High-precision biological perception method and system based on millimeter-wave radar

Abstract

The present application relates to the field of millimeter wave radar detection technology, and specifically to a high-precision biological perception method and system based on millimeter wave radar. It collects millimeter wave detection transmission signals and millimeter wave detection echo signals, and introduces a signal processing and analysis algorithm based on artificial intelligence and deep learning at the back end to decompose and analyze the signal intrinsic modes of the millimeter wave detection transmission signal and the millimeter wave detection echo signal, so as to capture the key semantic aggregation features of the frequency domain difference of the signal components between the millimeter wave detection transmission and the echo signal, and then express the perception results based on the aggregated representation of the frequency domain difference features of the signal components, and use the perception results as the recognition results of biological perception, thereby improving the accuracy of biological perception using millimeter wave radar detection.

Description

Technical Field

[0001] The present application relates to the field of millimeter wave radar detection technology, and in particular to a high-precision biological perception method and system based on millimeter wave radar. Background Art

[0002] Millimeter-wave radar is a radar that works in the millimeter-wave band. It has the characteristics of high resolution, wide working frequency band, large value Doppler frequency response, short wavelength, easy to obtain target detail features and clear contour imaging, etc., which is particularly suitable for target classification and identification. Biosensing covers the comprehensive perception and analysis of the state and activities of organisms. Biomonitoring is an important part of biosensing. It refers to the real-time monitoring and analysis of the physiological state and movement state of the human body through various technical means to obtain the biological information of the detected person. By detecting biological information through millimeter-wave radar, the characteristics of organisms can be perceived, which can greatly improve the recognition effect of the detected biological characteristics. Biosensing technology using millimeter-wave radar has been used in many fields. For example, millimeter-wave radar is used to detect pedestrians around in autonomous driving technology, and millimeter-wave radar is used to monitor patient falls, breathing conditions, and sleep conditions in medical treatment.

[0003] For example, a Chinese patent with application number 202211466508.1 discloses a system and method for biometric identification based on millimeter-wave radar, including: a millimeter-wave radar biometric micro-motion detection subsystem: collecting basic data of biometric micro-motion through millimeter-wave radar detection, and obtaining basic data of biometric micro-motion detected by millimeter-wave radar; a biometric identification data model generation subsystem: based on the basic data of biometric micro-motion detected by millimeter-wave radar, the trend of the whole system of biometric movements and the standard classification of the subtle movements of the biometric movement are performed, and a standardized data model of the whole system of biometric micro-motion is established; a feature analysis subsystem of the whole system standardized data model: inputting the data of the biometric to be identified into the standardized big data model of the whole system of biometric micro-motion, and analyzing the features of the whole system of micro-motion of the biometric to be identified; a follow-up detection of the whole system of micro-motion biometric identification subsystem: based on the features of the whole system of micro-motion of the biometric to be identified, the micro-motion of the biometric is detected by millimeter-wave radar, and the whole system of micro-motion biometric identification is performed by millimeter-wave radar detection. Similar to the above-mentioned technologies, based on millimeter-wave detection technology, the main focus is on the processing of biological recognition data models, the deduction of movement trends, etc., but lacks analysis of the detection data of millimeter-wave radar detection of biological bodies, that is, the analysis of millimeter-wave radar transmission signals and echo signals is insufficient, which will directly affect the accuracy of subsequent analysis results and reduce the effectiveness of millimeter-wave radar perception of biological results.

[0004] Therefore, a high-precision biological perception method and system based on millimeter-wave radar is needed to solve the above technical problems. Summary of the invention

[0005] This application is made in consideration of the above problems, and one purpose of this application is to provide a high-precision biological perception method and system based on millimeter-wave radar.

[0006] The embodiment of the present application provides a high-precision biological perception method based on millimeter wave radar, which includes:

[0007] Acquire millimeter wave detection transmission signal and millimeter wave detection echo signal;

[0008] Performing CEEMDAN decomposition on the millimeter wave detection transmission signal and the millimeter wave detection echo signal to obtain a set of IMF components of the millimeter wave detection transmission signal and a set of IMF components of the millimeter wave detection echo signal;

[0009] Performing signal component frequency domain feature extraction on the set of IMF components of the millimeter wave detection transmission signal and the set of IMF components of the millimeter wave detection echo signal to obtain a set of frequency domain feature vectors of the IMF components of the millimeter wave detection transmission signal and a set of frequency domain feature vectors of the IMF components of the millimeter wave detection echo signal;

[0010] Calculate the differential eigenvector between each corresponding group of the frequency domain eigenvectors of the IMF component of the millimeter wave detection transmission signal and the frequency domain eigenvectors of the IMF component of the millimeter wave detection echo signal in the set of the frequency domain eigenvectors of the IMF component of the millimeter wave detection transmission signal and the frequency domain eigenvectors of the IMF component of the millimeter wave detection echo signal to obtain a set of frequency domain differential eigenvectors of the millimeter wave echo-transmission signal component;

[0011] Inputting the set of frequency domain differential feature vectors of the millimeter wave echo-transmitted signal component into a cluster analysis network based on node feature significant marking to obtain a frequency domain differential aggregate representation vector of the millimeter wave echo-reflected signal;

[0012] The perception result is determined based on the frequency domain differential aggregation representation vector of the millimeter wave echo-reflection signal.

[0013] For example, according to the high-precision bio-perception method based on millimeter-wave radar according to an embodiment of the present application, signal component frequency domain feature extraction is performed on the set of millimeter-wave detection transmit signal IMF components and the set of millimeter-wave detection echo signal IMF components to obtain a set of millimeter-wave detection transmit signal IMF component frequency domain feature vectors and a set of millimeter-wave detection echo signal IMF component frequency domain feature vectors, including: inputting each millimeter-wave detection transmit signal IMF component in the set of millimeter-wave detection transmit signal IMF components and each millimeter-wave detection echo signal IMF component in the set of millimeter-wave detection echo signal IMF components into a signal component frequency domain feature extractor based on a 1D-CNN model respectively to obtain a set of millimeter-wave detection transmit signal IMF component frequency domain feature vectors and a set of millimeter-wave detection echo signal IMF component frequency domain feature vectors.

[0014] For example, according to the high-precision bio-sensing method based on millimeter wave radar in an embodiment of the present application, the set of frequency domain differential feature vectors of the millimeter wave echo-transmitted signal component is input into a cluster analysis network based on node feature significant marking to obtain a millimeter wave echo-reflected signal frequency domain differential aggregation representation vector, including:

[0015] Calculating a characteristic significance description factor of each millimeter wave echo-transmitted signal component frequency domain difference feature vector in the set of the millimeter wave echo-transmitted signal component frequency domain difference feature vectors, wherein the characteristic significance description factor is related to the mean and variance of each of the millimeter wave echo-transmitted signal component frequency domain difference feature vectors;

[0016] Constructing a characteristic significance attenuation factor of each of the millimeter wave echo-transmit signal component frequency domain differential eigenvectors based on the distance span between each of the millimeter wave echo-transmit signal component frequency domain differential eigenvectors in the set of the millimeter wave echo-transmit signal component frequency domain differential eigenvectors and the current millimeter wave echo-transmit signal component frequency domain differential eigenvector;

[0017] Modulating the characteristic significance description factor of each of the frequency domain difference feature vectors of the millimeter wave echo-transmitted signal component based on the characteristic significance attenuation factor of each of the frequency domain difference feature vectors of the millimeter wave echo-transmitted signal component to obtain a sequence of characteristic significance attenuation description factors;

[0018] Inputting the sequence of feature significance attenuation description factors into a gated mask module to obtain a sequence of masked feature significance attenuation weight factors;

[0019] Taking the sequence of the mask feature significance attenuation weight factors as the sequence of weights, the weighted sum of the set of frequency domain differential feature vectors of the millimeter wave echo-transmitted signal component is calculated to obtain the frequency domain differential aggregation representation vector of the millimeter wave echo-reflected signal.

[0020] For example, according to the high-precision bio-perception method based on millimeter wave radar in an embodiment of the present application, the characteristic significance description factor of each millimeter wave echo-transmitted signal component frequency domain differential feature vector in the set of the millimeter wave echo-transmitted signal component frequency domain differential feature vector is calculated, wherein the characteristic significance description factor is related to the mean and variance of each of the millimeter wave echo-transmitted signal component frequency domain differential feature vectors, including:

[0021] Respectively calculating the mean and variance of the frequency domain differential feature vector of the millimeter wave echo-transmitted signal component to obtain the frequency domain differential feature mean of the millimeter wave echo-transmitted signal component and the frequency domain differential feature variance of the millimeter wave echo-transmitted signal component;

[0022] After calculating the fourth power of the position difference between the frequency domain differential feature vector of the millimeter wave echo-transmitted signal component and the mean value of the frequency domain differential feature of the millimeter wave echo-transmitted signal component, the expected value of the fourth power modulation semantic deviation vector is calculated to obtain the expected factor of the frequency domain differential feature of the millimeter wave echo-transmitted signal component;

[0023] The position-based division between the expected factor of the frequency domain differential feature of the millimeter wave echo-transmitted signal component and the square of the frequency domain differential feature variance of the millimeter wave echo-transmitted signal component is calculated to obtain a feature significance description factor of the frequency domain differential feature vector of the millimeter wave echo-transmitted signal component.

[0024] For example, according to the high-precision bio-perception method based on millimeter wave radar of an embodiment of the present application, based on the distance span between each millimeter wave echo-transmitted signal component frequency domain differential feature vector in the set of the millimeter wave echo-transmitted signal component frequency domain differential feature vector and the current millimeter wave echo-transmitted signal component frequency domain differential feature vector, a characteristic significance attenuation factor of each of the millimeter wave echo-transmitted signal component frequency domain differential feature vectors is constructed, including:

[0025] Calculating the distance span between other millimeter wave echo-transmitted signal component frequency domain differential eigenvectors in the set of the millimeter wave echo-transmitted signal component frequency domain differential eigenvectors and the current millimeter wave echo-transmitted signal component frequency domain differential eigenvector to obtain an eigenvalue distance span;

[0026] Extracting the maximum value of the frequency domain differential feature vector of the current millimeter wave echo-transmitted signal component to obtain the maximum value of the frequency domain differential feature of the current millimeter wave echo-transmitted signal component;

[0027] Extracting the maximum value of the frequency domain differential feature vector of the other millimeter wave echo-transmitted signal component to obtain the maximum value of the frequency domain differential feature of the other millimeter wave echo-transmitted signal component;

[0028] The characteristic significance attenuation factor is obtained by calculating the difference between the maximum value of the frequency domain differential feature of the current millimeter wave echo-transmitted signal component and the maximum value of the frequency domain differential feature of other millimeter wave echo-transmitted signal components and dividing the difference by the characteristic value distance span.

[0029] For example, according to the high-precision bio-perception method based on millimeter-wave radar according to the embodiment of the present application, the characteristic significance descriptor of each of the frequency domain differential feature vectors of the millimeter-wave echo-transmitted signal component is modulated based on the characteristic significance attenuation factor of each of the frequency domain differential feature vectors of the millimeter-wave echo-transmitted signal component to obtain a sequence of characteristic significance attenuation description factors, including: calculating the positional product between the characteristic significance attenuation factor of each of the frequency domain differential feature vectors of the millimeter-wave echo-transmitted signal component and the characteristic significance descriptor of each of the frequency domain differential feature vectors of the millimeter-wave echo-transmitted signal component to obtain a sequence of characteristic significance attenuation description factors.

[0030] For example, according to the high-precision biometric perception method based on millimeter wave radar according to an embodiment of the present application, the sequence of feature significance attenuation description factors is input into the gated mask module to obtain a sequence of mask feature significance attenuation weight factors, including:

[0031] Taking the negative number of each feature significance attenuation description factor in the sequence of feature significance attenuation description factors as an exponential power, calculating an exponential function value with a natural constant e as a base to obtain a sequence of feature significance attenuation class support description factors;

[0032] Calculating the reciprocal of the sum of each feature significance attenuation class support description factor in the sequence of feature significance attenuation class support description factors and a constant one to obtain a sequence of feature significance attenuation weight factors;

[0033] Based on a predetermined threshold, the sequence of feature significance attenuation weight factors is masked to obtain a sequence of masked feature significance attenuation weight factors, wherein, in response to each feature significance attenuation weight factor in the sequence of feature significance attenuation weight factors being greater than the predetermined threshold, the feature significance attenuation weight factor greater than the predetermined threshold is taken as the original value, otherwise it is set to zero.

[0034] For example, according to the high-precision biometric perception method based on millimeter wave radar in an embodiment of the present application, the sequence of the mask feature significance attenuation weight factors is used as the sequence of weights, and the weighted sum of the set of frequency domain differential feature vectors of the millimeter wave echo-transmitted signal component is calculated to obtain the millimeter wave echo-reflected signal frequency domain differential aggregation representation vector, including:

[0035] Taking the sequence of the mask feature significance attenuation weight factors as the sequence of weights, calculating the weighted sum between each other millimeter wave echo-transmitted signal component frequency domain differential feature vector in the set of the millimeter wave echo-transmitted signal component frequency domain differential feature vector to obtain the historical millimeter wave echo-transmitted signal component frequency domain differential feature vector;

[0036] The position-based sum of the historical millimeter-wave echo-transmitted signal component frequency domain differential feature vector and the current millimeter-wave echo-transmitted signal component frequency domain differential feature vector is calculated to obtain the millimeter-wave echo-reflected signal frequency domain differential aggregation representation vector.

[0037] For example, according to the high-precision bio-perception method based on millimeter-wave radar according to an embodiment of the present application, the perception result is determined based on the frequency domain differential aggregation representation vector of the millimeter-wave echo-reflection signal, including: inputting the frequency domain differential aggregation representation vector of the millimeter-wave echo-reflection signal into a high-precision bio-perception module based on a decoder to obtain the perception result.

[0038] The embodiment of the present application also provides a high-precision biometric perception system based on millimeter wave radar, which includes:

[0039] A signal acquisition module is used to acquire millimeter wave detection transmission signals and millimeter wave detection echo signals;

[0040] A decomposition module, used for performing CEEMDAN decomposition on the millimeter wave detection transmission signal and the millimeter wave detection echo signal to obtain a set of IMF components of the millimeter wave detection transmission signal and a set of IMF components of the millimeter wave detection echo signal;

[0041] A signal component frequency domain feature extraction module, used to perform signal component frequency domain feature extraction on the set of IMF components of the millimeter wave detection transmission signal and the set of IMF components of the millimeter wave detection echo signal to obtain a set of frequency domain feature vectors of the IMF components of the millimeter wave detection transmission signal and a set of frequency domain feature vectors of the IMF components of the millimeter wave detection echo signal;

[0042] A differential eigenvector calculation module is used to calculate the differential eigenvector between each corresponding set of the frequency domain eigenvectors of the IMF component of the millimeter wave detection transmission signal and the frequency domain eigenvectors of the IMF component of the millimeter wave detection echo signal in the set of frequency domain eigenvectors of the IMF component of the millimeter wave detection transmission signal to obtain a set of frequency domain differential eigenvectors of the millimeter wave echo-transmission signal component;

[0043] A feature aggregation module, used for inputting the set of frequency domain differential feature vectors of the millimeter wave echo-transmission signal component into a cluster analysis network based on node feature significant marking to obtain a frequency domain differential aggregation representation vector of the millimeter wave echo-reflection signal;

[0044] The perception result determination module is used to determine the perception result based on the frequency domain differential aggregation representation vector of the millimeter wave echo-reflection signal.

[0045] The present application has at least the following technical effects: Compared with the prior art, the high-precision bio-perception method and system based on millimeter-wave radar of the embodiments of the present application collects millimeter-wave detection transmission signals and millimeter-wave detection echo signals, and introduces signal processing and analysis algorithms based on artificial intelligence and deep learning at the back end to decompose and analyze the signal intrinsic modes of the millimeter-wave detection transmission signals and the millimeter-wave detection echo signals, so as to capture the key semantic aggregation features of the frequency domain difference of the signal components between the millimeter-wave detection transmission and echo signals, and then express the perception results based on the aggregate representation of the frequency domain difference features of the signal components, and use the perception results as the recognition results of bio-perception, thereby improving the accuracy of bio-perception detection using millimeter-wave radar. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings of the embodiments of the present application are briefly introduced below. Obviously, the drawings described below only relate to some embodiments of the present application, and are not intended to limit the present application.

[0047] Figure 1 A schematic diagram of the application architecture of a high-precision biological perception method based on millimeter-wave radar in an embodiment of the present application is shown;

[0048] Figure 2 A flowchart of a high-precision biological perception method based on millimeter-wave radar in an embodiment of the present application is shown;

[0049] Figure 3 A flowchart of sub-step S550 of the high-precision biological perception method based on millimeter-wave radar in an embodiment of the present application is shown;

[0050] Figure 4A schematic diagram of the structure of a high-precision biometric perception system based on millimeter-wave radar in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0051] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work also fall within the scope of protection of the present application.

[0052] The terms used in this specification are those common terms currently widely used in the art in consideration of the functions of the present application, but these terms may vary according to the intentions of those skilled in the art, precedents, or new technologies in the art. In addition, specific terms may be selected, and in this case, their detailed meanings will be described in the detailed description of the present application. Therefore, the terms used in the specification should not be understood as simple names, but rather as a general description based on the meaning of the terms and the present application.

[0053] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules can be used and run on the user terminal and / or server. The modules are only illustrative, and different aspects of the system and method can use different modules.

[0054] Flowcharts are used in the present application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps may be processed in reverse order or simultaneously as required. At the same time, other operations may also be added to these processes, or a certain step or several steps of operations may be removed from these processes.

[0055] Figure 1 A schematic diagram of the application architecture of a high-precision biosensing method based on millimeter-wave radar in an embodiment of the present application is shown, including a server 100 and a terminal device 200.

[0056] The terminal device 200 and the server 100 can be connected via the Internet to achieve mutual communication. Optionally, the above-mentioned Internet uses standard communication technology and / or protocol. The Internet is usually the Internet, but it can also be any network, including but not limited to any combination of a local area network (LAN), a metropolitan area network (MAN), a wide area network (WAN), a mobile, wired or wireless network, a private network or a virtual private network. In some embodiments, the data exchanged through the network can be represented by technologies and / or formats including Hyper Text Markup Language (HTML), Extensible Markup Language (XML), etc. In addition, conventional encryption technologies such as Secure Socket Layer (SSL), Transport Layer Security (TLS), Virtual Private Network (VPN), Internet Protocol Security (IPsec), etc. can be used to encrypt all or some links. In other embodiments, customized and / or dedicated data communication technologies can also be used to replace or supplement the above-mentioned data communication technologies.

[0057] The server 100 can provide various network services for the terminal device 200, wherein the server 100 can be a single server, a server cluster consisting of several servers, or a cloud computing center. Specifically, the server 100 may include a processor 110 (Center Processing Unit, CPU), a memory 120, an input device 130, and an output device 140, etc. The input device 130 may include a keyboard, a mouse, a touch screen, etc., and the output device 140 may include a display device, such as a liquid crystal display (Liquid Crystal Display, LCD), a cathode ray tube (Cathode Ray Tube, CRT), etc.

[0058] The memory 120 may include a read-only memory (ROM) and a random access memory (RAM), and provides the processor 110 with program instructions and data stored in the memory 120. In the embodiment of the present application, the memory 120 may be used to store the program of the high-precision biometric sensing method based on millimeter-wave radar in the embodiment of the present application.

[0059] The processor 110 calls the program instructions stored in the memory 120, and the processor 110 is used to execute the steps of any high-precision biological perception method based on millimeter-wave radar in the embodiments of the present application according to the obtained program instructions.

[0060] In addition, the application architecture diagram in the embodiment of the present application is intended to more clearly illustrate the technical solution in the embodiment of the present application, and does not constitute a limitation on the technical solution provided in the embodiment of the present application. Of course, for other application architectures and business applications, the technical solution provided in the embodiment of the present application is also applicable to similar problems.

[0061] It should be noted that in this application, all actions to obtain signals, information or data are carried out in compliance with the relevant data protection laws and policies of the country where the device is located and with the authorization of the owner of the corresponding device.

[0062] The following is a non-restrictive description of the high-precision biosensing method based on millimeter-wave radar provided according to at least one embodiment of the present application through several examples or embodiments. As described below, different features in these specific examples or embodiments can be combined with each other without conflicting with each other to obtain new examples or embodiments, and these new examples or embodiments also fall within the scope of protection of the present application.

[0063] The emergence of millimeter wave radar technology provides a new solution for non-contact bio-sensing. Millimeter wave radar can emit high-frequency millimeter waves, which will be reflected back when encountering the target object, and the reflected millimeter waves will be received by the receiving antenna and converted into electrical signals for processing. In addition, millimeter wave radar also has strong penetration ability and can perform non-contact monitoring in complex environments. It is particularly suitable for non-sensing measurement of biological information, especially for real-time monitoring of the tiny movements of the target object. The application of this technology in the field of biological monitoring, especially in monitoring the physiological signals of the human body (such as breathing and heartbeat), has shown great potential. In bio-sensing applications, when the target object (such as the human body) has a tiny movement (such as breathing and heartbeat), the reflected millimeter waves will change slightly, which will be captured by the radar and converted into analyzable data (such as extracting the breathing rate and the changes in posture before and after falling), realizing non-contact biological perception and monitoring, and improving the accuracy of perception of the monitored object.

[0064] Based on this, in the technical solution of the present application, a high-precision bio-perception method based on millimeter-wave radar is proposed, which can collect millimeter-wave detection transmission signals and millimeter-wave detection echo signals, and introduce signal processing and analysis algorithms based on artificial intelligence and deep learning at the back end to decompose and analyze the signal intrinsic modes of the millimeter-wave detection transmission signals and the millimeter-wave detection echo signals, so as to capture the key semantic aggregation features of the frequency domain difference of the signal components between the millimeter-wave detection transmission and echo signals, and then express the perception results based on the aggregate representation of the frequency domain difference features of the signal components, and use the perception results as the recognition results of bio-perception, thereby improving the accuracy of bio-perception detection using millimeter-wave radar.

[0065] Figure 2 The flowchart of the high-precision biological perception method based on millimeter wave radar in the embodiment of the present application is shown. For example, the high-precision biological perception method based on millimeter wave radar can be executed by a server, which can be Figure 1 The server 100 shown in FIG. Figure 2As shown, according to the high-precision bio-sensing method based on millimeter-wave radar according to the embodiment of the present application, the steps include: S510, acquiring a millimeter-wave detection transmission signal and a millimeter-wave detection echo signal; S520, performing CEEMDAN decomposition on the millimeter-wave detection transmission signal and the millimeter-wave detection echo signal to obtain a set of IMF components of the millimeter-wave detection transmission signal and a set of IMF components of the millimeter-wave detection echo signal; S530, performing signal component frequency domain feature extraction on the set of IMF components of the millimeter-wave detection transmission signal and the set of IMF components of the millimeter-wave detection echo signal to obtain a set of frequency domain feature vectors of the IMF components of the millimeter-wave detection transmission signal and a set of frequency domain feature vectors of the IMF components of the millimeter-wave detection echo signal; S5 40, calculate the differential feature vector between each corresponding set of the frequency domain feature vectors of the IMF component of the millimeter wave detection transmission signal and the frequency domain feature vectors of the IMF component of the millimeter wave detection echo signal in the set of the frequency domain feature vectors of the IMF component of the millimeter wave detection transmission signal and the frequency domain feature vectors of the IMF component of the millimeter wave detection echo signal to obtain a set of frequency domain differential feature vectors of the millimeter wave echo-transmission signal component; S550, input the set of frequency domain differential feature vectors of the millimeter wave echo-transmission signal component into a clustering analysis network based on node feature significant marking to obtain a frequency domain differential aggregation representation vector of the millimeter wave echo-reflection signal; S560, determine the perception result based on the frequency domain differential aggregation representation vector of the millimeter wave echo-reflection signal. In one embodiment, the perception result is used to represent the respiratory frequency decoding value, which can be used to monitor the respiratory frequency of the detected person by representing the respiratory frequency decoding value, and can be specifically applied to non-contact respiratory monitoring of critically ill patients, and can also be used for sleep monitoring of the monitored person, etc. In another embodiment, the sensing result is used to monitor the position, posture, etc. of the guardian, and to determine the activity position of the monitored person, whether he or she has fallen, etc., so as to take timely countermeasures.

[0066] Specifically, in the technical solution of the present application, first, a millimeter wave detection transmission signal and a millimeter wave detection echo signal are obtained. Next, considering that millimeter wave radar signals usually have nonlinear and non-stationary characteristics, traditional signal processing methods (such as Fourier transform) may not be able to effectively extract useful information when processing such signals. CEEMDAN, as an adaptive signal processing method, can effectively process nonlinear and non-stationary signals. At the same time, CEEMDAN can decompose the original radar signal into a series of intrinsic mode functions (IMFs), which are the natural frequency components of the signal and represent the fluctuations on different time scales in the signal. Based on this, in the technical solution of the present application, the millimeter wave detection transmission signal and the millimeter wave detection echo signal are decomposed by CEEMDAN to obtain a set of IMF components of the millimeter wave detection transmission signal and a set of IMF components of the millimeter wave detection echo signal. By decomposing the millimeter wave detection transmission signal and the millimeter wave detection echo signal into multiple IMF components, each IMF component represents a different frequency component in the signal. This decomposition can more clearly analyze the various components of the signal, which is convenient for subsequent feature extraction and analysis. Specifically, during the decomposition process, CEEMDAN adds a certain degree of white noise, which helps to reduce mode aliasing and improve the stability and accuracy of signal decomposition. This method of adding noise can effectively suppress the random noise in the millimeter wave detection transmission signal and the millimeter wave detection echo signal, ensuring that the extracted IMF component is purer, which is helpful for subsequent signal feature analysis and detection.

[0067] Then, each millimeter wave detection transmission signal IMF component in the set of millimeter wave detection transmission signal IMF components and each millimeter wave detection echo signal IMF component in the set of millimeter wave detection echo signal IMF components are respectively input into the signal component frequency domain feature extractor based on the 1D-CNN model for feature mining, so as to respectively extract the signal IMF component frequency domain implicit feature information in each millimeter wave detection transmission signal IMF component and each millimeter wave detection echo signal IMF component, thereby obtaining a set of millimeter wave detection transmission signal IMF component frequency domain feature vectors and a set of millimeter wave detection echo signal IMF component frequency domain feature vectors. Through the processing of the signal component frequency domain feature extractor based on the 1D-CNN model, the frequency domain features of each IMF component in the millimeter wave detection transmission and echo signals can be respectively extracted, which helps to perform the subsequent correlation analysis of these component frequency domain features to highlight the semantic expression of specific key frequency components related to biological perception (such as breathing frequency), so as to better realize the detection task of biological perception (such as breathing frequency).

[0068] Correspondingly, in step S530, signal component frequency domain feature extraction is performed on the set of millimeter-wave detection transmit signal IMF components and the set of millimeter-wave detection echo signal IMF components to obtain a set of millimeter-wave detection transmit signal IMF component frequency domain feature vectors and a set of millimeter-wave detection echo signal IMF component frequency domain feature vectors, including: inputting each millimeter-wave detection transmit signal IMF component in the set of millimeter-wave detection transmit signal IMF components and each millimeter-wave detection echo signal IMF component in the set of millimeter-wave detection echo signal IMF components into a signal component frequency domain feature extractor based on a 1D-CNN model to obtain a set of millimeter-wave detection transmit signal IMF component frequency domain feature vectors and a set of millimeter-wave detection echo signal IMF component frequency domain feature vectors.

[0069] Furthermore, in order to be able to capture the relevant signal components and feature information about biological perception (such as breathing frequency), in the technical solution of the present application, the differential feature vectors between each group of corresponding millimeter wave detection transmission signal IMF component frequency domain feature vectors and the millimeter wave detection echo signal IMF component frequency domain feature vectors in the set of the millimeter wave detection transmission signal IMF component frequency domain feature vectors and the millimeter wave detection echo signal IMF component frequency domain feature vectors are further calculated to obtain a set of millimeter wave echo-transmission signal component frequency domain differential feature vectors. By calculating each group of corresponding differential feature vectors, the differences and changes in the frequency domain characteristics of the corresponding signal IMF components between the transmission signal and the echo signal can be highlighted. Such changes often contain important information related to physiological activities, such as signal changes caused by breathing or heartbeat, and differential features help to more clearly identify these dynamic changes. In other words, calculating differential feature vectors can help extract and enhance signal components related to breathing frequency, because the difference between the transmission signal and the echo signal is mainly reflected in the signal changes caused by human body micro-movements (such as breathing). By calculating the differential eigenvectors, these small changes can be highlighted, thereby improving the performance of subsequent breathing rate detection and biosensing.

[0070] It should be understood that since each millimeter wave echo-transmitted signal component frequency domain differential feature vector in the set of the millimeter wave echo-transmitted signal component frequency domain differential feature vectors respectively contains the differences and changes in the frequency domain features of each corresponding signal IMF component between the millimeter wave echo signal and the transmitted signal, and considering that not all signal IMF component frequency domain differential features and signal components have the same contribution to the subsequent biological perception (such as breathing rate) detection task. Therefore, in order to be able to use the semantic correlation between each millimeter wave echo-transmitted signal component frequency domain differential feature vector to highlight the features that are more important to the subsequent task, and to realize the dynamic weighted aggregation of information of different signal components, in the technical solution of the present application, the set of the millimeter wave echo-transmitted signal component frequency domain differential feature vectors is input into a clustering analysis network based on node feature significant marking to obtain a millimeter wave echo-reflected signal frequency domain differential aggregation representation vector.

[0071] Specifically, first, the feature significance descriptor factor of the frequency domain differential feature vector of each millimeter wave echo-transmitted signal component is calculated, which is determined by analyzing the mean and variance of the feature vector. The mean reflects the semantic center of the feature value in the vector, while the variance indicates the dispersion of the feature value. Combining the two can determine which features are statistically representative and discriminative, thereby evaluating the importance or significance of the feature. Specifically, the feature significance descriptor factor can help the model identify important signal components related to biological perception (such as breathing rate), so that these key features can be given priority in subsequent processing, thereby giving these features higher weights in the aggregation process. Next, the feature significance attenuation factor is constructed by measuring the distance between the frequency domain differential feature vectors of each millimeter wave echo-transmitted signal component and the current frequency domain differential feature vector of the millimeter wave echo-transmitted signal component. The construction of the feature significance attenuation factor allows the model to dynamically adjust the weight according to the distance between the features, thereby adjusting the influence of different feature vectors. Features that are closer to the current node feature will receive higher weights, which can strengthen the frequency domain features of local signal components related to biological perception (such as breathing rate). This method helps to capture small but important changes in the process of biological perception (such as breathing) and improve the model's sensitivity to these changes. Furthermore, the feature significance attenuation factor is used to modulate the feature significance descriptor to obtain the feature significance attenuation descriptor. This step enables the model to more accurately reflect the relative importance of each millimeter wave echo-transmitted signal component frequency domain differential feature vector in the sequence. Features that are more consistent or close to the current feature vector will receive a larger weight, while those that are less relevant will be appropriately weakened. Then, the sequence of feature significance attenuation descriptors is input into the gated mask module, which uses an activation function or an attention mechanism to further filter these descriptors. The function of the gated mask module is to dynamically adjust the significance weights of the frequency domain differential features of the millimeter wave echo-transmitted signal components of different signal components to ensure that the model can focus on features that are more important for the current biological perception (such as breathing rate) detection task. Through such screening, the signal components of noise or minor features that are not related to biological perception (such as breathing rate) will be further suppressed, while the semantics of the signal components of key features will be highlighted. Finally, the set of frequency domain differential feature vectors of the millimeter wave echo-transmitted signal component is weighted and summed using the feature significance attenuation weight factor sequence obtained after processing by the gated mask module. The generated millimeter wave echo-reflected signal frequency domain differential aggregate representation vector integrates the frequency domain differential feature information of all millimeter wave echo-transmitted signal components of different signal components after highlighting their importance, while emphasizing the frequency domain differential semantic features of those signal components with high significance and suppressing unimportant features.The ultimate goal is to generate a vector that is highly representative of the changing patterns of biological perceptions (such as breathing rate), thereby improving the accuracy and robustness of subsequent processing steps. In this way, the expression of signal components related to biological perceptions (such as breathing rate) can be highlighted and enhanced during the information aggregation process. This processing method helps improve the system's ability to detect and identify biological perceptions (such as breathing rate), especially in the presence of noise and other interfering factors.

[0072] Accordingly, in step S550, if Figure 3 As shown, the set of the frequency domain differential feature vectors of the millimeter wave echo-transmitted signal component is input into a cluster analysis network based on node feature significant marking to obtain a millimeter wave echo-reflected signal frequency domain differential aggregation representation vector, including: S551, calculating the feature significance description factor of each millimeter wave echo-transmitted signal component frequency domain differential feature vector in the set of the millimeter wave echo-transmitted signal component frequency domain differential feature vector, wherein the feature significance description factor is related to the mean and variance of each millimeter wave echo-transmitted signal component frequency domain differential feature vector; S552, based on the distance span between each millimeter wave echo-transmitted signal component frequency domain differential feature vector in the set of the millimeter wave echo-transmitted signal component frequency domain differential feature vector and the current millimeter wave echo-transmitted signal component frequency domain differential feature vector. Degree, construct the characteristic significance attenuation factor of each frequency domain differential feature vector of the millimeter wave echo-transmitted signal component; S553, modulate the characteristic significance description factor of each frequency domain differential feature vector of the millimeter wave echo-transmitted signal component based on the characteristic significance attenuation factor of each frequency domain differential feature vector of the millimeter wave echo-transmitted signal component to obtain a sequence of characteristic significance attenuation description factors; S554, input the sequence of characteristic significance attenuation description factors into the gated mask module to obtain a sequence of masked characteristic significance attenuation weight factors; S555, using the sequence of masked characteristic significance attenuation weight factors as a sequence of weights, calculate the weighted sum of the set of frequency domain differential feature vectors of the millimeter wave echo-transmitted signal component to obtain the frequency domain differential aggregation representation vector of the millimeter wave echo-reflected signal.

[0073] Wherein, in step S551, the characteristic significance description factor of each millimeter wave echo-transmit signal component frequency domain differential feature vector in the set of the millimeter wave echo-transmit signal component frequency domain differential feature vector is calculated, wherein the characteristic significance description factor is related to the mean and variance of each of the millimeter wave echo-transmit signal component frequency domain differential feature vectors, including: respectively calculating the mean and variance of the millimeter wave echo-transmit signal component frequency domain differential feature vector to obtain the millimeter wave echo-transmit signal component frequency domain differential feature mean and the millimeter wave echo-transmit signal component frequency domain differential feature variance ; After calculating the fourth power of the positional difference between the frequency domain differential feature vector of the millimeter wave echo-transmitted signal component and the mean of the frequency domain differential feature of the millimeter wave echo-transmitted signal component, calculate the expected value of the fourth power modulation semantic deviation vector to obtain the expected factor of the frequency domain differential feature of the millimeter wave echo-transmitted signal component; calculate the positional division between the expected factor of the frequency domain differential feature of the millimeter wave echo-transmitted signal component and the square of the variance of the frequency domain differential feature of the millimeter wave echo-transmitted signal component to obtain the feature significance description factor of the frequency domain differential feature vector of the millimeter wave echo-transmitted signal component.

[0074] Wherein, in step S552, based on the distance span between each millimeter wave echo-transmit signal component frequency domain differential feature vector in the set of the millimeter wave echo-transmit signal component frequency domain differential feature vector and the current millimeter wave echo-transmit signal component frequency domain differential feature vector, construct the characteristic significance attenuation factor of each millimeter wave echo-transmit signal component frequency domain differential feature vector, including: calculating the distance span between other millimeter wave echo-transmit signal component frequency domain differential feature vectors in the set of the millimeter wave echo-transmit signal component frequency domain differential feature vector and the current millimeter wave echo-transmit signal component frequency domain differential feature vector. degree to obtain the eigenvalue distance span; extract the maximum value in the frequency domain differential eigenvector of the current millimeter-wave echo-transmitted signal component to obtain the maximum value of the frequency domain differential feature of the current millimeter-wave echo-transmitted signal component; extract the maximum value in the frequency domain differential eigenvector of the other millimeter-wave echo-transmitted signal components to obtain the maximum value of the frequency domain differential feature of other millimeter-wave echo-transmitted signal components; calculate the difference between the maximum value of the frequency domain differential feature of the current millimeter-wave echo-transmitted signal component and the maximum value of the frequency domain differential feature of other millimeter-wave echo-transmitted signal components, and then divide it by the eigenvalue distance span to obtain the characteristic significance attenuation factor.

[0075] Wherein, in step S553, the characteristic significance descriptor of each of the frequency domain differential feature vectors of the millimeter-wave echo-transmitted signal components is modulated based on the characteristic significance attenuation factor of each of the frequency domain differential feature vectors of the millimeter-wave echo-transmitted signal components to obtain a sequence of characteristic significance attenuation descriptors, including: calculating the positional product between the characteristic significance attenuation factor of each of the frequency domain differential feature vectors of the millimeter-wave echo-transmitted signal components and the characteristic significance descriptor of each of the frequency domain differential feature vectors of the millimeter-wave echo-transmitted signal components to obtain a sequence of characteristic significance attenuation descriptors.

[0076] Wherein, in step S554, the sequence of feature significance attenuation descriptors is input into a gated mask module to obtain a sequence of masked feature significance attenuation weight factors, including: taking the negative number of each feature significance attenuation descriptor in the sequence of feature significance attenuation descriptors as an exponential power, calculating an exponential function value with a natural constant e as the base to obtain a sequence of feature significance attenuation class support descriptors; calculating the reciprocal of the sum of each feature significance attenuation class support descriptor in the sequence of feature significance attenuation class support descriptors and a constant one to obtain a sequence of feature significance attenuation weight factors; based on a predetermined threshold, masking the sequence of feature significance attenuation weight factors to obtain a sequence of masked feature significance attenuation weight factors, wherein, in response to each feature significance attenuation weight factor in the sequence of feature significance attenuation weight factors being greater than the predetermined threshold, the feature significance attenuation weight factor greater than the predetermined threshold is taken to have an original value, otherwise it is set to zero.

[0077] Wherein, in step S555, the sequence of the mask feature significance attenuation weight factors is used as the sequence of weights to calculate the weighted sum of the set of the millimeter-wave echo-transmitted signal component frequency domain differential feature vectors to obtain the millimeter-wave echo-reflected signal frequency domain differential aggregation representation vector, including: using the sequence of the mask feature significance attenuation weight factors as the sequence of weights, calculating the weighted sum between each other millimeter-wave echo-transmitted signal component frequency domain differential feature vector in the set of the millimeter-wave echo-transmitted signal component frequency domain differential feature vector to obtain the historical millimeter-wave echo-transmitted signal component frequency domain differential feature vector; calculating the positional sum between the historical millimeter-wave echo-transmitted signal component frequency domain differential feature vector and the current millimeter-wave echo-transmitted signal component frequency domain differential feature vector to obtain the millimeter-wave echo-reflected signal frequency domain differential aggregation representation vector.

[0078] In a specific example, the set of frequency domain differential feature vectors of the millimeter wave echo-transmitted signal component is input into a cluster analysis network based on node feature significant marking to obtain a frequency domain differential aggregation representation vector of the millimeter wave echo-reflected signal, including: inputting the set of frequency domain differential feature vectors of the millimeter wave echo-transmitted signal component into the cluster analysis network based on node feature significant marking and processing them with the following feature aggregation formula to obtain the frequency domain differential aggregation representation vector of the millimeter wave echo-reflected signal; wherein the feature aggregation formula is:

[0079]

[0080]

[0081]

[0082]

[0083]

[0084]

[0085]

[0086] in, is a set of frequency domain differential eigenvectors of the millimeter wave echo-transmitted signal component, is the first in the set of frequency domain differential eigenvectors of the millimeter wave echo-transmitted signal component The frequency domain differential eigenvectors of the millimeter wave echo-transmitted signal component, and are respectively the first and The frequency domain differential eigenvectors of the millimeter wave echo-transmitted signal component, Indicates taking The maximum value in Indicates taking The minimum value in express and The eigenvalue distance span between For the said The characteristic significance attenuation factor corresponding to the frequency domain differential eigenvector of the millimeter wave echo-transmitted signal component, For the said The mean of the frequency domain difference eigenvectors of the millimeter wave echo-transmitted signal components, For the said The variance of the frequency domain differential eigenvector of the millimeter wave echo-transmitted signal component, For the said The frequency domain differential eigenvector of the millimeter wave echo-transmitted signal component The eigenvalues ​​at the positions, represents the computational expectation, For the said The characteristic significance attenuation description factor corresponding to the frequency domain differential eigenvector of the millimeter wave echo-transmitted signal component, For the said The characteristic significance attenuation weight factor of the frequency domain difference eigenvector of the millimeter wave echo-transmitted signal component, For masking operation, is the predetermined threshold, For the said The mask feature significance attenuation weight factor of the frequency domain difference feature vector of the millimeter wave echo-transmitted signal component, is the number of eigenvectors in the set of frequency domain differential eigenvectors of the millimeter wave echo-transmitted signal component, is a frequency domain differential aggregation representation vector of the millimeter wave echo-reflection signal.

[0087] Then, the frequency domain differential aggregate representation vector of the millimeter wave echo-reflection signal is input into the high-precision bio-perception module based on the decoder to obtain the perception result. In other words, the frequency domain differential aggregate representation feature between the millimeter wave echo signal and the reflection signal is used for decoding regression, so as to perform detection based on bio-perception (such as breathing frequency), and then the perception result is expressed based on the aggregate representation of the frequency domain difference feature of the signal component, and the perception result is used as the recognition result of bio-perception, thereby improving the accuracy of bio-perception detection using millimeter wave radar.

[0088] Accordingly, in step S560, the perception result is determined based on the millimeter-wave echo-reflection signal frequency domain differential aggregation representation vector; including: inputting the millimeter-wave echo-reflection signal frequency domain differential aggregation representation vector into a high-precision biological perception module based on a decoder to obtain the perception result; in one embodiment, the perception result is used to represent the respiratory rate decoding value, and by representing the respiratory rate decoding value, it can be used to monitor the respiratory rate of the person being detected, specifically, it can be applied to non-contact respiratory monitoring of critically ill patients, and can also be used for sleep monitoring of the monitored person, etc.

[0089] That is, considering that each millimeter-wave echo-transmitted signal component frequency domain differential feature vector in the set of millimeter-wave echo-transmitted signal component frequency domain differential feature vectors respectively represents the semantic characteristics of the frequency domain distribution differences of each IMF component of the millimeter-wave detection transmitted signal and each IMF component of the millimeter-wave detection echo signal, when further performing feature aggregation based on local significance gating of feature energy, the significance difference of the differential semantic feature representation of the single sample domain signal in the entire sample domain will cause differences in the aggregation space structure between sample domains, affecting the convergence consistency of the decoder, and thus affecting the accuracy of the perception results obtained by the high-precision bio-perception module based on the decoder.

[0090] In a preferred example, inputting the millimeter wave echo-reflection signal frequency domain differential aggregation representation vector into a decoder-based high-precision biometric sensing module to obtain a sensing result includes the following steps:

[0091] The absolute value sum and the square root of the square sum of all eigenvalues ​​of the millimeter wave echo-reflection signal frequency domain differential aggregation representation vector are calculated to obtain the first millimeter wave echo-reflection signal frequency domain differential aggregation spatial structure value and the second millimeter wave echo-reflection signal frequency domain differential aggregation spatial structure value, that is:

[0092]

[0093]

[0094] in, represents the frequency domain differential aggregation representation vector of the millimeter wave echo-reflection signal eigenvalues, represents the spatial structure value of the frequency domain differential aggregation of the first millimeter wave echo-reflection signal, represents the frequency domain differential aggregation spatial structure value of the second millimeter wave echo-reflection signal;

[0095] Determine the number of eigenvalues ​​of all eigenvalues ​​of the frequency domain differential aggregation representation vector of the millimeter wave echo-reflection signal , that is, the length of the frequency domain differential aggregation representation vector of the millimeter wave echo-reflection signal;

[0096] For each eigenvalue of the frequency domain differential aggregation representation vector of the millimeter wave echo-reflection signal, the first millimeter wave echo-reflection signal frequency domain differential aggregation spatial structure value minus the product of the eigenvalue and the number of eigenvalues ​​is calculated to obtain the first millimeter wave echo-reflection signal frequency domain differential aggregation long-range dependency value ,in, represents the spatial structure value of the frequency domain differential aggregation of the first millimeter wave echo-reflection signal, represents the frequency domain differential aggregation representation vector of the millimeter wave echo-reflection signal eigenvalues, The number of eigenvalues ​​representing all eigenvalues ​​of the frequency domain differential aggregation representation vector of the millimeter wave echo-reflection signal, represents the long-range dependency value of the frequency domain differential aggregation of the first millimeter wave echo-reflection signal;

[0097] Calculate the second millimeter wave echo-reflection signal frequency domain differential aggregation long-range dependency value obtained by multiplying the square root of the number of eigenvalues ​​by the product of the eigenvalues ​​minus the second millimeter wave echo-reflection signal frequency domain differential aggregation spatial structure value ,in, represents the frequency domain differential aggregation spatial structure value of the second millimeter wave echo-reflection signal, represents the frequency domain differential aggregation representation vector of the millimeter wave echo-reflection signal eigenvalues, The number of eigenvalues ​​representing all eigenvalues ​​of the frequency domain differential aggregation representation vector of the millimeter wave echo-reflection signal, represents the second millimeter wave echo-reflection signal frequency domain differential aggregation long-range dependency value;

[0098] The index value calculated by taking the first millimeter wave echo-reflection signal frequency domain differential aggregation long-range dependence value as the exponent of the natural constant and the inverse of the second millimeter wave echo-reflection signal frequency domain differential aggregation long-range dependence value are weighted summed to obtain the optimized eigenvalue corresponding to each eigenvalue ,in, represents the long-range dependency value of the frequency domain differential aggregation of the first millimeter-wave echo-reflection signal, represents the second millimeter wave echo-reflection signal frequency domain differential aggregation long-range dependency value, represents a natural constant, and represents the weighted hyperparameter, Indicates the optimized eigenvalue corresponding to each eigenvalue;

[0099] The optimized eigenvalues ​​are combined into an optimized millimeter wave echo-reflection signal frequency domain differential aggregation representation vector;

[0100] The optimized millimeter wave echo-reflection signal frequency domain differential aggregation representation vector is input into the decoder-based high-precision bio-sensing module to obtain the perception result.

[0101] That is, in view of the possible spatial structure loss of the feature set of the millimeter-wave echo-reflection signal frequency domain differential aggregation representation vector in the high-dimensional space, which causes the decoder weight to implicitly infer the spatial structure information based on the features, resulting in inconsistent convergence. A long-distance feature dependency is established based on the spatial structure representation of the millimeter-wave echo-reflection signal frequency domain differential aggregation representation vector relative to the overall feature scale of the millimeter-wave echo-reflection signal frequency domain differential aggregation representation vector, so as to establish the feature local connectivity of the millimeter-wave echo-reflection signal frequency domain differential aggregation representation vector, and to capture the spatial ambiguity information of the object feature value through the unstructured feature value point prediction of the millimeter-wave echo-reflection signal frequency domain differential aggregation representation vector, thereby improving the spatial inductive deviation perception ability of the feature set of the millimeter-wave echo-reflection signal frequency domain differential aggregation representation vector, improving the convergence consistency of the decoder, and improving the accuracy of the perception result obtained by the millimeter-wave echo-reflection signal frequency domain differential aggregation representation vector through the high-precision bio-perception module based on the decoder, so as to more accurately detect bio-perception (such as breathing rate), realize non-contact bio-perception, and improve the intelligence level of bio-perception and data analysis.

[0102] Based on the above embodiments, see Figure 4The figure is a schematic diagram of the structure of a high-precision bio-sensing system 800 based on millimeter-wave radar in an embodiment of the present application. The high-precision bio-sensing system 800 based on millimeter-wave radar includes: a signal acquisition module 810, used to acquire a millimeter-wave detection transmission signal and a millimeter-wave detection echo signal; a decomposition module 820, used to perform CEEMDAN decomposition on the millimeter-wave detection transmission signal and the millimeter-wave detection echo signal to obtain a set of IMF components of the millimeter-wave detection transmission signal and a set of IMF components of the millimeter-wave detection echo signal; a signal component frequency domain feature extraction module 830, used to perform signal component frequency domain feature extraction on the set of IMF components of the millimeter-wave detection transmission signal and the set of IMF components of the millimeter-wave detection echo signal to obtain a set of frequency domain feature vectors of the IMF components of the millimeter-wave detection transmission signal and a set of frequency domain feature vectors of the IMF components of the millimeter-wave detection echo signal; a differential feature vector calculation module 830; a signal component frequency domain feature extraction ... Module 840 is used to calculate the differential feature vectors between each corresponding group of frequency domain feature vectors of the millimeter wave detection transmission signal IMF component frequency domain feature vectors and the set of frequency domain feature vectors of the millimeter wave detection echo signal IMF component frequency domain feature vectors to obtain a set of frequency domain differential feature vectors of the millimeter wave echo-transmitted signal component; feature aggregation module 850 is used to input the set of frequency domain differential feature vectors of the millimeter wave echo-transmitted signal component into a clustering analysis network based on node feature significant markings to obtain a frequency domain differential aggregation representation vector of the millimeter wave echo-reflected signal; perception result determination module 860 is used to determine the perception result based on the frequency domain differential aggregation representation vector of the millimeter wave echo-reflected signal.

[0103] The specific functions and operations of each module in the millimeter wave radar based high-precision biological perception system 800 have been referred to above. Figure 2 to Figure 3 The present invention has been introduced in detail in the description of the high-precision biosensing method based on millimeter-wave radar, and therefore, its repeated description will be omitted.

[0104] Based on the above embodiments, another exemplary embodiment of an electronic device is also provided in the embodiments of the present application. In some possible implementations, the electronic device in the embodiments of the present application may include a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor can implement the steps of the high-precision biosensing method based on millimeter-wave radar in the above embodiments when executing the program.

[0105] For example, in the case of electronic equipment Figure 1 Taking the server 100 in the example as an example, the processor in the electronic device is the processor 110 in the server 100, and the memory in the electronic device is the memory 120 in the server 100.

[0106] The embodiment of the present application also provides a computer-readable storage medium, on which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, the high-precision bio-sensing method based on millimeter-wave radar according to the embodiment of the present application described with reference to the above figures can be executed. The computer-readable storage medium includes, but is not limited to, for example, volatile memory and / or non-volatile memory. The volatile memory may, for example, include a random access memory (RAM) and / or a cache memory (cache), etc. The non-volatile memory may, for example, include a read-only memory (ROM), a hard disk, a flash memory, etc.

[0107] The embodiment of the present application also provides a computer program product or a computer program, which includes computer executable instructions, and the computer executable instructions are stored in a computer readable storage medium. The processor of the computer device reads the computer executable instructions from the computer readable storage medium, and the processor executes the computer executable instructions, so that the computer device executes the high-precision biological perception method based on millimeter wave radar according to the embodiment of the present application.

[0108] Those skilled in the art will appreciate that the contents disclosed in this application may be subject to various variations and improvements. For example, the various devices or components described above may be implemented by hardware, or by software, firmware, or a combination of some or all of the three.

[0109] In addition, although the present application makes various references to certain units in the system according to embodiments of the present application, any number of different units can be used and run on the client and / or server. The units are only illustrative, and different aspects of the system and method can use different units.

[0110] Those skilled in the art will appreciate that all or part of the steps in the above method can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium, such as a read-only memory, a disk or an optical disk. Optionally, all or part of the steps in the above embodiment can also be implemented using one or more integrated circuits. Accordingly, each module / unit in the above embodiment can be implemented in the form of hardware or in the form of software function modules. The present application is not limited to any particular form of combination of hardware and software.

[0111] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by those of ordinary skill in the art to which this application belongs. It should also be understood that terms such as those defined in common dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology, and should not be interpreted in an idealized or extremely formal sense, unless explicitly defined as such herein.

[0112] The above is an explanation of the present application and should not be considered as limiting thereof. Although several exemplary embodiments of the present application have been described, those skilled in the art will readily appreciate that many modifications may be made to the exemplary embodiments without departing from the novel teachings and advantages of the present application.

Claims

1. A high-precision biological perception method based on millimeter-wave radar, characterized in that: include: Acquire millimeter wave detection transmission signal and millimeter wave detection echo signal; Performing CEEMDAN decomposition on the millimeter wave detection transmission signal and the millimeter wave detection echo signal to obtain a set of IMF components of the millimeter wave detection transmission signal and a set of IMF components of the millimeter wave detection echo signal; Performing signal component frequency domain feature extraction on the set of IMF components of the millimeter wave detection transmission signal and the set of IMF components of the millimeter wave detection echo signal to obtain a set of frequency domain feature vectors of the IMF components of the millimeter wave detection transmission signal and a set of frequency domain feature vectors of the IMF components of the millimeter wave detection echo signal; Calculate the differential eigenvector between each corresponding group of the frequency domain eigenvectors of the IMF component of the millimeter wave detection transmission signal and the frequency domain eigenvectors of the IMF component of the millimeter wave detection echo signal in the set of the frequency domain eigenvectors of the IMF component of the millimeter wave detection transmission signal and the frequency domain eigenvectors of the IMF component of the millimeter wave detection echo signal to obtain a set of frequency domain differential eigenvectors of the millimeter wave echo-transmission signal component; Inputting the set of frequency domain differential feature vectors of the millimeter wave echo-transmitted signal component into a cluster analysis network based on node feature significant marking to obtain a frequency domain differential aggregate representation vector of the millimeter wave echo-reflected signal; The perception result is determined based on the frequency domain differential aggregation representation vector of the millimeter wave echo-reflection signal.

2. The high-precision biological perception method based on millimeter-wave radar according to claim 1 is characterized in that: Signal component frequency domain feature extraction is performed on the set of millimeter wave detection transmission signal IMF components and the set of millimeter wave detection echo signal IMF components to obtain a set of millimeter wave detection transmission signal IMF component frequency domain feature vectors and a set of millimeter wave detection echo signal IMF component frequency domain feature vectors, including: inputting each millimeter wave detection transmission signal IMF component in the set of millimeter wave detection transmission signal IMF components and each millimeter wave detection echo signal IMF component in the set of millimeter wave detection echo signal IMF components into a signal component frequency domain feature extractor based on a 1D-CNN model to obtain a set of millimeter wave detection transmission signal IMF component frequency domain feature vectors and a set of millimeter wave detection echo signal IMF component frequency domain feature vectors.

3. The high-precision biological perception method based on millimeter-wave radar according to claim 2 is characterized in that: Inputting the set of frequency domain differential feature vectors of the millimeter wave echo-transmitted signal component into a cluster analysis network based on node feature significant marking to obtain a frequency domain differential aggregate representation vector of the millimeter wave echo-reflected signal, including: Calculating a characteristic significance description factor of each millimeter wave echo-transmitted signal component frequency domain difference feature vector in the set of the millimeter wave echo-transmitted signal component frequency domain difference feature vectors, wherein the characteristic significance description factor is related to the mean and variance of each of the millimeter wave echo-transmitted signal component frequency domain difference feature vectors; Constructing a characteristic significance attenuation factor of each of the millimeter wave echo-transmit signal component frequency domain differential eigenvectors based on the distance span between each of the millimeter wave echo-transmit signal component frequency domain differential eigenvectors in the set of the millimeter wave echo-transmit signal component frequency domain differential eigenvectors and the current millimeter wave echo-transmit signal component frequency domain differential eigenvector; Modulating the characteristic significance description factor of each of the frequency domain difference feature vectors of the millimeter wave echo-transmitted signal component based on the characteristic significance attenuation factor of each of the frequency domain difference feature vectors of the millimeter wave echo-transmitted signal component to obtain a sequence of characteristic significance attenuation description factors; Inputting the sequence of feature significance attenuation description factors into a gated mask module to obtain a sequence of masked feature significance attenuation weight factors; Taking the sequence of the mask feature significance attenuation weight factors as the sequence of weights, the weighted sum of the set of frequency domain differential feature vectors of the millimeter wave echo-transmitted signal component is calculated to obtain the frequency domain differential aggregation representation vector of the millimeter wave echo-reflected signal.

4. The high-precision biological perception method based on millimeter-wave radar according to claim 3 is characterized in that: Calculating a characteristic significance description factor of each millimeter wave echo-transmitted signal component frequency domain difference feature vector in the set of the millimeter wave echo-transmitted signal component frequency domain difference feature vectors, wherein the characteristic significance description factor is related to the mean and variance of each of the millimeter wave echo-transmitted signal component frequency domain difference feature vectors, including: Respectively calculating the mean and variance of the frequency domain differential feature vector of the millimeter wave echo-transmitted signal component to obtain the frequency domain differential feature mean of the millimeter wave echo-transmitted signal component and the frequency domain differential feature variance of the millimeter wave echo-transmitted signal component; After calculating the fourth power of the position difference between the frequency domain differential feature vector of the millimeter wave echo-transmitted signal component and the mean value of the frequency domain differential feature of the millimeter wave echo-transmitted signal component, the expected value of the fourth power modulation semantic deviation vector is calculated to obtain the expected factor of the frequency domain differential feature of the millimeter wave echo-transmitted signal component; The position-based division between the expected factor of the frequency domain differential feature of the millimeter wave echo-transmitted signal component and the square of the frequency domain differential feature variance of the millimeter wave echo-transmitted signal component is calculated to obtain a feature significance description factor of the frequency domain differential feature vector of the millimeter wave echo-transmitted signal component.

5. The high-precision biological perception method based on millimeter-wave radar according to claim 4 is characterized in that: Based on the distance span between each millimeter wave echo-transmit signal component frequency domain differential feature vector in the set of the millimeter wave echo-transmit signal component frequency domain differential feature vector and the current millimeter wave echo-transmit signal component frequency domain differential feature vector, constructing a characteristic significance attenuation factor of each of the millimeter wave echo-transmit signal component frequency domain differential feature vectors, including: Calculating the distance span between other millimeter wave echo-transmitted signal component frequency domain differential eigenvectors in the set of the millimeter wave echo-transmitted signal component frequency domain differential eigenvectors and the current millimeter wave echo-transmitted signal component frequency domain differential eigenvector to obtain an eigenvalue distance span; Extracting the maximum value of the frequency domain differential feature vector of the current millimeter wave echo-transmitted signal component to obtain the maximum value of the frequency domain differential feature of the current millimeter wave echo-transmitted signal component; Extracting the maximum value of the frequency domain differential feature vector of the other millimeter wave echo-transmitted signal component to obtain the maximum value of the frequency domain differential feature of the other millimeter wave echo-transmitted signal component; The characteristic significance attenuation factor is obtained by calculating the difference between the maximum value of the frequency domain differential feature of the current millimeter wave echo-transmitted signal component and the maximum value of the frequency domain differential feature of other millimeter wave echo-transmitted signal components and dividing the difference by the characteristic value distance span.

6. The high-precision biological perception method based on millimeter-wave radar according to claim 5 is characterized in that: Based on the characteristic significance attenuation factor of each of the frequency domain differential feature vectors of the millimeter wave echo-transmitted signal components, the characteristic significance description factor of each of the frequency domain differential feature vectors of the millimeter wave echo-transmitted signal components is modulated to obtain a sequence of characteristic significance attenuation description factors, including: calculating the positional product between the characteristic significance attenuation factor of each of the frequency domain differential feature vectors of the millimeter wave echo-transmitted signal components and the characteristic significance description factor of each of the frequency domain differential feature vectors of the millimeter wave echo-transmitted signal components to obtain a sequence of characteristic significance attenuation description factors.

7. The high-precision biological perception method based on millimeter-wave radar according to claim 6 is characterized in that: Inputting the sequence of feature significance attenuation description factors into a gated mask module to obtain a sequence of masked feature significance attenuation weight factors, including: Taking the negative number of each feature significance attenuation description factor in the sequence of feature significance attenuation description factors as an exponential power, calculating an exponential function value with a natural constant e as a base to obtain a sequence of feature significance attenuation class support description factors; Calculating the reciprocal of the sum of each feature significance attenuation class support description factor in the sequence of feature significance attenuation class support description factors and a constant one to obtain a sequence of feature significance attenuation weight factors; Based on a predetermined threshold, the sequence of feature significance attenuation weight factors is masked to obtain a sequence of masked feature significance attenuation weight factors, wherein, in response to each feature significance attenuation weight factor in the sequence of feature significance attenuation weight factors being greater than the predetermined threshold, the feature significance attenuation weight factor greater than the predetermined threshold is taken as the original value, otherwise it is set to zero.

8. The high-precision biological perception method based on millimeter-wave radar according to claim 7 is characterized in that: Taking the sequence of the mask feature significance attenuation weight factors as the sequence of weights, calculating the weighted sum of the set of frequency domain differential feature vectors of the millimeter wave echo-transmitted signal component to obtain the frequency domain differential aggregate representation vector of the millimeter wave echo-reflected signal, including: Taking the sequence of the mask feature significance attenuation weight factors as the sequence of weights, calculating the weighted sum between each other millimeter wave echo-transmitted signal component frequency domain differential feature vector in the set of the millimeter wave echo-transmitted signal component frequency domain differential feature vector to obtain the historical millimeter wave echo-transmitted signal component frequency domain differential feature vector; The position-based sum of the historical millimeter-wave echo-transmitted signal component frequency domain differential feature vector and the current millimeter-wave echo-transmitted signal component frequency domain differential feature vector is calculated to obtain the millimeter-wave echo-reflected signal frequency domain differential aggregation representation vector.

9. The high-precision biological perception method based on millimeter-wave radar according to claim 8 is characterized in that: Based on the millimeter wave echo-reflection signal frequency domain differential aggregation representation vector, determining the perception result includes: inputting the millimeter wave echo-reflection signal frequency domain differential aggregation representation vector into a high-precision biometric perception module based on a decoder to obtain the perception result.

10. High-precision biosensing system based on millimeter-wave radar, characterized in that: include: A signal acquisition module is used to acquire millimeter wave detection transmission signals and millimeter wave detection echo signals; A decomposition module, used for performing CEEMDAN decomposition on the millimeter wave detection transmission signal and the millimeter wave detection echo signal to obtain a set of IMF components of the millimeter wave detection transmission signal and a set of IMF components of the millimeter wave detection echo signal; A signal component frequency domain feature extraction module, used to perform signal component frequency domain feature extraction on the set of IMF components of the millimeter wave detection transmission signal and the set of IMF components of the millimeter wave detection echo signal to obtain a set of frequency domain feature vectors of the IMF components of the millimeter wave detection transmission signal and a set of frequency domain feature vectors of the IMF components of the millimeter wave detection echo signal; A differential eigenvector calculation module is used to calculate the differential eigenvector between each corresponding set of the frequency domain eigenvectors of the IMF component of the millimeter wave detection transmission signal and the frequency domain eigenvectors of the IMF component of the millimeter wave detection echo signal in the set of frequency domain eigenvectors of the IMF component of the millimeter wave detection transmission signal to obtain a set of frequency domain differential eigenvectors of the millimeter wave echo-transmission signal component; A feature aggregation module, used for inputting the set of frequency domain differential feature vectors of the millimeter wave echo-transmission signal component into a cluster analysis network based on node feature significant marking to obtain a frequency domain differential aggregation representation vector of the millimeter wave echo-reflection signal; The perception result determination module is used to determine the perception result based on the frequency domain differential aggregation representation vector of the millimeter wave echo-reflection signal.

Images