An integrated circuit structure for sensing, memory, and computing based on graphene / silicon charge-coupled devices.

Abstract

This invention discloses an integrated sensing, storage, and computing circuit structure based on graphene / silicon charge-coupled devices (CCDs). The circuit structure includes an array module, a gate voltage generation module, a weighted voltage generation module, a summation control module, and a summation operation module. The array module consists of n×n array units, each composed of a graphene / silicon CCD and a switching transistor. The source of each device is connected to the weighted voltage generation module, and its drain is connected to the switching transistor. The control signal terminals of the switching transistors are connected to the summation control module, which controls the switching on and off of the transistors. The other end of the switching transistor is connected to the summation operation module, which integrates all incoming currents and converts them into a voltage for output. This invention utilizes the photosensitive and storage characteristics of graphene / silicon CCDs, combining photosensitive and storage capabilities with computational circuitry. It can be applied to fields such as image recognition and image sensing, improving integration and speed while reducing cost and power consumption.

Description

Technical Field

[0001] This invention belongs to the fields of image sensing technology and integrated circuit technology, specifically relating to an integrated circuit structure of sensing, storage, and computing based on graphene / silicon charge-coupled devices. Background Technology

[0002] Image recognition refers to the technology of using computers to process, analyze, and understand images in order to identify targets and objects of various patterns. Currently, image recognition technology is generally divided into facial recognition and product recognition. Facial recognition is mainly used in security checks, identity verification, and mobile payments; product recognition is mainly used in the commodity circulation process. Image recognition technology plays an increasingly important role in many fields; therefore, research on image recognition technology is of great significance.

[0003] The traditional image recognition process consists of four steps: image acquisition, image preprocessing, feature extraction, and image recognition. During image acquisition, low-power, low-cost, and highly integrated CMOS image sensors are primarily used. After acquisition, image processing and recognition are performed. This process mainly involves algorithms calculating and processing the acquired information to ultimately obtain the image recognition result.

[0004] With technological advancements, the volume of image data to be processed is rapidly increasing, posing challenges to terminal computing power and transmission bandwidth. In traditional image recognition processes, sensors are only responsible for image acquisition and do not participate in image processing or recognition, lacking computational capabilities. Given the high power consumption and transmission latency issues caused by the large volume of data to be processed, the potential of integrated sensor architectures with edge computing capabilities remains to be explored. However, existing integrated sensor-memory-computing circuit structures are designed for photodiodes and lack designs specifically for conductive photoelectric devices. Summary of the Invention

[0005] To address the problems of existing technologies, this invention provides an integrated circuit structure for sensing, storage, and computing based on graphene / silicon charge-coupled devices.

[0006] The objective of this invention is achieved through the following technical solution: a circuit structure integrating inductance, storage, and computing based on graphene / silicon charge-coupled devices, the circuit structure comprising an array module, a gate voltage generation module, a weighted voltage generation module, a summation control module, and a summation operation module;

[0007] The array module consists of n×n array units, each array unit consisting of a graphene / silicon charge-coupled device and a switching transistor;

[0008] The graphene / silicon charge-coupled device is a conductive optoelectronic device. The source of each graphene / silicon charge-coupled device is connected to a weighted voltage generation module, and the drain is connected to a switching transistor. The control signal terminals of the switching transistors are all connected to a summation control module, which controls the opening and closing of the switching transistors. The other end of the switching transistor is connected to a summation calculation module, which is used to integrate all the incoming currents and convert them into voltage for output.

[0009] The gate voltage generation module generates a pulse voltage V. g It is added to the gate of a graphene / silicon charge-coupled device.

[0010] Furthermore, the graphene / silicon charge-coupled device is a three-terminal conductivity photoelectric sensor. When the gate voltage Vg is high, the sensor is in the working state; when the gate voltage Vg is low, the sensor is in the reset state. There is a graphene channel between the drain and source of the sensor. When the sensor is in the working state, if light shines into the sensor, the conductivity of the graphene decreases linearly. The rate of decrease in conductivity is proportional to the light intensity. When the light disappears, the conductivity of the graphene stops decreasing and remains unchanged, which is used to store light intensity information.

[0011] Furthermore, the weighted voltage generation module is connected to the source of the graphene / silicon charge-coupled device, and generates a corresponding weighted voltage based on the externally input weights. When the weights are positive, the generated voltage is less than V. ref The voltage generated when the weight is negative is greater than V. ref Furthermore, the larger the weight, the larger the absolute value of the voltage difference between the two voltages, and the larger the voltage applied across the device.

[0012] Furthermore, the summation control module controls the summation operation module by controlling the opening and closing of the switching transistors. The summation control module controls the switching transistors of the array unit. When the device does not need to perform summation operations, the switching transistors are turned off, the graphene / silicon charge-coupled device does not work, and no current flows into the summation operation module. If summation operations are needed, the switching transistors are turned on, and the current of the graphene / silicon charge-coupled device flows into the summation operation module. The drain of the graphene / silicon charge-coupled device is connected to the summation operation module. At this time, the source voltage of the graphene / silicon charge-coupled device is the weighting voltage, and the drain voltage is the reference voltage input to the summation operation module. The device is in working mode, and the current flowing through the graphene / silicon charge-coupled device will be output to the summation operation module.

[0013] Furthermore, the summation module includes an operational amplifier, a current-to-voltage circuit, and a sampling switch; the non-inverting input of the operational amplifier is connected to one end of the switching transistor of each array unit, and the inverting input is connected to the reference voltage; when the sampling switch is open, the current flowing into the non-inverting input of the operational amplifier is integrated on the capacitor, converted into voltage, and output; when the sampling switch is closed, the voltage output by the operational amplifier is the reference voltage.

[0014] The beneficial effects of this invention are as follows: This invention combines graphene / silicon charge-coupled devices with CMOS integrated circuits to realize an integrated sensing-memory-computing circuit. Compared with traditional in-memory computing circuits, this design integrates photoconductive sensing and in-memory computing, which can greatly improve the circuit's processing speed of sensor data, and has high integration, simple structure, and low power consumption. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of the integrated inductor-memory-computing circuit structure based on graphene / silicon charge-coupled devices;

[0016] Figure 2 This is a schematic diagram of the unit structure of a graphene / silicon charge-coupled device array module;

[0017] Figure 3 This is a schematic diagram of the summation operation circuit.

[0018] Figure 4 This is a schematic diagram showing the change of the conductivity G of a graphene / silicon charge-coupled device with time T under different light intensities.

[0019] Figure 5 This is a schematic diagram of the waveforms of each part of the integrated inductor-memory-computer circuit based on graphene / silicon charge-coupled devices during operation. Detailed Implementation

[0020] The technical solutions of the embodiments of the present invention will be fully described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] like Figure 1 The diagram shows the overall structure of an integrated inductor-memory-computing circuit based on graphene / silicon charge-coupled devices, including a weighted voltage generation module, a gate voltage generation module, an array module composed of graphene / silicon charge-coupled devices, a summation control module, and a summation operation module.

[0022] The weighted voltage generation module generates different weighted voltages V based on externally input weighted values. i ;

[0023] The gate voltage generation module generates a pulse voltage V. g Added to the gate of a graphene / silicon charge-coupled device;

[0024] The array module is an n×n array, and each array unit consists of a graphene / silicon charge-coupled device and a switching transistor. The graphene / silicon charge-coupled device is a conductive optoelectronic device. When a bias voltage is applied to the gate, if light shines into the device, the conductivity of the graphene between the source and drain of the device will decrease linearly, and the rate of decrease is proportional to the light intensity. At the same time, the device also has a storage function. When the light disappears, the conductivity of the graphene can maintain the conductivity before the light disappeared, so it can store the information of the light.

[0025] The summation control circuit is used to control the switching transistors of the array unit. When the device needs to perform a summation operation, the switching transistor is closed, and the current of the device flows into the summation operation module. If the device does not need to perform a summation operation, the switching transistor is opened.

[0026] The summation module integrates all incoming currents and outputs them as voltages.

[0027] like Figure 2 The diagram shows a schematic of a graphene / silicon charge-coupled device (CCD) array module. This module consists of a graphene / silicon CCD 1, a gate voltage generation module 2, a weighted voltage generation module 3, a switch 4, and a summation control module 5. When the gate voltage is high, the device is in the working state; when the gate voltage is low, the device is in the reset state. The weighted voltage generation module generates a corresponding voltage based on the externally input weight value. The switch is opened and closed by the summation control module. When the switch is closed, the device is working, and the current flowing through the device flows into the summation module. When the switch is open, the device is not working.

[0028] like Figure 3 The diagram shows the structure of the summation circuit. When sampling switch 7 is closed, the circuit is in a reset state, and the output voltage 9 is V. ref When sampling switch 7 is open, the circuit is in integration mode. The current flowing out of the device is integrated on the integrating capacitor 8, and the output voltage 9 is obtained at the output terminal.

[0029] like Figure 4 The figure shows a schematic diagram of the conductivity G of a graphene / silicon charge-coupled device changing with time T under different light intensities. Figure 4The upper part shows the waveform of the gate voltage of the graphene / silicon charge-coupled device over time, and the lower part shows the waveform of the graphene conductance over time. When the gate voltage of the graphene / silicon charge-coupled device is low, the device is in a reset state and does not work. When the gate voltage is high, the device is in an operating state. In the operating state, if there is no light, the graphene conductance remains unchanged. If there is light, the graphene conductance will decrease first, and the rate of decrease is proportional to the light intensity.

[0030] like Figure 5 The figure shows a schematic diagram of the waveforms of each part of the integrated circuit based on graphene / silicon charge-coupled devices during operation. Figure 5 The top image shows the waveform of the gate voltage of the graphene / silicon charge-coupled device over time; the middle image shows the waveform of the graphene conductance over time; and the bottom image shows the waveform of the sampling signal over time. Sampling is performed when the gate voltage is high. When the device gate voltage is low, the device is in a reset state, and the circuit is also in a reset state. When the device gate voltage is high, the summation control module controls the switching of each device. If the switch is open, the device does not work; if the switch is closed, the device source-drain voltage is equal to the weighted voltage V. i and reference voltage V ref The device is in operation. During one operating cycle, the summation module samples the incoming current twice, with each sampling time being T. S The difference between the two sampled output voltages is the final voltage value, which is the result of image preprocessing and can be applied to fields such as neuromorphic vision and image recognition. Each sampling time is very short, and the current flowing in during the sampling time can be considered a constant current. The integrating capacitor is C. int Assuming that the current from m devices flows into the summation module, the voltage output from the first sample is:

[0031]

[0032] The voltage output from the second sampling is:

[0033]

[0034] The difference between the two voltages is:

[0035]

[0036] The voltage difference is the result of a single convolution calculation, where the sampling time T S and integrating capacitor C int V is a constant. ref -V i For the weights, V ref >V i When the weight is positive, V ref <Vi When the weight is negative, V ref =V i The time weight is zero; G i1 -G i2 The change in the electrical conductivity of graphene reflects the intensity of light.

[0037] The above description is merely a preferred embodiment of the present invention. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make many possible variations and modifications to the technical solutions of the present invention using the methods and techniques disclosed above, or modify them into equivalent embodiments with equivalent changes, without departing from the scope of the technical solutions of the present invention. Therefore, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solutions of the present invention shall still fall within the protection scope of the technical solutions of the present invention.

Claims

1. A circuit structure for integrated inductance, storage, and computing based on graphene / silicon charge-coupled devices, characterized in that, The circuit structure includes an array module, a gate voltage generation module, a weighted voltage generation module, a summation control module, and a summation operation module. The array module consists of n×n array units, each array unit consisting of a graphene / silicon charge-coupled device and a switching transistor; The graphene / silicon charge-coupled device is a conductive optoelectronic device. The source of each graphene / silicon charge-coupled device is connected to a weighted voltage generation module, and the drain is connected to a switching transistor. The control signal terminals of the switching transistors are all connected to a summation control module, which controls the opening and closing of the switching transistors. The other end of the switching transistor is connected to a summation calculation module, which is used to integrate all the incoming currents and convert them into voltage for output. The gate voltage generation module generates pulse voltage. Added to the gate of a graphene / silicon charge-coupled device; The weighted voltage generation module is connected to the source of the graphene / silicon charge-coupled device (CCD) and generates a corresponding weighted voltage based on the externally input weights. For positive weights, the generated voltage is less than [value missing]. The voltage generated when the weight is negative is greater than Furthermore, the larger the weight, the larger the absolute value of the voltage difference between the two voltages, and the larger the voltage applied across the device. The summation control module controls the summation operation module by controlling the opening and closing of the switching transistors; the summation control module is used to control the switching transistors of the array units. The drain of the graphene / silicon charge-coupled device is connected to the summation module. At this time, the source voltage of the graphene / silicon charge-coupled device is the weighting voltage, and the drain voltage is the reference voltage input to the summation module. The device is in working mode, and the current flowing through the graphene / silicon charge-coupled device will be output to the summation module. The summation module includes an operational amplifier, a current-to-voltage circuit, and a sampling switch. The non-inverting input of the operational amplifier is connected to one end of the switching transistor of each array unit, and the inverting input is connected to the reference voltage. When the sampling switch is open, the current flowing into the non-inverting input of the operational amplifier is integrated on the capacitor, converted into voltage, and output. When the sampling switch is closed, the voltage output by the operational amplifier is the reference voltage.

2. The integrated inductor-memory-computing circuit structure based on graphene / silicon charge-coupled devices according to claim 1, characterized in that, The graphene / silicon charge-coupled device is a three-terminal conductivity type photoelectric sensor. When the gate voltage... When the gate voltage is high, the sensor is in working condition. When the level is low, the sensor is in a reset state; there is a graphene channel between the sensor's drain and source. When the sensor is in operation, if light shines into the sensor, the conductivity of the graphene decreases linearly. The rate of decrease in conductivity is proportional to the light intensity. When the light disappears, the conductivity of the graphene stops decreasing and remains constant, which is used to store light intensity information.

3. The integrated inductor-memory-computing circuit structure based on graphene / silicon charge-coupled devices according to claim 1, characterized in that, When the device does not need to perform a summation operation, the switch is turned off, the graphene / silicon charge-coupled device does not work, and no current flows into the summation module; if a summation operation is needed, the switch is turned on, and the current from the graphene / silicon charge-coupled device flows into the summation module.

Images