CN106937808B

CN106937808B - Data acquisition system of intelligent mattress

Info

Publication number: CN106937808B
Application number: CN201710235987.9A
Authority: CN
Inventors: 刘新宇; 郭延锐; 姚小慧; 张永钦
Original assignee: Foshan Liangnao Technology Co ltd
Current assignee: Shenzhen Feizhi Health Internet Of Things Technology Co ltd
Priority date: 2017-04-12
Filing date: 2017-04-12
Publication date: 2020-04-10
Anticipated expiration: 2037-04-12
Also published as: CN106937808A

Abstract

The invention relates to a data acquisition system of an intelligent mattress, which at least comprises a data acquisition unit with a plurality of acquisition channels and an intelligent terminal for data interaction with the data acquisition unit, wherein the data acquisition unit is arranged on the intelligent mattress, wherein, the data acquisition system of the intelligent mattress selects an acquisition channel for acquiring physiological signals of a user based on the classification of the user using the intelligent mattress, the data acquisition unit converts the acquired physiological signals into an AD voltage value through an A/D converter, pre-processes the AD voltage value and then sends the AD voltage value to an intelligent terminal appointed by the user and/or nearby the intelligent mattress through a communication device, and the intelligent terminal carries out calculation by screening out the physiological signals collected by the collection channels with the preprocessed AD voltage values meeting the set threshold value so as to obtain the physiological data of the user on the intelligent mattress.

Description

Data acquisition system of intelligent mattress

Technical Field

The invention relates to the technical field of intelligent home furnishing, in particular to a mattress integrated with intelligent hardware, and particularly relates to a data acquisition system of an intelligent mattress.

Background

The intelligent hardware is the core content in the intelligent home. Statistics show that one third of the life of a person is spent in bed. Therefore, it is very meaningful to combine intelligent hardware with a mattress for disease prediction or monitoring of the physical condition of a patient.

In addition, with the coming of the aging trend of the population in China, the health problem of the elderly living alone is increasingly prominent. The unique physiological characteristics of the elderly give special care, and even some elderly need to pay attention to their physiological changes and sleep states all the time. The existing sleeping posture and sleeping behavior testing methods mainly comprise a polysomnography method, a body movement recording graph analysis method and a camera shooting method. Although the polysomnography can obtain detailed information of a sleep structure, the method can be completed only in a laboratory and needs to place a multi-conductive electrode on a human body, so that the normal sleep is influenced to a certain degree. The body movement recording diagram analysis method is a good method for testing the body movement of a specific human body, but in order to obtain comprehensive sleep movement behaviors, body movement recording devices need to be worn at multiple positions of the body parts of the human body, so that the sleep is influenced to a certain degree. Although the camera shooting method can test the sleeping posture and the sleeping action behavior on the premise of not disturbing the sleep, the method does not respect the privacy of the testee, and meanwhile, cannot accurately test the slight action behavior and the action behavior covered by bedding. Although the product of the intelligent mattress in the prior art can monitor physiological changes and sleep states of the old, the accuracy of the intelligent mattress is too low to provide effective data support for the intelligent home system, and wrong prompt information often appears.

In order to solve the technical problem, a chinese patent (publication number CN204654369U) discloses an intelligent mattress based on the internet of things. The intelligent mattress based on the Internet of things comprises a mattress, wherein the mattress comprises a plurality of piezoelectric units arranged inside the mattress, the piezoelectric units are symmetrically arranged on the mattress, each piezoelectric unit further comprises 4 piezoelectric strips and binding posts, the piezoelectric strips are symmetrically arranged, the piezoelectric strips are connected with the binding posts through electric wires, and the binding posts of each piezoelectric unit are connected with an analysis module through electric wires; the analysis module comprises an addressor, a memory, a data transceiver, a processor and a sounder, wherein the addressor is used for allocating a unique address for each piezoelectric unit, the memory is used for storing information sent by the mattress, the data transceiver is used for receiving or sending the information to the server, the processor is used for processing signals of each piezoelectric unit, and the sounder is used for giving an alarm or language information.

The intelligent mattress of this patent has solved the technical problem that the accuracy of intelligent mattress among the prior art is low. However, the intelligent mattress of the patent has at least the following technical problems not considered:

(1) the signal that will gather piezoelectric unit is used for processing in order to obtain sign such as user's breathing, heartbeat not to carry out filtering process to the signal of gathering before, because interference signal's influence for analysis result's accuracy is low, and in addition, the signal of gathering a plurality of piezoelectric unit is all used for handling in order to obtain sign such as user's breathing, heartbeat, because data processing volume is big, makes data processing inefficiency.

(2) Under the condition that there is not communication connection, the intelligent mattress can't upload the user data who gathers to intelligent terminal and/or high in the clouds service management system and save and/or data processing, often causes data loss this moment easily, when the user abnormal state appears, also can't report to the police and remind the abnormal state of user based on the data of gathering.

(3) The battery capacity of the intelligent mattress is limited, and the intelligent mattress is in communication connection with the intelligent terminal and/or the cloud service management system for a long time and exchanges data for a long time, so that the battery power consumption is large, and the intelligent mattress is not beneficial to keeping the working state for a long time.

Disclosure of Invention

Aiming at the defects of the polysomnography, the body movement record chart analysis method and the camera shooting method in the prior art, the invention provides a test method for the sleep behavior of a user by detecting physiological signals of the user on an intelligent mattress by using a sensor and calculating physiological data of the user based on the acquired physiological signals. The invention particularly provides a data acquisition system of an intelligent mattress, and the sensor is arranged on the intelligent mattress, so that the sleep quality of a user is not affected.

Furthermore, most of the analysis methods in the prior art are to send the collected data to a background server for data analysis to obtain the sleep quality of the user. The acquired data are sent to the background server, and the data processing efficiency is low due to the large data storage and processing capacity of the background server and the long data transmission distance. Therefore, the invention provides a data acquisition system of an intelligent mattress. Preferably, the data acquisition system of the intelligent mattress at least comprises a data acquisition unit and a communication device which are arranged on the intelligent mattress, and an intelligent terminal which performs data interaction with the data acquisition unit through the communication device. The data acquisition unit with a plurality of acquisition channels is used for acquiring physiological signals of a user using the intelligent mattress and transmitting the acquired physiological signals to the intelligent terminal specified by the user and/or nearby the intelligent mattress through the communication device. The intelligent terminal carries out calculation by screening out physiological signals acquired by the acquisition channels with the acquired data meeting the set threshold value so as to obtain the physiological data of the user on the intelligent mattress. Preferably, the physiological signal acquired by the effective acquisition channel is the physiological signal acquired by the acquisition channel of which the physiological signal reaches a set threshold. The data acquired by the data acquisition unit is interacted with the data of the intelligent terminal near the intelligent mattress, so that the defect that the acquired data needs to be sent to a background server for processing is overcome. On the other hand, when the method is used for calculating the physiological data of the user, the effective data is extracted by a screening method in the face of a large amount of collected data, so that the analysis speed can be increased, and the analysis accuracy can be improved.

Furthermore, when the intelligent hardware faces a long-term continuous data collection task, the intelligent hardware and the intelligent terminal matched with the intelligent hardware have high requirements on computing capacity and storage space. In addition, the battery capacity of the intelligent mattress is limited, in order to reduce the requirement on hardware and save the battery power, so that the intelligent mattress can keep a working state for a long time, the invention provides a scheme for selecting an acquisition channel for acquiring the body sign signals of the user based on the judgment of the user. Preferably, the data acquisition unit has a plurality of acquisition channels, and the intelligent mattress system selects the acquisition channel for acquiring the body sign signal of the user based on the judgment of the user using the intelligent mattress. When the user uses intelligent mattress, if all gather the passageway and all open the health sign signal that is used for gathering the user, not only can cause the data collection volume big, cause data storage and/or analysis efficiency to reduce, but also can cause the collection passageway extravagant. The acquisition channels specified by the invention are positioned at different parts of the mattress and are used for acquiring sign signals of different body parts. According to a preferred embodiment of the present invention, the orientation of the face of the sleeping user is determined according to the magnitude of the sound collected by the collection channel for the head. In particular, the head-directed acquisition channel is activated whenever a turning action of the user in the sleep state is determined by another acquisition channel based on the user movement data for picking up the corresponding sound and further determining the face orientation of the user after the turning action is completed based on the received sound amplitude, wherein it is determined whether the user enters the sleep state based on the data collected by the plurality of acquisition channels, and the acquisition channel facing away from the user's face is disabled after the user enters the deep sleep state; and wherein a plurality, in particular all, acquisition channels are activated upon determining that a movement of the user who has entered a sleep state is occurring or is about to occur based on data collected by at least one acquisition channel. For a user entering a sleep state, when at least one acquisition channel towards the user collects step-like changes obviously different from an average signal level, verifying and determining whether the step-like changes are interference noise according to an acquisition system of at least one other acquisition channel; if the step change is judged to be originated from the user entering the sleep state according to at least two acquisition channels, comprehensively judging the source of the step change according to the data of a plurality of acquisition channels, activating a signal amplification mechanism of the corresponding acquisition channel, and simultaneously extracting and storing the step change by adopting a band-pass filtering means for medical diagnosis later.

Furthermore, the data acquisition unit acquires physiological signals of a user and also acquires more interference signals, and in order to remove the influence of the interference signals, the data acquisition unit also comprises a first filter, an amplifier and a second filter for preprocessing the physiological signals acquired by the data acquisition unit. Preferably, the physiological signal collected by the data collecting unit is converted into an AD voltage value by an A/D converter. After the physiological signals of the user are subjected to analog-to-digital conversion, the later-stage data calculation is facilitated. Preferably, the data acquisition system of the intelligent mattress selects one or more of the first filter, the amplifier and the second filter to preprocess the acquired signals based on the body movement condition of the user. Preferably, the first filter filters the AD voltage value to a signal amplitude less than or equal to 3V, the amplifier amplifies the AD voltage value by 0-300 times, and the second filter filters the AD voltage value to a signal amplitude less than or equal to 5V. Preferably, the amplification of the amplifier is adjustable according to different settings of the amplitude of the filtered signal. For example, the amplification of the amplifier may be 80 times, 100 times, 150 times, 170 times, 200 times, or 300 times. Preferably, the first filter is selected to perform filtering preprocessing on the physiological signals of the head of the user acquired by the acquisition channel. The physiological signal of the head is for example a breathing signal, a snoring signal or a head rotation signal of the user. According to the invention, after the acquired head signal is processed by the first filter, the influence of interference signals can be reduced, and the accuracy of an analysis result is improved. And selecting the amplifier for amplification pretreatment on the physiological signals of the chest and/or the pulse of the user acquired by the acquisition channel. The physiological signal of the chest is for example a heartbeat signal of the user. The invention can make the collected signals have more characteristics after the collected chest and/or pulse signals are processed by the amplifier, thereby being convenient for finding out characteristic waveforms. The amplifier can also be used for amplifying the collected vibration condition of the pregnant woman fetus. And selecting the second filter for filtering and preprocessing the physiological signals of the limbs or the back of the user, which are acquired by the acquisition channel. The physiological signal of the limbs or the back is, for example, a signal of a movement of the limbs of the user. The signals of the limbs or the back are obvious, and the collected signals of the limbs or the back are processed by the second filter and then filtered until the signal amplitude is less than or equal to 5V. The AD voltage value preprocessed by the first filter, the amplifier and the second filter is raised to be more than 0V, so that the hardware reading is facilitated.

Further, the signal acquired by the acquisition channel contains the sum of the heartbeat signal and the respiration signal, and the heartbeat signal and the respiration signal are modeled by using a sinusoidal signal and are respectively described by the following formulas:

S_h(t)＝A_h×Sin(2πf_h+θ_h) (1)

S_b(t)＝A_b×Sin(2πf_b+θ_b) (2)

S(t)＝S_h(t)+S_b(t)+A (3)

the formula (1) and the formula (2) are heartbeat and respiration voltage signals respectively, the formula (3) is a voltage signal output by the sensor, and the voltage signal comprises the sum of the heartbeat and the respiration signals and a direct current component A, wherein A_hAnd A_bAmplitude of heartbeat and respiration, respectively, f_hAnd f_bRespectively the heart beat and respiration rate, theta_hAnd theta_bThe initial phases of the heartbeat and respiration, respectively.

Furthermore, the intelligent terminal carries out AD conversion on the physiological signals collected by the screened effective collection channel and carries out fast Fourier transform on the signal time domain frequency spectrum obtained after preprocessing so as to obtain a signal frequency domain frequency spectrum, the number that the peak value output of the signal frequency domain frequency spectrum obtained after processing exceeds a preset threshold value in a preset respiratory frequency range is used for obtaining the respiratory frequency of the user, and the number that the peak value output of the signal frequency domain frequency spectrum obtained after processing exceeds the preset threshold value in a preset heartbeat frequency range is used for obtaining the heartbeat frequency. Preferably, the preset respiratory frequency range is 0-0.5 Hz, and the preset heartbeat frequency range is 0.6-2.7 Hz. Preferably, in order to improve the calculation efficiency, the invention calculates the breathing frequency and the heartbeat frequency in 30S to obtain the breathing frequency and the heartbeat frequency of the user. For example, the number of breaths of the user in 30S is calculated as 14, and the breathing frequency of the user is 28/min by multiplying the number by 2.

Further, a fourier transform formula of the voltage signal s (t) output by the sensor of the acquisition channel is as follows:

and the voltage signal s (t) output by the sensor of the acquisition channel can also be represented by fourier change, and the expression formula is as follows:

wherein, the formula (4) is fourier transform, the formula (5) is inverse fourier transform, and the voltage signal s (t) output by the sensor can be transformed from time domain to frequency domain for analysis by fourier transform.

Further, the intelligent mattress system selects the acquisition scheme of the user to acquire the physical sign signals of the user based on the judgment of the age, the sex and/or the physical condition of the user using the intelligent mattress, or the intelligent mattress system acquires the physical sign signals of the user based on the acquisition scheme input by the user using the intelligent mattress through the intelligent terminal. The acquisition scheme includes, but is not limited to, an acquisition channel for acquiring the body sign signal of the user, a set threshold value for screening an effective acquisition channel by the intelligent terminal, and a user physiological data type to be calculated based on the body sign signal. The user of different ages, sex and/or health condition needs to gather the health sign signal differently, and the collection scheme of user preset according to different demands is prestored in intelligent mattress system, when the user uses intelligent mattress, selects the collection scheme through the preliminary judgement to the user, so not only can satisfy different users' demand, but also can avoid gathering the waste of passageway.

Furthermore, under the condition of communication interruption or no communication condition, the data acquisition unit cannot perform data interaction with the intelligent terminal and/or the cloud service management system, and in order to avoid data loss, the data acquisition system of the intelligent mattress further comprises a temporary storage unit. Preferably, each acquisition channel of the data acquisition unit acquires physiological signals of a user using the smart mattress using one or more sensors of a pressure sensor, a humidity sensor, a temperature sensor and a heart rate sensor and stores the acquired physiological signals in the temporary storage unit. The data acquisition unit of the invention stores the acquired data locally, thereby avoiding data loss caused by communication interruption.

Furthermore, the data volume collected by the data collection unit is large, the relationship among the data is complicated, and in order to accurately and quickly store and/or analyze the data, the physiological signals collected by the data collection unit are temporarily stored in the temporary storage unit according to the collection channels corresponding to the physiological signals and the collection time. The temporary storage unit can respond to the successful matching of the communication device and the intelligent terminal and push a data set which is temporarily stored in the temporary storage unit and at least consists of the acquisition channel, the acquisition time and the physiological signals acquired by the acquisition channel to the intelligent terminal. The invention adopts the form of data set to store and/or analyze data, not only can reduce the operation load of the intelligent terminal, but also can improve the speed of storage and/or analysis and the accuracy of analysis, ensure that the data acquisition system of the intelligent mattress can run stably and smoothly for a long time, and improve the user experience.

Furthermore, the storage capacity and the data processing capacity of the intelligent terminal are limited, and in order to further improve the data analysis and processing capacity of the intelligent mattress system, the intelligent mattress system further comprises a cloud service management system. On the other hand, the battery capacity of the intelligent mattress is limited, and the intelligent mattress is in communication connection with the intelligent terminal and/or the cloud service management system for a long time and exchanges data for a long time, so that the battery power consumption is large, and the intelligent mattress is not beneficial to keeping the working state for a long time. To this end, in response to the successful matching between the communication device and the intelligent terminal, the temporary storage unit pushes the data set acquired by the data acquisition unit and temporarily stored in the temporary storage unit to the intelligent terminal successfully paired with the intelligent mattress, and forwards the data set to the cloud service management system by the intelligent terminal. The cloud service management system responds to the communication connection with the intelligent terminal to acquire the data set temporarily stored in the temporary storage unit by the data acquisition unit. Under the condition that the battery capacity of the intelligent mattress is limited, the intelligent mattress does not need to communicate with an intelligent terminal at any time, and only needs to continuously complete a large number of data storage tasks. When the intelligent mattress is in communication connection with the intelligent terminal regularly, data are exchanged in a short time, so that the electric quantity of the battery can be saved, and the continuous working time of the intelligent mattress can be effectively prolonged.

Furthermore, when the posture of the user on the intelligent mattress changes, the data collected by the data collection unit also changes correspondingly, and the posture of the user on the bed is difficult to predict. Preferably, when the change of the user physiological signal acquired by the acquisition channel reaches a set threshold, the intelligent terminal and/or the cloud service management system calculates to obtain the physiological data after the user state change based on the changed physiological signal acquired by the acquisition channel. When the change of the physiological signals acquired by the acquisition channel does not reach a set threshold value, the intelligent terminal and/or the cloud service management system calculates to obtain the physiological data of the user after the state change based on the physiological signals acquired by the acquisition channel before the change. The data collected by the collecting channel changes due to the change of the posture of the user on the intelligent mattress, and when the change reaches a set threshold value, the changed data is selected to analyze the physiological data of the user, so that the analysis accuracy can be ensured; when the change does not reach the set threshold value, the data before the change is selected to analyze the physiological data of the user, so that the data loss caused by channel switching can be reduced.

Drawings

FIG. 1 is a schematic diagram of a preferred embodiment of a data acquisition system for a smart mattress of the present invention;

FIG. 2 is a schematic diagram of a preferred embodiment of the data acquisition process of the present invention;

FIG. 3 is a graph of the AD voltage values of the preferred 6 acquisition channels of example 2;

FIG. 4 is a graph of AD voltage values of 6 acquisition channels after pretreatment in example 2;

FIG. 5 is a graph showing the AD voltage values of the sampling channels screened for respiration determination in example 2;

FIG. 6 is a statistical chart for calculation of respiratory rate in example 2;

FIG. 7 is the graph of the AD voltage values of the collecting channels when the user is in the turning-over state in the embodiment 2; and

FIG. 8 is a graph of the AD voltage values of the acquisition channels in the out-of-bed condition of the user in example 2.

List of reference numerals

10: the intelligent mattress 20: intelligent terminal

30: cloud service management system 101: data acquisition unit

101 a: the a/D converter 101 b: first filter

101 c: the amplifier 101 d: second filter

102: the communication device 103: temporary storage unit

104: the data processing unit 105: early warning unit

106: image acquisition unit

Detailed Description

The following detailed description is made with reference to the accompanying drawings and examples.

Fig. 1 shows a schematic diagram of a preferred embodiment of the data acquisition system of the smart mattress of the present invention. As shown in fig. 1, the data acquisition system of the smart mattress at least includes a smart mattress 10, a smart terminal 20 and a cloud service management system 30. The intelligent terminal 20 and the cloud service management system 30 can perform data interaction with the intelligent mattress 10. Preferably, the communication mode of the smart mattress 10 and the smart terminal 20 and/or the cloud service management system 30 includes, but is not limited to, 2G, 3G, 4G, 5G and 3GPP communication. Preferably, the smart terminal 20 includes, but is not limited to, a mobile phone, a tablet computer, and a smart band. All mobile devices which can be connected to the cloud service management system can be regarded as intelligent terminals. Referring again to fig. 1, the smart mattress 10 includes at least a data acquisition unit 101, an a/D converter 101a, a first filter 101b, an amplifier 101c, a second filter 101D, a communication device 102, a temporary storage unit 103, a data processing unit 104, an early warning unit 105, and an image acquisition unit 106. Preferably, the data acquisition unit 101 is configured to acquire physiological signals of a user using the smart mattress 10, convert the acquired physiological signals into AD voltage values through the a/D converter 101a, pre-process the AD voltage values, and transmit the AD voltage values to the smart terminal 20 specified by the user and/or in the vicinity of the smart mattress 10 via the communication device 102. Preferably, the smart terminal 20 near the smart mattress 10 refers to a smart terminal 20 attached to a user using the smart mattress 10 or capable of performing near field communication with the smart mattress 10. The intelligent terminal 20 and/or the cloud service management system 30 performs calculation by screening out the physiological signals acquired by the acquisition channels with the pre-processed AD voltage values meeting the set threshold value to obtain the physiological data of the user on the intelligent mattress 10.

According to a preferred embodiment, before the body sign signal of the user is acquired, the data acquisition system of the intelligent mattress selects an acquisition channel for acquiring the body sign signal of the user based on the judgment of the user. The data acquisition unit 101 has a plurality of acquisition channels that cover portions of the smart mattress 10 that may be touched by a user. When the user uses the smart mattress 10, the age, sex and/or physical condition of the user are primarily judged, and unnecessary collecting channels are closed according to the judgment result. For example, when the data acquisition system of the intelligent mattress determines that the user is an infant, more acquisition channels are provided on the intelligent mattress 10, which are not accessible by the infant during sleep, and the data acquisition system of the intelligent mattress closes the acquisition channels that are not accessible by the infant, and the data acquisition system of the intelligent mattress mainly opens the acquisition channels for acquiring the turn-over frequency of the infant according to the sleep characteristics of the infant. For another example, when the data acquisition system of the intelligent mattress determines that the user is an elderly person, the data acquisition system mainly opens an acquisition channel for acquiring the breathing and heart rate of the elderly person according to the sleep characteristics of the elderly person. Also for example, based on the difference in sleep characteristics between adult male and adult female, a corresponding acquisition scheme may be set.

According to a preferred embodiment, the data acquisition system of the smart mattress can also acquire the physical sign signals of the user based on the acquisition scheme input by the user through the smart terminal 20. Preferably, the acquisition scheme includes, but is not limited to, an acquisition channel for acquiring the physical sign signal of the user, a set threshold value for the smart terminal 20 to screen the effective acquisition channel, and a type of physiological data of the user that needs to be calculated based on the physical sign signal. For users with special requirements, the acquisition scheme can also be input through the intelligent terminal 20. For example, for pregnant women, in order to monitor the activities of the fetus at night, the vibration condition of the fetus can be collected by setting a required pressure collecting channel. The duration and the number of vibrations of the fetus are recorded, so that the condition of the fetus at night can be obtained. For another example, for a person with a heart disease, the heartbeat condition of the user can be acquired by setting a required heartbeat acquisition channel. The abnormal condition of the heartbeat of the user at each time is recorded, and the method can be used for assisting diagnosis and treatment of doctors.

According to a preferred embodiment, a bluetooth module is provided on the communication device 102. The Bluetooth module is connected with the STM32l151 chip through a serial port line uart2 (serial port 2). The Bluetooth module is provided with a packaging module. And the packaging module packages the data to be sent according to a packaging protocol through a packaging function. The bluetooth module sends the packed data to the bluetooth transmission module of the intelligent terminal 20. The intelligent terminal 20 sends the configuration information of the wifi to the circuit board or the single chip microcomputer on the intelligent mattress 10 through the Bluetooth transmission module by configuring the SSID and the password of the wifi, and the wifi module is configured by the chip transmission serial port data. Preferably, a wifi module is disposed on the communication device 102. The wifi module is connected with the stm32L151 chip through the serial port 3. And the wifi module sends the packaged data to a wireless router connected with the Internet. The wireless router sends the data to the cloud service management system 30.

According to a preferred embodiment, the present embodiment collects physiological signals of the user through multiple channels because it is difficult to predict in what posture the user worked on the smart mattress 10. If all the data collected by the multiple collection channels are used for calculation, the calculation load of the intelligent terminal 20 and/or the cloud service management system 30 is increased, and the calculation speed is reduced. For this reason, the system of the intelligent mattress 10 of the present embodiment selects the physiological signals acquired by the acquisition channels whose physiological signals reach the set threshold value to perform calculation to obtain the physiological data of the user. For example, a user lies down on the smart mattress 10, the sensors on all channels start to collect pressure signal changes at the same time, after a certain period of time (e.g., 6s), the channels start to be screened, and when the number of points in the channels where the data collected is below/above a certain threshold is the largest, the channel is selected for data processing. Preferably, the data acquisition unit 101 has a plurality of acquisition channels. The plurality of acquisition channels are distributed in a manner that uniformly covers the portion of the human body that may be contacted on the smart mattress 10. Preferably, the surface of the smart mattress 10 is divided into a plurality of regions according to a rectangular grid, each region having an identical rectangular shape. Each region has one or more acquisition channels.

According to a preferred embodiment, each acquisition channel comprises one or more of a pressure sensor, a humidity sensor, a temperature sensor and a heart rate sensor. Preferably, the pressure sensor is one or more of a ceramic piezoelectric sensor, a semiconductor piezoresistance sensor, an electrostatic capacity type pressure sensor, and a diffused silicon pressure transmitter. The humidity sensor is one or more of a resistance type lithium chloride hygrometer, a dew point type lithium chloride hygrometer, a carbon humidity sensitive type hygrometer, an alumina hygrometer and a ceramic humidity sensor. The temperature sensor is a contact or non-contact thermometer sensor. The heart rate sensor is one or more of an infrared pulse sensor, a heart rate pulse sensor, a photoelectric pulse sensor, a digital pulse sensor, a heart sound pulse sensor and an integrated pulse sensor. The data acquisition system of this embodiment intelligence mattress gathers user's physiological data through multiple sensor, not only can realize the comprehensive monitoring to user's physiological sign, and the data signal cross combination who gathers through multiple sensor uses moreover, can obtain more accurate user's state on intelligent mattress 10.

According to a preferred embodiment, the a/D converter 101a is used to convert the analog signals collected by the sensors into digital signals. The first filter 101a, the amplifier 101b, and the second filter 101c are used to preprocess the AD voltage value converted by the a/D converter 101 a. Preferably, the data acquisition system of the smart mattress selects one or more of the first filter 101b, the amplifier 101c and the second filter 101d to preprocess the acquired signals based on the body movement condition of the user. The first filter 101b filters the AD voltage value until the signal amplitude is less than or equal to 3V. The amplifier 101c amplifies the AD voltage value by 0 to 300 times. The second filter 101d filters the AD voltage value until the signal amplitude is less than or equal to 5V. Preferably, the body movement condition of the user refers to the vibration condition of each part of the user collected by the collecting channel. For example, the pulse signal of the user is not obvious and needs to be preprocessed by the amplifier 101 b. The user's breathing signal is susceptible to the snoring signal and requires filtering using the first filter 101 a. The moving signals of the four limbs of the user are not easily interfered by other signals, and the second filter 101c can be used for filtering processing. Preferably, the signals before and after being pre-processed by the first filter 101a, the amplifier 101b and the second filter 101c are stored in a correlated manner, and a user can extract the correlated signals for checking the sleep condition or use the correlated signals for assisting the doctor in diagnosis and treatment.

According to a preferred embodiment, the intelligent terminal 20 and/or the cloud service management system 30 performs AD conversion on the physiological signals acquired by the screened effective acquisition channels and performs fast fourier transform on the signal time domain spectrum obtained after preprocessing to obtain a signal frequency domain spectrum, obtains the number of processed signal frequency domain spectrum peaks exceeding a preset threshold in a preset respiratory frequency range to obtain the respiratory frequency of the user, and obtains the number of processed signal frequency domain spectrum peaks exceeding a preset threshold in a preset heartbeat frequency range to obtain the heartbeat frequency. Preferably, the normal breathing frequency of the human body is not more than 30 times/minute, and the breathing frequency range is set to be 0-0.5 Hz. The heart rate range of the human body is 40-160 times/minute, and the range of the heartbeat frequency band is set to be 0.6-2.7 Hz.

According to a preferred embodiment, in the absence of a communication connection, the smart mattress 10 cannot upload the collected data to the smart terminal 20 and/or the cloud service management system 30, and in order to avoid data loss, the smart mattress 10 of the present embodiment further includes a temporary storage unit 103. The intelligent mattress 10 collects physiological signals of a user through the data collecting unit 101, and stores the collected data locally when the intelligent mattress 10 is in an off-line state. Once the smart mattress 10 is in communication with the smart terminal 20, the smart mattress 10 automatically uploads the stored and/or collected data to the smart terminal 20. The physiological signals of the user are uploaded to the cloud service management system 30 through the intelligent terminal 20 for storage and/or analysis. The data acquisition system of the intelligent mattress of this embodiment can monitor user's physiological signal in real time, even when intelligent mattress 10 is in the off-line state, still can save several days even several weeks's data, intelligent mattress 10 through the data to local storage carry out the analysis so that send out warning feedback information when the user appears unusually, avoid can not in time salvage the user when the unexpected condition takes place.

According to a preferred embodiment, the temporary storage unit 103 stores the physiological signals acquired by the data acquisition unit 101 in a manner correlated with their respective acquisition channels and acquisition times. The temporary storage unit 103 can push the stored data set composed of at least the acquisition channel, the acquisition time, and the physiological signal acquired by the acquisition channel to the smart terminal 20 in response to a successful matching of the communication device 102 with the smart terminal 20. The intelligent terminal 20 forwards the data set to the cloud service management system 30. In the embodiment, the data is stored and/or analyzed in a data set form, so that the operation load of the intelligent terminal 20 and/or the cloud service management system 30 can be reduced, the storage and/or analysis speed and the analysis accuracy are improved, the data acquisition system of the intelligent mattress can be ensured to run stably and smoothly for a long time, and the user experience is improved.

According to a preferred embodiment, the data collected by each sensor of the data collection unit 101 changes with the change of the user status, and when the change of the physiological signal of the user collected by the collection channel reaches a set threshold, the intelligent terminal 20 and/or the cloud service management system 30 calculates to obtain the physiological data after the change of the user status based on the changed physiological signal collected by the collection channel. When the change of the physiological signal acquired by the acquisition channel does not reach the set threshold, the intelligent terminal 20 and/or the cloud service management system 30 calculates to obtain the physiological data after the state of the user changes based on the physiological signal before the change acquired by the acquisition channel. For example, when the user changes his/her sleeping posture from lying down to left, some channels distributed on the smart mattress 10 will generate a large signal change, and the system will make a channel selection again. For another example, when the user moves the body slightly, the channel for calculating the physiological data of the user does not change greatly, and the channel switching is not performed at this time, so as to reduce the data loss caused by the channel switching.

According to a preferred embodiment, the data processing unit 104 determines the state of the user on the smart mattress 10 based on the physiological data of the user calculated from the physiological signals of the user collected by the data collecting unit 101. The data processing unit 104 corrects the state judgment of the user on the smart mattress 10 based on the normal physiological data range provided by the user using the smart mattress 10 through the smart terminal 20 and/or the cloud service management system 30. Based on the calculated physiological data of the user, whether the user has an abnormal state, such as short-time apnea, turning over or getting out of bed, can be preliminarily judged. However, the physiological data of users in different ages have great difference, and if the sleep quality of the user is judged only according to the processed physiological data, the judgment is not accurate, and false alarm is easy to occur. Preferably, the present embodiment corrects the user's condition on the smart mattress 10 in conjunction with the normal physiological data ranges provided by the user. Taking the heartbeat as an example, the normal heartbeats of users in different age groups are different, the normal heartbeat of an infant is about 120 times/minute, and the heartbeat of an adult is about 60-80 times/minute. Therefore, if the user using the smart mattress 10 is an infant, when the heartbeat frequency is about 120 times/minute, the user is easily classified into an abnormal state according to the normal heartbeat range of an adult, which results in an analysis error.

According to a preferred embodiment, the smart mattress 10 further comprises an early warning unit 105. Preferably, the early warning unit 105 sends an alarm message when the intelligent terminal 20 and/or the cloud service management system 30 analyzes that the user has an abnormal state on the intelligent mattress 10 based on the calculated physiological data of the user. The abnormal state of the user may be, for example, a user's poor sleep quality, a user's disease, a user getting out of bed, etc. The alarm message is classified into different classes based on the abnormal state of the user on the intelligent mattress 10 analyzed by the intelligent terminal 20 and/or the cloud service management system 30 and the abnormal state class classified according to the normal physiological data range of the user. The alarm messages are divided into different grades, and the rescue personnel and/or departments can take corresponding measures according to the alarm messages of different grades. Preferably, while the smart mattress 10 is in the off-line state, the smart mattress 10 analyzes the state of the user on the smart mattress 10 based on the locally stored physiological signals and/or physiological data of the user, and issues an alarm message when an abnormality occurs to the user.

According to a preferred embodiment, the smart mattress 10 comprises an image acquisition unit 106. Preferably, the image acquisition unit 106 is a camera. Preferably, the image acquisition unit 106 is activated based on an early warning by the early warning unit 105 and/or remote control of personnel and/or departments associated with the user using the smart mattress 10. For example, when the signals acquired by the sensors are all zeroed for a certain time, such as 10S, the camera can be automatically turned on for identification, if the user is not found, the user is determined to be out of bed, otherwise, the user is determined to have an emergency, such as sudden respiratory arrest or sudden cardiac arrest. For another example, when the user is in a rolling state, it is determined that the user needs to observe, the cloud service management system 30 sends information to the rescuer, and the rescuer may remotely turn on a camera at the smart terminal 20 and/or the cloud service management system 30 to observe the video.

Example 2

This embodiment is a further modification of embodiment 1, and only the modified portion will be described.

Fig. 2 shows a schematic diagram of a preferred embodiment of the data acquisition process of the present invention. As shown in fig. 2, the intelligent mattress 10 collects physiological signals of a user through the data collecting unit 101, and when the intelligent terminal 20 communicates with the intelligent mattress 10, the intelligent mattress 10 actively transmits stored data to the intelligent terminal 20. When the smart terminal 20 communicates with the cloud service management system 30, the cloud service management system 30 receives the physiological signal of the user and the normal physiological data range of the user, which is input in advance through the smart terminal 20. Preferably, the user may also directly input the normal physiological data range of the user in advance in the cloud service management system 30.

Further, the smart terminal 20 and/or the cloud service management system 30 calculates physiological data of the user based on the received physiological signals of the user and determines whether an abnormal state occurs on the smart mattress 10 by combining a normal physiological data range provided by the user. Preferably, the smart terminal 20 and/or the cloud service management system 30 classify the abnormal state into different grades based on the analyzed abnormal state of the user on the smart mattress 10 and the normal physiological data range of the user. For example, the user's abnormal state level may be classified as poor sleep, disease prevention, need for rescue, and the like. When the user is just turning over, it can be judged that the health state of the user is good, but the sleep is not good. When the user rolls over, the user can be judged to need to observe.

According to a preferred embodiment, the intelligent terminal 20 and/or the cloud service management system 30 preliminarily classifies the physiological data of the user calculated based on the physiological signal of the user acquired by the data acquisition unit 101 into a state with poor sleep quality, observation requirement, disease prevention requirement, rescue requirement or bed leaving requirement by using a normalization method. Preferably, the present embodiment performs normalization processing on the physiological data of the user by: the sensor samples physiological signals of n points of a user, wherein n is a positive integer greater than or equal to 1. The intelligent terminal 20 and/or the cloud service management system 30 calculates the user physiological data corresponding to the signals of the n points, and then calculates the proportion of the physiological data in the sum of the physiological data of the n points, so as to obtain the physiological data after normalization processing. After the physiological data is subjected to normalization processing, analysis errors caused by errors of physiological signals detected by the sensor can be reduced, and the accuracy of analysis is improved.

According to a preferred embodiment, since it is difficult to predict the posture of the user when he or she works in bed, a multi-channel acquisition of the physiological signals of the user is required, and the system is selected to obtain the channel most suitable for data analysis. When the user changes the sleep posture, a corresponding algorithm is required to change the channel for data processing. Preferably, the sensor for acquiring the physiological signal of the user in the embodiment is a ceramic piezoelectric sensor. Preferably, the smart terminal 20 and/or the cloud service management system 30 determine whether the user is in an abnormal state in the following manner.

When a user is on the intelligent mattress 10, the sensors on all channels start to collect physiological signals of the user at the same time, after a period of time, for example 6S, all channels start to be screened, and when the signals collected in the channels meet the condition that the number of points lower/higher than a set threshold is the maximum, the channels are selected for data processing to obtain the physiological data of the user. Preferably, the smart terminal 20 and/or the cloud service management system 30 calculate to obtain the breathing frequency and/or the heartbeat frequency of the user based on the physiological signal of the user. Preferably, the breathing frequency and/or the heartbeat frequency of the user is calculated by: the method comprises the steps that AD conversion is carried out on physiological signals collected by the screened effective collection channels, fast Fourier transform is carried out on signal time domain frequency spectrums obtained after preprocessing to obtain signal frequency domain frequency spectrums, the number that the peak value output of the processed signal frequency domain frequency spectrums exceeds a preset threshold value in a preset respiration frequency range is obtained to obtain the respiration frequency of a user, and the number that the peak value output of the processed signal frequency domain frequency spectrums exceeds the preset threshold value in a preset heartbeat frequency range is obtained to obtain the heartbeat frequency. Preferably, the preset respiratory frequency range is 0-0.5 Hz. The preset heartbeat frequency range is 0.6-2.7 Hz.

Preferably, before the acquired physiological signals of the user are subjected to Fourier transform, the data acquired by the selected channel are subjected to AD conversion, clutter filtering, signal amplification and clutter filtering again. Preferably, a ceramic piezoelectric sensor is connected to the a/D converter 101 a. The a/D converter 101a is connected to the first filter 101 b. The first filter 101b is connected to the amplifier 101 c. The amplifier 101c is connected to the second filter 101 d. Preferably, the first filter 101b is used to filter the AD voltage value until the signal amplitude is less than or equal to 3V. The amplifier 101c is used for amplifying the signal filtered by the first filter 101b by 0-300 times. The second filter 101d is used to filter the signal amplified by the amplifier 101c to a signal amplitude less than or equal to 3V. The user physiological signal collected by the embodiment can be filtered by the first filter to remove obvious clutter, and the signal is processed by the amplifier and the second filter to improve the signal-to-noise ratio.

According to a preferred embodiment, the heart rate and/or the breathing rate of the user is obtained as follows. The heartbeat and respiration actions of the human body can be regarded as vibration, and are also vibration signals after being converted into voltage signals through the sensor, so that the heartbeat and respiration signals are modeled by using sinusoidal signals. Preferably, the heartbeat and respiration signals are described using the following equations, respectively:

S_h(t)＝A_h×Sin(2πf_h+θ_h) (1)

S_b(t)＝A_b×Sin(2πf_b+θ_b) (2)

S(t)＝S_h(t)+S_b(t)+A (3)

the expressions (1) and (2) are the heartbeat and respiration voltage signals, respectively. Wherein A is_hAnd A_bThe amplitude of the heartbeat and respiration, respectively. f. of_hAnd f_bRespectively the heart beat and the breathing frequency. Theta_hAnd theta_bThe initial phases of the heartbeat and respiration, respectively. Equation (3) is the voltage signal output by the sensor, which contains the sum of the heartbeat and respiration signals, and may also contain a dc component a.

Preferably, the heartbeat and respiration signals are similar to sine wave signals, so fourier transforms are used for the measurement of frequency for signal processing. Preferably, the fourier transform formula of the voltage signal s (t) output by the sensor is as follows:

similarly, the signal s (t) can be represented by fourier transform, and the expression is as follows:

it can be seen that the signal S (t) of this embodiment can be represented as a sum of different single-frequency signals, and the amplitude of the single-frequency signal is the fourier transform value S (j2 pi f). The above expressions (4) and (5) are referred to as a fourier transform pair, where expression (4) is a fourier transform and expression (5) is an inverse fourier transform. The fourier transform may transform the signal from the time domain to the frequency domain for analysis. For the signal output by the sensor, the signal consists of three single-frequency signals of a heartbeat signal, a respiration signal and a direct current signal, and the corresponding frequencies are f_h、f_bAnd 0 Hz. In the time domain, because a plurality of frequency components are present and noise and interference are added, the heartbeat and the respiratory frequency are difficult to extract, but after the heartbeat and the respiratory frequency are converted into the frequency domain through Fourier transform, the interference can be separated, so that the heartbeat and the respiratory frequency can be extracted.

According to a preferred embodiment, the heart rate and/or the breathing rate of the user is obtained as follows. Take the calculation of the breathing frequency of the user as an example. Fig. 3 shows a graph of AD voltage values for the preferred 6 acquisition channels of the present embodiment. As shown in fig. 3, the abscissa is time in seconds and the ordinate is the voltage value count of the a/D converter 101 a. The collecting channel 1, the collecting channel 2, the collecting channel 3, the collecting channel 4, the collecting channel 5 and the collecting channel 6 are arranged from top to bottom in sequence. The a/D converter 101a monitors the AD voltage value converted from the pressure data in each signal channel for 30 seconds at intervals of 1 second. The signals collected by the sensor are subjected to preprocessing by the first filter 101b, the amplifier 101c and the second filter 101d, and then smooth filtering is completed, so that noise in the signals is removed. Fig. 4 shows the graph of the AD voltage values of the 6 acquisition channels after the preprocessing in the present embodiment. As shown in fig. 4, after the denoising process, the waveforms of the channels are more obvious, which is convenient for the subsequent calculation. The sensor can generate voltage change due to pressure change, the sensor can generate negative voltage, but the hardware cannot read out the negative value, so that the voltage is raised to 0, and after the data acquisition unit is amplified and the voltage is raised, the change threshold of the voltage is raised to (0V, + 3V). 2048 corresponds to 1.5V, and 4096 corresponds to 3V. The numerical meaning such as 2048 or 4096 is the AD voltage value converted by the AD converter 101a for the sensor sample data. When the sensor is in a static state, the voltage collected by the system is 1.5V, namely, around 2048 points. Therefore, in the present embodiment, a channel corresponding to the maximum value of the peak or trough distance 2048 is selected from the waveform after the denoising processing as a channel for determining respiration. Based on the above analysis, the present embodiment selects the channel 5 for breathing judgment. Fig. 5 is a graph showing the AD voltage values of the acquisition channels screened for respiration determination in the present embodiment. After amplification, the wave crests and wave troughs of the wave form of the channel 5 are more obvious. Fig. 6 shows a statistical map for the calculation of the breathing frequency for the present embodiment. With 2048 as a line, if 2048 is worn on each peak of the channel 5, it is marked as 1, which indicates that the user breathes once; if the user wears 2048, it is noted as 0, indicating that the user is not breathing. In this manner, the number of breaths in the user 30S is recorded and multiplied by 2 to obtain a breath per minute value. The breathing rate of the user shown in fig. 6 is 14 × 2 ═ 28 times/min.

Take the user turning over as an example. When the user changes the sleeping posture, the physiological signals of the user collected by the channels distributed on the intelligent mattress 10 will generate larger signal changes, and at the moment, the system can select the effective collecting channels again. If the physiological signals of the user collected by the channels distributed on the intelligent mattress 10 do not change obviously, the channels are not switched. Preferably, when the change of the physiological signal of the user acquired by the acquisition channel reaches a set threshold, the intelligent terminal calculates to obtain the physiological data after the state of the user changes based on the changed physiological signal acquired by the acquisition channel. When the change of the physiological signals collected by the collecting channel does not reach the set threshold value, the intelligent terminal calculates to obtain the physiological data after the state of the user changes based on the physiological signals before the change collected by the collecting channel.

Preferably, in the channel selection process, the squares of the differences between the sampled values of all the channels within 5 seconds and 2048 are calculated, the squares of the differences are sorted, the largest two channels are selected as the currently selected channels, and if the channel switching indicates that the object is turned over once. Fig. 7 shows a graph of the AD voltage values of the acquisition channels when the user is in a turning state in the present embodiment. As shown in fig. 7, the abscissa is time in seconds and the ordinate is the voltage value count of the a/D converter 101 a. The selected channel is in the dashed line frame, and the original 2 channels are switched into completely 2 new channels, which shows that the object turns over once, and the turning over action is really generated compared with the actual situation.

Take the user out of bed as an example. When the user is no longer on the intelligent mattress 10, the physiological signals of the user collected by the sensors change, and when the change waveform meets the preset conditions, the user is recorded in a bed leaving state, and the times of the user leaving the bed are recorded. Preferably, the preset condition may be that all sensor signals originally having a signal slowly return to zero. Unlike the roll-over state, the change in the out-of-bed sensor signal is the speed of zeroing and all sensor signals that originally had a signal are zeroed. Preferably, the system determines that the user is out of bed, and if the user is a group unable to take care of oneself, the system sends out an alarm signal after the out-of-bed state reaches a certain time period, such as 10 minutes.

Preferably, when the signals of all channels are monitored to trend to 2048, i.e., no physiological signal is sensed, the test subject is considered to be out of bed. Fig. 8 shows a graph of the AD voltage values of the acquisition channels when the user is in the out-of-bed state in the present embodiment. As shown in fig. 8, the abscissa is time in seconds and the ordinate is the voltage value count of the a/D converter 101 a. After 7 seconds for the voltage curve, all signal values trended to 2048, and the subject was considered to be out of bed.

It should be noted that the above-mentioned embodiments are exemplary, and that those skilled in the art, having benefit of the present disclosure, may devise various arrangements that are within the scope of the present disclosure and that fall within the scope of the invention. It should be understood by those skilled in the art that the present specification and figures are illustrative only and are not limiting upon the claims. The scope of the invention is defined by the claims and their equivalents.

Claims

1. A data acquisition system of an intelligent mattress, in particular to a data acquisition system of an intelligent mattress for acquiring physiological signals of a user, which is characterized by at least comprising a data acquisition unit (101) which is arranged on the intelligent mattress (10) and is provided with a plurality of acquisition channels and an intelligent terminal (20) which can perform data interaction with the data acquisition unit (101), wherein,

the data acquisition system of the intelligent mattress selects an acquisition scheme of a user using the intelligent mattress (10) to acquire physiological signals of the user based on the judgment of the age, the sex and/or the physical condition of the user, or the data acquisition system of the intelligent mattress acquires the physiological signals of the user based on the acquisition scheme input by the user using the intelligent mattress (10) through the intelligent terminal (20), wherein the acquisition scheme comprises but is not limited to an acquisition channel for acquiring the physiological signals of the user, a set threshold value for screening effective acquisition channels by the intelligent terminal (20), and the type of the physiological data of the user needing to be calculated based on the physiological signals,

the data acquisition unit (101) converts the acquired physiological signals into AD voltage values through an A/D converter (101a), pre-processes the AD voltage values and sends the AD voltage values to the intelligent terminal (20) appointed by the user and/or nearby the intelligent mattress (10) through a communication device (102), and the intelligent terminal (20) analyzes the physiological signals acquired by screening out the pre-processed acquisition channels with the AD voltage values meeting set threshold values to obtain the physiological data of the user on the intelligent mattress (10),

the system selects physiological signals acquired by acquisition channels with physiological signals reaching a set threshold value to calculate so as to obtain physiological data of a user, the data acquisition unit (101) is provided with a plurality of acquisition channels, the distribution mode of the acquisition channels uniformly covers the part of a human body which is possibly contacted on the intelligent mattress (10),

the data acquisition system of the smart mattress further comprises a first filter (101b), an amplifier (101c) and a second filter (101d) for preprocessing the physiological signals acquired by the data acquisition unit (101), and the data acquisition system of the smart mattress selects one or more of the first filter (101b), the amplifier (101c) and the second filter (101d) to preprocess the acquired signals based on the physical movement condition of the user,

the system obtains a channel most suitable for data analysis by selecting, when a user changes the sleep posture, the channel for data processing needs to be changed by a corresponding algorithm, a sensor for collecting physiological signals of the user is a pressure sensor,

when a user is positioned on the intelligent mattress (10), a sensor on a channel which is most suitable for data analysis is selected to simultaneously start to collect physiological signals of the user, after a period of time, all channels are re-screened, when the number of points of the signals collected in the channels which are lower than/higher than a set threshold value is the largest, the channel is selected for data processing to obtain the physiological data of the user, and the intelligent terminal (20) and/or the cloud service management system (30) calculate the physiological data of the user based on the received physiological signals of the user and judge whether the abnormal state of the user occurs on the intelligent mattress (10) or not by combining a normal physiological data range provided by the user.

2. The data acquisition system of a smart mattress according to claim 1, wherein the signal acquired by the acquisition channel comprises a sum of heartbeat and respiration signals, and the heartbeat and respiration signals are modeled using sinusoidal signals, described respectively by the following equations:

S_h(t)＝A_h×Sin(2πf_h+θ_h) (1)

S_b(t)＝A_b×Sin(2πf_b+θ_b) (2)

S(t)＝S_h(t)+S_b(t)+A (3)

the formula (1) and the formula (2) are heartbeat and respiration voltage signals respectively, the formula (3) is a voltage signal output by the sensor, and comprises the sum of the heartbeat and the respiration signals and a direct current component A, wherein,

A_hand A_bAmplitude of heartbeat and respiration, respectively, f_hAnd f_bRespectively the heart beat and respiration rate, theta_hAnd theta_bThe initial phases of the heartbeat and respiration, respectively.

3. The data acquisition system of the intelligent mattress according to claim 1, wherein the intelligent terminal (20) performs AD conversion on the physiological signals acquired by the screened effective acquisition channels and performs fast fourier transform on the signal time domain spectrum obtained after preprocessing to obtain a signal frequency domain spectrum, obtains the number of peak outputs of the processed signal frequency domain spectrum exceeding a preset threshold value within a preset respiratory frequency range to obtain the respiratory frequency of the user, and obtains the number of peak outputs of the processed signal frequency domain spectrum exceeding the preset threshold value within a preset heartbeat frequency range to obtain the heartbeat frequency, wherein the preset respiratory frequency range is 0 to 0.5Hz, and the preset heartbeat frequency range is 0.6 to 2.7 Hz.

4. The data acquisition system for a smart mattress according to one of the preceding claims, wherein the fourier transform formula of the voltage signal s (t) output by the sensor of the acquisition channel is as follows:

5. The data acquisition system of an intelligent mattress according to claim 1, further comprising a temporary storage unit (103), wherein each acquisition channel of the data acquisition unit (101) acquires physiological signals of a user using the intelligent mattress (10) using a pressure sensor and stores the acquired physiological signals in the temporary storage unit (103) in a manner correlated with the acquisition channel corresponding thereto and acquisition time.

6. The data acquisition system of a smart mattress according to claim 5, characterized in that the temporary storage unit (103) is capable of pushing a data set consisting of at least the acquisition channel, the acquisition time and the physiological signals acquired by the acquisition channel temporarily stored in the temporary storage unit (103) to the smart terminal (20) in response to a successful matching of the communication device (102) with the smart terminal (20),

wherein the content of the first and second substances,

the intelligent terminal (20) calculates by screening out the physiological signals acquired by the acquisition channels with the AD voltage values meeting the set threshold value after being preprocessed so as to obtain the physiological data of the user on the intelligent mattress (10); or the intelligent terminal (20) forwards the data set to a cloud service management system (30), and the cloud service management system (30) analyzes the physiological signals collected by the collection channels which screen out the preprocessed AD voltage values and meet the set threshold value to obtain the physiological data of the user on the intelligent mattress (10).

7. The data acquisition system for a smart mattress of claim 1, wherein the data acquisition system for a smart mattress determines an orientation of the user's current face based on a comparison of physiological signals acquired by at least two acquisition channels and activates each acquisition channel respectively adjacent to the user's face, back and abdomen based on the orientation,

when the signals acquired by the acquisition channels respectively close to the face, the back and the abdomen of the user exceed a specific threshold value and the user can be judged not to turn over based on other acquisition channels, the data acquisition system activates an amplifier (101c) to pointedly amplify the physiological signals of the acquisition channel with the largest signal variation amplitude in the later time, and simultaneously activates a first filter (101b) and a second filter (101d) to inhibit the signals of other acquisition channels except the acquisition channel with the largest signal variation amplitude.

8. The data acquisition system of the smart mattress according to claim 1, wherein when the state of the user on the smart mattress (10) changes and the physiological signal variation of the user acquired by the acquisition channel reaches a set threshold, the smart terminal (20) and/or the cloud service management system (30) calculates based on the changed physiological signal acquired by the acquisition channel to obtain the physiological data after the state of the user changes;

or when the state of the user on the intelligent mattress (10) changes and the change of the physiological signals acquired by the acquisition channel does not reach a set threshold value, the intelligent terminal (20) and/or the cloud service management system (30) calculate to obtain the physiological data after the state of the user changes based on the physiological signals acquired by the acquisition channel before the change.