Intelligent mattress control method based on piezoelectric film sensor

Abstract

The invention discloses an intelligent mattress control method based on piezoelectric film sensors. According to the intelligent mattress control method, X rows and Y columns of piezoelectric film sensors arranged in an intelligent mattress, a signal processing module, X * Y registers, a lifting mechanism and a control module are included; each piezoelectric film sensor is electrically connected with the signal processing module, the signal processing module is electrically connected with the control module and each register, and the control module is electrically connected with the lifting mechanism; according to the invention, the action and breathing mode of the user can be automatically identified, and the corresponding control signal is generated, so that the user gets rid of dependence on a remote controller or a button, and the use convenience is increased; the possibility of false triggering is reduced, and the safety and the stability are enhanced.

Description

Technical Field

[0001] This invention relates to the field of smart bed technology, and in particular to a widely applicable and easy-to-operate smart mattress control method based on a piezoelectric thin film sensor. Background Technology

[0002] Smart mattresses are an important component of the next generation of home IoT and a significant indicator of the development of the home information age. Currently, smart mattresses on the market primarily use built-in sensors and intelligent control methods to monitor users' sleep patterns and adjust the mattress. These sensors can detect physiological indicators such as sleeping posture, heart rate, respiratory rate, and body movement. Intelligent algorithms analyze sleep quality and adjust parameters such as the mattress's tilt angle, firmness, and temperature based on the user's needs and sleep state, thereby providing a more comfortable sleep environment.

[0003] However, current smart mattress control methods on the market have some inconveniences, mainly in terms of user operation. Smart mattresses typically require remote control or buttons for operation, and users need to locate the control device when needed, which undoubtedly brings additional trouble, especially for users with limited mobility, making operation even more difficult. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of smart mattresses, which usually require remote control or buttons for operation and control, and to provide a widely applicable and easy-to-operate smart mattress control method based on piezoelectric thin film sensors.

[0005] To achieve the above objectives, the present invention adopts the following technical solution:

[0006] A smart mattress control method based on piezoelectric thin film sensors includes X rows and Y columns of piezoelectric thin film sensors installed in the smart mattress, a signal processing module, X×Y registers, a lifting mechanism and a control module installed on the smart mattress; each piezoelectric thin film sensor is electrically connected to the signal processing module, the signal processing module is electrically connected to the control module and each register respectively, and the control module is electrically connected to the lifting mechanism.

[0007] Step 100: The user lies supine on the smart mattress, and each thin-film sensor collects the user's analog electrical signals and transmits the analog electrical signals to the signal processing module;

[0008] Step 200: The signal processing module amplifies, samples, and performs analog-to-digital conversion on the X×Y analog electrical signals to obtain the X×Y composite signal S_C[i], where C = 1, 2, 3, ..., X×Y; i = 1, 2, 3, ..., N; and N is the data length of 60 seconds of S_C[i]. Each S_C[i] is stored in a corresponding register.

[0009] Step 300: The signal processing module performs zero-mean processing, normalization processing, and filtering processing on S_C[i] in each register to obtain the respiratory signal BR_C[i] and the action signal A_C[i].

[0010] Step 400: The signal processing module constructs thresholds for A_C[i] and BR_C[i] over 60 seconds to obtain the resting-state threshold MESB of the respiratory signal for each channel S_C[i]. C and the resting state threshold MESA of the action signal C ;

[0011] Step 500, Real-time Detection:

[0012] Step 510: The signal processing module uses the X×Y channel 1-second data New[t] acquired at the current time t to repeat steps 200-400 to calculate the total energy of the respiratory signal EB1_C and the total energy of the action signal EA1_C.

[0013] Step 520: The control module connects EA1_C and EB1_C to MESA respectively. C and MESB C Compare;

[0014] If EA1_C>MESA C ×th

[0015] Or EB1_C>MESB C ×th, th∈[1.0,1.3]

[0016] The control module determines that the user is breathing forcefully or tapping / pressing the mattress at time t, and sets the state parameter Con[t] at time t to 1;

[0017] If EA1_C≤MESA C ×th

[0018] Or EB1_C≤MESB C ×th, th∈[1.0,1.3]

[0019] Then set Con[t] at time t to 0 and return to step 510;

[0020] Step 530: After the detection continues for a period of time, if Con[t] = 1 remains for more than 3 seconds, and Con[t] is not continuously equal to 1 for the next 3 seconds, proceed to step 600; otherwise, return to step 510.

[0021] Step 600, enter control mode:

[0022] The user inputs a 2-second status parameter in the form of 1+1, and the control module generates a corresponding control signal and transmits it to the lifting mechanism. The lifting mechanism then controls the raising and lowering of the smart mattress according to the control signal.

[0023] If the user sets the state parameter Con[t] = 1 again for 3 consecutive seconds, the process returns to step 510. The piezoelectric film sensor converts a series of signals, including the user's cardiac impact signal, motion signal, breathing signal, and noise signal, into analog electrical signals. The signal processing module processes the data, and the processed signal is converted into control information and transmitted to the control module; the control module then converts the control information into specific mattress adjustment operations.

[0024] This invention can recognize the user's movements and breathing patterns and generate corresponding control signals, freeing the user from dependence on remote controls or buttons, increasing the convenience of actual use, and making the user more comfortable and free when using the mattress; the piezoelectric film sensors are arranged in an array, which reduces the possibility of false triggering compared with traditional integrated mattresses.

[0025] It not only achieves high-precision multi-point sensing and touch control, but also enhances safety and stability; users can control the smart mattress through various methods such as breathing, hand tapping, and leg impact; making it more convenient for users with limited mobility and the elderly to operate;

[0026] Users no longer need to rely on remote controls or buttons for control; instead, they can adjust the mattress's position and environment through simple movements or breathing patterns, greatly improving their quality of life and autonomy.

[0027] Preferably, step 400 includes the following specific steps:

[0028] Step 410, calculate the total energy EA_C of A_C[i] over 60 seconds, and calculate the total energy EB_C of BR_C[i] over 60 seconds:

[0029]

[0030] Step 420: The signal processing module calculates the instantaneous energy ESA_C[d] of A_C[i] over 60 seconds and the instantaneous energy ESB_C[d] of BR_C[i] over 60 seconds, where d = 1, 2, 3, ..., 60; the instantaneous interval is 1 second.

[0031]

[0032] Step 430: The signal processing module calculates the relative instantaneous energy ratio ESPA_C[d] of A_C[i] over 60 seconds and the relative instantaneous energy ratio ESPB_C[d] of BR_C[i] over 60 seconds.

[0033] ESPA_C[d] = ESA_C[d] / EA_C

[0034] ESPB_C[d] = ESB_C[d] / EB_C

[0035] Step 440: The signal processing module compares the instantaneous energy ratios ESPA_C[d] and ESPB_C[d] in each register, sorts the 60 instantaneous action energy ratios of ESPA_C[d] for each synthesized signal, and sorts the 60 instantaneous respiratory energy ratios of ESPB_C[d] for each synthesized signal. It then removes the largest and smallest 15% of data from each sorted sequence, obtaining the remaining 42 ESPA_C[d1] and 42 ESPB_C[d1], where d1 = 1, 2, 3, ..., 42; and calculates the resting-state threshold MESB of the respiratory signal. C and the resting state threshold MESA of the action signal C :

[0036]

[0037] Preferably, step 300 includes the following specific steps:

[0038] Step 310: The signal processing module performs zero-mean processing on S_C[i] in each register to obtain the zero-mean signal S_C. rm [i]:

[0039]

[0040] Step 320, the signal processing module processes S_C rm [i] Perform normalization to obtain the normalized signal S_C nm [i]::

[0041]

[0042] Where std() is the standard deviation function;

[0043] Step 330, the signal processing module processes S_C nm [i] is filtered to obtain the respiratory signal BR_C[i]:

[0044] BR_C[i] = filter1(S_C nm [i]);

[0045] Where filter1() is the breathing filter function;

[0046] Step 340, the signal processing module processes S_C nm[i] is filtered to obtain the action signal A_C[i]:

[0047] A_C[i] = filter2(S_C nm [i])

[0048] Here, filter2() is the action filtering function.

[0049] Preferably, the piezoelectric thin film sensor is a PVDF piezoelectric thin film sensor.

[0050] PVDF piezoelectric thin film sensors are soft, lightweight, and possess excellent flexibility, high sensitivity, and a wide frequency response range, enabling operation over a broad frequency spectrum. They also exhibit good impedance matching, high mechanical strength, and significantly higher working elasticity and tensile strength yield points compared to piezoelectric ceramics. This results in extremely high-speed response characteristics, stable operation under various environmental conditions, and the ability to be processed using existing equipment. Large-area sensing elements can be fabricated, and the relatively low permittivity is beneficial for signal processing. Suitable for applications requiring low thermal conductivity, these sensors can be manufactured into extremely thin films, making them ideal for applications requiring very thin materials. Cost-effective and inexpensive, they are suitable for large-scale production and application.

[0051] Preferably, the piezoelectric thin film sensor is located inside the smart mattress at a distance of 5-10 cm from the mattress surface.

[0052] The piezoelectric film sensor is located inside the smart mattress, 5-10 cm away from the mattress surface, which makes the piezoelectric film sensor highly sensitive and not easily damaged.

[0053] Preferably, the lower surface of the smart mattress is provided with a longitudinally extending left groove and a right groove; the lifting mechanism includes two symmetrically arranged left and right rotating mechanisms, the left rotating mechanism includes a left longitudinal rotating shaft in the left groove, a reclining left L-shaped support frame connected to the left longitudinal rotating shaft, a left insert plate connected to the left longitudinal rotating shaft, and a left servo motor on the bed for placing the smart mattress; the left insert plate is inserted into the smart mattress on the left side of the left groove, the left L-shaped support frame is connected to the lower surface of the smart mattress on the left side of the left groove, and the rotating shaft of the left servo motor is connected to one end of the left longitudinal rotating shaft.

[0054] When the foot of the bed needs to be raised, the control module controls the left servo motor to drive the left longitudinal shaft to rotate. The left longitudinal shaft drives the left L-shaped support frame and the left insert plate to rotate upward, and the left L-shaped support frame and the left insert plate raise the foot of the smart mattress.

[0055] Preferably, the right rotation mechanism includes a right longitudinal rotating shaft disposed in the right groove, a right L-shaped support frame that lies down connected to the right longitudinal rotating shaft, a right insert plate connected to the right longitudinal rotating shaft, and a right servo motor disposed on the bed; the right insert plate is inserted into the smart mattress on the right side of the right groove, the right L-shaped support frame is connected to the lower surface of the smart mattress on the right side of the right groove, and the rotating shaft of the right servo motor is connected to one end of the left longitudinal rotating shaft.

[0056] When the head of the bed needs to be raised, the control module controls the right servo motor to drive the right longitudinal shaft to rotate. The right longitudinal shaft drives the right L-shaped support frame and the right insert plate to rotate upward, and the right L-shaped support frame and the right insert plate raise the head of the smart mattress.

[0057] Preferably, the device also includes an electric blanket mounted on the smart mattress, with the control module electrically connected to the electric blanket. In winter, the user can transmit control information via breathing or motion signals to allow the control module to turn the electric blanket on or off.

[0058] Therefore, the present invention has the following beneficial effects:

[0059] It can automatically recognize the user's movements and breathing patterns and generate corresponding control signals, freeing the user from dependence on remote controls or buttons and increasing ease of use; it reduces the possibility of accidental triggering and enhances safety and stability; users no longer need to rely on remote controls or buttons for control, greatly improving the user's quality of life and autonomy. Attached Figure Description

[0060] Figure 1 This is a schematic diagram illustrating the use of the present invention;

[0061] Figure 2 This is a cross-sectional view of the smart mattress of the present invention;

[0062] Figure 3 yes Figure 2 A magnified view of a specific area;

[0063] Figure 4 This is a correspondence diagram of respiratory signals, action signals, and state parameter encoding according to the present invention. Detailed Implementation

[0064] The present invention will now be further described with reference to the accompanying drawings and specific embodiments.

[0065] like Figures 1-3The illustrated embodiment is a smart mattress control method based on piezoelectric thin film sensors, including 3 rows and 3 columns of piezoelectric thin film sensors 2 disposed in the smart mattress 1, a signal processing module, 9 registers, a lifting mechanism and a control module disposed on the smart mattress; each piezoelectric thin film sensor is electrically connected to the signal processing module, the signal processing module is electrically connected to the control module and each register respectively, and the control module is electrically connected to the lifting mechanism; the piezoelectric thin film sensor in this embodiment is a PVDF piezoelectric thin film sensor.

[0066] like Figure 2 , Figure 3 As shown, the lower surface of the smart mattress has a longitudinally extending left groove 1 and a right groove 2; the lifting mechanism includes two symmetrically arranged left and right rotating mechanisms, the left rotating mechanism includes a left longitudinal rotating shaft 31 located in the left groove, a reclining left L-shaped support frame 32 connected to the left longitudinal rotating shaft, a left insert plate 33 connected to the left longitudinal rotating shaft, and a left servo motor located on the bed for placing the smart mattress; the left insert plate is inserted into the left insert plate slot in the smart mattress on the left side of the left groove, the left L-shaped support frame is connected to the lower surface of the smart mattress on the left side of the left groove, and the rotating shaft of the left servo motor is connected to one end of the left longitudinal rotating shaft.

[0067] The right rotation mechanism includes a right longitudinal rotating shaft 41 located in the right groove, a right L-shaped support frame 42 connected to the right longitudinal rotating shaft and lying down, a right insert plate 43 connected to the right longitudinal rotating shaft, and a right servo motor located on the bed. The right insert plate is inserted into the right insert plate slot in the smart mattress on the right side of the right groove. The right L-shaped support frame is connected to the lower surface of the smart mattress on the right side of the right groove. The rotating shaft of the right servo motor is connected to one end of the left longitudinal rotating shaft. The piezoelectric film sensor is located inside the smart mattress, 6 cm away from the mattress surface, and is located on the upper side of the left and right insert plates.

[0068] Step 100: The user lies supine on the smart mattress, and each thin-film sensor collects the user's analog electrical signals and transmits the analog electrical signals to the signal processing module;

[0069] The initialization time is set to 60 seconds. During this time, users should try to keep their bodies still and avoid large movements to ensure they are in a calm, lying position.

[0070] Step 200: The signal processing module amplifies, samples, and performs analog-to-digital conversion on the X×Y analog electrical signals to obtain the X×Y composite signal S_C[i], where C = 1, 2, 3, ..., 9; i = 1, 2, 3, ..., 7500; and each S_C[i] is stored in a corresponding register.

[0071] Each integrated signal is assigned a register with a storage space containing 7500 points, or 60 seconds of signal length. The register uses a first-in, first-out (FIFO) storage rule, meaning that when new input data exceeds the register's length, the oldest input data is deleted, and all data is shifted forward one position.

[0072] Step 300: The signal processing module performs zero-mean normalization, normalization, and filtering on S_C[i] in each register to obtain the respiratory signal BR_C[i] and the action signal A_C[i].

[0073] Step 310: The signal processing module performs zero-mean processing on S_C[i] in each register to obtain the zero-mean signal S_C. rm [i]:

[0074]

[0075] Step 320, the signal processing module processes S_C rm [i] Perform normalization to obtain the normalized signal S_C nm [i]::

[0076]

[0077] Where std() is the standard deviation function;

[0078] Step 330, the signal processing module processes S_C nm [i] is filtered to obtain the respiratory signal BR_C[i]:

[0079] BR_C[i] = filter1(S_C nm [i]);

[0080] Where filter1() is the breathing filter function;

[0081] filter1 uses a fourth-order Butterworth bandpass filter with a bandwidth of 0.1 Hz to filter S_C. nm [i] Perform filtering to remove S_C nm [i] non-breathing signal frequency band signal data, and remove noise.

[0082] Step 340, the signal processing module processes S_C nm [i] is filtered to obtain the action signal A_C[i]:

[0083] A_C[i] = filter2(S_C nm [i])

[0084] Where filter2() is the action filtering function;

[0085] filter2 uses a fourth-order Butterworth low-pass filter with a cutoff frequency of 0.1Hz for S_C. nm [i] Perform filtering to remove S_C nm [i] non-action signal frequency band signal data, and remove noise.

[0086] Step 400: The signal processing module constructs thresholds for A_C[i] and BR_C[i] over 60 seconds to obtain the resting-state threshold MESB of the respiratory signal for each channel S_C[i]. C and the resting state threshold MESA of the action signal C :

[0087] Step 410, calculate the total energy EA_C of A_C[i] over 60 seconds, and calculate the total energy EB_C of BR_C[i] over 60 seconds:

[0088]

[0089] Step 420: The signal processing module calculates the instantaneous energy ESA_C[d] of A_C[i] over 60 seconds and the instantaneous energy ESB_C[d] of BR_C[i] over 60 seconds, where d = 1, 2, 3, ..., 60; the instantaneous interval is 1 second.

[0090]

[0091]

[0092] Step 430: The signal processing module calculates the relative instantaneous energy ratio ESPA_C[d] of A_C[i] over 60 seconds and the relative instantaneous energy ratio ESPB_C[d] of BR_C[i] over 60 seconds.

[0093] ESPA_C[d] = ESA_C[d] / EA_C

[0094] ESPB_C[d] = ESB_C[d] / EB_C

[0095] Step 440: The signal processing module compares the instantaneous energy ratios ESPA_C[d] and ESPB_C[d] in each register, sorts the 60 instantaneous action energy ratios of ESPA_C[d] for each synthesized signal, and sorts the 60 instantaneous respiratory energy ratios of ESPB_C[d] for each synthesized signal. It then removes the largest and smallest 15% of data from each sorted sequence, obtaining the remaining 42 ESPA_C[d1] and 42 ESPB_C[d1], where d1 = 1, 2, 3, ..., 42; and calculates the resting-state threshold MESB of the respiratory signal.C and the resting state threshold MESA of the action signal C :

[0096]

[0097] Step 500, Real-time Detection:

[0098] Step 510: The signal processing module uses the X×Y channel 1-second data New[t] acquired at the current time t to repeat steps 200-400 to calculate the total energy of the respiratory signal EB1_C and the total energy of the action signal EA1_C.

[0099] Step 520: The control module connects EA1_C and EB1_C to MESA respectively. C and MESB C Compare;

[0100] If EA1_C>MESA C ×th

[0101] Or EB1_C>MESB C ×th, th∈[1.0,1.3]

[0102] The control module determines that the user is breathing forcefully or tapping / pressing the mattress at time t, and sets the state parameter Con[t] at time t to 1;

[0103] If EA1_C≤MESA C ×th

[0104] Or EB1_C≤MESB C ×th, th∈[1.0,1.3]

[0105] Then set Con[t] at time t to 0 and return to step 510;

[0106] Step 530: After the detection continues for a period of time, if Con[t] = 1 remains for more than 3 seconds, and Con[t] is not continuously equal to 1 for the next 3 seconds, proceed to step 600; otherwise, return to step 510.

[0107] Step 600, enter control mode:

[0108] User inputs a 2-second status parameter in the form of 1+1:

[0109] like Figure 4 As shown, respiratory and motor signals have been converted into energy forms. Figure 4 It records the encoding of four state parameters: 0s~2s, 2s~4s, 4s~6s, and 6s~8s. Figure 4The upper part represents the user's breathing energy signal. Among them, 0s to 2s is the user's normal breathing state, 2s to 4s is the user's breathing state of breathing normally for one second and then taking a deep breath for one second, 4s to 6s is the user's breathing state of breathing deeply for one second and then breathing normally for one second, and 6s to 8s is the user's breathing state of taking two consecutive deep breaths.

[0110] Figure 4 The lower part represents the user's action energy signal, where 0s~2s is the user's no-tapping state, 2s~4s is the user's action state of first no tapping for one second and then tapping for one second, 4s~6s is the user's action state of first tapping the mattress for one second and then stopping tapping for one second, and 6s~8s is the user's action state of continuously tapping for two seconds.

[0111] As shown in Table 1, the state parameters in this embodiment are encoded in a 1+1 format, that is, a set of 2-second state parameter data is used as the encoding method. The correspondence between the state parameter encoding and the specific operation is as follows:

[0112] Specific operations State parameter encoding Raise the head of the bed 00 Lower the head of the bed 01 Raise the foot of the bed 10 Lower the foot of the bed 11

[0113] Table 1. Correspondence between Status Parameter Codes and Specific Operations

[0114] In this embodiment, the state parameter code is 00. The specific operation corresponding to this is that when the head of the bed is raised, the control module controls the right servo motor to drive the right longitudinal shaft to rotate. The right longitudinal shaft drives the right L-shaped support frame and the right insert plate to rotate upwards by a certain angle. The right L-shaped support frame and the right insert plate drive the head of the smart mattress to be raised.

[0115] If the user sets the state parameter Con[t] = 1 again for 3 consecutive seconds, return to step 510.

[0116] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A smart mattress control method based on a piezoelectric thin-film sensor, characterized in that it includes: The smart mattress (1) includes an X-row Y-column piezoelectric film sensor (2), a signal processing module, X×Y registers, a lifting mechanism and a control module on the smart mattress; each piezoelectric film sensor is electrically connected to the signal processing module, the signal processing module is electrically connected to the control module and each register, and the control module is electrically connected to the lifting mechanism. Step 100: The user lies supine on the smart mattress, and each thin-film sensor collects the user's analog electrical signals and transmits the analog electrical signals to the signal processing module; Step 200: The signal processing module amplifies, samples, and performs analog-to-digital conversion on the X×Y analog electrical signals to obtain the X×Y composite signal S_C[i], where C = 1, 2, 3, ..., X×Y; i = 1, 2, 3, ..., N; and N is the data length of 60 seconds of S_C[i]. Each S_C[i] is stored in a corresponding register. Step 300: The signal processing module performs zero-mean processing, normalization processing, and filtering processing on S_C[i] in each register to obtain the respiratory signal BR_C[i] and the action signal A_C[i]. Step 400: The signal processing module constructs thresholds for A_C[i] and BR_C[i] over 60 seconds to obtain the resting-state threshold MESB of the respiratory signal for each channel S_C[i]. C and the resting state threshold MESA of the action signal C ; Step 500, Real-time Detection: Step 510: The signal processing module uses the X×Y channel 1-second data New[t] acquired at the current time t to repeat steps 200-400 to calculate the total energy of the respiratory signal EB1_C and the total energy of the action signal EA1_C. Step 520: The control module connects EA1_C and EB1_C to MESA respectively. C and MESB C Compare; if EA1_C > MESA C ×th Or EB1_C>MESB C ×th, th∈[1.0,1.3] The control module determines that the user is breathing forcefully or tapping / pressing the mattress at time t, and sets the state parameter Con[t] at time t to 1; If EA1_C≤MESA C ×th Or EB1_C≤MESB C ×th, th∈[1.0,1.3] Then set Con[t] at time t to 0 and return to step 510; Step 530: After the detection continues for a period of time, if Con[t] = 1 remains for more than 3 seconds, and Con[t] is not continuously equal to 1 for the next 3 seconds, proceed to step 600; Otherwise, return to step 510; Step 600, enter control mode: The user inputs a 2-second status parameter in the form of 1+1. The control module generates a corresponding control signal and transmits it to the lifting mechanism. The lifting mechanism controls the raising and lowering of the smart mattress according to the control signal. When the user makes the status parameter Con[t] = 1 again for 3 consecutive seconds, the system returns to step 510.

2. The intelligent mattress control method based on a piezoelectric thin-film sensor according to claim 1, characterized in that, Step 400 includes the following specific steps: Step 410, calculate the total energy EA_C of A_C[i] over 60 seconds, and calculate the total energy EB_C of BR_C[i] over 60 seconds: Step 420: The signal processing module calculates the instantaneous energy ESA_C[d] of A_C[i] over 60 seconds and the instantaneous energy ESB_C[d] of BR_C[i] over 60 seconds, where d = 1, 2, 3, ..., 60; the instantaneous interval is 1 second. Step 430: The signal processing module calculates the relative instantaneous energy ratio ESPA_C[d] of A_C[i] over 60 seconds and the relative instantaneous energy ratio ESPB_C[d] of BR_C[i] over 60 seconds. ESPA_C[d] = ESA_C[d] / EA_C ESPB_C[d] = ESB_C[d] / EB_C Step 440: The signal processing module compares the instantaneous energy ratios ESPA_C[d] and ESPB_C[d] in each register, sorts the 60 instantaneous action energy ratios of ESPA_C[d] for each synthesized signal, and sorts the 60 instantaneous respiratory energy ratios of ESPB_C[d] for each synthesized signal. It then removes the largest and smallest 15% of data from each sorted sequence, obtaining the remaining 42 ESPA_C[d1] and 42 ESPB_C[d1], where d1 = 1, 2, 3, ..., 42; and calculates the resting-state threshold MESB of the respiratory signal. C and the resting state threshold MESA of the action signal C :

3. The intelligent mattress control method based on a piezoelectric thin-film sensor according to claim 1, characterized in that, Step 300 includes the following specific steps: Step 310: The signal processing module performs zero-mean processing on S_C[i] in each register to obtain the zero-mean signal S_C. rm [i]: Step 320, the signal processing module processes S_C rm [i] Perform normalization to obtain the normalized signal S_C nm [i]:: Where std() is the standard deviation function; Step 330, the signal processing module processes S_C nm [i] is filtered to obtain the respiratory signal BR_C[i]: BR_C[i]=filter1(S_C nm [i]); Where filter1() is the breathing filter function; Step 340, the signal processing module processes S_C nm [i] is filtered to obtain the action signal A_C[i]: A_C[i]=filter2(S_C nm [i]) Here, filter2() is the action filtering function.

4. The intelligent mattress control method based on a piezoelectric thin-film sensor according to claim 1, characterized in that, The piezoelectric thin film sensor is a PVDF piezoelectric thin film sensor.

5. The intelligent mattress control method based on a piezoelectric thin-film sensor according to claim 1, characterized in that, The piezoelectric thin film sensor is located inside the smart mattress, 5-10 cm away from the mattress surface.

6. The intelligent mattress control method based on a piezoelectric thin-film sensor according to claim 1, characterized in that, The smart mattress has a longitudinally extending left groove (1) and a right groove (2) on its lower surface; the lifting mechanism includes two symmetrically arranged left and right rotating mechanisms, the left rotating mechanism includes a left longitudinal rotating shaft (31) in the left groove, a reclining left L-shaped support frame (32) connected to the left longitudinal rotating shaft, a left insert plate (33) connected to the left longitudinal rotating shaft, and a left servo motor on the bed for placing the smart mattress; the left insert plate is inserted into the smart mattress on the left side of the left groove, the left L-shaped support frame is connected to the lower surface of the smart mattress on the left side of the left groove, and the rotating shaft of the left servo motor is connected to one end of the left longitudinal rotating shaft.

7. The intelligent mattress control method based on a piezoelectric thin-film sensor according to claim 6, characterized in that, The right rotation mechanism includes a right longitudinal rotating shaft (41) located in the right groove, a right L-shaped support frame (42) connected to the right longitudinal rotating shaft, a right insert plate (43) connected to the right longitudinal rotating shaft, and a right servo motor located on the bed; the right insert plate is inserted into the smart mattress on the right side of the right groove, the right L-shaped support frame is connected to the lower surface of the smart mattress on the right side of the right groove, and the rotating shaft of the right servo motor is connected to one end of the left longitudinal rotating shaft.

8. The smart mattress control method based on a piezoelectric thin film sensor according to claim 1, 2, 3, 4, 5, 6, or 7, characterized in that, It also includes an electric blanket installed on the smart mattress, with the control module electrically connected to the electric blanket.

Images