Pressure recognition system, pressure recognition method, and computer-readable storage medium

Abstract

The application discloses a pressure recognition system, a pressure recognition method and a computer readable storage medium. The pressure recognition system comprises a control and peripheral circuit, a pressure sensing circuit, an analog front-end circuit and an analog calculation module connected with the control and peripheral circuit respectively; the control and peripheral circuit is used for controlling the circuit working logic and timing of the pressure sensing circuit, the analog front-end circuit and the analog calculation module; the pressure sensing circuit is used for acquiring and storing the pressure signal of the outside world; the analog front-end circuit is used for converting the pressure signal from a resistance value into an analog voltage value; and the analog calculation module is used for processing the analog voltage value to obtain a pressure signal recognition result. The pressure recognition system of the application embodiment improves the information transmission efficiency and energy efficiency, reduces the power consumption, saves the cost and energy consumption, reduces the signal transmission delay and avoids the quantization noise introduced in the analog-digital conversion process since the system of the application embodiment does not need to use an analog-digital converter and a register.

Description

Technical Field

[0001] This application relates to the field of Internet of Things (IoT) technology, specifically to a pressure recognition system, a pressure recognition method, and a computer-readable storage medium. Background Technology

[0002] Pressure detection plays a crucial role in the Internet of Things (IoT) ecosystem, finding widespread application in smart homes, wearable devices, smart healthcare, and bionic prosthetics. The pressure signal from a pressure sensor is volatile; when external pressure is removed, the sensor returns to its initial state, and the previous pressure signal is lost. Therefore, it's necessary to store the pressure sensor signal for subsequent signal processing. Traditional pressure recognition systems typically use an analog-to-digital converter (ADC) to convert the analog signal from the pressure sensor into a digital signal, then store the digital signal in a register to buffer the sensing information. However, this process introduces additional costs, high energy consumption, significant signal transmission delays, and substantial quantization noise during the ADC conversion. Therefore, traditional pressure recognition systems are increasingly unable to meet the growing demands of the IoT for pressure sensing systems. Summary of the Invention

[0003] The purpose of this application is to provide a pressure recognition system, a pressure recognition method, and a computer-readable storage medium to overcome the technical shortcomings of existing technologies, such as high cost, high energy consumption, large signal transmission delay, and large quantization noise introduced during analog-to-digital conversion. To provide a basic understanding of some aspects of the disclosed embodiments, a brief summary is given below. This summary is not intended as a general description, nor is it intended to identify key / important components or describe the scope of protection of these embodiments. Its sole purpose is to present some concepts in a simple form as a prelude to the detailed description that follows.

[0004] According to one aspect of the embodiments of this application, a pressure recognition system is provided, including control and peripheral circuits, and a pressure sensing circuit, an analog front-end circuit, and an analog computing module respectively connected to the control and peripheral circuits;

[0005] The control and peripheral circuits are used to control the circuit operation logic and timing of the pressure sensing circuit, the analog front-end circuit, and the analog calculation module;

[0006] The pressure sensing circuit is used to acquire and store external pressure signals;

[0007] The analog front-end circuit is used to convert the pressure signal from a resistance value to an analog voltage value;

[0008] The simulation calculation module is used to process the simulation voltage value to obtain the pressure signal identification result.

[0009] In some embodiments of this application, the control and peripheral circuitry includes a controller and a power management module, a clock source, and a clock tree, all connected to the controller. The clock source is connected to the clock tree, and both the clock source and the clock tree are connected to the power management module. The controller is used to control the circuit operation logic and timing of the pressure sensing circuit, the analog front-end circuit, and the analog computing module.

[0010] The power management module is used to manage the power supply to the pressure sensing circuit, the analog front-end circuit, and the analog computing module; the clock source is used to provide clock pulse signals; and the clock tree is used to manage the circuit timing of the clock pulse signals.

[0011] In some embodiments of this application, the pressure sensing circuit includes word line and bit line driving circuits, and a plurality of pressure sensing units respectively connected to the word line and bit line driving circuits; the word line and bit line driving circuits are used to select one or more of the pressure sensing units under the control of the control and peripheral circuits, so as to detect and store external pressure signals through the selected pressure sensing units.

[0012] In some embodiments of this application, the pressure sensing unit includes a pressure sensor unit and an RRAM unit connected in series. The pressure sensor unit is used to detect external pressure signals, and the RRAM unit is used to store pressure signals from the pressure sensor unit.

[0013] In some embodiments of this application, the analog front-end circuit includes an interconnected resistor-to-voltage circuit and a programmable gain amplifier;

[0014] The resistor-to-voltage circuit is used to convert the pressure signal from a resistance value to an analog voltage value, and the programmable gain amplifier is used to amplify the analog voltage value output by the resistor-to-voltage circuit and output the amplified analog voltage value.

[0015] In some embodiments of this application, the analog computing module includes an analog computing circuit; the analog computing circuit includes at least one layer of SRAM analog computing array, the SRAM analog computing array includes multiple SRAM analog computing units, the SRAM analog computing unit includes an SRAM cell, a capacitor, a transistor, a first switch, a second switch, and a third switch, the SRAM cell stores the weights of the analog computing, the SRAM cell is connected to the gate of the transistor; the upper plate of the capacitor is connected to the second switch, and the lower plate of the capacitor is grounded; the drain of the transistor is connected to the second switch, and the source of the transistor is grounded; the first terminal of the first switch is used to connect to the input voltage, and the second terminal of the first switch is connected to the second switch, the third switch, and the capacitor.

[0016] In some embodiments of this application, the analog computing module further includes a signal post-processing module connected to the analog computing circuit. The signal post-processing module is used to post-process the output of the analog computing circuit to obtain a pressure signal identification result. The post-processing includes any one or more combinations of quantization, threshold judgment, logical judgment, and nonlinear activation.

[0017] According to another aspect of the embodiments of this application, a pressure recognition method is provided, implemented by any of the pressure recognition systems described above; the pressure recognition method includes:

[0018] The control and peripheral circuits control the circuit operation logic and timing of the pressure sensing circuit, the analog front-end circuit, and the analog calculation module;

[0019] The pressure sensing circuit acquires and stores external pressure signals;

[0020] The analog front-end circuit converts the pressure signal from a resistance value to an analog voltage value;

[0021] The simulation calculation module processes the simulated voltage value to obtain the pressure signal identification result.

[0022] In some embodiments of this application, the pressure sensing unit includes a pressure sensor unit and an RRAM unit connected in series; the pressure sensing circuit acquires and stores external pressure signals, including:

[0023] When an external pressure signal is detected, the resistance value of the pressure sensor unit changes, the voltage drop across the RRAM unit increases, and the resistance of the RRAM unit changes, thus storing the pressure signal.

[0024] In some embodiments of this application, the analog computing module includes interconnected analog computing circuits and a signal post-processing module; the analog computing circuit includes at least one layer of SRAM analog computing array; the analog computing module processes the analog voltage value to obtain a pressure signal identification result, including:

[0025] Each pixel of the voltage distribution map is arranged into a voltage vector according to a preset order and input into the first layer of SRAM analog computing array for processing, and a voltage vector is output.

[0026] Each intermediate layer array receives a voltage vector from the previous layer array, and the output voltage vector is used as the input to the next layer array; the intermediate layer array is an SRAM analog computing array between the first layer and the last layer.

[0027] The voltage vector of the last layer SRAM analog computing array is processed by the signal post-processing module to obtain the pressure signal recognition result. The post-processing includes any one or more combinations of quantization, threshold judgment, logical judgment and nonlinear activation.

[0028] In some embodiments of this application, the SRAM analog computing array includes multiple SRAM analog computing units, each of which includes an SRAM cell, a capacitor, a transistor, a first switch, a second switch, and a third switch.

[0029] The SRAM analog computing array processes the voltage vector as follows:

[0030] When a voltage vector from the upper-level array is input to the SRAM analog computing unit, the first switch is closed, the second switch and the third switch are opened, and the capacitor is charged.

[0031] After charging is complete, control to close the second switch and control to open the first switch and the third switch;

[0032] Control the closing of the third switch, control the opening of the first and second switches, and output a voltage vector;

[0033] Wherein, the voltage of the voltage vector is equal to the total charge divided by the total capacitance; the total charge is the sum of the charges of all the capacitors in the SRAM simulation array, and the total capacitance is the sum of the capacitances of all the capacitors in the SRAM simulation array.

[0034] According to another aspect of the embodiments of this application, a computer-readable storage medium is provided having a computer program stored thereon, the computer program being executed by a processor to implement the pressure identification method described in any of the preceding claims.

[0035] One aspect of the technical solution provided in this application embodiment may include the following beneficial effects:

[0036] The pressure recognition system provided in this application improves information transmission efficiency and energy efficiency, reduces power consumption, and saves cost and energy consumption because the system does not require the use of analog-to-digital converters and registers. It also reduces signal transmission delay, avoids quantization noise introduced during analog-to-digital conversion, and overcomes the technical defects of existing technologies, such as high cost, high energy consumption, large signal transmission delay, and large quantization noise introduced during analog-to-digital conversion.

[0037] Other features and advantages of this application will be set forth in the following description and will be apparent in part from the description, or some features and advantages may be inferred from the description or determined without doubt, or may be learned by practicing embodiments of this application. Attached Figure Description

[0038] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0039] Figure 1 A block diagram of a pressure recognition system according to an embodiment of this application is shown.

[0040] Figure 2 A block diagram of a specific example pressure recognition system is shown.

[0041] Figure 3 A block diagram of the control and peripheral circuitry in an embodiment of this application is shown.

[0042] Figure 4 An equivalent circuit diagram of the pressure sensing unit in an embodiment of this application is shown.

[0043] Figure 5 An equivalent circuit diagram of the pressure sensing circuit in an embodiment of this application is shown.

[0044] Figure 6 A schematic diagram illustrating the process of converting resistance values ​​into analog voltages in an embodiment of this application is shown.

[0045] Figure 7 A circuit diagram of the analog computing circuit in an embodiment of this application is shown.

[0046] Figure 8 A schematic diagram of the structure of the SRAM analog computing array in an embodiment of this application is shown.

[0047] Figure 9 A flowchart of a pressure recognition method according to an embodiment of this application is shown.

[0048] Figure 10 A schematic diagram of a computer-readable storage medium according to another embodiment of this application is shown.

[0049] The purpose, functional features, and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0050] To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application. All other embodiments obtained by those skilled in the art based on the embodiments in this application without inventive effort are within the scope of protection of this application.

[0051] It will be understood by those skilled in the art that, unless otherwise defined, all terms used herein (including technical and scientific terms) have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains. It should also be understood that terms such as those defined in general dictionaries should be understood to have a meaning consistent with their meaning in the context of the prior art, and should not be interpreted in an idealized or overly formal sense unless specifically defined as herein.

[0052] refer to Figure 1 As shown, one embodiment of this application provides a pressure recognition system, including control and peripheral circuitry, and a pressure sensing circuit, an analog front-end circuit, and an analog computing module respectively connected to the control and peripheral circuitry; the control and peripheral circuitry is used to control the circuit operation logic and timing of the pressure sensing circuit, the analog front-end circuit, and the analog computing module; the pressure sensing circuitry is used to acquire and store external pressure signals; the analog front-end circuit is used to convert the pressure signals from resistance values ​​to analog voltage values; the analog computing module is used to process the analog voltage values ​​to obtain pressure signal recognition results.

[0053] In one embodiment, the control and peripheral circuitry includes a controller and a power management module, a clock source, and a clock tree, all connected to the controller. The clock source is connected to the clock tree, and both the clock source and the clock tree are connected to the power management module. The controller controls the circuit operation logic and timing of the pressure sensing circuit, the analog front-end circuit, and the analog computing module. The power management module manages the power supply to the pressure sensing circuit, the analog front-end circuit, and the analog computing module. The clock source provides clock pulse signals, and the clock tree manages the circuit timing of the clock pulse signals.

[0054] In one embodiment, the pressure sensing circuit includes word line and bit line driving circuits, and a plurality of pressure sensing units respectively connected to the word line and bit line driving circuits; the word line and bit line driving circuits are used to select one or more of the pressure sensing units under the control of the control and peripheral circuits, so as to detect and store external pressure signals through the selected pressure sensing units.

[0055] In one embodiment, the pressure sensing unit includes a pressure sensor unit and an RRAM unit connected in series. The pressure sensor unit is used to detect external pressure signals, and the RRAM unit is used to store the pressure signals from the pressure sensor unit.

[0056] In one embodiment, the analog front-end circuit includes a resistor-to-voltage circuit and a programmable gain amplifier connected to each other; the resistor-to-voltage circuit is used to convert the pressure signal from a resistance value to an analog voltage value, and the programmable gain amplifier is used to amplify the analog voltage value output by the resistor-to-voltage circuit and output the amplified analog voltage value.

[0057] In one embodiment, the analog computing module includes interconnected analog computing circuitry and a signal post-processing module, with the analog computing circuitry connected to a programmable gain amplifier. The analog computing module may further include word line and bit line driving circuitry connected to the analog computing circuitry. The signal post-processing module is used to post-process the output of the analog computing circuitry to obtain a pressure signal recognition result. The post-processing includes a combination of one or more of quantization, threshold judgment, logical judgment, and nonlinear activation. Exemplarily, the analog computing circuit includes at least one layer of SRAM analog computing array, which includes multiple SRAM analog computing units. Each SRAM analog computing unit includes an SRAM cell, a capacitor, a transistor, a first switch, a second switch, and a third switch. The SRAM cell stores the weights of the analog computing calculation and is connected to the gate of the transistor. The upper plate of the capacitor is connected to the second switch, and the lower plate of the capacitor is grounded. The drain of the transistor is connected to the second switch, and the source of the transistor is grounded. A first terminal of the first switch is used to connect to an input voltage, and a second terminal of the first switch is connected to the second switch, the third switch, and the capacitor.

[0058] refer to Figure 2 As shown, a specific example of the pressure recognition system provided in this application includes a pressure sensing circuit, an analog front-end circuit, an analog computing module, and control and peripheral circuits. The pressure sensing circuit includes multiple pressure sensing units, and optionally, may also include word line and bit line driving circuits. The pressure sensing circuit can also be referred to as a pressure sensing integrated group.

[0059] The analog front-end circuitry includes a resistor-to-voltage converter and, optionally, a programmable gain amplifier. The analog computing module includes interconnected analog computing circuitry and a signal post-processing module, and optionally, word line and bit line driver circuitry connected to the analog computing circuitry.

[0060] refer to Figure 3As shown, the control and peripheral circuitry includes a power management module, a clock source, a clock tree, and a controller. The power management module, clock source, and clock tree are all connected to the controller. The clock source is connected to the clock tree, and both the clock source and clock tree are connected to the power management module. The controller is used to control the power management module, clock source, and clock tree. The power management module manages the power supply to each module. The clock source provides a stable and level-matched square wave clock pulse signal. The clock signal output by the clock source needs to be managed by the clock tree in the clock domain to ensure good circuit timing.

[0061] The controller is also used to control the circuit operation logic and timing of the pressure sensing circuit, analog front-end circuit, and analog computing module.

[0062] The pressure sensing circuit consists of M×N pressure sensing units. Each pressure sensing unit is composed of a pressure sensor unit and an RRAM unit connected in series. The equivalent circuit of the pressure sensing unit is two resistors connected in series, as shown below. Figure 4 As shown. The equivalent circuit diagram of the pressure sensing circuit is as follows. Figure 5 As shown.

[0063] The word line and bit line drive circuits in the control and peripheral circuitry control the pressure sensing and storage circuitry to detect and store the pressure signal through a pressure sensing and storage unit in the pressure sensing and storage circuitry via address gating.

[0064] When the pressure sensor unit in the pressure sensing memory unit detects an external pressure signal, the resistance of the pressure sensor decreases to varying degrees depending on the pressure magnitude. This causes the voltage drop across the RRAM to increase, resulting in a change in the resistance of the RRAM and thus storing the pressure signal. Then, the resistance value of each pressure sensing memory unit or RRAM unit is sent to the resistor-to-voltage conversion circuit in the analog front-end circuit to be converted into an analog voltage value. Next, the control and peripheral circuits continue to control the word line and bit line drive circuits to select each pressure sensing memory unit in the pressure sensing memory circuit in row and column order, repeating the above "resistance value to analog voltage" process until the resistance values ​​of all pressure sensing memory units are converted into corresponding analog voltage signals, resulting in an analog voltage value distribution diagram, such as... Figure 6As shown, the voltage distribution map reflects the spatial distribution of pressure. Each pixel in the voltage distribution map reflects the pressure response of the corresponding pressure sensor, and the depth of each pixel reflects the magnitude of the pressure. Therefore, different external pressure distributions can be converted into different analog voltage distribution patterns. Optionally, the analog voltage is amplified by a programmable gain amplifier, and then the analog voltage distribution map is input into an analog computing circuit for simulation calculation. The result calculated by the analog computing circuit is post-processed by a signal post-processing module to obtain the final classification or recognition result, enabling intelligent detection such as gesture recognition and health detection. The signal post-processing module is used to post-process the output result of the analog computing circuit to obtain the pressure signal recognition result. The post-processing includes any one or more combinations of quantization, threshold judgment, logical judgment, and nonlinear activation.

[0065] Optionally, the post-processing method can be quantization, threshold judgment, logical judgment, nonlinear activation, or various combinations thereof.

[0066] In addition, the system's control and peripheral circuits consist of power management, clock source, clock tree, and controller, providing clock signals, power supply voltage, and control functions to various modules of the system.

[0067] This embodiment of the application stores pressure sensing signals by connecting a pressure sensor unit and an RRAM unit in series to form a pressure sensing memory unit. When the pressure sensor unit detects an external pressure signal, its resistance changes, causing a change in the voltage divider on the RRAM resistive random access memory (RRAM). This results in a resistance change in the RRAM unit, thus storing the volatile sensing signal from the pressure sensor unit into the RRAM unit. This process can store the sensing signal without going through an analog-to-digital converter (ADC) and registers, thereby improving information transmission efficiency and energy efficiency, reducing power consumption, and saving costs and energy consumption because the system in this embodiment does not require an ADC and registers. It also reduces signal transmission delay and avoids quantization noise introduced by the analog-to-digital conversion process.

[0068] This application embodiment stores the pressure signal received in the external space through a pressure sensing and storage unit group, and obtains a voltage value distribution map reflecting the spatial pressure distribution through a resistor-to-voltage circuit. Through neural network training and then configuring chip weights, it can be used for more advanced intelligent detection such as posture recognition and health detection.

[0069] Since the embodiments of this application store analog signals, analog calculations can be performed. In some non-high-precision application scenarios, it has the advantages of low power consumption and space saving.

[0070] In this embodiment, the pressure sensing unit is composed of a pressure sensor unit and an RRAM unit connected in series, and the pressure sensor unit is a resistive pressure sensor. In some embodiments, the pressure sensor is a MEMS pressure sensor based on micro / nano fabrication technology, which can be integrated with RRAM and CMOS circuits. RRAM is a non-volatile memory that can be embedded in the metal layer of CMOS back-end process, thereby achieving high integration with logic circuits, and the fabrication process is highly compatible with CMOS process, which is conducive to mass production; at the same time, RRAM has excellent storage performance such as simple structure, high switching ratio, low operating voltage, low power consumption, and high read and write speed. Under the action of an external electric field, it can achieve reversible conversion between high resistance state and low resistance state, thereby realizing data storage. RRAM has a sandwich structure, consisting of a top electrode, a resistive switching layer, and a bottom electrode, with the resistive switching layer sandwiched between the bottom electrode and the top electrode; the resistive switching layer can be TaO. x HfO2, Al2O3, TiO x Any material. The bottom and top electrodes can be made of any of the following materials: TiN, WN, TaN, platinum, tungsten, and palladium.

[0071] In some embodiments of this application, the analog computing circuit used is as follows: Figure 7 As shown, the analog computing circuit consists of at least one layer of SRAM analog computing array, which in turn consists of multiple SRAM analog computing units. Each SRAM analog computing unit consists of an SRAM cell, a capacitor, a transistor, and three switches, namely, a first switch S1, a second switch S2, and a third switch S3. The SRAM cell stores the weights of the analog computing and is connected to the gate of the transistor. The weights of each SRAM cell are written through word line and bit line driving circuits. The upper plate of the capacitor is connected to the second switch S2, and the lower plate is grounded. The drain of the transistor is connected to the second switch S2, and the source is grounded. One end of the first switch S1 is connected to the input voltage, and the other end is connected to the second switch S2, the third switch S3, and the capacitor.

[0072] SRAM analog computing arrays can execute fully connected neural network algorithms, arranging each pixel of the voltage distribution image into a voltage vector V in a specific order. in1 V in2 ...V inmThe input layer of the first layer is fed into one end of the first-layer SRAM analog computing array, and the selection of the SRAM analog computing units is controlled by word line and bit line driver circuits. When the voltage vector is input to the SRAM analog computing unit, the first switch S1 is closed, and the second and third switches S2 and S3 are opened to charge the capacitor. After charging, the voltage value of the upper plate of the capacitor is almost equal to the input voltage value. Then, the second switch S2 is closed, and the first and third switches S1 and S3 are opened. The weights stored in the SRAM unit control the switching of the transistor. If the transistor is on, the corresponding capacitor will discharge, so the charge on the capacitor plate is 0. If the transistor is off, the charge on the capacitor remains unchanged. Finally, the third switch S3 is closed, and the first and second switches S1 and S2 are opened. At this time, the capacitors in each column are connected in parallel, and the total charge is equal to the sum of the charges of the capacitors in each SRAM analog computing unit. The total capacitance is equal to the sum of all the capacitors. The output voltage of the column is equal to the total charge divided by the total capacitance. This process completes the multiplication and accumulation of the analog computing. After each line of calculation, a voltage vector is obtained. This voltage vector becomes the input to the second-layer SRAM analog computing array. The same calculation method is used to obtain the voltage vector of the second-layer SRAM analog computing array. This process is repeated until the voltage vector of the last layer SRAM analog computing array is calculated. The voltage vector is then input to the signal post-processing module for post-processing to obtain the final detection result.

[0073] There are several ways to implement a resistor-to-voltage circuit, one of which is as follows: Figure 8 As shown, the resistor of the pressure sensing unit or RRAM unit ( Figure 8 The diagram only illustrates the conversion of the RRAM cell's resistance value into a voltage value, and the reference resistor is connected in series between the input and output of the operational amplifier.

[0074] Among them, V out It is the output voltage, V ref R1 is the reference voltage, and R2 is a fixed resistor.

[0075] The pressure recognition system provided in this application improves information transmission efficiency and energy efficiency, reduces power consumption, and saves cost and energy consumption because the system does not require the use of analog-to-digital converters and registers. It also reduces signal transmission delay, avoids quantization noise introduced during analog-to-digital conversion, and overcomes the technical defects of existing technologies, such as high cost, high energy consumption, large signal transmission delay, and large quantization noise introduced during analog-to-digital conversion.

[0076] Another embodiment of this application provides a pressure recognition method, implemented by the pressure recognition system of any of the above embodiments; the pressure recognition method includes:

[0077] The control and peripheral circuits control the operating logic and timing of the pressure sensing circuit, the analog front-end circuit, and the analog calculation circuit.

[0078] The pressure sensing circuit acquires and stores external pressure signals;

[0079] The analog front-end circuit converts the pressure signal from a resistance value to an analog voltage value;

[0080] The simulation calculation module processes the simulated voltage value to obtain the pressure signal identification result.

[0081] In some embodiments of this application, the pressure sensing unit includes a pressure sensor unit and an RRAM unit connected in series; the pressure sensing circuit acquires and stores external pressure signals, including:

[0082] When an external pressure signal is detected, the resistance value of the pressure sensor unit changes, the voltage drop across the RRAM unit increases, and the resistance of the RRAM unit changes, thus storing the pressure signal.

[0083] In one embodiment, the analog computing circuit includes at least one layer of SRAM analog computing array; the analog computing module processes the analog voltage value to obtain a pressure signal identification result, including:

[0084] Each pixel of the voltage distribution map is arranged into a voltage vector according to a preset order and input into the first layer of SRAM analog computing array for processing, and a voltage vector is output.

[0085] Each intermediate layer array receives a voltage vector from the previous layer array, and the output voltage vector is used as the input to the next layer array; the intermediate layer array is an SRAM analog computing array between the first layer and the last layer.

[0086] The voltage vector of the last layer SRAM analog computing array is processed by the signal post-processing module to obtain the pressure signal recognition result. The post-processing includes any one or more combinations of quantization, threshold judgment, logical judgment and nonlinear activation.

[0087] In one embodiment, the SRAM analog computing array includes a plurality of SRAM analog computing units, each of which includes an SRAM cell, a capacitor, a transistor, a first switch, a second switch, and a third switch.

[0088] The SRAM analog computing array processes the voltage vector as follows:

[0089] When a voltage vector from the upper-level array is input to the SRAM analog computing unit, the first switch is closed, the second switch and the third switch are opened, and the capacitor is charged.

[0090] After charging is complete, control to close the second switch and control to open the first switch and the third switch;

[0091] Control the closing of the third switch, control the opening of the first and second switches, and output a voltage vector;

[0092] Wherein, the voltage of the voltage vector is equal to the total charge divided by the total capacitance; the total charge is the sum of the charges of all the capacitors in the SRAM simulation array, and the total capacitance is the sum of the capacitances of all the capacitors in the SRAM simulation array.

[0093] In a specific example of the stress identification method in this application, such as Figure 9 As shown, firstly, the pressure sensor units in the pressure storage unit detect external pressure signals, and the resistance value of each sensor unit changes. Then, the voltage across the RRAM unit increases, causing a change in resistance in the RRAM, thus storing the sensor signals. The resistance value of the pressure storage unit or RRAM unit is converted into a voltage value, resulting in a voltage distribution map. This voltage distribution map is then input into the analog calculation module for algorithm recognition, calculating the action posture or health detection recognition results. Action posture recognition can be gesture recognition, motion recognition, gait recognition, etc. In this embodiment, for health monitoring and recognition, the pressure storage circuit senses and stores pulse signals from different locations on the human body. Then, through image recognition training algorithms and weight configuration, the cardiovascular health status of the human body can be detected and judged.

[0094] Another embodiment of this application provides a computer-readable storage medium having a computer program stored thereon, the computer program being executed by a processor to implement the pressure identification method described in any of the above embodiments. Reference Figure 10 As shown, the computer-readable storage medium is an optical disc 20, on which a computer program (i.e., a program product) is stored. When the computer program is run by a processor, it executes the methods provided in any of the aforementioned embodiments.

[0095] It should be noted that examples of computer-readable storage media may also include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other optical and magnetic storage media, which will not be elaborated here.

[0096] The computer-readable storage medium provided in the above embodiments of this application and the method provided in the embodiments of this application are based on the same inventive concept and have the same beneficial effects as the methods adopted, run or implemented by the applications stored therein.

[0097] It should be noted that:

[0098] It should be understood that although the steps in the flowcharts of the accompanying figures are shown sequentially as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the accompanying figures may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the sub-steps or stages of other steps.

[0099] The above embodiments merely illustrate the implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of this application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A pressure recognition system, characterized in that, It includes control and peripheral circuits, as well as a pressure sensing circuit, an analog front-end circuit, and an analog calculation module, which are respectively connected to the control and peripheral circuits. The control and peripheral circuits are used to control the circuit operation logic and timing of the pressure sensing circuit, the analog front-end circuit, and the analog calculation module; The pressure sensing circuit is used to acquire and store external pressure signals; The analog front-end circuit is used to convert the pressure signal from a resistance value to an analog voltage value; The simulation calculation module is used to process the simulation voltage value to obtain the pressure signal identification result; The pressure sensing circuit includes word line and bit line driving circuits, and multiple pressure sensing units respectively connected to the word line and bit line driving circuits; the word line and bit line driving circuits are used to select one or more of the pressure sensing units under the control of the controller in the control and peripheral circuits, so as to detect and store external pressure signals through the selected pressure sensing units; the pressure sensing unit includes a pressure sensor unit and an RRAM unit connected in series, the pressure sensor unit is used to detect external pressure signals, and the RRAM unit is used to store pressure signals from the pressure sensor unit.

2. The system according to claim 1, characterized in that, The control and peripheral circuitry includes a controller and a power management module, a clock source, and a clock tree, all connected to the controller. The clock source is connected to the clock tree, and both the clock source and the clock tree are connected to the power management module. The controller is used to control the circuit operation logic and timing of the pressure sensing circuit, the analog front-end circuit, and the analog calculation module. The power management module is used to manage the power supply to the pressure sensing circuit, the analog front-end circuit, and the analog computing module; the clock source is used to provide clock pulse signals; and the clock tree is used to manage the circuit timing of the clock pulse signals.

3. The system according to claim 1, characterized in that, The analog front-end circuit includes an interconnected resistor-to-voltage circuit and a programmable gain amplifier. The resistor-to-voltage circuit is used to convert the pressure signal from a resistance value to an analog voltage value, and the programmable gain amplifier is used to amplify the analog voltage value output by the resistor-to-voltage circuit and output the amplified analog voltage value.

4. The system according to claim 1, characterized in that, The analog computing module includes an analog computing circuit; the analog computing circuit includes at least one layer of SRAM analog computing array, the SRAM analog computing array includes multiple SRAM analog computing units, the SRAM analog computing unit includes an SRAM cell, a capacitor, a transistor, a first switch, a second switch, and a third switch, the SRAM cell stores the weights of the analog computing, the SRAM cell is connected to the gate of the transistor; the upper plate of the capacitor is connected to the second switch, the lower plate of the capacitor is grounded; the drain of the transistor is connected to the second switch, the source of the transistor is grounded; the first terminal of the first switch is used to connect to the input voltage, and the second terminal of the first switch is connected to the second switch, the third switch, and the capacitor.

5. The system according to claim 4, characterized in that, The analog computing module further includes a signal post-processing module connected to the analog computing circuit. The signal post-processing module is used to post-process the output of the analog computing circuit to obtain the pressure signal recognition result. The post-processing includes any one or more of the following methods: quantization, threshold judgment, logical judgment, and nonlinear activation.

6. A pressure recognition method, characterized in that, This is achieved through the pressure recognition system according to any one of claims 1-5; The pressure recognition method includes: The control and peripheral circuits control the circuit operation logic and timing of the pressure sensing circuit, the analog front-end circuit, and the analog calculation module; The pressure sensing circuit acquires and stores external pressure signals; The analog front-end circuit converts the pressure signal from a resistance value to an analog voltage value; The simulation calculation module processes the simulated voltage value to obtain the pressure signal identification result.

7. The method according to claim 6, characterized in that, The pressure sensing and storage unit includes a pressure sensor unit and an RRAM unit connected in series; the pressure sensing and storage circuit acquires and stores external pressure signals, including: When an external pressure signal is detected, the resistance value of the pressure sensor unit changes, the voltage drop across the RRAM unit increases, and the resistance of the RRAM unit changes, thus storing the pressure signal.

8. The method according to claim 6, characterized in that, The analog computing module includes interconnected analog computing circuits and a signal post-processing module; the analog computing circuit includes at least one layer of SRAM analog computing array; The analog calculation module processes the analog voltage value to obtain the pressure signal identification result, including: Each pixel of the voltage distribution map is arranged into a voltage vector according to a preset order and input into the first layer of SRAM analog computing array for processing, and a voltage vector is output. Each intermediate layer array receives a voltage vector from the previous layer array, and the output voltage vector is used as the input to the next layer array; the intermediate layer array is an SRAM analog computing array between the first layer and the last layer. The voltage vector of the last layer SRAM analog computing array is processed by the signal post-processing module to obtain the pressure signal recognition result. The post-processing includes any one or more combinations of quantization, threshold judgment, logical judgment and nonlinear activation.

9. The method according to claim 8, characterized in that, The SRAM analog computing array includes multiple SRAM analog computing units, each of which includes an SRAM cell, a capacitor, a transistor, a first switch, a second switch, and a third switch. The SRAM analog computing array processes the voltage vector as follows: When the voltage vector from the upper-level array is input to the SRAM analog computing unit, the first switch is closed, the second switch and the third switch are opened, and the capacitor is charged. After charging is complete, control the closing of the second switch and control the opening of the first switch and the third switch; Control the closing of the third switch, control the opening of the first and second switches, and output a voltage vector; Wherein, the voltage of the voltage vector is equal to the total charge divided by the total capacitance; the total charge is the sum of the charges of all the capacitors in the SRAM simulation array, and the total capacitance is the sum of the capacitances of all the capacitors in the SRAM simulation array.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, The computer program is executed by a processor to implement the pressure recognition method as described in any one of claims 6-9.

Images