Memristor sensing, storage and calculation integrated circuit structure applied to image recognition field

Abstract

The invention belongs to the technical field of semiconductor devices and integrated circuits, provides a memristor sensing, storage and calculation integrated circuit structure applied to the field of image recognition, and aims to realize a complete process of sensing, storage and calculation of image information by performing monolithic integration on a memristor with a photoelectric characteristic and a storage characteristic and a CMOS (complementary metal oxide semiconductor) integrated circuit. According to the invention, the photoelectric characteristic and the storage characteristic of the memristor are fully utilized, the memristor device is used as an image information sensing device and a storage unit, and a corresponding operation circuit module is combined, so that a circuit function module for simultaneously integrating sensing, storage and calculation on a single chip is realized; meanwhile, a diode and an MOS (Metal Oxide Semiconductor) transistor are combined to optimize the circuit structure. In conclusion, the memristor sensing, storage and calculation integrated circuit structure applied to the field of image recognition can be applied to the application fields of image sensing, image recognition and the like, and has the advantages of being small in size, high in speed, low in power consumption, high in integration level, low in cost and high in anti-jamming capability.

Description

technical field

[0001] The invention belongs to the technical field of semiconductor devices and integrated circuits, and in particular relates to a circuit structure for monolithic integration of a memristor with photoelectric and storage characteristics and a CMOS integrated circuit, and specifically provides a memristor sensor applied in the field of image recognition Storage and calculation integrated circuit structure. Background technique

[0002] At present, the development trend of artificial intelligence technology is very rapid, and image recognition based on artificial intelligence technology has also made great progress. Image recognition technology refers to the technology that uses computers to process, analyze and understand images to identify targets and objects in various patterns. This technology has been applied in many fields, such as automatic driving, face recognition, digital recognition etc.; it can be seen that image recognition technology, as an im...

Examples

Embodiment 1

[0057] This embodiment provides a memristor sensor-storage-computing integrated circuit structure applied in the field of image recognition, such as figure 1 Shown; specifically include: memristor array module, external erasing circuit module, computing circuit module, wherein,

[0058] The memristor array module includes two positive-weight memristor array modules and negative-weight memristor array modules with the same structure and the same position in space, which are respectively used to store external Input positive weight and negative weight; the positive weight memristor array module includes M×N memristors, wherein the memristor The first terminal is connected to the input row line RL m The upper and second ends are connected to the output column line on, enter the row line RL m Connect the input voltage V in,m ; n=1,2,...,N, m=1,2,...,M; The negative weight memristor array module includes M×N memristors, wherein the memristor The first terminal is connected ...

Embodiment 2

[0100] This embodiment provides a memristor sensor-storage-computing integrated circuit structure applied in the field of image recognition, such as figure 2 shown; the structure acts as figure 1 The improvement of the shown circuit structure, compared with the circuit structure in Embodiment 1, on the basis of each memristor unit structure, a diode is connected in series, that is, the cathode of the diode is connected to the first end of the memristor, and the diode The anode of is connected to the row line RL; the structure of its memristor unit is as follows Figure 9 shown.

[0101] Because the circuit structure in Embodiment 1 has the problem of crosstalk between lines, when the memristor is in a low-impedance state, the voltage signal on a row line RL will be transmitted to the connected column line CL through the memristor. When multiple memristors connected to the same column line CL maintain a low resistance state, the voltage on the same column line CL will be aff...

Embodiment 3

[0103] This embodiment provides a memristor sensor-storage-computing integrated circuit structure applied in the field of image recognition, such as image 3 shown; the structure acts as figure 1 The improvement of the shown circuit structure, compared with the circuit structure in embodiment 1, on the basis of each memristor unit structure, a MOS transistor is connected in series, that is, the source of the MOS transistor is connected to the first memristor The terminal, the drain of the MOS transistor are connected to the row line RL, and the gate of the MOS transistor is connected to the control signal; the structure of the memristor unit is as follows Figure 10 shown.

[0104] In the above circuit structure, when the control signal of the MOS transistor is high level, the MOS tube is turned on, and the memristor works normally; and when the control signal of the MOS transistor is low level, the MOS tube is turned off, so the crosstalk between lines can also be reduced T...

technical field

[0001] The invention belongs to the technical field of semiconductor devices and integrated circuits, and in particular relates to a circuit structure for monolithic integration of a memristor with photoelectric and storage characteristics and a CMOS integrated circuit, and specifically provides a memristor sensor applied in the field of image recognition Storage and calculation integrated circuit structure. Background technique

[0002] At present, the development trend of artificial intelligence technology is very rapid, and image recognition based on artificial intelligence technology has also made great progress. Image recognition technology refers to the technology that uses computers to process, analyze and understand images to identify targets and objects in various patterns. This technology has been applied in many fields, such as automatic driving, face recognition, digital recognition etc.; it can be seen that image recognition technology, as an im...

Examples

Embodiment 1

[0057] This embodiment provides a memristor sensor-storage-computing integrated circuit structure applied in the field of image recognition, such as figure 1 Shown; specifically include: memristor array module, external erasing circuit module, computing circuit module, wherein,

[0058] The memristor array module includes two positive-weight memristor array modules and negative-weight memristor array modules with the same structure and the same position in space, which are respectively used to store external Input positive weight and negative weight; the positive weight memristor array module includes M×N memristors, wherein the memristor The first terminal is connected to the input row line RL m The upper and second ends are connected to the output column line on, enter the row line RL m Connect the input voltage V in,m ; n=1,2,...,N, m=1,2,...,M; The negative weight memristor array module includes M×N memristors, wherein the memristor The first terminal is connected ...

Embodiment 2

[0100] This embodiment provides a memristor sensor-storage-computing integrated circuit structure applied in the field of image recognition, such as figure 2 shown; the structure acts as figure 1 The improvement of the shown circuit structure, compared with the circuit structure in Embodiment 1, on the basis of each memristor unit structure, a diode is connected in series, that is, the cathode of the diode is connected to the first end of the memristor, and the diode The anode of is connected to the row line RL; the structure of its memristor unit is as follows Figure 9 shown.

[0101] Because the circuit structure in Embodiment 1 has the problem of crosstalk between lines, when the memristor is in a low-impedance state, the voltage signal on a row line RL will be transmitted to the connected column line CL through the memristor. When multiple memristors connected to the same column line CL maintain a low resistance state, the voltage on the same column line CL will be aff...

Embodiment 3

[0103] This embodiment provides a memristor sensor-storage-computing integrated circuit structure applied in the field of image recognition, such as image 3 shown; the structure acts as figure 1 The improvement of the shown circuit structure, compared with the circuit structure in embodiment 1, on the basis of each memristor unit structure, a MOS transistor is connected in series, that is, the source of the MOS transistor is connected to the first memristor The terminal, the drain of the MOS transistor are connected to the row line RL, and the gate of the MOS transistor is connected to the control signal; the structure of the memristor unit is as follows Figure 10 shown.

[0104] In the above circuit structure, when the control signal of the MOS transistor is high level, the MOS tube is turned on, and the memristor works normally; and when the control signal of the MOS transistor is low level, the MOS tube is turned off, so the crosstalk between lines can also be reduced T...

Images