Detection device, training method and device, and computer-readable storage medium

Abstract

A detection device, comprising: a plurality of MEMS sensors (1), wherein the MEMS sensors (1) are configured to acquire, in real time, signals to undergo detection, and output sensing signals on the basis of the signals to undergo detection; and a processing module (2), wherein the processing module (2) is configured to receive the sensing signals from the plurality of MEMS sensors (1), and on the basis of sensing signals of the plurality of MEMS sensors (1) at each sampling time point, determine a detection result of the detection device; wherein there is a predetermined interval duration between two adjacent sampling time points, and a sensing signal of each MEMS sensor (1) at a current sampling time point is related to a signal to undergo detection at the current sampling time point, a sensing signal at a previous sampling time point, and an attenuation coefficient. The invention further relates to a training method for the detection device, a training device for the detection device, and a computer-readable storage medium.

Description

A detection device, training method and device, and computer-readable storage medium Technical Field

[0001] The present disclosure relates to the field of display technology, and in particular to a detection device, a training method, a training device, and a computer-readable storage medium. Background Art

[0002] MEMS (Micro Electro Mechanical System) is an emerging, interdisciplinary, high-tech research field. Pressure sensors manufactured using MEMS technology are now commonly used for pressure measurement in the consumer, automotive, aerospace, petrochemical, and biomedical industries due to their advantages such as small size, ease of integration, reliable performance, and ability to convert non-electrical signals into electrical signals.

[0003] Summary of the Invention

[0004] The present disclosure provides a detection device, comprising:

[0005] A plurality of MEMS sensors, wherein the MEMS sensors are configured to collect a signal to be detected in real time and output a sensing signal according to the signal to be detected;

[0006] a processing module, the processing module being configured to receive sensing signals from the plurality of MEMS sensors and determine a detection result of the detection device according to the sensing signals from the plurality of MEMS sensors at each sampling moment;

[0007] There is a predetermined interval between two adjacent sampling moments, and the sensing signal of each MEMS sensor at the current sampling moment is related to the signal to be detected at the current sampling moment, the sensing signal at the previous sampling moment, and the attenuation coefficient.

[0008] In some embodiments, the MEMS sensor includes: a MEMS pressure sensor.

[0009] In some embodiments, the MEMS sensor includes a pressure-sensitive cavity and a pressure-sensitive layer located on one side of the pressure-sensitive cavity along its depth direction, the maximum deformation of the pressure-sensitive layer is greater than 1 / 5 of the thickness of the pressure-sensitive layer; the maximum stress of the pressure-sensitive layer is less than 1 / 5 of the yield strength of the pressure-sensitive layer.

[0010] In some embodiments, the MEMS sensor comprises:

[0011] A first component includes the pressure-sensitive layer and a plurality of piezoresistors disposed on the pressure-sensitive layer, wherein the plurality of piezoresistors are connected in series to form a Wheatstone bridge;

[0012] a second component, the second component being bonded to the first component and defining the pressure-sensing cavity;

[0013] The piezoresistor and its orthographic projection on the pressure-sensing layer are located within the orthographic projection range of the pressure-sensing cavity on the pressure-sensing layer; and the Wheatstone bridge is electrically connected to the processing module through the second component to transmit the sensing signal of the pressure sensor to the processing module.

[0014] In some embodiments, the first component further includes a supporting portion, and the pressure-sensitive layer is located on a side of the supporting portion facing the second component, wherein the supporting portion and the pressure-sensitive layer are connected to form an integral structure.

[0015] In some embodiments, the first component further includes: a plurality of connectors located on a side of the pressure-sensitive layer facing the second component; and the plurality of piezoresistors are connected in series through the plurality of connectors.

[0016] In some embodiments, the connecting member includes: a connecting wire and a pad, the connecting wire connects two adjacent varistors, and the pad is located on a side of the connecting wire close to the second component.

[0017] In some embodiments, the first component further includes a first substrate, the piezoresistor is disposed on the first substrate, the first substrate is shared by the plurality of MEMS sensors, and a first power line and a second power line are further disposed on the first substrate;

[0018] The multiple connecting parts include: a first connecting part, a second connecting part and a third connecting part; the second connecting part is electrically connected to the first power line, the third connecting part is electrically connected to the second power line, and the first connecting part is electrically connected to the processing module through the second component.

[0019] In some embodiments, the second component includes:

[0020] a second substrate bonded to the first component, wherein a first via hole is formed on the second substrate;

[0021] a first conductive member located on a side of the second substrate away from the first component, the first conductive member being electrically connected to the first connecting member via a first transmission member; at least a portion of the first transmission member being located within the first via hole;

[0022] Wherein, the first conductive member electrically connects the first connecting member to the processing module.

[0023] In some embodiments, the plurality of MEMS sensors include a master sensor and a plurality of slave sensors, and the second substrate of the master sensor is further provided with a second via hole and a third via hole;

[0024] The second component of the main sensor also includes:

[0025] a second conductive member, located on a side of the second substrate away from the first component and electrically connected to the second connecting member through a second transmission member; at least a portion of the second transmission member is located in the second via hole;

[0026] a third conductive member, located on a side of the second substrate away from the first component and electrically connected to the third connecting member through a third transmission member, wherein at least a portion of the third transmission member is located in the third via hole;

[0027] The second conductive member is electrically connected to the first power supply end, and the third conductive member is electrically connected to the second power supply end.

[0028] In some embodiments, the MEMS sensor includes a second connector and a third connector, the second connector is used to receive a first power signal, and the third connector is used to receive a second power signal;

[0029] The multiple MEMS sensors include a master sensor and multiple slave sensors, and the master sensor and the multiple slave sensors are arranged and connected according to a topological structure. In the topological structure, the second connecting parts of any two connected sensors are connected, and the third connecting parts of any two connected MEMS sensors are connected.

[0030] In some embodiments, the topology includes any one of a star topology, a ring topology, a tree topology, and a bus topology;

[0031] The plurality of slave sensors in the star topology are located on one side of a first centerline of the master sensor and are arranged in a row; the plurality of slave sensors in the ring topology are arranged around the master sensor; the plurality of slave sensors in the tree topology are located on one side of a first centerline of the master sensor and are arranged in multiple rows; in two adjacent rows, the number of slave sensors in the row closer to the master sensor is smaller than the number of slave sensors in the row farther from the master sensor; the plurality of slave sensors in the bus topology are located on the same side of a second centerline of the master sensor and are distributed on opposite sides of the first centerline of the master sensor;

[0032] Wherein, the first center line intersects with the second center line.

[0033] In some embodiments, the processing module includes:

[0034] a signal amplifying submodule, configured to amplify output signals of the plurality of MEMS sensors to obtain amplified signals of the plurality of MEMS sensors;

[0035] an analog-to-digital conversion submodule, configured to perform analog-to-digital conversion on the output signal of the signal amplification module to obtain analog-to-digital conversion results of the multiple MEMS sensors;

[0036] The analysis module is configured to determine the detection result of the detection device according to the analog-to-digital conversion results of the multiple MEMS sensors.

[0037] In some embodiments, the analysis module includes:

[0038] A monitor configured to determine a target signal range based on the analog-to-digital conversion results of the multiple MEMS sensors; and determine whether there is an error result in the analog-to-digital conversion results of the multiple MEMS sensors based on the target signal range;

[0039] The algorithm processor is configured to correct the error result if the error result exists; and perform analysis and processing based on the correction result and the analog-to-digital conversion results of the remaining MEMS sensors to determine the detection result of the detection device.

[0040] The present disclosure provides a training method for a detection device as described in any one of the above embodiments, comprising:

[0041] creating a model of the detection device;

[0042] Acquire a training set, wherein the training set includes a plurality of sample detection signals;

[0043] Inputting the sample detection signal into the model to obtain the sample output signal;

[0044] According to the difference between the sample output signal and the target output signal, the excitation condition of the MEMS sensor in the model and the predetermined interval duration of the processor in the model are adjusted until the target training condition is met.

[0045] In some embodiments, the excitation condition includes a first power supply voltage and a second power supply voltage provided to the MEMS sensor.

[0046] The present disclosure provides a training device for a detection device, comprising:

[0047] A model creation module is configured to create a model of the detection device;

[0048] A training set acquisition module is configured to acquire a training set, wherein the training set includes a plurality of sample detection signals;

[0049] The training module is configured to input the sample detection signal into the model to obtain the sample output signal; and adjust the excitation conditions of the MEMS sensor in the model and the predetermined interval duration of the processor in the model according to the difference between the sample output signal and the target output signal until the target training conditions are met.

[0050] The present disclosure provides a computer-readable storage medium having a computer program stored thereon. When the program is executed by a processor, the method described in the above embodiments is implemented. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] The accompanying drawings are used to provide a further understanding of the present disclosure and constitute a part of the specification. Together with the following detailed description, they are used to explain the present disclosure but do not constitute a limitation of the present disclosure. In the accompanying drawings:

[0052] FIG1 is a schematic structural diagram of a detection device provided in an embodiment of the present disclosure.

[0053] FIG2A is a schematic diagram of input and output of a detection device provided in an embodiment of the present disclosure.

[0054] FIG2B is a schematic diagram of the working principle of a sensor provided in an embodiment of the present disclosure.

[0055] FIG3 is a schematic structural diagram of a sensor provided in an embodiment of the present disclosure.

[0056] FIG4A is a cross-sectional view of a first component provided in an embodiment of the present disclosure.

[0057] FIG4B is a top view of a first component provided in an embodiment of the present disclosure.

[0058] FIG4C is a cross-sectional view of another first component provided in an embodiment of the present disclosure.

[0059] FIG4D is a top view of another first component provided in an embodiment of the present disclosure.

[0060] FIG. 5A is a cross-sectional view of a second component provided in an embodiment of the present disclosure.

[0061] FIG5B is a top view of a second component provided in an embodiment of the present disclosure.

[0062] FIG6 is a schematic diagram of the bonding between a second component and a first component provided in an embodiment of the present disclosure.

[0063] FIG7 is a wiring diagram of multiple first components provided in an embodiment of the present disclosure.

[0064] FIG8 is a schematic diagram of a topological structure in which multiple sensors may be arranged according to an embodiment of the present disclosure.

[0065] FIG9 is a block diagram of a detection device provided in an embodiment of the present disclosure.

[0066] 10A to 10G are process flow charts of a sensor provided in an embodiment of the present disclosure.

[0067] FIG11 is a schematic structural diagram of a training device for a detection device provided in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0068] The following describes the specific embodiments of the present disclosure in detail with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are only used to illustrate and explain the present disclosure and are not intended to limit the present disclosure.

[0069] In this specification, for convenience, words and phrases indicating orientation or positional relationships, such as "middle," "upper," "lower," "front," "back," "vertical," "horizontal," "top," "bottom," "inside," and "outside," are used to illustrate the positional relationships of constituent elements with reference to the accompanying drawings. This is only for the purpose of facilitating the description of this specification and simplifying the description, and does not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operate in a specific orientation. Therefore, it should not be understood as a limitation of the present disclosure. The positional relationships of the constituent elements may be appropriately changed according to the direction in which each constituent element is described. Therefore, the words and phrases described in the specification are not limited and may be appropriately replaced according to the circumstances.

[0070] Unless otherwise defined, the technical terms or scientific terms used in the embodiments of the present disclosure should have the usual meanings understood by people with ordinary skills in the field to which the present disclosure belongs. The words "first", "second" and similar words used in the present disclosure do not indicate any order, quantity or importance, but are only used to distinguish different components. Similarly, words such as "include" or "comprise" mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. The terms "install", "connect" and "connect" should be understood in a broad sense. For example, it can be fixedly connected, or detachably connected, or connected in one piece; it can be mechanically connected or connected; it can be directly connected, or indirectly connected through an intermediate piece, or the internal communication of two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present disclosure can be understood according to specific circumstances.

[0071] With the development of the Internet of Things (IoT), research is increasingly underway both domestically and internationally on using various sensors to simulate neural network algorithms, aiming to provide better solutions for edge computing. Academic research on intelligent MEMS sensors has yielded significant results, effectively combining the technological advantages of MEMS, such as small size, light weight, and low power consumption. This has initially demonstrated the feasibility of using MEMS smart sensors based on neural network algorithms in tasks such as pattern recognition and time series prediction.

[0072] However, MEMS sensors are mostly designed as single, independent devices, and the nonlinear mapping effects they can achieve are often only suitable for relatively simple tasks. This significantly restricts the further development of the technical advantages of MEMS sensors, the realization of stronger computing power, the potential for greater market applications, and the provision of strong technical support for the future "Internet of Everything" era of the Internet of Things.

[0073] FIG1 is a schematic diagram of the structure of a detection device provided in an embodiment of the present disclosure. As shown in FIG1 , the detection device includes: multiple sensors 1 and a processing module 2. Sensors 1 are configured to acquire a signal to be detected in real time and output a sensing signal based on the signal to be detected. The signal to be detected can be a signal to be detected in nature, such as a pressure signal, sound signal, or other signal. Sensors 1 are MEMS sensors.

[0074] The processing module 2 is configured to receive sensing signals from the plurality of sensors 1 and determine a detection result of the detection device according to the sensing signals from the plurality of sensors 1 at each sampling moment.

[0075] Sensor 1 can collect the signal to be detected in real time, and processing module 2 does not need to process all the sensing signals generated by sensor 1. Instead, it can process the sensing signals generated by sensor 1 at multiple sampling moments. There is a predetermined interval between two adjacent sampling moments. The sensing signal of each sensor 1 at the current sampling moment is related to the signal to be detected at the current sampling moment, the sensing signal at the previous sampling moment, and the attenuation coefficient. The attenuation coefficient is related to the inherent parameters of the sensor 1 device.

[0076] Among them, the device intrinsic parameters of sensor 1 are parameters related to the device structure of sensor 1 itself. For example, if sensor 1 is a pressure sensor, the device intrinsic parameters of sensor 1 may include parameters such as the thickness of the pressure membrane of the pressure sensor and the sensing area.

[0077] Among them, the attenuation coefficient can characterize the nonlinearity of sensor 1. The so-called "nonlinearity" is a concept opposite to "linearity". "Linearity" means that the sensing signal (i.e., output signal) of sensor 1 at the current sampling moment is only related to the current signal to be detected (i.e., input signal), and has nothing to do with the output signal at the previous sampling moment. "Nonlinearity" means that the sensing signal of sensor 1 at the current sampling moment is not only related to the signal to be detected at the current sampling moment, but also to the sensing signal at the previous sampling moment. The larger the attenuation coefficient, the greater the influence of the sensing signal at the previous sampling moment on the sensing signal at the current sampling moment, and the higher the nonlinearity of sensor 1; the smaller the attenuation coefficient, the less the influence of the sensing signal at the previous sampling moment on the sensing signal at the current sampling moment, and the lower the nonlinearity of sensor 1.

[0078] The detection results of the detection device may include the current usage of the detection device. For example, using a pressure detection device as an example, the processing module can obtain the sensing signals of multiple pressure sensors to obtain the pressure applied to each pressure sensor, thereby determining the pressure state of the pressure detection device. For example, the pressure detection device can be a device worn on the human body, and the processing module 2 can determine the posture and movement of the human body by performing force analysis. The processing module 2 can determine the detection results based on a classification algorithm.

[0079] The detection device provided in the disclosed embodiments includes multiple sensors 1. A processing module 2 processes the sensing signals output by the multiple sensors 1 to determine a detection result. This allows for the fusion of the sensing signals from the multiple sensors 1 to obtain a more reliable and accurate perception result, thereby improving the fault tolerance and environmental adaptability of the detection device. Because the multiple sensors 1 in the detection device utilize a parallel structure, the computational speed of the detection device for detecting signals to be detected can also be increased.

[0080] In addition, the sensing signal of any sensor 1 among the multiple sensors 1 at the current sampling moment is not only related to the signal to be detected at the current sampling moment, but is also affected by the sensing signal at the previous sampling moment. Therefore, the sensors 1 provided in the present disclosure all have a high degree of nonlinearity and can be used as neurons in a neural network. The mutual coupling between the multiple sensors 1 can make the nonlinear mapping of the signal to be detected stronger, thereby improving the classification and recognition ability of the neural network, and thus making the perception effect of the detection device better, the accuracy higher, and achieving stronger computing power.

[0081] Figure 2A is a schematic diagram of the input and output principles of the detection device provided in an embodiment of the present disclosure, and Figure 2B is a schematic diagram of the working principle of a sensor 1 provided in an embodiment of the present disclosure, wherein the multiple sensors in Figures 2A and 2B can improve the detection effect of the detection device by realizing the hardware of the neural network.

[0082] For example, when the neural network is a recurrent neural network, as shown in Figures 2A and 2B, X = [X1, X2, X3...Xm] represents the signal to be detected by the detection device, that is, the input layer of the recurrent neural network. S = [S1, S2, S3...Sn] represents the detection device, that is, the hidden layer of the neural network. Among them, S1, S2, S3...Sn represent multiple sensors 1 in the detection device, that is, neurons in the recurrent neural network. O = [O1, O2, O3...Om] represents the detection result of the detection device, that is, the output layer of the recurrent neural network.

[0083] As shown in Figure 2B, S t =f(U*X t +W*X t-1 ) represents the sensing signal of the sensor 1 of the detection device at the sampling time t. The sensing signal St at the current sampling time is generated based on the signal to be detected at the current sampling time, the sensing signal at the previous sampling time, and the attenuation coefficient. t =g(V*S t ) represents the detection result of the processing module 2 at the sampling time t. W represents the attenuation coefficient, that is, the feedback weight in the recurrent neural network.

[0084] Among them, for the input signal at a certain moment, multiple sensors can realize multi-point data collection and mapping, and the sensing signal of each sensor is not only affected by the signal to be measured at the current sampling moment, but also by the state accumulation of this sensor, thereby realizing more complex signal transmission and nonlinear state coupling, and improving the overall high-dimensional mapping capability of the detection device.

[0085] Figure 3 is a schematic diagram of the structure of a sensor 1 provided in an embodiment of the present disclosure. As shown in Figure 3, in this embodiment of the present disclosure, the sensor 1 provided in the embodiment of the present disclosure includes a MEMS pressure sensor. A MEMS pressure sensor is a micro-electromechanical system that can measure pressure and convert it into an electrical signal for output. MEMS pressure sensors are characterized by small size, light weight, low cost, high reliability, suitability for mass production, ease of integration, and intelligent implementation.

[0086] The MESM pressure sensor includes a first component 101 and a second component 102, wherein the second component 102 can be an integrated circuit (ACIS) chip. The first component 101 is used to convert the sensed pressure into an electrical signal, and the second component 102 is used to output the electrical signal output by the first component 101 to the processing module 2. The first component 101 and the second component 102 are typically electrically connected using anodic bonding.

[0087] In some embodiments, as shown in FIG3 , the sensor 1 includes a pressure-sensing cavity 3 and a pressure-sensing layer 1010 located on one side of the pressure-sensing cavity 3 along its depth direction. The maximum deformation of the pressure-sensing layer 1010 is greater than 1 / 5 of its thickness. The maximum stress of the pressure-sensing layer 1010 is less than 1 / 5 of its yield strength.

[0088] It should be noted that the maximum deformation refers to the deformation of the pressure-sensitive layer 1010 when the maximum pressure value of the detection range of the sensor 1 is applied to the pressure-sensitive layer 1010. The deformation is specifically the displacement of the pressure-sensitive layer 1010 in its thickness direction.

[0089] The yield strength of the pressure-sensitive layer 1010 refers to the ability of the pressure-sensitive layer 1010 to resist external forces. The yield strength of the pressure-sensitive layer 1010 is also used to measure the ability of the material to withstand pressure without deformation under the action of external forces.

[0090] In the disclosed embodiment, the maximum deformation of the pressure-sensing layer 1010 is greater than 1 / 5 of its thickness. This ensures that the pressure-sensing layer 1010 cannot return to its natural state (i.e., unstressed state) from its maximum deformation during the interval between two adjacent sampling moments. This ensures that the sensing signal of the sensor 1 at the current sampling moment is affected by the sensing signal at the previous sampling moment. Furthermore, the maximum stress of the pressure-sensing layer 1010 is less than 1 / 5 of its yield strength, preventing damage to the sensing layer 1010.

[0091] Figure 4A is a cross-sectional view of a first component provided in an embodiment of the present disclosure, and Figure 4B is a top view of a first component provided in an embodiment of the present disclosure. As shown in Figures 3 to 4B, in some embodiments, the sensor 1 includes a first component 101 and a second component 102. The first component 101 also includes a support portion 1011, and a pressure-sensitive layer 1010 is located on the side of the support portion 1011 facing the second component 102. The support portion 1011 and the pressure-sensitive layer 1010 are connected to form an integral structure, which is recorded as the first substrate 101a.

[0092] As shown in FIG3 to FIG4B , a groove Va is formed on the side of the first substrate 101a away from the second component 102 , wherein at least a portion of the pressure-sensitive layer 1010 is located between the surface of the first substrate 101a facing the second component 102 and the bottom surface of the groove Va.

[0093] The first substrate 101a may be a silicon substrate.

[0094] As shown in Figures 3 to 4B , a plurality of piezoresistors 101d are provided on the pressure-sensing layer 1010, and the plurality of piezoresistors 101d are connected in series to form a Wheatstone bridge. The piezoresistors 101d can be shaped like strips or curves, depending on actual needs. In one example, the shapes of the plurality of piezoresistors 101d can be different; for example, two piezoresistors 101d can be strips, while the remaining two piezoresistors 101d can be bent.

[0095] 4A and 4B , the first component 101 further includes a plurality of connectors S. The plurality of connectors S are located on a side of the first substrate 101 a facing the second component 102. The plurality of varistors 101 d are connected in series via the plurality of connectors S.

[0096] In some embodiments, as shown in FIG. 4A and FIG. 4B , the connector S includes a connecting wire 101 c and a pad 101 d . The connecting wire 101 c connects two adjacent varistors 101 d , and the pad 101 d is located on a side of the connecting wire 101 c close to the second component 102 .

[0097] In some examples, as shown in FIG3 , the first component 101 may further include a lightly doped layer 101b, with the varistor 101d and the connector S both located on the side of the lightly doped layer 101b facing the second component 102. The lightly doped layer 101b may be formed by lightly doping a silicon substrate, and the connecting wire 101c may be formed by heavily doping a silicon substrate and then performing a patterning process.

[0098] In one example, the pad 101d may be a metal pad. Specifically, a metal layer may be formed by a sputtering process and then subjected to a patterning process to form the metal pad.

[0099] In FIG. 4A to FIG. 4B , the varistor 101 d and the connecting line 101 c may be provided in the same layer, and both are formed by heavy doping and patterning processes.

[0100] FIG4C is a cross-sectional view of another first assembly provided in an embodiment of the present disclosure, and FIG4D is a top view of another first assembly provided in an embodiment of the present disclosure. The first assemblies shown in FIG4C and FIG4D are similar to those in FIG4A and FIG4B , except that in FIG4C and FIG4D , the varistor 101d may be a PN junction. Different connecting wires 101c and different PN junctions are separated by a lightly doped layer 101b.

[0101] FIG7 is a schematic diagram illustrating the wiring of multiple first components provided in an embodiment of the present disclosure. In some embodiments, as shown in FIG7 , multiple sensors 1 share a common first substrate 101a. A first power line K1 and a second power line K2 are also provided on the first substrate 101a. One of the first power line K1 and the second power line K2 is a drive signal line, and the other is a ground line.

[0102] As shown in FIG3 , the plurality of connectors S include a second connector Sa, a third connector Sb, and a first connector Sc. The second connector Sa is electrically connected to the first power line K1, the third connector Sb is electrically connected to the second power line K2, and the first connector Sc is electrically connected to the processing module 2 via the second component 102 to transmit the output signal of the pressure sensor 1 to the processing module 2.

[0103] It should be noted that the second connector Sa is connected to the first power line K1. This can be done by connecting the connecting wire 101d in the second connector Sa or by connecting the pad 101c in the second connector Sa. Alternatively, both the connecting wire 101d and the pad 101c in the second connector Sa are connected to the first power line K1. The first power line K1 can be on the same layer as the connecting wire 101d or the pad 101c. Alternatively, the first power line K1 can include two sublayers, each of which is on the same layer as the connecting wire 101d and the pad 101c.

[0104] Similarly, the third connector Sb is connected to the second power line K2. This can be done by connecting the connecting wire 101c in the third connector Sb or by connecting the pad 101d in the third connector Sb. Alternatively, both the connecting wire 101c and the pad 101d in the third connector Sb are connected to the second power line K2. The second power line K2 can be on the same layer as the connecting wire 101c or the pad 101d. Alternatively, the second power line K2 can include two sublayers, each of which is on the same layer as the connecting wire 101c and the pad 101d.

[0105] It should be noted that "same-layer arrangement" in the present disclosure means that the two structures are formed by the same material layer through a composition process, so the two are in the same layer in terms of stacking relationship; but this does not mean that the distance between the two and the same reference plane must be the same.

[0106] FIG5A is a cross-sectional view of a second assembly provided in an embodiment of the present disclosure, and FIG5B is a top view of the second assembly provided in an embodiment of the present disclosure. As shown in FIG5A and FIG5B , in some embodiments, the second assembly 102 includes a second substrate 102a and a first conductive member 102e. The first conductive member 102e may be a solder ball having good conductivity. The second substrate 102a is bonded to the first assembly 101, and a first via V1 is defined in the second substrate 102a.

[0107] The first conductive member 102e is located on a side of the second substrate 102a away from the first component 101. The first conductive member 102e is electrically connected to the first connector Sc via the first transmission member 102b. At least a portion of the first transmission member 102b is located within the first via V1. The first conductive member 102e electrically connects the first connector Sc to the processing module 2.

[0108] Among them, the part of the first transmission member 102b located in the first via hole V1 can be a metal material, and can be located on the side wall of the first via hole V1, but does not completely fill the first via hole V1. In this case, a filling part 102c can be set in the first via hole V1, wherein the filling part 102c can include but is not limited to conductive materials such as metal and glass paste, and of course, it can also be an insulating material.

[0109] FIG6 is a schematic diagram illustrating the bonding of a second component to a first component according to an embodiment of the present disclosure. As shown in FIG3 and FIG6 , the second component 102 is bonded to the first component 101 to define a pressure-sensing cavity 3. The orthographic projection of the piezoresistor 101d on the pressure-sensing layer 1010 is located within the orthographic projection of the pressure-sensing cavity 3 on the pressure-sensing layer 1010.

[0110] As shown in FIG. 5B and FIG. 7 , in some embodiments, the plurality of sensors 1 include a master sensor 1A and a plurality of slave sensors 1B. The second substrate 102 a of the master sensor 1A is further provided with a second via hole V2 and a third via hole V3 .

[0111] In addition to the first conductive element 102e, the second component 102 of the main sensor 1A also includes a second conductive element 102f and a third conductive element 102g. The second conductive element 102f is located on a side of the second substrate 102a away from the first component 101 and is electrically connected to the second connector Sa via a second transmission element 102m. At least a portion of the second transmission element 102m is located within the second via V2.

[0112] As shown in Figures 5A, 5B, and 7, the third conductive member 102g is located on a side of the second substrate 102a away from the first component 101 and is electrically connected to the third connector Sb via a third transmission member 102n. At least a portion of the third transmission member 102n is located within the third via V3. The second conductive member 102f is electrically connected to the first power supply terminal, and the third conductive member 102g is electrically connected to the second power supply terminal.

[0113] The first power supply terminal and the second power supply terminal are ports for providing power signals to the multiple sensors 1. For example, the multiple sensors 1 and the processing module 2 can be integrated on the same printed circuit board (PCB), and the first power supply terminal and the second power supply terminal can be power supply ports provided on the PCB.

[0114] In the disclosed embodiment, the master sensor 1A may include a first conductive element 102e, a second conductive element 102f, and a third conductive element 102g, while the slave sensor 1B may include the first conductive element 102e but not the second conductive element 102f and the third conductive element 102g. In this case, the signal from the first power supply terminal can be provided to the master sensor 1A via the second conductive element 102f and then to each slave sensor 1B via the first power line K1. The signal from the second power supply terminal can be provided to the master sensor 1A via the third conductive element 102g and then to each slave sensor 1B via the second power line K2. This configuration eliminates the need for the second conductive element 102f and the third conductive element 102g in the slave sensor 1B, thereby simplifying the structure of the sensor 1, optimizing the wiring layout, and improving the fatigue resistance and reliability of the sensor 1.

[0115] In some embodiments, the master sensor 1 and multiple slave sensors 1 can be arranged as needed, for example, they can be arranged in an array (i.e., arranged in multiple rows and columns), or they can be arranged and connected according to a topological structure. In the topological structure, the second connecting parts Sa of any two connected sensors 1 can be connected through the first power line K1, and the third connecting parts Sb of any two connected sensors 1 can be connected through the second power line K2.

[0116] FIG8 is a schematic diagram of a topological structure in which multiple sensors provided in an embodiment of the present disclosure can be arranged. As shown in FIG8 , in some embodiments, the topological structure includes: any one of a star topology, a ring topology, a tree topology, and a bus topology. Specifically, as shown in FIG8 (a), the multiple sensors in the detection device can be arranged in a star topology. One sensor 1 is selected from the multiple sensors as the main sensor 1A, and the remaining sensors 1 are all used as slave sensors 1B, and the multiple slave sensors 1B are located on one side of the first center line L1 of the main sensor 1A and are arranged in a row or a column. Among them, all the slave sensors 1B are connected to the main sensor 1A through the first power line K1 and the second power line K2 respectively. Among them, the first center line L1 is a straight line passing through the center of the main sensor 1A and parallel to the pressure-sensitive layer 1010.

[0117] As shown in FIG8( b ), the multiple sensors 1 in the detection device can be arranged in a ring topology. Among the multiple sensors 1 , one sensor 1 serves as a master sensor 1A, and the remaining sensors 1 serve as slave sensors 1B. The master sensor 1A is surrounded by the multiple slave sensors 1B, and all the slave sensors 1B are connected to the master sensor 1A via a first power line K1 and a second power line K2 , respectively.

[0118] Specifically, as shown in FIG8(c), the multiple sensors 1 in the detection device can be arranged in a tree-like topology. One of the multiple sensors 1 serves as a master sensor 1A, and the remaining sensors 1 serve as slave sensors 1B. The multiple slave sensors 1B in the tree-like topology are located on one side of a first centerline L1 of the master sensor 1A and arranged in multiple rows. In two adjacent rows, the number of slave sensors 1B in the row closer to the master sensor 1A is smaller than the number of slave sensors 1B in the row farther from the master sensor 1A.

[0119] Among them, the master sensor 1A is connected to each slave sensor 1B in the first row through the first power line K1 and the second power line K2; among the multiple slave sensors 1B, each slave sensor 1B in the upper row is connected to at least two slave sensors 1B in the lower row through the first power line K1 and the second power line K2.

[0120] As shown in Figure 8 (d), the multiple sensors 1 in the detection device can be arranged in a bus topology. Specifically, the multiple slave sensors 1B in the bus topology are located on the same side of the second centerline L2 of the master sensor 1A and are distributed on opposite sides of the first centerline L1 of the master sensor 1A. The first centerline L1 intersects the second centerline L2. For example, the first centerline L1 is perpendicular to the second centerline L2. The multiple sensors 1 in the detection device of the present disclosure can be arranged in various topologies, which are not specifically limited here.

[0121] It should be noted that in the multiple topologies shown in FIG8 , each circular or square structure represents a sensor, and the interconnection lines between the circular and square structures represent RDL (ReDistribution Layer) routing. The circular structure represents a master sensor (including drive, ground, and two signal outputs) that is connected to an external electrical signal through the second component 102 , while the square structure represents a slave sensor. The slave sensors can achieve electrical signal interconnection with the master sensor through different connection methods (topologies), eliminating the need for each slave sensor to be connected to an external drive power signal and ground signal through the second component 102 .

[0122] In practical applications, a specific arrangement structure can be selected based on the application scenario of the detection device, thereby facilitating improved detection accuracy of the processing module. For example, the detection device can be used in a handwriting tablet, in which case the multiple sensors 1 can be arranged in an array. Another example is that the detection device can be used in a wearable device, in which case the multiple sensors 1 can be arranged according to a topological structure, and the specific topological structure can be selected based on the wearable device's wearing position.

[0123] The present disclosure provides a method for manufacturing a sensor, comprising the following steps:

[0124] S1. Make the first component.

[0125] S2. Make a second component and bond the first component to the second component.

[0126] The first component 101 and the second component 102 may be bonded by anodic bonding.

[0127] FIG10A to FIG10G are process flow charts of a sensor provided in an embodiment of the present disclosure. As shown in FIG10A to FIG10G , in some embodiments, step S1 includes the following steps:

[0128] S10, as shown in FIG10A, provides a first substrate 101a. The material of the first substrate 101a can be single crystal silicon, which is a high-purity silicon material. The first substrate 101a made of single crystal silicon has the advantages of small thermal expansion coefficient and high mechanical strength.

[0129] S11 , lightly doping the first substrate 101 a with n-type doping to form an n-type lightly doped layer 101 b .

[0130] Among them, the impurity concentration in the n-type lightly doped layer is low. Under this doping method, the carrier mobility in the doped layer is high and the electron velocity distribution is more uniform, so it has better conductivity and lower resistivity.

[0131] S12, as shown in FIG10B, forming a plurality of connecting wires 101c. The n-type lightly doped layer may be heavily doped, and a patterning process is performed to form a plurality of connecting wires 101c and a plurality of varistors 101d.

[0132] S13. As shown in FIG10B , pads are formed on the connection wires 101c. Each connection wire 101c and the corresponding pad 101d constitute a connection member.

[0133] S14 , as shown in FIG10C , etching the surface of the first substrate 101 a away from the connecting line to form a groove Va.

[0134] In some embodiments, step S2 includes the following steps:

[0135] S20, as shown in FIG10D, provides a second substrate 102a. The material of the second substrate 102a can be glass. The locations where the via holes and recessed structures are to be formed on the second substrate 102 are processed using a laser modification process.

[0136] S21 . As shown in FIG. 10E , a via hole and a recessed structure Vb are formed on the second substrate 102 a by using a wet etching process.

[0137] For the master sensor 1A, the plurality of via holes may include a first via hole V1 , a second via hole, and a third via hole; and for the slave sensor 1B, the formed via holes may only include the first via hole V1 .

[0138] In some embodiments, CMP (Chemical Mechanical Polishing) and other processes can also be used to achieve thinning and flattening of the second substrate 102a in the second component 102, thereby ensuring the performance of the second component 102 and facilitating subsequent bonding between the first component 101 and the second component 102.

[0139] S22. As shown in FIG10F , the second substrate 102a is bonded to the first component 101. After bonding, the recessed structure Vb is formed into a vacuum cavity.

[0140] After step S22, as shown in FIG10G, a first transmission element 102b, a second transmission element, and a third transmission element are formed. At least a portion of the first transmission element 102b is located within the first via V1 and connected to the first connector among the multiple connectors; at least a portion of the second transmission element is located within the second via and connected to the second connector among the multiple connectors; and at least a portion of the third transmission element is located within the third via and connected to the third connector among the multiple connectors. Furthermore, an insulating layer 102d and a first conductive element 102e, a second conductive element, and a third conductive element are formed on the second substrate 102a. The first conductive element 102e, the second conductive element, and the third conductive element are all located on the side of the insulating layer 102d away from the second substrate 102a. The first conductive element 102e is connected to the first transmission element 102b via a via in the insulating layer 102d. The second conductive element is connected to the second transmission element via a via in the insulating layer 102d. The third conductive element is connected to the third transmission element via a via in the insulating layer.

[0141] The positional relationship of the second transmission member, the third transmission member, the second connecting member, the third connecting member, the second conductive member, and the third conductive member can be found in the above description of FIG5B and FIG7. In addition, the second transmission member, the third transmission member, the second conductive member, and the third conductive member may not be required for the slave sensor.

[0142] FIG9 is a block diagram of a detection device provided in an embodiment of the present disclosure. As shown in FIG9 , in some embodiments, the processing module 2 includes a signal amplification submodule 201, an analog-to-digital conversion submodule 202, and an analysis module 203. The signal amplification submodule 201 is configured to amplify the output signals of the plurality of sensors 1 to obtain amplified signals of the plurality of sensors 1. The analog-to-digital conversion submodule 202 is configured to perform analog-to-digital conversion on the output signals of the signal amplification module 201 to obtain analog-to-digital conversion results of the plurality of sensors 1. The analysis module is configured to determine the detection result of the detection device based on the analog-to-digital conversion results of the plurality of sensors 1.

[0143] In some embodiments, the analysis module 203 includes a monitor and an algorithm processor. The monitor is configured to determine a target signal range based on the analog-to-digital conversion results of the multiple sensors 1; and based on the target signal range, determine whether there are any errors in the analog-to-digital conversion results of the multiple sensors 1. The algorithm processor is configured to correct the error results if any; and analyze and process the correction results and the analog-to-digital conversion results of the remaining sensors 1 to determine the detection results of the detection device.

[0144] When the analog-to-digital conversion result of a sensor 1 exceeds the target signal range, the result is considered an error, and the corresponding sensor is marked as an error sensor. In this case, the error result can be corrected based on the analog-to-digital conversion results of the remaining sensors 1. For example, the average of the analog-to-digital conversion results of the remaining sensors can be used as the correction result. Alternatively, the analog-to-digital conversion results of the remaining sensors can be filtered according to predetermined filtering parameters to obtain the correction result. For example, if a sensor 1 is continuously marked as an error sensor for a period of time, the error sensor can be latched and isolated, and a fault message can be issued to alert the user.

[0145] It should be understood that when there is no error result, analysis and processing can be performed directly based on the analog-to-digital conversion results of the multiple sensors 1.

[0146] In some embodiments, the processing module 2 may further include a register (not shown). After determining that an error result exists, the monitor may send the error result to the register for storage.

[0147] The processing module 2 may further include a power supply submodule (not shown) for supplying power to the analysis submodule 203 , the mode conversion submodule 202 , and the signal amplification submodule 201 .

[0148] The present disclosure provides a training method for a detection device as described in any one of the above embodiments, the training method comprising the following steps:

[0149] S1. Create a model of the detection device.

[0150] S2. Obtain a training set, where the training set includes multiple sample detection signals.

[0151] S3. Input the sample detection signal into the model to obtain a sample output signal. Based on the difference between the sample output signal and the target output signal, adjust the excitation conditions of the sensor 1 in the model and the predetermined interval duration of the processor in the model until the target training conditions are met.

[0152] In some embodiments, the excitation condition includes a first supply voltage and a second supply voltage provided to the sensor 1 .

[0153] It should be understood that, given other conditions, the longer the processor's predetermined interval, the less the sensor is affected by the previous sampling moment, i.e., the lower the sensor's nonlinearity. Furthermore, given other conditions, differences in the first and second power supply voltages will also result in differences in the sensor's sensing signal. Therefore, by adjusting the first and second power supply voltages, as well as the predetermined interval, the nonlinearity of the multiple sensors can be optimized, thereby optimizing the parameters of the neural network and, in turn, making the detection structure of the detection device more accurate.

[0154] It should be noted that the target training condition may be that the loss function determined based on the difference between the sample output signal and the target output signal is less than a preset value, or that the number of training times reaches a predetermined number.

[0155] It should also be noted that the model can be trained multiple times using multiple sample detections in the training set, and the sample detection signals used in each training process are different.

[0156] Figure 11 is a structural schematic diagram of a training device for a detection device provided in an embodiment of the present disclosure. As shown in Figure 11, the present disclosure provides a training device 500 for a detection device, and the training device 500 includes a model creation module 100, a training set acquisition module 200, and a training module 300.

[0157] The model creation module 100 is configured to create a model of the detection device.

[0158] The training set acquisition module 200 is configured to acquire a training set, where the training set includes a plurality of sample detection signals.

[0159] The training module 300 is configured to input the sample detection signal into the model to obtain a sample output signal; based on the difference between the sample output signal and the target output signal, adjust the excitation conditions of the sensor 1 in the model and the predetermined interval duration of the processor in the model until the target training conditions are met.

[0160] The present disclosure provides a computer-readable storage medium having a computer program stored thereon. When the program is executed by a processor, the method in the above embodiment is implemented.

[0161] Computer-readable media may include computer storage media (or non-transitory media) and communication media (or temporary media). As is well known to those of ordinary skill in the art, the term computer storage media includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules or other data). Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and can be accessed by a computer. In addition, it is well known to those of ordinary skill in the art that communication media typically contains computer-readable instructions, data structures, program modules or other data in a modulated data signal such as a carrier wave or other transmission mechanism, and may include any information delivery medium.

[0162] It will be understood that the above embodiments are merely exemplary embodiments for illustrating the principles of the present invention, and the present invention is not limited thereto. Those skilled in the art will appreciate that various modifications and improvements can be made without departing from the spirit and substance of the present invention, and such modifications and improvements are also considered to be within the scope of protection of the present invention.

Claims

1. A detection device, comprising: A plurality of MEMS sensors, wherein the MEMS sensors are configured to collect signals to be detected in real time and output sensing signals according to the signals to be detected; A processing module, wherein the processing module is configured to receive sensing signals from the multiple MEMS sensors and determine a detection result of the detection device according to the sensing signals from the multiple MEMS sensors at each sampling moment; There is a predetermined interval between two adjacent sampling moments, and the sensing signal of each MEMS sensor at the current sampling moment is related to the signal to be detected at the current sampling moment, the sensing signal at the previous sampling moment, and the attenuation coefficient.

2. The detection device according to claim 1, wherein: The MEMS sensor includes: a MEMS pressure sensor.

3. The detection device according to claim 1 or 2, wherein: The MEMS sensor includes a pressure-sensitive cavity and a pressure-sensitive layer located on one side of the pressure-sensitive cavity along the depth direction thereof, wherein the maximum deformation of the pressure-sensitive layer is greater than 1 / 5 of the thickness of the pressure-sensitive layer; and the maximum stress of the pressure-sensitive layer is less than 1 / 5 of the yield strength of the pressure-sensitive layer.

4. The detection device according to claim 3, wherein: The MEMS sensor comprises: A first component includes the pressure-sensitive layer and a plurality of varistors disposed on the pressure-sensitive layer, wherein the plurality of varistors are connected in series to form a Wheatstone bridge; a second component, the second component is bonded to the first component and defines the pressure sensing cavity; The piezoresistor and its orthographic projection on the pressure-sensing layer are located within the orthographic projection range of the pressure-sensing cavity on the pressure-sensing layer; and the Wheatstone bridge is electrically connected to the processing module through the second component to transmit the sensing signal of the pressure sensor to the processing module.

5. The detection device according to claim 4, wherein: The first component further includes a supporting portion, and the pressure-sensitive layer is located on a side of the supporting portion facing the second component, wherein the supporting portion and the pressure-sensitive layer are connected to form an integral structure.

6. The detection device according to claim 4, wherein: The first component further includes: a plurality of connectors located on a side of the pressure-sensitive layer facing the second component; and the plurality of varistors are connected in series through the plurality of connectors.

7. The detection device according to claim 6, wherein the connecting member comprises: A connecting wire and a pad, wherein the connecting wire connects two adjacent varistors, and the pad is located on a side of the connecting wire close to the second component.

8. The detection device according to claim 6, wherein: The first component further includes a first substrate, the piezoresistor is arranged on the first substrate, the first substrate of the plurality of MEMS sensors is shared, and a first power line and a second power line are also arranged on the first substrate; The multiple connectors include: a first connector, a second connector and a third connector; the second connector is electrically connected to the first power line, the third connector is electrically connected to the second power line, and the first connector is electrically connected to the processing module through the second component.

9. The detection device according to claim 8, wherein: The second component comprises: A second substrate, bonded to the first component, and having a first via hole formed in the second substrate; a first conductive member, located at a side of the second substrate away from the first component, the first conductive member being electrically connected to the first connecting member through a first transmission member; at least a portion of the first transmission member being located in the first via hole; Wherein, the first conductive member electrically connects the first connecting member to the processing module.

10. The detection device according to claim 9, wherein: The multiple MEMS sensors include a master sensor and multiple slave sensors, and the second substrate of the master sensor is further provided with a second via hole and a third via hole; The second component of the main sensor also includes: A second conductive member is located on a side of the second substrate away from the first component and is electrically connected to the second connecting member through a second transmission member; at least a portion of the second transmission member is located in the second via hole; A third conductive member is located at a side of the second substrate away from the first component and is electrically connected to the third connecting member through a third transmission member, and at least a portion of the third transmission member is located in the third via hole; Wherein, the second conductive member is electrically connected to the first power supply end, and the third conductive member is electrically connected to the second power supply end.

11. The detection device according to any one of claims 1 to 10, wherein: The MEMS sensor comprises a second connecting member and a third connecting member, wherein the second connecting member is used to receive a first power supply signal, and the third connecting member is used to receive a second power supply signal; The multiple MEMS sensors include a master sensor and multiple slave sensors, and the master sensor and the multiple slave sensors are arranged and connected according to an array structure or a topological structure. In the array structure or the topological structure, the second connecting members of any two connected sensors are connected, and the third connecting members of any two connected MEMS sensors are connected.

12. The detection device according to claim 11, wherein: The topology structure includes: any one of a star topology structure, a ring topology structure, a tree topology structure and a bus topology structure; The plurality of slave sensors in the star topology are located on one side of the first center line of the master sensor and arranged in a row; the plurality of slave sensors in the ring topology are arranged in a row. The slave sensors in the tree topology are arranged around the master sensor; the multiple slave sensors in the tree topology are located on one side of the first center line of the master sensor and are arranged in multiple rows; in two adjacent rows, the number of the slave sensors in the row close to the master sensor is less than the number of the slave sensors in the row far from the master sensor; the multiple slave sensors in the bus topology are located on the same side of the second center line of the master sensor and are distributed on opposite sides of the first center line of the master sensor; Wherein, the first center line intersects with the second center line.

13. The detection device according to any one of claims 1 to 12, wherein: The processing module comprises: a signal amplification submodule, configured to amplify output signals of the plurality of MEMS sensors to obtain amplified signals of the plurality of MEMS sensors; an analog-to-digital conversion submodule, configured to perform analog-to-digital conversion on the output signal of the signal amplification module to obtain analog-to-digital conversion results of the multiple MEMS sensors; The analysis module is configured to determine the detection result of the detection device according to the analog-to-digital conversion results of the multiple MEMS sensors.

14. The detection device according to claim 13, wherein: The analysis module comprises: A monitor configured to determine a target signal range according to the analog-to-digital conversion results of the plurality of MEMS sensors; and to determine whether there is an error result in the analog-to-digital conversion results of the plurality of MEMS sensors according to the target signal range; The algorithm processor is configured to correct the error result when the error result exists; and to analyze and process the correction result and the analog-to-digital conversion results of the remaining MEMS sensors to determine the detection result of the detection device.

15. A training method for a detection device according to any one of claims 1 to 14, comprising: Creating a model of the detection device; Acquire a training set, wherein the training set includes a plurality of sample detection signals; Inputting the sample detection signal into the model to obtain the sample output signal; According to the difference between the sample output signal and the target output signal, the excitation condition of the MEMS sensor in the model and the predetermined interval duration of the processor in the model are adjusted until the target training condition is met.

16. The training method according to claim 15, wherein: The excitation conditions include a first supply voltage and a second supply voltage provided to the MEMS sensor.

17. A training device for a detection device, comprising: A model creation module is configured to create a model of the detection device; A training set acquisition module is configured to acquire a training set, wherein the training set includes a plurality of sample detection signals; The training module is configured to input the sample detection signal into the model to obtain the sample output signal; and adjust the excitation conditions of the MEMS sensor in the model and the predetermined interval duration of the processor in the model according to the difference between the sample output signal and the target output signal until the target training conditions are met.

18. A computer-readable storage medium having a computer program stored thereon, wherein the program implements the method of claim 15 or 16 when executed by a processor.

Images