CN106667435A - Intelligent sensing mattress for monitoring sleep - Google Patents

Abstract

The invention belongs to the technical field of sleep monitoring, in particular to an intelligent sensor mattress for sleep monitoring. The present invention adopts two kinds of flexible pressure sensors, the first kind of sensor is used to make the pressure sensing array, which is evenly distributed in the whole mattress, and is used to detect the stress status of most areas of the mattress; the second kind of sensor is distributed in the mattress Several key positions are used to collect the pressure changes caused by the user's breathing, body movement and heart rhythm signals, and convert the pressure change signals into electrical signals; the subsequent hardware will amplify, filter, and transmit the obtained electrical signals, and then pass them through the PC. Feature extraction and pattern recognition processing are performed on the terminal to obtain pressure distribution images and sleep analysis. In the present invention, the sensor and its electrodes are all made of ultra-thin flexible materials, and the mattress is thin, light in weight, and easy to carry; because it is non-invasive, it can perform sleep monitoring without the user's awareness, completely restoring the mattress. The user's natural sleep is a powerful supplement to home care.

Description

Sleep monitoring smart sensor mattress

technical field

The invention belongs to the technical field of sleep monitoring, and in particular relates to a non-invasive sleep monitoring device.

Background technique

At present, the gold standard for clinical sleep monitoring of patients is polysomnography monitoring technology, but the structure of polysomnography monitoring technology is complex and inconvenient to operate, which causes a certain psychological burden on patients and seriously affects their normal sleep. Therefore, sleep monitoring is developing in a non-invasive direction, and strives to restore natural sleep to the greatest extent. In current clinical practice, whole-night polysomnography (PSG) is the gold standard for sleep monitoring. PSG measurements are usually only available in hospital sleep laboratories. PSG includes multiple physiological signals such as ECG, EEG, EMG, eye movement tracking, and blood oxygen saturation. The system uses many gel electrodes and bulky equipment directly connected to the human body through numerous wires, which greatly interferes with the natural way of sleeping, and is uncomfortable and inconvenient, so user acceptance is very low. For example, when a patient needs to go to the bathroom, he or she has to disconnect all the connected sensors one by one, come back, reconnect them, and activate them, which is a fairly complicated process. Moreover, due to the interference with natural sleep, the measurement of PSG cannot fully represent the real sleep condition, so there are errors in the diagnosis and treatment based on it.

At present, the sensor mattresses that can perform non-invasive sleep monitoring on the market mainly use embedded pressure sensors to detect the pressure signals of the user's body through the sensors embedded in the mattress, and perform body motion sensing, Bed-leaving sensing, respiration detection and heart rate detection, the above physiological signals are analyzed and processed to obtain the sleep condition of the user. At the same time, there are articles that show that the device for generating human body pressure distribution images can be made into the form of mattresses, but the sensor mattresses on the market cannot generate pressure distribution images, and cannot obtain the user's sleeping posture and pressure-concentrated parts. The image resolution of the pressure distribution is low, and only the sleeping posture and pressure concentration of the human body can be roughly seen; and the embedded sensors are relatively concentrated, and the quality of the analysis results depends on the relative position of the sensor and the user, so the mattress Installation and use require professional guidance. In recent years, many research results on "weakly invasive" or mobile sleep monitoring systems have been reported, for example: smart mattresses realize BCG monitoring [1]; force-sensitive textile fabrics are used to make pressure The sensor array is implanted into the bed sheet to realize the human posture change and respiration monitoring; the smart watch or wristband measures the PPG signal [2]; the peripheral arterial sound is monitored at the fingertips [3]; the sleep level analysis is realized through the analysis of the cardiopulmonary activity recording data and classification [4]; application of forehead near-infrared technology to detect apnea during sleep [5]; wearable devices launched by Xiaomi, Actiwatch, Fitbit and other companies to record body movements during sleep [6, 7] and so on. However, these systems often only measure some of the parameters required for sleep analysis, and do not provide a comprehensive data measurement solution to meet the clinical diagnosis of sleep diseases. Moreover, there are still some technical difficulties in the above-mentioned various new technologies, such as: reliable and stable acquisition of standard ECG and respiratory signals, how to improve the robustness of detection under the interference of motion artifacts, and how to eliminate the noise introduced by different sleeping postures. error. At present, there is no research that can realize the system to be developed in this project to be "completely non-intrusive to daily sleep", "suitable for comfortable sleep", and to achieve high reliability and stability of signal acquisition.

[1] Kortelainen, et al. Sleep staging based on signals acquiredthrough bed sensor. IEEE TITB., 2010.

[2] Sady, et al. Automatic sleep staging from ventilator signals in non-invasive ventilation. Comput. Biol. Med., 2013.

[3] J. Hedner, et al. Sleep staging based on autonomic signals: a multi-center validation study. J. Clin. Sleep Med., 2011.

[4] Sleep profiler, available from http://www.advancedbrainmonitoring.com/sleep-profiler/

[5] PortaNIRS Screening for Sleep Apnea, available from http://www.portanirs.com/

[6] Xiaomi smart bracelet, www.mi.com/shouhuan

[7] Fitbit ONE Wireless Activity and Sleep Tracker. Available from https://www.fitbit.com.

Related companies are: Shanghai Maidong Medical Instrument Co., Ltd.; American Jesse Medical Co., Ltd.

Contents of the invention

The purpose of the present invention is to provide a non-invasive sleep monitoring device with high reliability and good stability.

The non-invasive sleep monitoring device provided by the present invention is in the form of a mattress, which is an intelligent sensor mattress, in which two kinds of flexible pressure sensors are used, the first sensor is used to make a pressure sensing array, and the first The sensor is a square sheet with a side length of 0.6-0.9cm, and the distance between each sensor is 0.3-0.45cm, which is evenly distributed throughout the mattress, so the force status of most areas of the mattress can be obtained, and according to each The pressure on each sensor is converted into a single mapped electrical signal, and the electrical signal is converted into a matrix color block corresponding to the position of the sensor on the pressure distribution image. The chromaticity of the matrix color block increases with the pressure, so The pressure distribution image can not only obtain the posture of the user, but also the concentrated part of the pressure of the user; the second sensor is distributed in several key positions of the mattress, such as the position of the heart corresponding to the mattress when a person is lying flat, and the position of the heart relative to the mattress. The symmetrical position of the spine, the central position of the human torso and the corresponding position of the human waist on the mattress facilitate the collection of pressure changes caused by the user's breathing, body movement and heart rhythm signals, and convert these pressure change signals into electrical signals. Subsequent circuits and processors amplify, filter, and transmit the obtained electrical signals, and then perform feature extraction, pattern recognition, and other processing through the PC to obtain pressure distribution images and sleep analysis. The thickness of the flexible material adopted in the present invention is all in the order of microns, and the expandability is extremely strong; the comfort can be improved by adjusting the thickness according to the actual situation.

The structure of the smart sensor mattress provided by the present invention includes: an upper buffer layer and a lower base layer, and in the middle of the two are the placement layer of the second pressure sensor, the first pressure sensor array layer, or the second pressure sensor array layer. The mixed layer of the pressure sensor and the first pressure sensor array; in addition, it also includes: the upper and lower electrode layers of the first pressure sensor array layer, and the isolation layer between the two pressure sensor layers.

In the present invention, the buffer layer is used to prevent the pressure sensor from being in direct contact with the outside world, and properly buffer the pressure from the human body.

In the present invention, the base layer is used to prevent the lower electrode from directly contacting with the outside world and play a protective role.

In the present invention, the isolation layer is used to prevent the structure of the upper layer from contacting the pressure sensor layer due to gravity, so as to avoid affecting the accuracy of the signal.

In the present invention, a protective layer is provided between the pressure sensor layer and the corresponding electrode layer to prevent the pressure sensor from contacting the electrode layer to generate interference signals.

The electrical schematic diagram of the smart sensor mattress designed by the present invention is shown in Figure 4. Also includes: components related to the first sensor (pressure sensor array): cycle scan acquisition module, amplifier and filter circuit, and components related to the second sensor: amplifier and filter circuit; in addition, processing of subsequent circuits Devices, power modules, clock modules, storage modules, data transmission devices, PC terminals, digital signal processors, etc. The processor controls the cyclic scanning acquisition module to select the embedded first sensor array, scan the pressure sensor array signal, and collect the pressure signal of each position of the human body, and the signal is sent to the processor after being amplified by the amplifier and filtered by the filter circuit; the processor The second sensor is controlled to work, and the electrical signal obtained by the second sensor is sent to the processor through an amplification circuit and a filter circuit. The processor performs analog-to-digital conversion on the analog signal processed by the amplification circuit and the filter circuit through the built-in analog-to-digital converter to obtain a digital signal; the processor puts the obtained digital signal into the storage module, and can transmit the digital signal through data transmission The module is transmitted to the PC by wireless transmission; the data transmission module is composed of an integrated WIFI module and peripheral circuits; the power supply module of the processor is responsible for the power supply of the entire circuit system, and converts the 220V mains power into the voltage required by the system, including Modules such as rectification and voltage stabilization; the clock module of the processor is responsible for the timing control of the entire circuit system, ensuring that the hardware circuit can smoothly work according to the frequency and timing set by the software program.

On the PC side, the signal is pre-processed such as filtering, noise reduction, and waveform smoothing, and then the signal of the entire night is segmented in time sequence, for example, every 30-60 seconds is a segment, and feature extraction is performed on each segment of the signal. The features include body movement features, breathing features, heart rhythm features and posture features; and sleep states are classified according to the features:

First, perform the first step of classification, that is, classify all segment signals according to the obtained body motion characteristics, and specifically divide them into two categories: if the body motion characteristics meet the corresponding requirements, it is classified into the "night awake + light sleep" category; if Those who are not satisfied are divided into "rapid eye movement sleep + deep sleep";

Then, the second step of classification is carried out, that is, pattern recognition is performed on the existing two categories according to breathing characteristics, heart rate characteristics and posture characteristics, and further divided into "night awake", "light sleep", "deep sleep" and "rapid eye sleep". "Active sleep", and then perform sleep analysis according to the proportion of the four sleep stages, or choose an appropriate wake-up time, etc.

The sleep analysis results obtained from the final monitoring will be displayed to the user in the form of a report through the mobile APP or PC software to help users improve their sleep quality.

The intelligent sensor mattress designed by the present invention can not only perform body motion sensing, bed-leaving sensing, breathing detection and heart rate detection, but also can obtain the user's sleep posture image and pressure concentration part, and the resolution of the image is higher than that in the literature (the sensor arrangement rate per unit area directly affects the resolution of the pressure distribution image, while the sensor arrangement rate in the literature is 2621/m ² , the sensor arrangement rate of the present invention is 5056-11378/m ² ), which can display the user's pressure distribution and sleeping posture more clearly; at the same time, it can ensure that the user can obtain clear physiological signals and accurate sleep analysis no matter where the user is in the bed; and the present invention is light and thin, foldable and easy to use or install and use by themselves.

For example, postoperative personnel can use the mattress to prevent the pressure from being concentrated on the surgical wound during sleep and ensure the wound to heal as soon as possible. When the mattress is used by the elderly, the pressure alarm device can be used to prevent bedsores. At the same time, its better sleep structure analysis and early warning of some sleep-related diseases can enable users to get further care.

In the present invention, a flexible sensor array is combined with a flexible conductive fabric to obtain a sensor structure with a thickness of micron. The sensor structure can be spread throughout the entire mattress, and the close combination of the sensor or conductive material and the flexible fabric uses high-precision equipment, so it can be Arrange more sensors with higher precision. When the user is lying on the bed, the high-precision flexible sensor can perceive the physiological signals of the human body, and send the user's sleep status (sleep structure, sleep posture, stress concentration, and whether there are related diseases, etc.) Own. Among them, the remote analysis applies the latest algorithm in the field of sleep monitoring, which can obtain higher precision and operation speed. Since the sensors and electrodes are made of flexible materials and combined with fabrics, they can be folded and cleaned, and are easy to install and carry.

Technical effect:

1. The pressure sensor array made of the first sensor (piezoresistive sensor) can generate a higher-resolution pressure distribution image, but its structure can reduce the number of leads as much as possible, thereby reducing the weight of the mattress and facilitating production and production;

2. Since the physiological signals collected by the second sensor (piezoelectric sensor) can be screened and the data with the best signal quality can be selected for analysis, the sleep monitoring is more accurate;

3. Due to the ultra-thin and flexible materials used in the structure of the sensor and its electrodes, the mattress is thin, light in weight, and easy to carry, and can be used in various occasions such as homes and institutions;

4. Due to its non-invasive characteristics, it can perform sleep monitoring without the user's awareness, completely restore the user's natural sleep, and is more easily accepted by people. It is a powerful supplement to home care.

Description of drawings

Figure 1 The first structure of the smart mattress.

Figure 2 The second structure of the smart mattress.

Fig. 3 Schematic diagram of the mixed structure layer of two kinds of pressure sensors.

Fig. 4 Schematic diagram of smart sensor mattress.

Fig. 5 Schematic diagram of the cycle scanning acquisition module.

Figure 6 Data flow diagram.

Figure 7. The physical picture of the sample.

Fig. 8 Image of pressure distribution of human body posture.

detailed description

The intelligent sensor mattress provided by the present invention can be designed into two typical structures.

The first is a 9-layer structure, from top to bottom: the first layer is a buffer layer, which prevents the pressure sensor from directly contacting the outside world and properly buffers the pressure from the human body. The second layer is the placement layer of the second type of pressure sensor, and the second type of pressure sensor is fixed at different positions of the placement layer according to the design to ensure the normal acquisition of signals. The third layer is a protective layer, which is mainly used to prevent the second pressure sensor from contacting with the electrode layer to generate interference signals. The fourth layer is the upper electrode layer, which is used to gate the first pressure sensor array. The fifth layer is an isolation layer, which is used to prevent the structure of the upper layer from contacting the pressure sensor array due to gravity, which will affect the accuracy of the signal. The sixth layer is the first pressure sensor array layer, which fixes a large number of pressure sensors. The seventh layer is an isolation layer, which is used to prevent the structure of the upper layer from contacting the pressure sensor array due to gravity, which will affect the accuracy of the signal. The eighth layer is the lower electrode layer, which is used to gate the first pressure sensor array. The ninth layer is the base layer, which is used to prevent the lower electrode from directly contacting with the outside world and play a protective role. The first structure is shown in Figure 1.

The second structure is an 8-layer structure, from top to bottom: the first layer is a buffer layer, which prevents the pressure sensor from directly contacting the outside world and properly buffers the pressure from the human body. The second layer is a protective layer, which is mainly used to prevent the second pressure sensor from contacting with the electrode layer to generate interference signals. The third layer is the upper electrode layer, which is used to gate the first pressure sensor array. The fourth layer is an isolation layer, which is used to prevent the structure of the upper layer from contacting the pressure sensor array due to gravity, which will affect the accuracy of the signal. The fifth layer is a mixed structure layer of two types of pressure sensors. The thin strip-shaped second pressure sensor is tightly bonded to the insulating gap in the first pressure sensor array with sol, which can simplify the overall structure and facilitate the cleaning and replacement of other layers. The gap of the first type of pressure sensor at the position where the second type of pressure sensor is distributed is larger than that of other places, which facilitates the bonding of the second type of pressure sensor, as shown in FIG. 3 . Because the first pressure sensor adopts such a sparse distribution only where the second pressure sensor is distributed, if 6 second pressure sensors are distributed throughout the mattress, then the second structure is 30 less than the first structure For the first type of pressure sensor, the resolution drops by 2.6‰, which hardly affects the image accuracy. The sixth layer is an isolation layer, which is used to prevent the structure of the upper layer from contacting the pressure sensor array due to gravity, which will affect the accuracy of the signal. The seventh layer is the lower electrode layer, which is used to gate the first pressure sensor array. The eighth layer is the base layer, which is used to prevent the lower electrode from directly contacting the outside world and play a protective role. The second structure is shown in Figure 2.

Taking the first structure of the present invention as an example, its schematic diagram is shown in Figure 4. The two boxes in the figure are the hardware parts of the present invention, and the upper box (red box) is the first pressure sensor array And its peripheral hardware part, the 4th, 5th, 6th, 7th, 8th, 9th layer structure of the 9-layer structure belongs to this frame; the lower frame (blue box) is the second pressure sensor and its peripheral hardware part , the 1st, 2nd, and 3rd layers of the 9-layer structure belong to the frame.

When the user is lying on the mattress, the processor controls the cyclic scanning acquisition module to select the embedded first sensor array, and the electrical signals of all 11378 sensors pass through the amplification circuit and the filter circuit in turn, and then are sent to the processor . The amplifying circuit is composed of an amplifier chip and peripheral electronic devices, and is responsible for amplifying the weak voltage signal to facilitate accurate identification of subsequent circuits. The filter circuit is composed of an amplifier chip and peripheral electronic devices, forming a bandpass filter, which is responsible for initially filtering the noise introduced by the 50Hz mains and analog circuits, and reducing the difficulty of digital signal processing.

At the same time, the processor controls the second sensor to work continuously, passes the electrical signal obtained by the second sensor through the amplification circuit and the filter circuit, and then sends it to the processor. The amplifying circuit and filter circuit connected to the second sensor are similar in principle to the amplifying circuit and filter circuit of the first sensor, except that the chips and electronic devices selected are different.

After the analog signal processed by the amplification circuit and the filter circuit, the analog-to-digital conversion is performed by the built-in analog-to-digital converter of the processor to obtain a 10-bit digital signal. The processor puts the obtained digital signal into the storage module and can store the digital signal. Through the data transmission module, it is transmitted to the PC terminal by wireless transmission. The data transmission module is composed of an integrated WIFI module and peripheral circuits. The power supply module of the processor is responsible for the power supply of the entire circuit system, and converts the 220V mains power into the voltage required by the system, including rectification, voltage stabilization and other modules. The clock module of the processor is responsible for the timing control of the entire circuit system, ensuring that the hardware circuit can work smoothly according to the frequency and timing set by the software program.

The cyclic scanning acquisition module adopted in the present invention is shown in Figure 5. The present invention has a total of 128*64=8192 pressure sampling points. For the sensor matrix formed by the first sensor with 128 rows and 64 columns, the upper electrode layer has 128 conductive The electrodes are connected to Vcc through a multi-channel analog switch, and the lower electrode layer has 64 conductive electrodes, which are grounded through a pull-down resistor and connected to an analog-to-digital converter through a multi-channel analog switch. When the user lies on the mattress, the pressure will make the upper electrode, the sensor, and the lower electrode closely fit together, forming a part of the circuit loop, which can collect pressure signals. When the cyclic scanning acquisition circuit is working, firstly, by controlling the first-stage 4-way multi-channel analog switch of the lower electrode layer, the 4 16-way multi-channel analog switches of the second stage are selected in turn, and then the 4 16-way multi-channel analog switches of the second stage are selected in turn. Each column of electrodes is gated by analog switches in turn. During the time when the electrode signal of a certain column is selected, by controlling the first-stage 8-channel multi-channel analog switch on the upper electrode layer, the 8 16-channel multi-channel analog switches of the second stage are selected in turn. Analog switch, and then eight 16-way multi-channel analog switches in the second stage strobe each row of electrodes in turn, and connect Vcc. A total of 8192 times of gating process are needed to collect all the sampling point voltage values, and then form each frame of pressure distribution image. With this method, the circuit only needs 128+64 electrodes, and the system only needs one analog-to-digital converter for scanning acquisition of a matrix sequence of 128*64.

Fig. 6 is a data flow chart of the present invention, wherein the electrical signals generated by the first sensor and the second sensor are subjected to analog-to-digital conversion after being amplified and filtered by the analog circuit, and the converted digital signal is stored in the processor. And transmission, sent to the PC terminal for pre-processing such as filtering, noise reduction, smoothing waveform, etc., and then segment the signal of the whole night in chronological order, for example, every 30 seconds is a segment, and feature extraction is performed on each segment of the signal. The features include Body movement characteristics, breathing characteristics, heart rhythm characteristics and posture characteristics.

First, perform the first step of classification, that is, classify all segment signals according to the obtained body motion characteristics, and specifically divide them into two categories: if the body motion characteristics meet the corresponding requirements, it is classified into the "night awake + light sleep" category; if Those who are not satisfied are divided into "rapid eye movement sleep + deep sleep";

Then, the second step of classification is carried out, that is, pattern recognition is performed on the existing two categories according to breathing characteristics, heart rate characteristics and posture characteristics, and further divided into "night awake", "light sleep", "deep sleep" and "rapid eye sleep". "Active sleep", and then perform sleep analysis according to the proportion of the four sleep stages, or choose an appropriate wake-up time, etc.

The sleep analysis results obtained from the final monitoring will be displayed to the user in the form of a report through the mobile APP or PC software to help users improve their sleep quality.

Selected references:

[1]Liu J J, Xu W, Huang M C, et al. Sleep posture analysis using a densepressure sensitive bedsheet[J]. Pervasive and Mobile Computing, 2014, 10: 34-50.

[2]Samy L, Huang M C, Liu J J, et al. Unobtrusive sleep stage identification using a pressure-sensitive bed sheet[J]. IEEE Sensors Journal, 2014, 14(7): 2092-2101.

[3]Mora G G, Kortelainen J M, Hernández E R P, et al. Evaluation of pressure bed sensor for automatic SAHS screening[J]. IEEE Transactions on Instrumentation and Measurement, 2015, 64(7): 1935-1943.

[4]Watanabe K, Watanabe T, Watanabe H, et al. Noninvasive measurement of heartbeat, respiration, snoring and body movements of a subject in bed via apneumatic method[J]. IEEE transactions on biomedical engineering, 2005, 52(12): 2100-2107.

[5] Borbely et al. Sleep homeostasis and models of sleep regulation. J. Biol. Rhythms., 14(6), 1999.

[6] Krueger et al. Sleep as a fundamental property of neuronal assemblies. Nat. Rev. Neurosci., 2008.

[7] R. Stickgold. Sleep-dependent memory consolidation. Nature, 437(7063): 1272-1278, 2005.

[8] Gami, et al. Daynight pattern of sudden death in obstructive sleepapnea. N. Engl. J. Med., 2005

[9] E. Tasali, et al. Slow-wave sleep and the risk of type 2 diabetes inhumans. Natl. Acad. Sci., 2008.

Claims (4)

1. a kind of sleep monitor intelligent sensing mattress, it is characterised in that using two kinds of pliable pressure sensors, the first sensor For making pressure sensing array, the first sensor is the square sheets shape of length of side 0.6-0.9cm, between each sensor Distance be 0.3-0.45cm, be evenly distributed in whole mattress, for detect mattress major part region force-bearing situation, and according to Pressure size suffered by each sensor, changes into the signal of telecommunication of single mapping, and the signal of telecommunication is then converted into sensor in pressure point The matrix color lump of correspondence position on cloth image, the colourity of matrix color lump increases as pressure increases, so as to obtain user Sleep attitude, and the pressure concentrated part situation of user；Second sensor is distributed in a few place's key positions of mattress, bag Include the position of the correspondence mattress of heart when people lies low, and heart is with regard to the symmetrical position of spinal column, the center of trunk with And the position that human body waist is corresponding on mattress, for the breathing for gathering user, the pressure that body is moved and rhythm signal is caused Power changes, and these pressure change signals are changed into into the signal of telecommunication；Subsequent conditioning circuit and processor are put the signal of telecommunication for obtaining Greatly, filtering, transmission, then feature extraction, pattern recognition process are carried out by PC ends, obtain pressure distribution image and sleep point Analysis.

2. sleep monitor intelligent sensing mattress according to claim 1, it is characterised in that concrete structure includes：Above Cushion and following basal layer, are the placed layer of second pressure transducer, the first pressure transducer successively in the middle of both Array layer, or the mixed layer of second pressure transducer and the first array of pressure sensors；Additionally, also including：The first Sealing coat between the upper/lower electrode layer of array of pressure sensors layer, and two kinds of pressure transducer layers；Wherein：

Cushion is used to prevent pressure transducer from directly contacting with the external world, and carries out appropriate buffering to the pressure from human body；

Basal layer is used to prevent lower electrode from directly contacting with the external world, and plays a protective role；

Sealing coat is used to prevent the structure on upper strata because gravity is contacted with pressure transducer layer, it is to avoid affect the essence of signal Degree；

Matcoveredn is set between pressure transducer layer and corresponding electrode layer, for preventing pressure transducer and electrode layer Contact produces interference signal.

3. sleep monitor intelligent sensing mattress according to claim 2, it is characterised in that also include：With the first sensing Device is the related part of array of pressure sensors：Scan round acquisition module, amplifier and filter circuit, and pass with second The related part of sensor：Amplifier and filter circuit；Additionally, also including processor, power module, the clock mould of subsequent conditioning circuit Block, memory module, data transmission device, PC ends, digital signal processor；Wherein：

The first embedded sensor array of processor control scan round acquisition module gating, to array of pressure sensors signal Scanning, gather the pressure signal of human body each position, signal amplifies through amplifier, be delivered to process after filter circuit filtering Device；Processor also controls second working sensor, and the signal of telecommunication that second sensor is obtained is by amplifying circuit and filter Wave circuit, is delivered to processor；Processor is by built-in analog-digital converter to processing through amplifying circuit and filter circuit Analogue signal afterwards carries out analog digital conversion, obtains digital signal；The digital signal for obtaining is put into memory module by processor, and By digital signal data transmission module transmitting by way of being wirelessly transferred to PC ends；Data transmission module is by integrated WIFI Module and peripheral circuit are constituted；Power module is responsible for the power supply of whole circuit system, and 220V civil powers are changed into needed for system The voltage wanted；Clock module is responsible for the sequencing contro of whole circuit system, it is ensured that hardware circuit can smoothly according to software journey The frequency of sequence setting is operated with sequential.

4. sleep monitor intelligent sensing mattress according to claim 3, it is characterised in that at PC ends, signal is filtered Ripple, noise reduction, smooth waveform process, are then in chronological order segmented the signal of All Through The Night, and each segment signal is carried out Feature extraction, the feature of extraction includes body dynamic feature, respiratory characteristic, rhythm of the heart feature and posture feature；And according to feature to sleeping Dormancy state is classified：

First, first step classification is carried out, i.e., all segment signals is classified according to the dynamic feature of body for obtaining, be specifically divided into two Class：If the dynamic feature of body meet it is corresponding require, be divided into " night clear-headed+shallow sleep " class, be divided into if being unsatisfactory for " quick REM sleep+deep sleep " class；

Then, second step classification is carried out, i.e., existing two class is entered respectively according to respiratory characteristic, heart rate feature and posture feature Row mode identification, is further divided into " night regains consciousness ", " shallow sleep ", " deep sleep " and " rapid-eye-movement sleep (REM sleep) ", Ran Hougen Sleep analysis are carried out according to the ratio shared by four Sleep stages, or selects the suitable wake-up time；

Finally monitor the sleep analysis result for obtaining to be shown to and use by mobile phone A PP or PC softwares in the form of reporting Person, helps user to improve the health care of sleep quality.