present invention can be combined and / or combined in various ways, even if such combinations or combinations are not explicitly described in the present invention. In particular, without departing from the spirit and teachings of the present invention, the features described in the various embodiments of the present invention can be combined and / or combined in various ways. All such combinations and / or combinations fall within the scope of the present invention.

[0184] The embodiments of the present invention have been described above. However, these embodiments are for illustrative purposes only and are not intended to limit the scope of the present invention. Although the embodiments have been described separately above, this does not mean that the measures in each embodiment cannot be used advantageously in combination. Without departing from the scope of the present invention, those skilled in the art can make various substitutions and modifications, and all such substitutions and modifications should fall within the scope of the present invention.

Claims

1. A non-contact vital sign monitoring method, characterized in that, The method includes: Based on the authorization of the target user, the channel state information of the target user is collected in real time through non-contact detection technology, and feature extraction is performed on the channel state information in the time-frequency domain to obtain wireless time-frequency features; Using a pre-trained multi-dimensional vital sign adaptive dictionary, the wireless time-frequency features, and a preset threshold to construct a multi-signal sparse coefficient representation model based on a second-order regularization paradigm, where the pre-trained multi-dimensional vital sign adaptive dictionary is obtained through iterative convex optimization operations, and the multi-dimensional vital sign adaptive dictionary is a learnable template library with multi-dimensional physiological feature information; Using the alternating direction multiplier algorithm to perform distributed iterative processing on the multi-signal sparse coefficient representation model to obtain a multi-signal sparse coefficient matrix representing vital signs, and processing the multi-signal sparse coefficient matrix representing vital signs to obtain the multi-dimensional vital sign information of the target user.

2. The method according to claim 1, wherein The multi-dimensional vital sign information includes respiratory rate and heart rate; Among them, the pre-trained multi-dimensional vital sign adaptive dictionary includes a pre-trained respiratory rate adaptive dictionary and a pre-trained heart rate adaptive dictionary; Among them, the multi-signal sparse coefficient matrix includes a respiratory rate sparse coefficient matrix and a heart rate sparse coefficient matrix.

3. The method according to claim 1, wherein The pre-trained multi-dimensional vital sign adaptive dictionary is obtained through iterative convex optimization operations, including: By introducing a sine function based on physiological prior information and performing non-linear constraints on the joint optimization objective function for the multi-dimensional vital sign adaptive dictionary, a joint optimization objective function with prior constraints is obtained, where the physiological prior information includes the periodic physiological signals of organisms; Randomly selecting multiple wireless time-frequency samples from the wireless time-frequency feature samples related to the vital signs to initialize the multi-dimensional vital sign adaptive dictionary to obtain the initialized multi-dimensional vital sign adaptive dictionary; Fixing the parameters of the initialized multi-dimensional vital sign adaptive dictionary, and using the joint optimization objective function to perform a convex optimization operation on the training sparse coefficient matrix corresponding to the multi-dimensional vital sign adaptive dictionary by introducing a weighted quadratic diagonal matrix; Numerically fixing the convex-optimized training sparse coefficient matrix, and using the joint optimization objective function with prior constraints to perform gradient descent processing based on Lagrange multipliers on the initialized multi-dimensional vital sign adaptive dictionary to obtain the processed multi-dimensional vital sign adaptive dictionary; Iteratively perform convex optimization operations and gradient descent processing operations until the preset training conditions are met to obtain the pre-trained multi-dimensional vital sign adaptive dictionary.

4. The method according to claim 3, characterized in that Performing a convex optimization operation on the training sparse coefficient matrix corresponding to the multi-dimensional vital sign adaptive dictionary by introducing a weighted quadratic diagonal matrix includes: Calculating the Euclidean norm of each column vector in the training sparse coefficient matrix; Obtaining the second-order norm of the training sparse coefficient matrix based on the Euclidean norm of each column vector; Performing a twice-differentiable simulation on the second-order norm of the training sparse coefficient matrix by introducing a weighted quadratic diagonal matrix; Convert the joint optimization objective function into a convex optimization objective function by using the results of twice-differentiable simulation; Perform function expansion and derivative processing on the convex optimization objective function to obtain the trained sparse coefficient matrix after convex optimization.

5. The method according to claim 3, characterized in that, Perform gradient descent processing based on Lagrange multipliers on the initialized multi-dimensional vital sign self-adaptive dictionary by using the joint optimization objective function with prior constraints, and the processed multi-dimensional vital sign adaptive dictionary obtained includes: After introducing Lagrange multipliers for each prior constraint term in the joint optimization objective function with prior constraints, expand the joint optimization objective function with prior constraints into a Lagrangian function; Fix the Lagrange multipliers, calculate the gradient information corresponding to the initialized multi-dimensional vital sign self-adaptive dictionary by using the Lagrangian function, and update the parameters of the initialized multi-dimensional vital sign self-adaptive dictionary by using the gradient information; Fix the parameters of the updated multi-dimensional vital sign self-adaptive dictionary and update the Lagrange multipliers; Iteratively perform gradient calculation, adaptive dictionary parameter update, and Lagrange multiplier update until a preset convergence condition is met to obtain the processed multi-dimensional vital sign adaptive dictionary.

6. The method according to claim 5, characterized in that, The preset convergence condition is that the norm between the multi-dimensional vital sign adaptive dictionary updated in the current stage and the multi-dimensional vital sign adaptive dictionary updated in the previous stage is less than a preset value.

7. The method according to claim 1, characterized in that Perform distributed iterative processing on the multi-signal sparse coefficient representation model by using the alternating direction multiplier algorithm, and the multi-signal sparse coefficient matrix representing vital signs obtained includes: Convert the multi-dimensional optimization objective function in the multi-signal sparse coefficient representation model into an augmented Lagrangian function by introducing an auxiliary processing signal, the penalty parameter of the alternating direction multiplier algorithm, and a Lagrange multiplier matrix; Use the second-order regularization term optimization algorithm to perform distributed alternating iterative updates on the auxiliary processing signal, the penalty parameter, and the Lagrange multiplier matrix by using the augmented Lagrangian function until a preset iteration termination condition is met to obtain the multi-signal sparse coefficient matrix representing vital signs.

8. The method according to claim 1, characterized in that, Process the multi-signal sparse coefficient matrix representing vital signs to obtain the multi-dimensional vital sign information of the target user, including: Traverse the non-zero value elements in the multi-signal sparse coefficient matrix representing vital signs to obtain the position information of the non-zero value elements; Based on the position information of the non-zero value elements, extract the frequency information corresponding to the non-zero value elements from the pre-trained multi-dimensional vital sign self-adaptive dictionary; Obtain the multi-dimensional vital sign information of the target user by screening the frequency information corresponding to the non-zero value elements.

9. An electronic device, comprising: One or more processors; A memory for storing one or more computer programs, Characterized in that the one or more processors execute the one or more computer programs to implement the steps of the method according to any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program or instructions stored thereon, characterized in that, The computer program or instruction, when executed by the processor, implements the steps of the method according to any one of claims 1 to 8.

Images