A TaO-based x Compact artificial neurons with multifunctional devices

Abstract

This invention belongs to the field of artificial neuromorphology and discloses a method based on TaO. x A compact artificial neuron for a multifunctional device, consisting of two structurally identical TaO2 neurons. x It consists of multifunctional devices, which are connected in series back-to-back; among which TaO x The multifunctional device exhibits a charge trapping and releasing mechanism when a voltage is applied in a first direction, providing analog volatile characteristics; while when a voltage is applied in a second direction, it exhibits a conductive filament mechanism, providing digital volatile characteristics. This invention also discloses the aforementioned TaO. x Structural design of a multifunctional device. Through this invention, it is possible to utilize only two identical TaO... x Multifunctional devices can be connected back-to-back to achieve the required neuronal functions without the need for additional capacitors, resistors, or other components. The circuit structure is compact and easy to manufacture, effectively improving the problems of low integration and high manufacturing difficulty of existing artificial neurons. Therefore, it is particularly beneficial for the large-scale integration of artificial neural networks.

Description

Technical Field

[0001] This invention belongs to the field of artificial neuromorphology, and more specifically, relates to a method based on TaO2. x Compact artificial neurons with multiple functions. Background Technology

[0002] Spiking Neural Networks (SNNs) belong to the third generation of neural network models. Unlike second-generation neural networks based on multilayer perceptrons, they achieve a higher level of bio-neural simulation. The most significant difference lies in the neurons. Second-generation neurons have fixed activation functions and only possess zero-order dynamics, while neurons in spiking neural networks possess richer dynamics, enabling them to better process spatiotemporal information. These artificial neurons come in various models, ranging from the most accurate biophysical neuron models to their simplest firing models. Among them, the leak-integral-ignition based neuron model, due to its simple setup and efficient encoding capabilities, strikes the best balance between biological rationality and practicality, and is currently one of the most commonly used artificial neuron models.

[0003] In existing technologies, when implementing the functions of a leak-integration-ignition neuron in hardware, two different units are generally used to implement the membrane potential function of leak-integration and the threshold switching function of ignition, respectively. The membrane potential function can be implemented using units with time-dynamic characteristics, such as resistor-capacitor circuits (RC circuits) and dynamic memristors, while the threshold function can be implemented using threshold circuits and threshold devices such as metal conductive wire memristors and metal-insulator switching devices.

[0004] However, further research has revealed the following shortcomings in the existing solutions: they all require different types of hardware combinations; the circuitry of a single leaky-integral-ignition artificial neuron is complex, and its capacitance is difficult to miniaturize; in large-scale artificial neural network circuits, the complex neuron circuitry exponentially increases the difficulty of fabrication and makes it difficult to improve the system's integration level. Accordingly, research and improvements are urgently needed in this field to address the technical challenges of low integration and high fabrication difficulty in existing artificial neurons. Summary of the Invention

[0005] To address one or more of the above-mentioned deficiencies or improvement needs of the existing technology, the present invention provides a TaO-based... xThe compact artificial neuron of multifunctional devices, by combining the working characteristics and specific requirements of artificial neurons, and redesigning the specific structure and working mechanism of its key components, can obtain a more compact artificial neuron. It not only eliminates the need for additional components such as capacitors and resistors, but also simplifies the manufacturing process and eliminates the need to consider resistance matching issues during series connection. Therefore, it can effectively solve the problems of low integration and complex process of existing artificial neurons, and is particularly beneficial for the large-scale integrated application of artificial neural networks.

[0006] To achieve the above objectives, according to the present invention, a TaO-based... x The compact artificial neuron of the multifunctional device is characterized by:

[0007] This artificial neuron consists of two structurally identical TaO2 neurons. x It consists of multifunctional devices, which are connected in series back-to-back; among them,

[0008] The TaO x When a voltage is applied in the first direction, the multifunctional device exhibits a charge trapping and releasing mechanism to provide analog volatile characteristics; while when a voltage is applied in the second direction, it exhibits a conductive filament mechanism to provide digital volatile characteristics.

[0009] When the aforementioned compact neurons are working, one of the TaO neurons... x Multifunctional device, also known as the first TaO x The resistive switching mechanism of multifunctional devices is a charge trapping and releasing mechanism, and another TaO x Multifunctional device, also known as the second TaO x The resistive switching mechanism of the multifunctional device is a conductive wire mechanism, which is used to realize the leakage-integration-ignition function of artificial neurons.

[0010] As a further preferred embodiment of the present invention, for the TaO x For a multifunctional device, it is preferably constructed by stacking a first electrode layer, a functional layer, a barrier layer, and a second electrode layer sequentially from bottom to top. The first electrode layer and the second electrode layer are made of inert metal and active metal materials, respectively, to form different electrodes. The functional layer is made of TaO. x The material is made with x less than 2.5 so that oxygen vacancies exist in the functional layer to form trap energy levels; the barrier layer is made of an inert metal material to prevent metal atoms from the second electrode layer from entering the functional layer.

[0011] As a further preferred embodiment of the present invention, the first electrode layer and the barrier layer are respectively made of one of the following inert metal materials: Pt, Pd, Au, Ni, ITO, Ti, W, Al, Ta, Hf, TaN, TiN and TiW; the second electrode layer is made of the following active metal materials: Ag, Cu or other similar materials.

[0012] As a further preferred embodiment of the present invention, the thickness of the first electrode layer is designed to be 5nm to 2000nm, the thickness of the functional layer is designed to be 2nm to 100nm, the thickness of the barrier layer is designed to be 2nm to 10nm, and the thickness of the second electrode layer is designed to be 10nm to 2000nm.

[0013] As a further preferred embodiment of the present invention, for the first TaO x In the case of multifunctional devices, when the first electrode layer is grounded and a negative voltage is applied to the second electrode layer, it exhibits analog volatile characteristics, which can be used to realize the leakage and integration functions of neurons.

[0014] As a further preferred embodiment of the present invention, for the second TaO x In the case of a multifunctional device, when the first electrode layer is grounded and the second electrode layer is initialized with a positive bias voltage, a positive voltage is applied, exhibiting digital volatile characteristics. This characteristic can be used to realize the ignition and leakage functions of a neuron. As a further preferred embodiment of the present invention, in order to realize the leakage-integration-ignition function of the above-mentioned artificial neuron, the first TaO... x Multifunctional device, second TaO x Multifunctional devices can be connected in series back-to-back via their respective first electrode layers, and wherein the first TaO x A pulse in the negative bias direction is applied to the second electrode layer of the multifunctional device, while the second TaO x The second electrode layer of the multifunctional device is grounded.

[0015] As a further preferred embodiment of the present invention, when the first TaO x When the amplitude of the negative bias pulse applied to the second electrode layer of the multifunctional device is greater than the preset amplitude, the neuron achieves the ignition function, and the number of pulses required decreases as the amplitude or frequency of the applied pulse increases.

[0016] In summary, the technical solutions conceived by this invention have the following main technical advantages compared with the prior art:

[0017] (1) In the artificial neurons of this application, there are only two identical TaO2 neurons. xMultifunctional devices can be connected in series back-to-back, and different voltage directions can be applied to give them different electrical characteristics to achieve different functions. When a negative voltage is applied, the device's resistive switching mechanism is a charge trapping and releasing mechanism, exhibiting simulated volatile characteristics, which can be used to realize the integration and leakage functions in neurons. When a positive voltage is applied, the device's resistive switching mechanism is a conductive filament mechanism, exhibiting digital volatile characteristics, which can be used to realize the ignition and leakage functions in neurons. In this way, not only can more compact products be provided, but the integration, leakage, and ignition functions of artificial neurons can also be realized without additional components such as capacitors and resistors.

[0018] (2) This application further addresses TaO x The specific structural composition and key processing parameters of the multifunctional device have been optimized. Accordingly, there is no need to consider the resistance matching problem in the series connection process. Moreover, two devices can be fabricated in one photolithography process. Therefore, the compact artificial neuron of this application effectively improves the problems of low integration and high process difficulty of existing artificial neurons. It can also improve the integration of large-scale neuromorphic computing circuits while reducing their process complexity. Attached Figure Description

[0019] Figure 1 This is a preferred embodiment of the invention based on TaO x A schematic diagram of a compact artificial neuron with multiple functions;

[0020] Figure 2 yes Figure 1 The TaO shown x Schematic diagram of the multifunctional device;

[0021] Figure 3 This is a specific embodiment of the present invention (Pt / TaO) x Voltage-current curves obtained by applying 100 consecutive negative bias scans to a Ti / Ag multifunctional device;

[0022] Figure 4 This is a specific embodiment of the present invention (Pt / TaO) x The pulse test response diagram of the / Ti / Ag multifunctional device under negative voltage direction, where the amplitude of the applied pulse is -6V and the frequency is 5Hz;

[0023] Figure 5 This is a specific embodiment of the present invention (Pt / TaO) x The voltage-current graph of the / Ti / Ag multifunctional device is obtained by applying a 1uA current limit and a 0-0.5V initialization operation, followed by 100 consecutive forward bias scans.

[0024] Figure 6An exemplary test circuit diagram for the compact artificial neuron used in this invention is shown;

[0025] Figure 7 The response diagram is obtained after applying continuous pulses to the compact neuron of the present invention, wherein the applied pulse amplitude is -6V and the frequency is 6.25Hz;

[0026] Figure 8 The diagram shows the response of the compact artificial neuron of the present invention under pulse testing at different amplitudes, wherein the applied pulse amplitudes are -5.8V, -5.9V, -6V, -6.1V, and -6.2V, and the frequency is 6.66Hz.

[0027] Figure 9 The diagram shows the response of the compact artificial neuron of the present invention under pulse testing at different frequencies, wherein the amplitude of the applied pulse is -6.1V and the frequencies are 5Hz, 5.56Hz and 7.14Hz, respectively. Detailed Implementation

[0028] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0029] It should be understood that expressions such as "comprising" and "may include" as used in this application indicate the existence of the disclosed functions, operations, or constituent elements, and do not limit one or more additional functions, operations, and constituent elements. In this application, terms such as "comprising" and / or "having" may be interpreted as indicating a specific characteristic, number, operation, constituent element, component, or combination thereof, but should not be interpreted as excluding the existence or possibility of adding one or more other characteristics, numbers, operations, constituent elements, components, or combinations thereof.

[0030] It should be understood that the terms “center,” “upper,” “lower,” “front,” “rear,” “left,” “right,” “vertical,” “horizontal,” “inner,” “outer,” “clockwise,” “counterclockwise,” “axial,” “radial,” and “circumferential” indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application.

[0031] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0032] In this application, unless otherwise expressly specified and limited, the terms "installation," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection between two components. Those skilled in the art can understand the specific meaning of the above terms in this application according to the specific circumstances.

[0033] Figure 1 This is a preferred embodiment of the invention based on TaO x A schematic diagram of a compact artificial neuron with multiple functions. The following will combine... Figure 1 To explain the invention in more detail.

[0034] like Figure 1 As shown, the present invention is based on TaO x The compact artificial neuron of the multifunctional device consists of two structurally identical TaO2 neurons. x It consists of multifunctional devices, and they are connected in series back-to-back; wherein, when TaO x When a voltage is applied to the second electrode layer of the multifunctional device, and the first electrode layer is grounded, a charge trapping and releasing mechanism is exhibited when a negative voltage is applied, providing analog volatile characteristics; after initialization, a conductive filament mechanism is exhibited when a positive voltage is applied, providing digital volatile characteristics; when the aforementioned compact neuron is working, one of the TaO... x The multifunctional device (shown as device A in the figure) is also the first TaO x Multifunctional device, its resistive switching mechanism is a charge trapping and releasing mechanism, another TaO x The multifunctional device (shown as device B in the figure) is also known as the second TaO. x Multifunctional device, its resistive switching mechanism is a conductive wire mechanism.

[0035] See Figure 2 According to a preferred embodiment of the present invention, for the TaO xFor a multifunctional device, it is preferably constructed by stacking a first electrode layer, a functional layer, a barrier layer, and a second electrode layer sequentially from bottom to top. The first electrode layer and the second electrode layer are made of inert metal and active metal materials, respectively, to form different electrodes. The functional layer is made of TaO. x The material is made with x less than 2.5 so that oxygen vacancies exist in the functional layer to form trap energy levels; the barrier layer is made of an inert metal material to prevent metal atoms from the second electrode layer from entering the functional layer.

[0036] According to another preferred embodiment of the present invention, the first electrode layer and the barrier layer are respectively made of one of the following inert metal materials: Pt, Pd, Au, Ni, ITO, Ti, W, Al, Ta, Hf, TaN, TiN and TiW; the second electrode layer is made of the following active metal materials: Ag, Cu or other similar materials.

[0037] According to another preferred embodiment of the present invention, the thickness of the first electrode layer is designed to be 5nm to 2000nm, the thickness of the functional layer is designed to be 2nm to 100nm, the thickness of the barrier layer is designed to be 2nm to 10nm, and the thickness of the second electrode layer is designed to be 10nm to 2000nm.

[0038] In a specific application scenario, for the first TaO x In the case of a multifunctional device, if its first electrode layer is grounded and a negative voltage is applied to the second electrode layer, the resistive switching mechanism of the device is a charge trapping and releasing mechanism, which has simulated volatile characteristics, and can thus be used to realize the leakage and integration functions of neurons.

[0039] In another specific application scenario, for the second TaO x In the case of a multifunctional device, after it performs the initialization operation of current limiting positive voltage, if its first electrode layer is grounded and a positive voltage is applied to the second electrode layer, the resistive switching mechanism of the device is a conductive wire mechanism, which has digital volatile characteristics, and can thus be used to realize the ignition and leakage functions of neurons.

[0040] In another specific application scenario, when the aforementioned compact artificial neuron is working, the first TaO x The first electrode layer and the second TaO of the multifunctional device x The first electrode layer of the multifunctional device is connected, and the first TaO x A negative bias pulse is applied to the second electrode layer of the multifunctional device, and the second TaO x The second electrode layer of the multifunctional device is grounded. When the amplitude of the applied negative bias pulse is greater than the preset amplitude, the neuron can achieve the ignition function, and the number of pulses required decreases as the amplitude or frequency of the applied pulse increases.

[0041] exist Figure 2 In the specific application scenario shown, the arrows indicate the direction of current between the two devices A and B when the compact artificial neuron is working; among them, the two TaO x The multifunctional devices, namely devices A and B, are connected in series through the first electrode layer. Furthermore, a negative voltage pulse is input to the second electrode layer of device A, while the second electrode layer of device B is grounded. Accordingly, device A utilizes its analog volatile characteristics under negative bias to achieve integration and leakage functions within the neuron; device B, after initialization with a current limit of 1uA and a voltage of 0–0.5V, utilizes its digital volatile characteristics under positive bias to achieve ignition and leakage functions within the neuron.

[0042] The following are some specific embodiments to more clearly explain the present invention.

[0043] Example 1

[0044] In this embodiment, the provided TaO x The multifunctional device comprises a first electrode layer, a functional layer, a barrier layer, and a second electrode layer stacked sequentially from bottom to top; wherein the functional layer is TaO. x The thin film has x = 2.042; the material of the first electrode layer is Pt; the material of the barrier layer is Ti; the material of the second electrode layer is Ag; in addition, the thickness of the first electrode layer is 100 nm, the thickness of the functional layer is 10 nm, the thickness of the barrier layer is 5 nm, and the thickness of the second electrode layer is 100 nm.

[0045] Example 2

[0046] In this embodiment, the provided TaO x The multifunctional device comprises a first electrode layer, a functional layer, a barrier layer, and a second electrode layer stacked sequentially from bottom to top; wherein the functional layer is TaO. x The thin film has x = 2.042; the material of the first electrode layer is Au; the material of the barrier layer is Pt; the material of the second electrode layer is Ag; in addition, the thickness of the first electrode layer is 5 nm, the thickness of the functional layer is 2 nm, the thickness of the barrier layer is 2 nm, and the thickness of the second electrode layer is 10 nm.

[0047] Example 3

[0048] In this embodiment, the provided TaO x The multifunctional device comprises a first electrode layer, a functional layer, a barrier layer, and a second electrode layer stacked sequentially from bottom to top; wherein the functional layer is TaO. xThe thin film has x = 2.042; the material of the first electrode layer is ITO; the material of the barrier layer is TaN; the material of the second electrode layer is Cu; in addition, the thickness of the first electrode layer is 30 nm, the thickness of the functional layer is 10 nm, the thickness of the barrier layer is 3 nm, and the thickness of the second electrode layer is 300 nm.

[0049] Example 4

[0050] In this embodiment, the provided TaO x The multifunctional device comprises a first electrode layer, a functional layer, a barrier layer, and a second electrode layer stacked sequentially from bottom to top; wherein the functional layer is TaO. x The thin film has x = 2.042; the material of the first electrode layer is TiN; the material of the barrier layer is W; the material of the second electrode layer is Cu; in addition, the thickness of the first electrode layer is 2000 nm, the thickness of the functional layer is 100 nm, the thickness of the barrier layer is 10 nm, and the thickness of the second electrode layer is 2000 nm.

[0051] Example 5

[0052] In this embodiment, the provided TaO x The multifunctional device comprises a first electrode layer, a functional layer, a barrier layer, and a second electrode layer stacked sequentially from bottom to top; wherein the functional layer is TaO. x The thin film has x = 2.042; the material of the first electrode layer is Ni; the material of the barrier layer is TaN; the material of the second electrode layer is Ag; in addition, the thickness of the first electrode layer is 1000 nm, the thickness of the functional layer is 70 nm, the thickness of the barrier layer is 7 nm, and the thickness of the second electrode layer is 1000 nm.

[0053] The following is an analysis of the Pt / TaO obtained in Example 1. x Results obtained from a series of tests on the / Ti / Ag multifunctional device.

[0054] like Figure 3 As shown, this illustrates the process of adding Pt / TaO. x The voltage-current curves of the / Ti / Ag multifunctional device obtained by applying 100 consecutive negative bias scans. The Pt electrode is grounded, and the Ag electrode is subjected to a negative scan bias voltage of 0 to -3 to 0V. Tests show that this device exhibits simulated volatile characteristics and can realize the function of neuron leakage.

[0055] like Figure 4 As shown, this illustrates the process of adding Pt / TaO. xThe response diagram of the / Ti / Ag multifunctional device under negative voltage is shown in the pulse test diagram. The Pt electrode is grounded, and ten consecutive pulses with a negative bias are applied to the Ag electrode, each with an amplitude of -6V and a frequency of 5Hz. The test results show that the device's response increases with the number of pulses, enabling it to perform neuron integration.

[0056] like Figure 5 As shown, this illustrates the process of adding Pt / TaO. x The / Ti / Ag multifunctional device was initialized with a current limit of 1uA and a voltage range of 0–0.5V. The current-voltage curves were obtained by continuously scanning with a forward bias voltage of 0–0.3–0V. The Pt electrode was grounded, and the Ag electrode was subjected to a voltage. Tests showed that the device exhibits digital volatile characteristics and can realize neuron ignition and leakage functions.

[0057] Based on the above electrical tests, it can be found that: (1) the multifunctional device Pt / TaO x When a negative bias is applied to the Ag electrode in / Ti / Ag, it exhibits simulated volatile characteristics and can be used to realize integration and leakage processes; (2) Multifunctional device Pt / TaO x After initialization by applying a 1uA current-limiting forward voltage to the Ag electrode in / Ti / Ag, a forward bias is applied, exhibiting digital volatile characteristics, which can be used to achieve ignition and leakage functions.

[0058] Figure 6 An exemplary test circuit diagram for the compact artificial neuron used in this invention is shown. Point M is... Figure 5 The first electrode layer connecting device A and device B is used to represent the change in membrane potential in biological neurons.

[0059] Figure 7 This is a response diagram obtained after applying continuous pulses to the compact neuron of the present invention, wherein the applied pulse amplitude is -6V and the frequency is 6.25Hz. As shown in the figure, as the number of applied pulses increases, device A integrates the input pulses, and the potential at point M continuously increases until it reaches the device's resistance threshold voltage. Device B then changes from a high-resistance state to a low-resistance state, thus realizing the ignition function.

[0060] Figure 8 The figure shows the response of the compact artificial neuron of the present invention under pulse testing at different amplitudes, where the applied pulse amplitudes are -5.8V, -5.9V, -6V, -6.1V, and -6.2V, and the frequency is 6.66Hz. As shown in the figure, the compact artificial neuron can only ignite when the applied pulse amplitude is greater than -5.8V, and the number of pulses required decreases as the applied pulse amplitude increases.

[0061] Figure 9This is a response graph of the compact artificial neuron of the present invention under pulse testing at different frequencies, where the amplitude of the applied pulses is -6.1V and the frequencies are 5Hz, 5.56Hz, and 7.14Hz, respectively. As shown in the figure, the number of pulses required for the compact neuron to achieve ignition decreases as the applied pulse frequency increases.

[0062] Based on the above electrical measurements, it can be found that (1) the compact artificial neuron can realize the functions of leakage, integration and ignition in the neuron; (2) the response of the compact artificial neuron can be controlled by adjusting the input pulse frequency and amplitude; wherein, the compact artificial neuron needs to apply a pulse amplitude greater than the threshold to realize ignition, and the number of pulses required to realize ignition decreases as the applied pulse amplitude increases; the number of pulses required to realize ignition of the compact artificial neuron decreases as the applied pulse frequency increases.

[0063] In summary, the compact artificial neuron system according to the present invention can rely on only two identical TaOs x Multifunctional devices can achieve integral ignition by connecting back to back without the need for additional resistors, capacitors, or other components. The circuit structure is simple and conducive to the large-scale integration of artificial neural networks. Therefore, it is particularly suitable for large-scale integration applications of artificial neural networks and has good practical value and application prospects.

[0064] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A TaO-based x The compact artificial neuron of the multifunctional device is characterized by: This artificial neuron consists of two structurally identical TaO2 neurons. x It consists of multifunctional devices, which are connected in series back-to-back; among them, Each of the TaO x The multifunctional device comprises, from bottom to top, a first electrode layer, a functional layer, a barrier layer, and a second electrode layer, wherein the first electrode layer and the second electrode layer are made of inert metal and active metal materials, respectively, to form different electrodes; the functional layer is made of TaO. x The material is made with x less than 2.5 so that oxygen vacancies exist in the functional layer to form trap energy levels; the barrier layer is made of an inert metal material to prevent metal atoms from the second electrode layer from entering the functional layer; The TaO x When a voltage is applied in the first direction, the multifunctional device exhibits a charge trapping and releasing mechanism to provide analog volatile characteristics; while when a voltage is applied in the second direction, it exhibits a conductive filament mechanism to provide digital volatile characteristics. When the aforementioned compact neurons are working, one of the TaO neurons... x Multifunctional device, also known as the first TaO x The resistive switching mechanism of multifunctional devices is a charge trapping and releasing mechanism, and another TaO x Multifunctional device, also known as the second TaO x The resistive switching mechanism of the multifunctional device is a conductive wire mechanism, which is used to realize the leakage-integration-ignition function of artificial neurons.

2. The compact artificial neuron as described in claim 1, characterized in that, The first electrode layer and the barrier layer are made of one of the following inert metal materials: Pt, Pd, Au, Ni, ITO, Ti, W, Al, Ta, Hf, TaN, TiN and TiW; the second electrode layer is made of the following active metal material: Ag or Cu.

3. The compact artificial neuron as described in claim 2, characterized in that, The thickness of the first electrode layer is designed to be 5nm~2000nm, the thickness of the functional layer is designed to be 2nm~100nm, the thickness of the barrier layer is designed to be 2nm~10nm, and the thickness of the second electrode layer is designed to be 10nm~2000nm.

4. The compact artificial neuron according to any one of claims 1 to 3, characterized in that, For the first TaO x In the case of multifunctional devices, when the first electrode layer is grounded and a negative voltage is applied to the second electrode layer, it exhibits analog volatile characteristics, which are used to realize the leakage and integration functions of neurons.

5. The compact artificial neuron as described in claim 4, characterized in that, For the second TaO x In the case of multifunctional devices, when the first electrode layer is grounded and the second electrode layer is initialized with a positive bias voltage, a positive voltage is applied, exhibiting digital volatile characteristics. This characteristic is used to realize the ignition and leakage functions of neurons.

6. The compact artificial neuron as described in claim 5, characterized in that, To achieve the leakage-integration-ignition function of the artificial neuron described above, the first TaO x Multifunctional device, second TaO x The multifunctional devices are connected in series back-to-back through their respective first electrode layers, and wherein the first TaO x A pulse in the negative bias direction is applied to the second electrode layer of the multifunctional device, while the second TaO x The second electrode layer of the multifunctional device is grounded.

7. The compact artificial neuron as described in claim 6, characterized in that, When the first TaO x When the amplitude of the negative bias pulse applied to the second electrode layer of the multifunctional device is greater than the preset amplitude, the neuron realizes the ignition function, and the number of pulses required decreases as the amplitude or frequency of the applied pulse increases.

Images