Non-contact sleep monitoring method and system

Abstract

The invention discloses a non-contact sleep monitoring method, which comprises the following steps of: acquiring echo data of a human body by using a radar, analyzing the echo data, and calculating the point number of a Doppler dimension; judging whether the user has body movement or not according to the point number of the Doppler dimension, and recording the number of body movement times when the user has body movement; calculating a sleep index according to the recorded body movement times, and judging whether the user is in a sleep state or a waking state according to the sleep index; calculating the interval time between two adjacent body movements of the user in the sleep state so as to judge whether the user is in a deep sleep state or a light sleep state; the method is suitable for long-term sleep monitoring.

Description

Technical Field

[0001] The present invention relates to the technical field of signal processing, and in particular to a non-contact sleep monitoring method and system. Background Art

[0002] Long-term poor sleep quality can cause health problems such as metabolic imbalance, cardiovascular disease, and respiratory disease. Monitoring the sleep quality of middle-aged and elderly people and analyzing the factors affecting their sleep quality are of great practical significance for improving the physical health of the middle-aged and elderly population.

[0003] The development of information technology can help improve the sleep quality of middle-aged and elderly people. Health monitoring equipment can collect physiological data of the human body in real time, analyze the health status of the human body through data processing models, and give targeted health advice. For middle-aged and elderly people with chronic diseases such as heart disease and hypertension, sleep health monitoring technology can help them understand their own sleep health status.

[0004] The aging population has brought huge challenges to the healthcare system. The development of sleep health monitoring products for use in home scenarios and the goal of achieving convenience and ease of use are current research hotspots in the field of health monitoring. Although the sleep monitoring products currently used in professional medical institutions can achieve accurate monitoring, they are expensive and difficult to operate, and lack comfort during use. For example, polysomnography equipment requires wet electrodes with conductive paste to be attached to the surface of human skin during use, which can easily cause discomfort to users during use.

[0005] In addition, in the past two years, there has been a lot of research on using millimeter radar to monitor sleep quality. For example, the Chinese invention patent with publication number CN116269298A discloses a non-contact sleep breathing monitoring method and system based on millimeter wave radar. The radar sensor is integrated to scan the upper body of the tester to obtain radar raw data; the radar raw data is preprocessed to obtain heartbeat data, breathing data and breathing and heart rate parameters for each window time; according to the waveform changes of the radar data, abnormal events are located, relevant features are extracted, and respiratory obstruction judgment and body movement event classification judgment are performed; however, it still has the defect of being too complicated to implement, for example, it needs to train a body movement sleep action classification model for auxiliary judgment. Summary of the invention

[0006] In order to solve the problems existing in the prior art, an object of the present invention is to provide a non-contact sleep monitoring method and system, which is suitable for long-term sleep monitoring.

[0007] To achieve the above object, the technical solution adopted by the present invention is: a non-contact sleep monitoring method, comprising the following steps:

[0008] Step 1: Collect human echo data using radar, analyze the echo data, and calculate the number of Doppler points;

[0009] Step 2: judging whether the user has body movement according to the number of Doppler points, and if there is body movement, recording the number of body movements;

[0010] Step 3: Calculate the sleep index according to the recorded body movement times, and determine whether the user is in a sleeping state or awake state according to the sleep index;

[0011] Step 4: Calculate the interval between two adjacent body movements of the user in the sleep state, so as to determine whether the user is in a deep sleep state or a light sleep state.

[0012] As a further improvement of the present invention, in step 1, analyzing the echo data and calculating the number of Doppler dimension points specifically includes:

[0013] Firstly, the echo data is processed by 2D-FFT to convert the distance and Doppler information of the echo data from the time domain to the frequency domain to obtain the range-Doppler heat map;

[0014] Secondly, the minimum constant false alarm detection algorithm is applied to the range-Doppler heat map in the Doppler dimension to extract the distance and speed information of the effective target;

[0015] Then perform azimuth FFT processing on the data after minimum constant false alarm detection, and perform peak search and beamforming technology on the azimuth;

[0016] Finally, FFT processing is performed on the pitch angle to obtain the angle information of the target, and the Cartesian coordinates are transformed into the three-dimensional spherical coordinate point cloud coordinates of the target to obtain a three-dimensional point cloud in a rectangular coordinate system, and the number of multi-dimensional points is calculated.

[0017] As a further improvement of the present invention, in step 2, when the number of Doppler dimension points is greater than a threshold value T1, it is defined as body motion, otherwise it is considered as no body motion.

[0018] As a further improvement of the present invention, in step 3, the sleep index is calculated as follows:

[0019]

[0020] Among them, D is the sleep index, a 1 , a 2 ,......,a n Indicates the number of body movements at each moment, a v is the arithmetic mean.

[0021] As a further improvement of the present invention, in step 3, if the sleep index D<1, it is considered that the user is in a sleeping state, otherwise the user is in an awake state.

[0022] As a further improvement of the present invention, in step 4, if the interval time Time between two adjacent body movements of the user in the sleeping state is greater than the threshold T2, it is determined that the user has entered a deep sleep state, otherwise it is determined to be a light sleep state.

[0023] The present invention also provides a non-contact sleep monitoring system, comprising: a millimeter wave radar sensor module, a data interface conversion and distribution module and an edge terminal module; wherein:

[0024] Millimeter-wave radar sensor module, used for collecting and transmitting raw data;

[0025] The data interface conversion and distribution module is used to package a large amount of raw data and convert the data interface into a general Ethernet link interface to facilitate processing and reception by the edge module;

[0026] The edge module is used for signal processing and human-computer interaction to realize system configuration, edge computing and data communication.

[0027] The beneficial effects of the present invention are:

[0028] The present invention determines the user's body movement by the number of Doppler points, and designs a body movement index based on the body movement data. The body movement index is used to determine whether the user is asleep or awake, and the interval between two adjacent body movements in the sleeping state is calculated to determine whether the user is in deep sleep or light sleep, so that the user can have a clearer understanding of his or her sleeping condition. The method is non-contact, does not affect comfort, and has very high detection accuracy. The sensor size is very small, the installation is very simple, and the power consumption is very low, which is suitable for long-term sleep monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 This is a flowchart of sleep monitoring in an embodiment of the present invention;

[0030] Figure 2 4 is a system block diagram of a monitoring system in an embodiment of the present invention. DETAILED DESCRIPTION

[0031] The embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0032] Example

[0033] like Figure 1 As shown, a non-contact sleep monitoring method comprises the following steps:

[0034] 1) Fix the radar at the head of the bed to collect human body echo data;

[0035] 2) Analyze the echo data and calculate the number of Doppler points;

[0036] 3) If the number of points is greater than the threshold T1, it is defined as body movement, otherwise it is considered as no body movement and the number of body movements is recorded;

[0037] 4) Determine the sleep index D based on the body movement data;

[0038] 5) If D<1, the user is considered to be in a sleeping state, otherwise he is in a wakeful state;

[0039] 6) Calculate the time interval between two consecutive body movements in the sleep state;

[0040] 7) If Time is greater than the threshold T2, the user is determined to be in a deep sleep state, otherwise it is determined to be in a light sleep state.

[0041] In the step 2), the analysis of echo data and calculation of the number of points in the Doppler dimension include the following steps: first, performing 2D-FFT processing on the echo data, converting the distance and Doppler information of the echo data from the time domain to the frequency domain to obtain a range-Doppler heat map (RDM), then performing a minimum constant false alarm detection algorithm on the RDM map in the Doppler dimension to extract the distance and speed information of the effective target, then performing azimuth FFT processing on the data after SO-CFAR detection, performing peak retrieval and beamforming technology on the azimuth, and finally performing FFT processing on the pitch angle to obtain the angle information of the target, and performing Cartesian coordinate conversion on the three-dimensional spherical coordinate point cloud coordinates of the target to obtain a three-dimensional point cloud in a rectangular coordinate system, and calculating the number of points in the Doppler dimension.

[0042] In step 4), the sleep index D refers to:

[0043]

[0044] Among them, a 1 , a 2 ,......,a n , represents the number of body movements at each moment, a v is the arithmetic mean.

[0045] This embodiment also provides a non-contact sleep monitoring system, which consists of three modules: a millimeter wave radar sensor module, a data interface conversion and distribution module, and an edge terminal module.

[0046] Millimeter wave radar sensor module, used for data collection and return. IWR6843AOPEVM radar sensor is selected, and the antenna mode of 2 transmission and 4 reception is configured. LVDS transmitter is used for data return.

[0047] The data interface conversion and distribution module is responsible for packaging a large amount of raw data and converting the data interface into a general Ethernet link interface to facilitate edge processing and reception. The DCA1000EVM data interface conversion and distribution board is used, and the signal interface between it and the above-mentioned radar sensor IWR6843AOPEVM uses a 60-pin Samtec high-density connector. The board uses a Lattice FPGA core with DDR3L to complete the packaging of millimeter-wave radar return data, and sends the packaged data packets to the high-speed Ethernet link interface in LVDS format. At the receiving end, users can receive millimeter-wave radar ADC raw data in LVDS format by providing a good application or a custom programming interface.

[0048] The edge module is responsible for software modules such as signal processing calculation, human-computer interaction, etc., and makes full use of the computing resources of the edge to realize software modules such as configuration module, edge computing module, and data communication module. Because the sleep data is long and large, the edge device needs to provide GPU-level computing power. The present invention uses the NVIDIA Jetson Xavier NX mobile development platform (NX) under NVIDIA Corporation. The development module is a cutting-edge product of the NVIDIA Jetson series. The design purpose of this product is to strengthen edge computing power and promote the marginalization of artificial intelligence algorithms.

[0049] This embodiment can realize sleep monitoring and help users understand their sleep status. The method is non-contact, does not affect comfort, and has very high detection accuracy, which can reach <1mm detection accuracy. The sensor size is very small, the installation is very simple, the power consumption is very low, and it is suitable for long-term monitoring and has important value.

[0050] The above-mentioned embodiments only express the specific implementation of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the present invention. It should be pointed out that for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention.

Claims

1. A non-contact sleep monitoring method, characterized in that: The following steps are involved: Step 1: Collect human echo data using radar, analyze the echo data, and calculate the number of Doppler points; Step 2: judging whether the user has body movement according to the number of Doppler points, and if there is body movement, recording the number of body movements; Step 3: Calculate the sleep index according to the recorded body movement times, and determine whether the user is in a sleeping state or awake state according to the sleep index; Step 4: Calculate the interval between two adjacent body movements of the user in the sleep state, so as to determine whether the user is in a deep sleep state or a light sleep state.

2. The non-contact sleep monitoring method according to claim 1, characterized in that: In step 1, the echo data is analyzed and the number of Doppler dimension points is calculated, which specifically includes: Firstly, the echo data is processed by 2D-FFT to convert the distance and Doppler information of the echo data from the time domain to the frequency domain to obtain the range-Doppler heat map; Secondly, the minimum constant false alarm detection algorithm is applied to the range-Doppler heat map in the Doppler dimension to extract the distance and speed information of the effective target; Then perform azimuth FFT processing on the data after minimum constant false alarm detection, and perform peak search and beamforming technology on the azimuth; Finally, FFT processing is performed on the pitch angle to obtain the angle information of the target, and the Cartesian coordinates are transformed into the three-dimensional spherical coordinate point cloud coordinates of the target to obtain a three-dimensional point cloud in a rectangular coordinate system, and the number of multi-dimensional points is calculated.

3. The non-contact sleep monitoring method according to claim 1, characterized in that: In step 2, when the number of Doppler dimension points is greater than a threshold value T1, it is defined as body motion, otherwise it is considered as no body motion.

4. The non-contact sleep monitoring method according to claim 1, characterized in that: In step 3, the sleep index is calculated as follows: Among them, D is the sleep index, a1, a2, ..., a n Indicates the number of body movements at each moment, a v is the arithmetic mean.

5. The non-contact sleep monitoring method according to claim 1 or 4, characterized in that: In step 3, if the sleep index D<1, the user is considered to be in a sleeping state, otherwise the user is considered to be in an awake state.

6. The non-contact sleep monitoring method according to claim 1, characterized in that: In step 4, if the interval time Time between two adjacent body movements of the user in the sleeping state is greater than the threshold value T2, it is determined that the user has entered a deep sleep state, otherwise it is determined to be a light sleep state.

7. A non-contact sleep monitoring system, characterized in that: include: Millimeter wave radar sensor module, data interface conversion and distribution module and edge terminal module; among which: Millimeter-wave radar sensor module, used for collecting and transmitting raw data; The data interface conversion and distribution module is used to package a large amount of raw data and convert the data interface into a general Ethernet link interface to facilitate processing and reception by the edge module; The edge module is used for signal processing and human-computer interaction to realize system configuration, edge computing and data communication.

Images