Sensing device and electronic equipment

Abstract

The invention provides a sensing device and electronic equipment, and relates to the technical field of micro electro mechanical systems, and the sensing device comprises a resonator which comprises a movable beam structure, and the initial motion state of the movable beam structure is a first resonant motion; the mechanical sensor is used for responding to force from an object and generating a target sensing signal, the target sensing signal is used for converting first resonant motion of the movable beam structure into second resonant motion, so that the resonator outputs a target resonant signal, and the target resonant signal is used for indicating displacement of the movable beam structure; and the signal analyzer is used for determining object attribute information according to the target resonance signal, and the object attribute information comprises the type and / or weight of the object. Based on the scheme, the resonator and the mechanical sensor are used in cooperation, the neural network based on the simulation domain is constructed based on the response result of the resonator to the sensing signal, the requirement for computing power is relatively low, the implementation cost is relatively low, and therefore the sensing device integrating sensing and calculation is achieved.

Description

Technical Field

[0001] The present application relates to the technical field of micro-electromechanical systems, and more particularly, to sensing devices and electronic devices. Background Art

[0002] With the development of sensor technology, smart terminal devices are gradually moving towards a more intelligent and autonomous direction. In this context, how to integrate micro-electro-mechanical systems (MEMS) sensors with artificial intelligence (AI) processing chips to build smart sensor devices has become an important topic in sensor technology at this stage. However, the smart sensor devices proposed at this stage have problems such as insufficient computing power and high cost. Summary of the invention

[0003] The present application provides a sensing device and an electronic device. By using a resonator in combination with a mechanical sensor, a neural network based on an analog domain is constructed based on the response of the resonator to the sensing signal. The computing power requirement is relatively low and the implementation cost is relatively low, thereby realizing a sensing device that integrates perception and computing.

[0004] In a first aspect, a sensing device is provided, which includes: a resonator, including a movable beam structure, wherein the initial motion state of the movable beam structure is a first resonant motion; a mechanical sensor, which is used to respond to a force from an object and generate a target sensing signal, wherein the target sensing signal is used to convert the movable beam structure from the first resonant motion to a second resonant motion, so that the resonator outputs a target resonant signal, wherein the target resonant signal is used to indicate the displacement of the movable beam structure; and a signal analyzer, which is used to determine object attribute information based on the target resonant signal, wherein the object attribute information includes the type and / or weight of the object.

[0005] In combination with the first aspect, in some implementations of the first aspect, the above-mentioned sensing device also includes: an AC power supply, the AC power supply is used to input an excitation signal to the resonator, and the excitation signal is used to drive the movable beam structure to perform a first resonant motion.

[0006] In combination with the first aspect, in some implementations of the first aspect, the resonant frequency of the first resonant motion is related to a response speed of the resonator.

[0007] In combination with the first aspect, in certain implementations of the first aspect, the above-mentioned resonator is specifically used to: generate a target resonance signal based on a high-dimensional mapping function, an excitation signal and a target sensor signal, and the high-dimensional mapping function is used to indicate a mapping relationship between the target resonance signal and the excitation signal and the target sensor signal.

[0008] In combination with the first aspect, in some implementations of the first aspect, the signal analyzer includes at least one classifier, and the classifier is trained based on different target resonance signals and object attribute information corresponding to the different target resonance signals.

[0009] In combination with the first aspect, in certain implementations of the first aspect, the above-mentioned sensing device includes: a base layer; a circuit layer, used to set a mechanical sensor and a resonator; a buried oxide (BOX) layer, the BOX layer is located between the base layer and the circuit layer, and the BOX layer includes holes, and the holes are opposite to the movable beam structure of the resonator; a packaging layer, the packaging layer is located on the side of the circuit layer away from the base layer, the packaging layer includes a first vacuum cavity and a second vacuum cavity, the first vacuum cavity covers the movable beam structure, and the second vacuum cavity covers the force sensitive element of the mechanical sensor.

[0010] In combination with the first aspect, in certain implementations of the first aspect, the base layer includes a back cavity, the vertical projection of the force sensitive element onto the projection plane is located within the vertical projection of the back cavity onto the projection plane, and the projection plane is parallel to the lower surface of the base layer.

[0011] In combination with the first aspect, in certain implementations of the first aspect, the force sensitive element includes: a pressure-sensitive film, which is disposed on a side of the BOX layer away from the base layer; and a varistor, which is disposed on a side of the pressure-sensitive film away from the base layer.

[0012] In combination with the first aspect, in certain implementations of the first aspect, the above-mentioned packaging layer is a wafer-level packaging structure, and the above-mentioned packaging layer also includes: a glass cover cap, a plurality of through holes, a plurality of leads and a plurality of solder balls, wherein the plurality of through holes penetrate the glass cover cap, the first end of the through hole is connected to the side of the mechanical sensor or resonator away from the base layer, the second end of the through hole is connected to the solder ball, and the through hole is used to set the lead.

[0013] In combination with the first aspect, in certain implementations of the first aspect, the mechanical sensor is a MEMS pressure sensor, and the resonator is a MEMS resonator.

[0014] In a second aspect, an electronic device is provided, comprising a sensor device in any possible implementation manner of the sensor device design of the first aspect. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 is a schematic diagram of the architecture of a sensor device 100 proposed in an embodiment of the present application;

[0016] Figure 2 is a schematic diagram of the architecture of another sensor device 100 proposed in an embodiment of the present application;

[0017] Figure 3is a circuit structure diagram of a resonator 110 proposed in an embodiment of the present application;

[0018] Figure 4 is a circuit structure diagram of another resonator 110 proposed in an embodiment of the present application;

[0019] Figure 5 is a structural schematic diagram of a sensor device 100 proposed in an embodiment of the present application;

[0020] Figure 6 It is a schematic diagram of an application of a smart fishing float 600 proposed in an embodiment of the present application. DETAILED DESCRIPTION

[0021] The technical solution in this application will be described below in conjunction with the accompanying drawings.

[0022] The embodiments of the present application will present various aspects, embodiments or features around a system including multiple devices, components, modules, etc. It should be understood and appreciated that each system may include additional devices, components, modules, etc., and / or may not include all devices, components, modules, etc. discussed in conjunction with the figures. In addition, combinations of these schemes may also be used.

[0023] In addition, in the embodiments of the present application, words such as "exemplary" and "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "exemplary" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of the word "exemplary" is intended to present concepts in a concrete way.

[0024] The business scenarios described in the embodiments of the present application are intended to more clearly illustrate the technical solutions of the embodiments of the present application, and do not constitute a limitation on the technical solutions provided in the embodiments of the present application. Ordinary technicians in this field can know that with the evolution of technology and the emergence of new business scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

[0025] References to "one embodiment" or "some embodiments" etc. described in this specification mean that a particular feature, structure or characteristic described in conjunction with the embodiment is included in one or more embodiments of the present application. Thus, the phrases "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear at different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0026] In the embodiments of the present application, "at least one" refers to one or more, and "plurality" refers to two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can be represented by: including the existence of A alone, the existence of A and B at the same time, and the existence of B alone, where A and B can be singular or plural. The character " / " generally indicates that the previous and subsequent associated objects are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b, or c can be represented by: a, b, c, ab, ac, bc, or abc, where a, b, c can be single or multiple.

[0027] In the description of the embodiments of the present application, the directions or positional relationships indicated by terms such as "up", "down", "left", "right", "inside", "outside", "vertical" and "horizontal" are defined relative to the directions or positions of the components schematically placed in the drawings. It should be understood that these directional terms are relative concepts. They are used for relative description and clarification, rather than indicating or implying that the device or component referred to must have a specific direction, or be constructed and operated in a specific direction. They may change accordingly according to changes in the directions of the components placed in the drawings, and therefore cannot be understood as limitations on the present application.

[0028] In the embodiments of the present application, the same reference numerals are used to represent the same components or parts. For the same parts in the embodiments of the present application, only one of the parts or parts may be marked with a reference numeral in the figure. It should be understood that the reference numerals are also applicable to other identical parts or parts. In addition, the various parts in the drawings are not drawn to scale, and the sizes and dimensions of the parts shown in the drawings are only exemplary and should not be understood as limiting the present application.

[0029] With the development of sensor technology, smart terminal devices are gradually moving towards a more intelligent and autonomous direction. In this context, how to integrate MEMS sensors with AI processing chips to build smart sensing devices has become an important topic in sensor technology at this stage.

[0030] However, the sensor device design schemes proposed at this stage are usually based on the construction of neural networks in the digital domain. However, the neural network built in the digital domain has high requirements for processor computing power during operation, which directly increases the manufacturing cost of the sensor device and increases the structural complexity of the sensor device. In some special scenarios, it is impossible to achieve the integration of perception and computing.

[0031] In view of this, the embodiments of the present application propose a sensing device and an electronic device. By using a resonator in combination with a mechanical sensor, a neural network based on the analog domain is constructed based on the response of the resonator to the sensing signal. The computing power requirement is relatively low and the implementation cost is relatively low, thereby realizing a sensing device that integrates perception and computing.

[0032] Figure 1 Schematic diagram of the structure of a sensor device 100 proposed in an embodiment of the present application.

[0033] refer to Figure 1 As shown, the sensor device 100 includes:

[0034] The resonator 110 includes a movable beam structure 111, and the initial motion state of the movable beam structure 111 is a first resonant motion;

[0035] The mechanical sensor 120 is used to generate a target sensing signal in response to a force from an object, wherein the target sensing signal is used to convert the movable beam structure 111 from the first resonant motion to a second resonant motion, so that the resonator 110 outputs a target resonant signal, wherein the target resonant signal is used to indicate a displacement of the movable beam structure 111;

[0036] The signal analyzer 130 is used to determine object attribute information according to the target resonance signal, where the object attribute information includes the type and / or weight of the object.

[0037] In some possible embodiments, reference Figure 1 As shown, the mechanical sensor 120 may include a force-sensitive element 121, which is used to sense the force applied by an external object and convert the mechanical quantity into a change in a physical quantity that can be more easily measured, such as a change in current.

[0038] In some possible embodiments, the force sensitive element 121 can convert the change of force into an electrical signal, and the waveform of the electrical signal can be used to characterize the relevant characteristics of the force applied by the object to the mechanical sensor 120, such as the magnitude of the force, the trend of force change, and even the direction of the force.

[0039] By way of example, the force-sensitive element 121 may include a resistance strain gauge, a piezoelectric crystal, a capacitive sensor, an inductive sensor, and the like.

[0040] It should be understood that when the sensor device 100 proposed in the embodiment of the present application extracts the object property information of the object based on the force applied by the object, it is necessary to obtain the force continuously applied by the object over a period of time, and the duration of this period needs to be calibrated in combination with the response speed of the resonator 110, so that the sensor device 100 can be dedicated to the specific type and / or weight of a specific type of object. For example, the sensor device 100 can be used to detect specific types of fish, such as grass carp, carp, etc., in which case the sensor device 100 can be used as the core device of an intelligent fishing float, or the sensor device 100 can be used to detect specific types of vegetables, such as cabbage, potatoes, etc., in which case the sensor device 100 can be used as the core device of an intelligent electronic scale, and the response speed of the resonator 110 may be different for different detection objects.

[0041] In some possible embodiments, reference Figure 1 As shown, the above-mentioned sensing device 100 may also include a hardware control circuit, which includes: a driving circuit 101 and a control circuit 102, wherein the driving circuit 101 is used to realize basic functions such as driving and signal detection of the above-mentioned resonator 110, and the control circuit 102 is used to realize signal threshold control of the mechanical sensor 120, that is, the target sensing signal amplitude of the mechanical sensor 120 needs to be greater than or equal to the preset threshold value, and the target sensing signal is transmitted to the resonator 110 only when the target sensing signal amplitude is greater than or equal to the preset threshold value, which is equivalent to a gate.

[0042] In some possible embodiments, reference Figure 1 As shown, the sensor device 100 may further include an AC power supply 140, wherein the AC power supply 140 is used to input an excitation signal to the resonator 110, and the excitation signal is used to drive the movable beam structure 111 to perform the first resonant motion.

[0043] It should be understood that when the movable beam structure 111 performs the first resonant motion, a resonant signal may be output accordingly, and the resonant signal matches the excitation signal.

[0044] In some possible embodiments, the AC power supply 140 may belong to the driving circuit 101 .

[0045] In some possible embodiments, the response speed of the resonator 110 is related to the size, shape, material of the resonator 110, and the frequency of the above-mentioned excitation signal, wherein the size, shape and material of the resonator 110 directly affect the natural frequency of the resonator 110, and when the frequency of the excitation signal matches the natural frequency of the resonator 110, the resonator 110 will resonate, and at this time the response speed of the resonator 110 is the fastest.

[0046] In addition, the damping coefficient of the resonator 110 also directly affects the response speed of the resonator 110 . The smaller the damping coefficient, the faster the response speed of the resonator 110 .

[0047] It can be seen that the resonant frequency of the first resonant motion is related to the response speed of the resonator 110 .

[0048] Therefore, when the sensing device 100 is used in different scenarios, the response speed of the resonator 110 can be calibrated in advance, that is, the physical parameters of the resonator 110 itself (including the movable beam structure 111), such as size, shape, material, damping coefficient, etc., and the working parameters of the AC power supply, such as the frequency of the excitation signal, can be determined.

[0049] For example, when the sensing device 100 is used to identify the physical property information of some objects that move at a faster speed, the frequency of the excitation signal can be set to a higher level so that the resonator 110 has a faster response speed; conversely, when the sensing device 100 is used to identify the physical property information of some objects that move at a slower speed, the frequency of the excitation signal can be appropriately lowered to adaptively reduce the response speed of the resonator 110 and reduce the overall operating overhead of the sensing device 100.

[0050] It should be understood that the movable beam structure 111 of the resonator 110 is an elastic structure that can vibrate under external excitation.

[0051] In some possible embodiments, the movable beam structure 111 may be made of a material with a high elastic modulus and good mechanical properties, such as silicon, metal alloy, etc. The selection of these materials is intended to ensure that the movable beam can maintain stable performance and reliability during long-term vibration.

[0052] In some possible embodiments, movable beam structures 111 with different designs can be applicable to the above-mentioned sensing device 100. For example, the shape of the movable beam structure 111 can be a beam with fixed support at both ends, a cantilever beam, a crab-claw beam, a bow beam, a right-angle beam, etc.

[0053] In some possible embodiments, the transducing mechanism of the resonator 110 may be a capacitive transducing mechanism, that is, utilizing the change of capacitance to realize energy conversion and transmission. When the movable beam structure 111 vibrates (such as the first resonant motion or the second resonant motion), it causes a change in capacitance, thereby converting mechanical energy into electrical energy (such as generating the target resonant signal).

[0054] It should be noted that the transducing mechanism of the resonator 110 is not limited to the capacitive type, and other forms such as piezoelectric type, electromagnetic type, etc. may also be adopted.

[0055] It should be understood that the first resonant motion refers to the natural response of the resonator 110 under the action of the excitation signal, but the target sensing signal, after being combined with the excitation signal, will directly affect the motion mode of the resonator 110, that is, under the action of the target sensing signal, the movable beam structure 111 is converted from the first resonant motion to the second resonant motion, and this conversion may involve changes in the motion frequency, amplitude or motion mode. In the second resonant motion, the vibration of the movable beam structure 111 is converted into an electrical signal (for example, through the piezoelectric effect), thereby generating a target resonant signal.

[0056] Since the target resonance signal is generated based on the second resonance motion, the target resonance signal will carry corresponding characteristic information. In the embodiment of the present application, the characteristic information may be the displacement of the movable beam structure 111 .

[0057] Since the target sensing signal is generated by the force exerted by the external object in response to the mechanical sensor 120, and the second resonant motion is carried out under the influence of the target sensing signal and generates a target resonance signal, the characteristic information carried by the target resonance signal can be used to describe some characteristics of the external object, which may include the above-mentioned object attribute information.

[0058] In some possible embodiments, the mechanical sensor 120 may be a MEMS pressure sensor, which can realize a high-sensitivity, high-resolution sensing function of the force applied by an external object.

[0059] In some possible embodiments, the resonator 110 may be a MEMS resonator, and the MEMS resonator has rich nonlinear dynamic characteristics, which enable the resonator 110 to be used as a nonlinear node in reservoir computing (RC). This is because the MEMS resonator exhibits a complex nonlinear response when subjected to external excitation (such as the target sensor signal mentioned above), and the nonlinear response can be mapped to the nonlinear node characteristics in the reservoir calculation, that is, the external excitation is mapped to a higher-dimensional computing space, and this process is also called high-dimensional mapping. Based on high-dimensional mapping, the processing power and generalization ability of RC can be further enhanced.

[0060] In some possible embodiments, the above RC is not limited to various variants under recurrent neural network (RNN).

[0061] Figure 2 It is a schematic diagram of the architecture of another sensor device 100 proposed in an embodiment of the present application.

[0062] refer to Figure 2As shown, the mechanical sensor 120 selected by the sensing device 100 is a MEMS pressure sensor, and the force sensitive element 121 of the MEMS pressure sensor can be Figure 2 The mechanical sensor 120 also has ports 21 and 22 for signal transmission on its surface. The ports 21 and 22 can be made into solder ball structures, wherein the port 21 is used to output a positive voltage signal corresponding to the pressure when the force-bearing surface of the MEMS pressure sensor receives pressure; the port 22 is used to output a negative voltage signal corresponding to the pressure when the force-bearing surface of the MEMS pressure sensor receives pressure.

[0063] In addition, the resonator 110 selected by the sensor device 100 is a MEMS resonator, and the movable beam structure 111 of the MEMS resonator can be Figure 2 The two-end clamped beam shown. Ports 11, 12 and 13 for signal transmission are also introduced on the surface of the resonator 110, and the ports 11 to 13 can be made into solder ball structures, wherein port 11 is used to receive a target sensing signal, and the target sensing signal is determined based on the voltage signal Vout output by the mechanical sensor 120 and the DC voltage Vac connected to the sensor device 100, wherein the voltage signal Vout is a voltage signal obtained by mixing the positive voltage signal and the negative voltage signal, and the target sensing signal is expressed in the following form: (Vac+Vout) / sinwx; port 12 is used to receive an AC voltage signal, so that the movable beam structure 111 can perform a first resonant motion based on the AC signal; after the resonator 110 obtains the target sensing signal received by port 11, based on the processing method mentioned in the above embodiment, the target sensing signal and the resonant signal output by the movable beam structure 111 are processed accordingly, and the target resonant signal is output through port 13, and finally the target resonant signal is transmitted to the signal analyzer 130 to determine the object attribute information of the force-applying object.

[0064] Based on the above technical solution, a resonator 110 is introduced into the sensor device 100, and the continuous response characteristics and nonlinear dynamic characteristics of the resonator 110 are utilized to use the resonator 110 as RC hardware to construct a neural network in the analog domain, and based on the neural network, the type and / or weight of the object that applies external force to the sensor device 100 can be identified, and based on the nonlinear characteristics of the resonator 110, the nonlinear activation function in the neural network can be realized, so that the nonlinear processing ability of the network can be enhanced, thereby improving the generalization ability and performance of the network. In addition, since the neural network in the analog domain can respond to the changes of the input signal in real time, it helps to increase the response speed of the sensor device 100 and reduce the signal processing delay; the power consumption of the analog circuit is lower than that of the digital circuit, so it helps to reduce the power consumption of the sensor device 100, and the component size of the analog circuit is usually smaller than that of the digital circuit, which helps to increase the integration of the sensor device 100, and the high integration helps to reduce the volume and weight of the sensor device 100, improve the portability of the equipment used by the sensor device 100, and help to reduce the manufacturing cost of the sensor device 100.

[0065] In some possible embodiments, the resonator 110 is specifically used to generate a target resonance signal based on a high-dimensional mapping function, an excitation signal and a target sensing signal, wherein the high-dimensional mapping function is used to indicate a mapping relationship between the target resonance signal and the excitation signal and the target sensing signal.

[0066] It should be understood that the above-mentioned target resonance signal can be understood as a resonance signal generated after high-dimensional mapping of the intermediate resonance signal generated by the movable beam structure 111 based on the second resonance motion.

[0067] In some possible embodiments, taking the resonator 110 based on a capacitive transducer mechanism as an example, the above high-dimensional mapping function can be expressed by the following formula (1):

[0068]

[0069] Among them, m e is used to represent the equivalent mass of the resonator 110, c is used to represent the equivalent damping of the resonator 110 (i.e., the damping coefficient mentioned above), and k m is used to represent the equivalent stiffness of the resonator 110, C0 is used to represent the initial capacitance value of the resonator 110, d is used to represent the initial capacitance spacing of the resonator 110, It is used to represent the acceleration of the movable beam structure 111, is used to indicate the speed of the movable beam structure 111, z is used to indicate the displacement of the movable beam structure 111, V dc Used to indicate DC voltage, V ac Used to indicate AC voltage, V outIt is used to represent the voltage of the output signal of the mechanical sensor 120, that is, the voltage of the target sensing signal.

[0070] It should be understood that the above-mentioned equivalent damping, equivalent mass and equivalent stiffness are physical parameters of the resonator 110, which are assigned based on the preparation process of the resonator 110 and are known and fixed physical quantities. In addition, the above-mentioned initial capacitance spacing and initial capacitance value of the resonator 110 are also physical quantities that can be confirmed in the design stage of the resonator 110, and can also be obtained. Moreover, the above-mentioned DC voltage, AC voltage and voltage of the target sensing signal are all physical quantities that can be obtained, so based on the continuous target sensing signal within a period of time and the above-mentioned formula (1), at least the displacement of the movable beam structure within the period of time can be solved, and further, the velocity and acceleration of the movable beam structure within the period of time can be solved. It can be seen from this that there is a mapping relationship between the displacement, velocity and acceleration and the waveform of the target sensing signal. Further, the target sensing signal is generated based on the external force applied to the mechanical sensor by an object with specified object attribute information, so the displacement, velocity and acceleration can establish a mapping relationship with the object attribute information of the object.

[0071] In addition, the high-dimensional mapping function is essentially a nonlinear dynamic differential equation of the resonator 110, and the nonlinear behavior of the MEMS resonator can simulate the nonlinear activation function in the neural network, such as ReLU, sigmoid, etc. It can be seen that based on the high-dimensional mapping function, the simulation domain neural network algorithm function can be realized.

[0072] Figure 3 1 is a circuit structure diagram of a resonator 110 proposed in an embodiment of the present application. Based on this circuit structure is the hardware implementation of a neural network.

[0073] The resonator 110 can realize real-time signal processing. The resonator 110 includes: a resonant circuit 1110, a transimpedance amplifier 112, a secondary amplifier 113, a bandpass filter 114, an analog-to-digital converter (ADC) 115, a processor 116, and a digital-to-analog converter (DAC) 117.

[0074] Among them, the resonant circuit 1110 includes the above-mentioned movable beam structure 111, the first capacitor plate 01 and the second capacitor plate 02, and the movable beam structure 111 is located between the first capacitor plate 01 and the second capacitor plate 02, wherein the first capacitor plate 01 is used to receive the above-mentioned target sensing signal, and the second capacitor plate 02 is used to output a resonant signal generated based on the excitation signal and the target sensing signal.

[0075] The transimpedance amplifier 112 has an input terminal electrically connected to the second capacitor plate 02 , and an output terminal electrically connected to an input terminal of the secondary amplifier 113 .

[0076] It should be understood that the resonant signal output by the resonant circuit 1110 is transmitted in the form of current, so the above-mentioned transimpedance amplifier 112 is used to convert the resonant signal transmitted in the form of current into a voltage signal according to a certain ratio to facilitate subsequent circuit processing, such as analog-to-digital conversion.

[0077] The output end of the secondary amplifier 113 is electrically connected to the input end of the bandpass filter 114 . The secondary amplifier 113 is used to further amplify the amplitude of the voltage signal from the transimpedance amplifier 112 to provide sufficient signal strength to drive the subsequent bandpass filter 114 .

[0078] The output end of the bandpass filter 114 is electrically connected to the input end of the ADC 115 . The bandpass filter 114 is used to extract a resonant signal of a specific frequency represented by a voltage, while filtering out noise and interference of other frequencies.

[0079] The output end of the ADC 115 is electrically connected to the input end of the processor 116 . The ADC 115 is used to convert the processed analog signal into a digital signal so that the processor 116 can perform digital signal processing.

[0080] An output terminal of the processor 116 is electrically connected to an input terminal of the DAC 117 .

[0081] In some possible embodiments, the processor 116 may be a microcontroller unit (MCU) or a field-programmable gate array (FPGA).

[0082] The output terminal of DAC117 is electrically connected to the first capacitor plate 01 to form a feedback loop for constructing a more complex neural network.

[0083] For some processors with limited computing power, it is impossible to perform real-time analysis on the real-time collected signals. This requires that the resonator 110 is provided with a delay feedback loop to delay the feedback of the resonance signal to the processor based on the signal delay mechanism.

[0084] Figure 4 It is a circuit structure diagram of another resonator 110 proposed in an embodiment of the present application.

[0085] Compared to Figure 3 The resonator 110 shown, Figure 4The resonator 110 shown further includes: a delay feedback loop circuit 118, the delay feedback loop circuit 118 including a phase shifter 1181, a limiter 1182 and a voltage divider 1183;

[0086] Among them, the input end of the phase shifter 1181 is the input end of the delay feedback loop circuit 118, which is electrically connected to the output end of the above-mentioned bandpass filter 114 and the input end of ADC115; the output end of the phase shifter 1182 is electrically connected to the input end of the limiter 1182.

[0087] The output end of the limiter 1182 is electrically connected to the input end of the voltage divider 1183;

[0088] The output end of the voltage divider 1183 is the output end of the delay feedback loop circuit 118 , and is electrically connected to the output end of the first capacitor plate 01 and the DAC 117 .

[0089] Based on the above circuit architecture, it is possible to construct multiple hidden layers of a neural network.

[0090] In some possible embodiments, refer to the above Figure 1 As shown, the signal analyzer 130 may include at least one classifier 131, which is trained based on different target resonance signals and object attribute information corresponding to the different target resonance signals, that is, the classifier 131 is pre-trained.

[0091] In some possible embodiments, the signal analyzer 130 may classify the data after the nonlinear high-dimensional mapping of the resonator 110, and the data after the high-dimensional mapping may be at least one of the displacement, velocity and acceleration of the movable beam structure 111 calculated based on the above formula (1) mentioned in the above embodiment. The classification processing algorithm includes but is not limited to linear regression, ridge regression, cable regression, etc.

[0092] Based on the above technical solution, by introducing the resonator 110 and the signal analyzer 130 into the sensor device 100, analog domain signal classification calculation is realized. The implementation scheme is relatively simple and does not involve the construction of cloud-based digital domain neural networks and massive data transmission. It can complete the nonlinear mapping of linear inseparable time series signals and utilize the high-dimensional mapping function of the nonlinear characteristics of the resonator 110 to achieve efficient linearly separable high-dimensional mapping of the target sensor signal, thereby improving the accuracy and calculation rate of object classification and recognition.

[0093] In addition, the embodiment of the present application also proposes a device structure design of the above-mentioned sensor device 100.

[0094] Figure 5 It is a schematic diagram of the structure of a sensor device 100 proposed in an embodiment of the present application.

[0095] refer to Figure 5 As shown, the above-mentioned sensor device 100 may include:

[0096] Base layer 210;

[0097] The circuit layer 220 is used to set the mechanical sensor 120 and the resonator 110;

[0098] A buried oxide BOX layer 230 , the BOX layer 230 is located between the base layer 210 and the circuit layer 220 , and the BOX layer 230 includes a hole 231 , and the hole 231 is opposite to the movable beam structure 111 of the resonator 110 ;

[0099] The packaging layer 240 is located on the side of the circuit layer 220 away from the base layer 210 . The packaging layer 240 includes a first vacuum cavity 2401 and a second vacuum cavity 2402 . The first vacuum cavity 2401 covers the movable beam structure 111 , and the second vacuum cavity 2402 covers the force sensitive element 121 of the mechanical sensor 120 .

[0100] In some possible embodiments, the base layer 210 may be made of silicon (Si) material.

[0101] In some possible embodiments, the lower surface of the base layer 210 is a receiving surface for the force of external objects.

[0102] In some possible embodiments, the thickness of each material layer is in the micrometer level. For example, the thickness of the base layer 210 is 200 μm, and the overall thickness of the packaging layer 240 and the circuit layer 220 is 200 μm.

[0103] In some possible embodiments, all the overall circuit elements of the sensor device 100 may be deployed in the circuit layer 220 .

[0104] It should be understood that the application of the first vacuum chamber 2401 is to prevent the resonance frequency of the resonator 110 from being affected by the number density of internal gas molecules, resulting in a decrease in the accuracy and stability of the resonance frequency. The application of the second vacuum chamber 2402 is to further improve the detection accuracy of the mechanical sensor 120, because when the mechanical sensor 120 is in a vacuum state, it can be considered that the output signal of the mechanical sensor 120 is only related to the input force.

[0105] In addition, the movable beam structure 111 of the resonator 110 is aligned with the hole 231 of the BOX layer in order to suspend the movable beam structure 111 in the air to avoid scratches with other components that may affect the resonant frequency of the resonator 110 .

[0106] In some possible embodiments, the encapsulation layer 240 may be prepared based on through glass via (TGV) technology to perform good sealing on the circuit layer 220 of the sensor device 100 .

[0107] In some possible embodiments, the circuit layer 220 can be packaged by silicon-silicon (Si-Si) bonding or silicon-glass (Si-glass) bonding. Based on this, it is possible to achieve multi-liquid protection for the sensor device 100, while achieving high sensitivity and anti-interference characteristics. In addition, the encapsulation layer 240 prepared based on the Si-Si bonding method or the Si-glass bonding method has strong process compatibility and can be compatible with a variety of semiconductor preparation processes, such as photolithography, etching, deposition, etc., which facilitates the integration of the preparation operation of the encapsulation layer 240 into the production line.

[0108] In some possible embodiments, the packaging layer 240 is a wafer-level packaging structure, and the packaging layer 240 includes: a glass cap 241, a plurality of through holes 242, a plurality of leads 243 and a plurality of solder balls 244, wherein the plurality of through holes 242 penetrate the glass cap 241, the first end of the through hole 242 is connected to the side of the mechanical sensor 120 or the resonator 110 away from the base layer 210, the second end of the through hole 242 is connected to the solder ball 244, and the through hole 242 is used to set the lead 243 so that the inside and the outside of the sensor device 100 have good conductivity.

[0109] In some possible embodiments, the radius of the solder ball 244 may be 100 μm.

[0110] It should be noted that the aforementioned embodiments provide geometric parameters of each material layer and electronic devices, but the geometric parameters of each material layer and electronic devices are not limited to the numerical values ​​in the above examples. In different preparation processes and different application scenarios of the sensor device 100, the geometric parameters of the above material layers and electronic devices can be adaptively adjusted.

[0111] Based on the above technical solution, the detection accuracy of the sensor device 100 can be effectively guaranteed, and it has good sealing, waterproofness and corrosion resistance, and enhances the durability and reliability of the sensor device, making the sensor device 100 suitable for a wider range of application scenarios.

[0112] In some possible embodiments, reference Figure 5 As shown, the above-mentioned base layer 210 may also include a back cavity 211 , and the vertical projection of the above-mentioned force sensitive element 121 to the projection plane is located within the vertical projection of the back cavity 211 to the projection plane, and the projection plane is parallel to the lower surface of the above-mentioned base layer 210 .

[0113] It should be understood that, due to the existence of the back cavity 211 , the force applied by the object to the lower surface of the base layer 210 can be more easily transmitted to the force-sensitive element 121 , so that the force applied by the object can be responded to more quickly and accurately.

[0114] In some possible embodiments, the force-sensitive element 121 may include:

[0115] A pressure-sensitive film 1211 , which is disposed on a side of the BOX layer 230 away from the base layer 210 ;

[0116] The varistor 1212 is disposed on a side of the pressure-sensitive film 1211 away from the base layer 210 .

[0117] Based on the above technical solution, the response speed and accuracy of the mechanical sensor can be further improved.

[0118] For ease of understanding, the object recognition mechanism of the sensor device 100 is described in detail below by taking a specific application scenario of the sensor device 100 as an example.

[0119] Figure 6 It is a schematic diagram of an application of a smart fishing float 600 proposed in an embodiment of the present application.

[0120] In this application scenario, the above-mentioned sensing device 100 is applied to the intelligent fishing float 600, so that the force of the fish when biting the hook is transmitted to the sensing device 100 through the fish hook. The sensing device 100 outputs a corresponding resonant signal based on the above-mentioned functional principles of the mechanical sensor 120 and the resonator 110, and is acquired by the signal analyzer 130, and the corresponding physical property information of the fish, such as the type and / or weight of the fish, is identified based on the classifier 131.

[0121] During the design stage of the sensor device 100, it has been confirmed that the application scenario of the sensor device 100 is an intelligent fishing float, and the object to be identified is fish, so parameters such as the equivalent damping, equivalent mass, and equivalent stiffness of the resonator 110 are set based on the application scenario, and the signal threshold of the control circuit 102 in the sensor device 100 can be set.

[0122] During the manufacturing stage of the sensor device 100, a data set of multiple samples can be constructed to perform optimal weight training and test calibration on the classifier 131. When the recognition accuracy rate is greater than 95%, it is determined that the classifier 131 has been trained.

[0123] In actual applications, the sensor device 100 is set at the fishing float position and floats on the water surface together with the fishing float. Due to the existence of the control circuit 102 in the sensor device 100, the amplitude of the target sensor signal formed by the force f exerted by the water surface waves on the sensor device 100 cannot reach the signal threshold, so the mechanical sensor 120 will not transmit the target sensor signal to the resonator 110.

[0124] In the initial working state of the sensor device 100 , the movable beam structure of the resonator 110 continues to perform the first resonant motion under the driving of the excitation signal of the AC power supply 140 .

[0125] When a fish of unknown species and size bites the hook, a large pulling force F is applied to the float. The amplitude of the target sensing signal generated by the mechanical sensor 120 based on the pulling force is greater than the signal threshold, and the target sensing signal is transmitted to the resonator 110. The movable beam structure 111 of the resonator 110 is converted from the first resonant motion to the second resonant motion based on the influence of the target sensing signal, and the target resonant signal is output accordingly. The target resonant signal is high-dimensionally mapped based on the high-dimensional mapping function to determine the displacement, velocity and acceleration of the movable beam structure 111 during the above-mentioned second resonant motion.

[0126] Since the classifier 131 of the signal analyzer is pre-trained based on the data set, after the classifier 131 obtains the target resonance signal, it can output the corresponding result, which is the object attribute information of the fish bitten by the hook. For example, in this scenario, the object attribute information can be: 1 kg carp.

[0127] In some possible embodiments, the above-mentioned smart fishing float can also be used in conjunction with other terminal devices (for example, smart phones), wherein the other terminal devices are equipped with client applications. After determining the above-mentioned object attribute information, the sensing device 100 can send the object attribute information to the terminal device, so that the object attribute information can be presented to the user through the client application, thereby giving the user effective information to assist the user in handling the fishing situation and increase the fish catching rate.

[0128] In addition, an embodiment of the present application further provides an electronic device, which may include any one of the sensor devices 100 provided in the embodiment of the present application.

[0129] For example, the electronic device may be the above-mentioned intelligent fishing float 600. Of course, the sensor device 100 proposed in the embodiment of the present application may also be applied to other scenarios, which is not limited in the embodiment of the present application.

[0130] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0131] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0132] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A sensing device, characterized in that: include: A resonator, comprising a movable beam structure, wherein an initial motion state of the movable beam structure is a first resonant motion; A mechanical sensor, configured to generate a target sensing signal in response to a force from an object, wherein the target sensing signal is used to convert the movable beam structure from the first resonant motion to a second resonant motion, so that the resonator outputs a target resonant signal, wherein the target resonant signal is used to indicate a displacement of the movable beam structure; A signal analyzer is used to determine object attribute information according to the target resonance signal, wherein the object attribute information includes the type and / or weight of the object.

2. The sensor device according to claim 1, characterized in that The sensor device also includes: An alternating current power supply, wherein the alternating current power supply is used to input an excitation signal to the resonator, and the excitation signal is used to drive the movable beam structure to perform the first resonant motion.

3. The sensor device according to claim 2, characterized in that The resonant frequency of the first resonant motion is related to the response speed of the resonator.

4. The sensor device according to claim 2 or 3, characterized in that: The resonator is specifically used for: The target resonance signal is generated based on a high-dimensional mapping function, the excitation signal and the target sensing signal, wherein the high-dimensional mapping function is used to indicate a mapping relationship between the target resonance signal and the excitation signal and the target sensing signal.

5. The sensor device according to any one of claims 1 to 4, characterized in that: The signal analyzer includes at least one classifier, and the classifier is trained based on different target resonance signals and the object attribute information corresponding to different target resonance signals.

6. The sensor device according to any one of claims 1 to 5, characterized in that: The sensing device comprises: Basal layer; A circuit layer, used for setting the mechanical sensor and the resonator; A buried oxide BOX layer is located between the base layer and the circuit layer, and the BOX layer includes a hole, and the hole is opposite to the movable beam structure of the resonator; The packaging layer is located on the side of the circuit layer away from the base layer, and the packaging layer includes a first vacuum cavity and a second vacuum cavity, the first vacuum cavity covers the movable beam structure, and the second vacuum cavity covers the force-sensitive element of the mechanical sensor.

7. The sensor device according to claim 6, characterized in that The base layer includes a back cavity, a vertical projection of the force sensitive element onto a projection plane is located within a vertical projection of the back cavity onto the projection plane, and the projection plane is parallel to a lower surface of the base layer.

8. The sensor device according to claim 6 or 7, characterized in that: The force sensitive element comprises: A pressure-sensitive film, the pressure-sensitive film being arranged on a side of the BOX layer away from the base layer; A varistor is disposed on a side of the pressure-sensitive film away from the base layer.

9. The sensor device according to any one of claims 6 to 8, characterized in that The packaging layer is a wafer-level packaging structure, and the packaging layer also includes: a glass cap, a plurality of through holes, a plurality of leads and a plurality of solder balls, wherein the plurality of through holes penetrate the glass cap, a first end of the through hole is connected to a side of the mechanical sensor or the resonator away from the base layer, a second end of the through hole is connected to the solder ball, and the through hole is used to set the lead.

10. The sensor device according to any one of claims 1 to 9, characterized in that: The mechanical sensor is a micro-electromechanical system (MEMS) pressure sensor, and the resonator is a MEMS resonator.

11. An electronic device, characterized in that: Comprising the sensing device as claimed in any one of claims 1 to 10.

Images