Human body breathing micro-motion feature space-time joint extraction method based on millimeter wave radar
By constructing a spatiotemporal joint extraction method for human respiratory micro-motion features using millimeter-wave radar, the problems of insufficient physiological relevance and missing database in the extraction of respiratory signs in existing technologies are solved, a comprehensive reflection of the respiratory micro-motion status and improved physiological interpretability are achieved, supporting intelligent monitoring and disease early warning.
Patent Information
- Application Number
- CN202510625957.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-05-15
- Publication Date
- 2025-09-05
AI Technical Summary
Existing respiratory sign extraction methods lack deep correlation with respiratory physiological mechanisms, making it difficult to fully reflect the spatiotemporal dynamics of chest and abdominal micro-movements and individual breathing pattern differences. In addition, there is a lack of a standardized sign database, resulting in low recognition accuracy and large computational complexity, making it difficult to meet real-time monitoring needs.
By constructing a spatiotemporal joint extraction method for human respiratory micro-motion features based on millimeter-wave radar, using the radar system to obtain 4D micro-motion waveforms, and combining biomedical mechanisms to construct a spatiotemporal cross-modal association model, the respiratory cycle is subdivided into 11 physiological stages and 4 breathing modes, and the time and space domain features are extracted to establish a standardized vital sign database.
It achieves a comprehensive reflection of respiratory micro-movement status and improves physiological interpretability, supports intelligent monitoring and disease early warning, provides high-quality data sets for machine learning, and promotes intelligent interpretation of respiratory health monitoring and early disease screening.
Smart Images
Figure CN120600332A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of biomedical engineering technology, and in particular to a method for spatiotemporal joint extraction of human respiratory micro-motion features based on millimeter-wave radar. Background Art
[0002] Respiratory diseases are common diseases that endanger human health. Their morbidity and mortality rates have continued to rise in recent years. There is an urgent need for non-contact, precise respiratory sign monitoring technology. Bioradar technology uses electromagnetic waves to sense the human body's micro-Doppler signals, and can realize non-contact detection of micro-motion information such as breathing and heartbeat. Among them, millimeter-wave bioradar has broad application prospects in home and community health monitoring due to its strong anti-interference ability, convenience and low cost.
[0003] Existing respiratory sign extraction methods mainly rely on manually designed features and black-box models. However, manually designed features are mostly based on signal processing experience and lack deep correlation with respiratory physiological mechanisms. They are difficult to fully reflect the spatiotemporal dynamics of chest and abdominal micro-movements and individual breathing pattern differences, resulting in limited recognition accuracy. The use of black-box models, although relying on powerful learning capabilities, has problems such as low interpretability, large computational complexity and high sample requirements, making it difficult to meet real-time monitoring needs.
[0004] At the same time, existing feature extraction methods also have significant defects. The time-domain-based feature calculation is simple but only reflects local micro-movements, and the time-frequency domain and complexity features have low interpretability and cannot effectively map physiological states.
[0005] Furthermore, the bioradar field lacks a cross-modal correlation model that integrates spatial and temporal respiratory physiological characteristics with radar signals. Furthermore, a standardized database of vital signs encompassing individual respiratory pattern differences has yet to be established, severely limiting the reliability and intelligence of respiratory health monitoring. Therefore, addressing the incompleteness of features, insufficient physiological correlation, and missing databases in existing technologies, and providing key technical support for accurate respiratory function assessment and disease early warning, remains an urgent challenge. To this end, we propose a method for the joint spatiotemporal extraction of human respiratory micro-motion features based on millimeter-wave radar. Summary of the Invention
[0006] The present invention aims to provide a method for spatiotemporal joint extraction of human respiratory micro-motion features based on millimeter-wave radar to solve the problems raised in the above background technology.
[0007] In order to achieve the above object, the present invention provides the following technical solutions: A method for spatiotemporal joint extraction of human respiratory micro-motion features based on millimeter-wave radar comprises the following steps: S1. Use a millimeter-wave radar system to obtain radar 4D micro-motion waveforms at each scattering point on the surface of the chest and abdominal cavity during a complete human respiratory cycle; S2. Construct a cross-modal correlation model between respiratory physiological characteristics and radar waveforms in the spatiotemporal dimension; S3. Based on the cross-modal correlation model between respiratory physiological characteristics and radar waveforms in the spatiotemporal dimension, respiratory signs are extracted and processed from the radar 4D micro-motion waveforms of each scattering point to obtain the time-domain and spatial-domain respiratory signs of each scattering point, and the human respiratory micro-motion characteristics are derived accordingly.
[0008] Preferably, the radar 4D micro-motion waveform includes three-dimensional spatial position information and time-dimensional micro-motion information.
[0009] Preferably, the cross-modal correlation model between the respiratory physiological characteristics in the spatiotemporal dimension and the radar waveform includes a cross-modal correlation model between the physiological state of the human body in each stage of breathing and the radar waveform and a cross-modal correlation model between the physiological state of the human body in each breathing mode and the radar waveform.
[0010] Preferably, the method for constructing a cross-modal correlation model between respiratory physiological characteristics and radar waveforms in spatiotemporal dimensions is as follows: Cross-modal correlation models between the physiological state of the human body at each stage of breathing and the radar waveform, as well as cross-modal correlation models between the physiological state of the human body and the radar waveform under each breathing mode, are constructed respectively. Based on this, a cross-modal correlation model between respiratory physiological characteristics and radar waveform in the spatiotemporal dimension is obtained. Based on biomedical principles, the human respiratory cycle is divided into 11 respiratory physiological stages. The corresponding relationship between the physiological state of each respiratory physiological stage and the radar waveform is obtained. Based on the corresponding relationship between the physiological state of each respiratory physiological stage and the radar waveform, a cross-modal correlation model between the physiological state of each respiratory stage and the radar waveform is constructed. Based on the principles of physiology and respiratory system dynamics, human breathing patterns are divided into four types. The corresponding relationship between the human physiological state and radar waveform under each breathing pattern is obtained, and a cross-modal correlation model between the human physiological state and radar waveform under each breathing pattern is constructed. The 11 respiratory physiological stages are pre-rest, initial inhalation, rapid inhalation, decelerated inhalation, end-inhalation, inhalation pause, initial exhalation, rapid exhalation, decelerated exhalation. , end-expiratory phase and expiratory pause; The four breathing modes are chest breathing, abdominal breathing, mixed breathing and other breathing modes.
[0011] Preferably, the time-domain respiratory signs include displacement-related signs, velocity-related signs, acceleration-related signs and cross-stage-related signs, and the displacement-related signs include displacement and displacement time accumulation , the speed-related signs include average speed and speed change , the acceleration-related signs include average acceleration and acceleration change , the cross-stage related signs include respiratory rate and amplitude .
[0012] Preferably, the method of performing respiratory sign extraction processing on the radar 4D micro-motion waveform of each scattering point to obtain the time-domain respiratory sign of each scattering point is as follows: According to the cross-modal correlation model between the physiological state of human breathing at each stage and the radar waveform, the radar 4D micro-motion waveform of each scattering point is calculated using the respiratory frequency formula and amplitude formula respectively to obtain the respiratory frequency corresponding to the radar 4D micro-motion waveform of each scattering point. and amplitude ,At the same time, the radar 4D micro-motion waveform of each scattering point is divided into 11 respiratory physiological stages, and the micro-motion waveform corresponding to each respiratory physiological stage is obtained; The respiratory rate formula is: ; in, is the respiratory rate, It is the total time of the inhalation band in the respiratory physiological stage of the radar 4D micro-motion waveform. The total time of the respiratory physiological stage in the radar 4D micro-motion waveform is the exhalation band. The inhalation band includes the initial inspiration, rapid inspiration, deceleration inspiration, the end of inspiration and inspiration pause. The exhalation band includes the initial exhalation, rapid exhalation, deceleration exhalation, the end of exhalation and exhalation pause. The amplitude formula is: ; in, It is the beginning of inhalation. It is the beginning of exhalation. is the amplitude, is the micro-displacement of the radar 4D micro-displacement waveform at the start of inhalation, is the micro-motion displacement of the radar 4D micro-motion waveform at the beginning of exhalation; The micro-motion waveforms corresponding to the respiratory physiological stages of initial inspiration, initial exhalation, end of inspiration and inspiratory pause are calculated by the displacement formula and the displacement time accumulation formula respectively, and the displacement corresponding to the initial inspiration, initial exhalation, end of inspiration and inspiratory pause are obtained. and displacement time accumulation ; The displacement formula is: ; in, is the displacement, It is the time when the respiratory physiological stage ends. It is the time when the respiratory physiological stage begins. is the micro-motion displacement of the radar 4D micro-motion waveform at the end of the respiratory physiological stage, is the micro-motion displacement of the radar 4D micro-motion waveform at the beginning of the respiratory physiological stage; The displacement time accumulation formula is: A= ; in, is the time accumulation of displacement, It is the time when the respiratory physiological stage ends. It is the time when the respiratory physiological stage begins. is the micro-motion displacement of the radar 4D micro-motion waveform at time t; The micro-motion waveforms corresponding to rapid inspiration and rapid exhalation in the respiratory physiological stage are calculated by the average speed formula and the speed change formula respectively, and the average speed corresponding to rapid inspiration and rapid exhalation are obtained. and speed change ; The average speed formula is: ; in, is the average speed, is the micro-motion displacement of the radar 4D micro-motion waveform at the end of the respiratory physiological stage, is the micro-motion displacement of the radar 4D micro-motion waveform at the beginning of the respiratory physiological stage, It is the time when the respiratory physiological stage ends. It is the time when the physiological phase of breathing begins; The speed change formula is: ; in, is the velocity change, is the micro-motion displacement of the radar 4D micro-motion waveform at the end of the respiratory physiological stage, is the micro-motion displacement of the radar 4D micro-motion waveform at the beginning of the respiratory physiological stage; The micro-motion waveforms corresponding to the decelerated inspiration and decelerated exhalation physiological stages of breathing are calculated using the average acceleration formula and the acceleration change formula respectively, and the average acceleration corresponding to the decelerated inspiration and decelerated exhalation are obtained. and acceleration change ; The average acceleration formula is: ; in, is the average acceleration, is the micro-motion displacement of the radar 4D micro-motion waveform at the end of the respiratory physiological stage, is the micro-motion displacement of the radar 4D micro-motion waveform at the beginning of the respiratory physiological stage, It is the time when the respiratory physiological stage ends. It is the time when the physiological phase of breathing begins; The acceleration change formula is: ; in, is the change in acceleration, is the micro-motion displacement of the radar 4D micro-motion waveform at the end of the respiratory physiological stage, is the micro-motion displacement of the radar 4D micro-motion waveform at the beginning of the respiratory physiological stage; Based on this, the time domain respiratory signs of each scattering point are obtained.
[0013] Preferably, the airspace respiratory signs include space sign parameters ,average value and variance .
[0014] Preferably, the method of extracting respiratory signs from the radar 4D micro-motion waveform of each scattering point to obtain the spatial respiratory signs of each scattering point is as follows: According to the cross-modal correlation model between the physiological state of the human body and the radar waveform in each breathing mode, the radar 4D micro-motion waveform of each scattering point is calculated by the spatial sign parameter formula, spatial average formula and spatial variance formula respectively, and the spatial sign parameters corresponding to the radar 4D micro-motion waveform of each scattering point are obtained. ,average value and variance ; The spatial sign parameter formula is: ; in, is the spatial sign parameter, It is the physical parameters including displacement, velocity and acceleration along with the three-dimensional space position information. The function of change, and It represents different spatial regions; The spatial average formula is: ; in, is the average value, It is the physical parameters including displacement, velocity and acceleration along with the three-dimensional space position information. The function of change, is a spatial region; The spatial variance formula is: ; in, is the variance, It is the physical parameters including displacement, velocity and acceleration along with the three-dimensional space position information. The function of change, is a spatial region, is the average value.
[0015] Preferably, the method for obtaining the human body's respiratory micro-motion characteristics is: Each scattering point is labeled with a unique corresponding ID number. At the same time, the time-domain respiratory signs and space-domain respiratory signs of each scattering point are associated with the corresponding ID number to obtain the time-domain and space-domain respiratory signs of each scattering point. Based on this, the human respiratory micro-motion characteristics containing the time-domain and space-domain respiratory signs of each scattering point are constructed.
[0016] Compared with the prior art, the present invention has the following beneficial effects: 1. This invention achieves joint spatiotemporal feature extraction to comprehensively reflect the state of respiratory micro-motion. It uses millimeter-wave radar to obtain 4D micro-motion waveforms containing three-dimensional spatial position and time dimensions. Combining biomedical mechanisms, it constructs a cross-modal correlation model between spatiotemporal respiratory physiological characteristics and radar waveforms. It simultaneously extracts time-domain and spatial-domain features, achieving a joint spatiotemporal analysis of the micro-motion state of each scattering point in the chest and abdominal cavity, comprehensively reflecting individual respiratory patterns and physiological differences.
[0017] 2. Driven by biomedical mechanisms, the present invention improves the physiological interpretability of features. To address the problem of disconnection between the features of traditional methods and respiratory physiological mechanisms, the present invention subdivides the respiratory cycle into 11 respiratory physiological stages based on the principles of respiratory system dynamics, clarifies the correspondence between the radar waveform characteristics of each stage and the lung ventilation status and the force of the respiratory muscles, and simultaneously establishes a cross-modal correlation model between the respiratory physiological status and the radar signal based on the spatial motion differences of the four respiratory modes. The extracted features can directly map the activity of the respiratory muscles and the ventilation pattern, significantly improving the physiological interpretability of the features and providing a basis for accurate assessment of respiratory function.
[0018] 3. This invention provides a foundation for the construction of a standardized vital sign database and supports intelligent expansion. The existing technology lacks a respiratory sign database that integrates spatiotemporal dimensions, which restricts the intelligent development of subsequent classification, identification, and disease early warning. This invention constructs a standardized vital sign database that includes respiratory stages, respiratory patterns, and physiological parameters by labeling and associating the spatiotemporal characteristics of each scattering point, providing a high-quality data set for machine learning model training, promoting the upgrade of respiratory health monitoring from signal processing to intelligent interpretation, and providing key technical support for early screening and intervention of respiratory diseases.
[0019] Figures in the specification In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.
[0020] Figure 1 The figure is a flow chart of the method steps of the present invention. DETAILED DESCRIPTION
[0021] To make the objectives, technical solutions, and advantages of the present invention more apparent, the technical solutions of the present invention will be described in detail below. Obviously, the embodiments described are only some of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other implementations obtained by those of ordinary skill in the art without inventive effort are within the scope of protection of the present invention.
[0022] Examples, such as Figure 1 As shown, a method for spatiotemporal joint extraction of human respiratory micro-motion features based on millimeter wave radar includes the following steps: S1. Use a millimeter-wave radar system to obtain radar 4D micro-motion waveforms at each scattering point on the surface of the chest and abdominal cavity during a complete human respiratory cycle; S2. Construct a cross-modal correlation model between respiratory physiological characteristics and radar waveforms in the spatiotemporal dimension; S3. Based on the cross-modal correlation model between respiratory physiological characteristics and radar waveforms in the spatiotemporal dimension, respiratory signs are extracted and processed from the radar 4D micro-motion waveforms of each scattering point to obtain the time-domain and spatial-domain respiratory signs of each scattering point, and the human respiratory micro-motion characteristics are derived accordingly.
[0023] Furthermore, the working principle of the present invention is described below by way of examples: A millimeter-wave radar system is used to obtain radar 4D micro-motion waveforms of each scattering point on the surface of the chest and abdominal cavity during a complete respiratory cycle of the human body. The millimeter-wave radar system uses a millimeter-wave radar with a center frequency of 77GHz, a transmission power of 10dBm, a bandwidth of 4GHz, a distance resolution of 0.0375m, an angle resolution of 1.5°, and a sampling frequency of 1kHz to ensure high-precision perception of micro-motion displacement on the surface of the chest and abdominal cavity. The subjects are placed in a supine position and remain in a resting state. The radar antenna array is aimed at the chest and abdominal cavity area, covering the range from the sternum to the pubic symphysis, and laterally covering the bilateral mid-axillary line. The acquisition time is 300 seconds, ensuring that at least 5 complete respiratory cycles are included, and each cycle contains 11 respiratory physiological stages. The radar signal is sampled at the intermediate frequency, pulse compressed, and processed by FFT to obtain the three-dimensional spatial position information of each scattering point. And the time dimension micro-displacement sequence, forming a 4D micro-displacement waveform, with matrix Indicates that is the spatial coordinate, For time series.
[0024] A cross-modal correlation model of respiratory physiological characteristics and radar waveforms in the spatiotemporal dimension was constructed. Based on biomedical principles, the respiratory cycle was divided into 11 respiratory physiological stages, and the physiological characteristics of each stage were defined. The physiological characteristics of each stage are as follows: initial inspiration is the initiation of diaphragm contraction and slow expansion of the chest and abdominal cavity; rapid inspiration is the coordinated rapid contraction of the diaphragm and intercostal muscles and a sharp increase in chest cavity volume; deceleration inspiration is the weakening of respiratory muscle contraction force and slowing of expansion speed; the terminal inspiration is the cessation of contraction and peak chest cavity volume; inspiratory pause is the brief stillness of respiratory muscles and maintenance of the inspiratory state; initial exhalation is the relaxation of the diaphragm and passive retraction of the chest cavity; rapid exhalation is the active contraction of the intercostal muscles and accelerated gas expulsion; deceleration exhalation is the weakening of contraction force and reduction of retraction speed; the terminal exhalation is the cessation of retraction and the minimum chest cavity volume; expiratory pause is the stillness of respiratory muscles and maintenance of the exhalation state. By synchronously collecting respiratory belt sensor signals and radar data, a cross-modal correlation model is established between the physiological state of the human body at each stage of breathing and the radar waveform, such as displacement, velocity and acceleration. For example, the rapid inhalation stage corresponds to the high velocity gradient of the radar waveform, and the deceleration exhalation stage corresponds to the decrease in the negative acceleration value. According to the principles of physiological medicine and respiratory system dynamics, breathing methods are divided into four categories, namely thoracic breathing, which is mainly based on intercostal muscle activity and has significant chest fluctuations; abdominal breathing, which is mainly based on diaphragm activity and has significant abdominal fluctuations; mixed breathing is the coordinated movement of the chest and abdomen and other breathing methods, such as abnormal breathing patterns. Data is collected from subjects with different breathing methods, and the spatial motion differences of scattering points in various regions of the chest and abdominal cavity are analyzed to establish a cross-modal correlation model between the physiological state of the human body and the radar waveform under each breathing method. For example, the displacement of scattering points in the abdominal area of abdominal breathing Significantly larger than the thoracic area.
[0025] Extract the time-domain respiratory signs and air-domain respiratory signs of each scattering point, construct the human respiratory micro-motion features, extract the time-domain respiratory signs, extract the cross-stage features including respiratory frequency and amplitude, and extract the specific features of each respiratory physiological stage of displacement-related signs, velocity-related signs or acceleration-related signs. For air-domain respiratory sign extraction, extract the spatial sign parameters Q, which are used to reflect the physical parameter gradients including displacement, velocity or acceleration of adjacent spatial regions such as the chest and abdominal cavities, as well as the mean E and variance V, which are used to describe the spatial distribution characteristics of the physical parameters of the scattering points in the entire chest and abdominal cavity. Assign a unique ID to each scattering point, such as Thorax_01 and Abdomen_05, and associate the time-domain features with the air-domain features according to the ID to form a respiratory sign vector containing time and space dimensions. ,Finally, the spatiotemporal joint respiratory micro-motion features including the scattering points of the entire chest and abdominal cavity are constructed and ,represented in matrix form.
[0026] The above-described embodiments are only used to illustrate the technical solutions of the present application, and not to limit them. Modifications to the technical solutions described in the aforementioned embodiments, or equivalent replacements of some of the technical features therein, do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application, and should all be included in the scope of protection of the present application.
Claims
1. A method for spatiotemporal joint extraction of human respiratory micro-motion features based on millimeter wave radar, characterized in that: The following steps are involved: S1. Use a millimeter-wave radar system to obtain radar 4D micro-motion waveforms at each scattering point on the surface of the chest and abdominal cavity during a complete human respiratory cycle; S2. Construct a cross-modal correlation model between respiratory physiological characteristics and radar waveforms in the spatiotemporal dimension; S3. Based on the cross-modal correlation model between respiratory physiological characteristics and radar waveforms in the spatiotemporal dimension, respiratory signs are extracted and processed from the radar 4D micro-motion waveforms of each scattering point to obtain the time-domain and spatial-domain respiratory signs of each scattering point, and the human respiratory micro-motion characteristics are derived accordingly.
2. The method for extracting human respiratory micro-motion characteristics in time and space based on millimeter wave radar according to claim 1, characterized in that: The radar 4D micro-motion waveform includes three-dimensional spatial position information and one-dimensional temporal micro-motion information.
3. The method for extracting human respiratory micro-motion characteristics in time and space based on millimeter wave radar according to claim 1, characterized in that: The cross-modal correlation model between the respiratory physiological characteristics in the spatiotemporal dimension and the radar waveform includes a cross-modal correlation model between the physiological state of the human body at each stage of breathing and the radar waveform and a cross-modal correlation model between the physiological state of the human body under each breathing mode and the radar waveform.
4. The method for extracting human respiratory micro-motion characteristics in time and space based on millimeter wave radar according to claim 3, characterized in that: The method for constructing a cross-modal correlation model between respiratory physiological characteristics and radar waveforms in the spatiotemporal dimension: Cross-modal correlation models between the physiological state of the human body at each stage of breathing and the radar waveform, as well as cross-modal correlation models between the physiological state of the human body and the radar waveform under each breathing mode, are constructed respectively. Based on this, a cross-modal correlation model between respiratory physiological characteristics and radar waveform in the spatiotemporal dimension is obtained. Based on biomedical principles, the human respiratory cycle is divided into 11 respiratory physiological stages. The corresponding relationship between the physiological state of each respiratory physiological stage and the radar waveform is obtained. Based on the corresponding relationship between the physiological state of each respiratory physiological stage and the radar waveform, a cross-modal correlation model between the physiological state of each respiratory stage and the radar waveform is constructed. Based on the principles of physiology and respiratory system dynamics, human breathing patterns are divided into four types. The corresponding relationship between the human physiological state and radar waveform under each breathing pattern is obtained, and a cross-modal correlation model between the human physiological state and radar waveform under each breathing pattern is constructed. The 11 respiratory physiological stages are pre-resting stage, initial inhalation, rapid inhalation, decelerated inhalation, end-inhalation stage, inhalation pause, initial exhalation, rapid exhalation, decelerated exhalation, end-expiratory phase and expiratory pause; The four breathing modes are chest breathing, abdominal breathing, mixed breathing and other breathing modes.
5. The method for extracting human respiratory micro-motion characteristics in time and space based on millimeter wave radar according to claim 1, characterized in that: The time domain respiratory signs include displacement related signs, velocity related signs, acceleration related signs and cross-stage related signs. The displacement related signs include displacement and displacement time accumulation , the speed-related signs include average speed and speed change , the acceleration-related signs include average acceleration and acceleration change , the cross-stage related signs include respiratory rate and amplitude .
6. The method for extracting human respiratory micro-motion characteristics in time and space based on millimeter wave radar according to claim 5, characterized in that: The method of extracting respiratory signs from the radar 4D micro-motion waveform of each scattering point to obtain the time-domain respiratory signs of each scattering point is as follows: According to the cross-modal correlation model between the physiological state of human breathing at each stage and the radar waveform, the radar 4D micro-motion waveform of each scattering point is calculated using the respiratory frequency formula and amplitude formula respectively to obtain the respiratory frequency corresponding to the radar 4D micro-motion waveform of each scattering point. and amplitude ,At the same time, the radar 4D micro-motion waveform of each scattering point is divided into 11 respiratory physiological stages, and the micro-motion waveform corresponding to each respiratory physiological stage is obtained; The micro-motion waveforms corresponding to the respiratory physiological stages of initial inspiration, initial exhalation, end of inspiration and inspiratory pause are calculated by the displacement formula and the displacement time accumulation formula respectively, and the displacement corresponding to the initial inspiration, initial exhalation, end of inspiration and inspiratory pause are obtained. and displacement time accumulation ; The micro-motion waveforms corresponding to rapid inspiration and rapid exhalation in the respiratory physiological stage are calculated by the average speed formula and the speed change formula respectively, and the average speed corresponding to rapid inspiration and rapid exhalation are obtained. and speed change ; The micro-motion waveforms corresponding to the decelerated inspiration and decelerated exhalation physiological stages of breathing are calculated using the average acceleration formula and the acceleration change formula respectively, and the average acceleration and acceleration change corresponding to the decelerated inspiration and decelerated exhalation are obtained. ; Based on this, the time domain respiratory signs of each scattering point are obtained.
7. The method for extracting human respiratory micro-motion characteristics in time and space based on millimeter wave radar according to claim 1, characterized in that: The airspace respiratory signs include space sign parameters ,average value and variance .
8. The method for extracting human respiratory micro-motion characteristics in time and space based on millimeter wave radar according to claim 7, characterized in that: The method of extracting respiratory signs from the radar 4D micro-motion waveform of each scattering point to obtain the spatial respiratory signs of each scattering point is as follows: According to the cross-modal correlation model between the physiological state of the human body and the radar waveform in each breathing mode, the radar 4D micro-motion waveform of each scattering point is calculated by the spatial sign parameter formula, spatial average formula and spatial variance formula respectively, and the spatial sign parameters corresponding to the radar 4D micro-motion waveform of each scattering point are obtained. ,average value and variance .
9. The method for extracting human respiratory micro-motion characteristics in time and space based on millimeter wave radar according to claim 1, characterized in that: The method for obtaining the micro-motion characteristics of human respiratory movement: Each scattering point is labeled with a unique corresponding ID number. At the same time, the time-domain respiratory signs and space-domain respiratory signs of each scattering point are associated with the corresponding ID number to obtain the time-domain and space-domain respiratory signs of each scattering point. Based on this, the human respiratory micro-motion characteristics containing the time-domain and space-domain respiratory signs of each scattering point are constructed.