A high-speed writing method for resistive memory in analog conductance state and its application

Abstract

This invention provides a high-speed method for writing analog conductance states to resistive memory, belonging to the fields of semiconductors, analog computing, and integrated circuits. The method is based on a feedback loop in the continuous time domain, consisting of an input module connected by electrical signals, a reference resistor module, a resistive memory module, and a feedback module. It can autonomously program the resistive memory to any analog conductance state within a continuous time period. Furthermore, the feedback signal intelligently determines whether the target conductance state has been reached and terminates the writing process, eliminating the need for multiple rounds of read verification. This improves the programming speed of the device and eliminates the need for analog-to-digital conversion and additional logic judgment circuits. It can significantly improve the speed and efficiency of memory array programming in matrix operation accelerators, fully leveraging the potential of analog computing and in-memory computing. This method has significant implications and broad development prospects for accelerating neural network inference and training scenarios that require programming a large number of devices.

Description

Technical Field

[0001] This invention belongs to the fields of semiconductors, analog computing, and integrated circuits, and relates to a high-speed writing method and application of analog conductivity programming for resistive memory. Background Technology

[0002] With the development of artificial intelligence technology and the surge in data volume, traditional digital computers, limited by their in-memory computing architecture and serial digital algorithms, are struggling to handle today's massive and complex computing tasks. Analog computing circuits based on resistive memory arrays can perform in-situ matrix operations, such as matrix-vector multiplication, matrix inversion, matrix pseudo-inversion, and matrix eigenvector generation. Benefiting from the elimination of data movement in in-memory computing architecture and the extremely high parallelism of analog computing, this method provides a strong guarantee for overcoming the von Neumann bottleneck and has been widely studied and applied as an accelerator for neural network training and inference.

[0003] As an acceleration core for matrix operations, memory arrays need to be accurately and efficiently programmed to target conductance states to map corresponding matrix elements. Due to the randomness of resistive memories, the currently widely used method is the write-verify method. This means that for each device, a read-verify operation is performed after each write pulse to determine the height and width of the next pulse. This typically requires dozens of rounds to program a device to the target state, consuming a significant amount of time. Furthermore, the additional analog-to-digital conversion and digital logic operation modules also generate substantial power consumption and chip area overhead, ultimately diminishing or even negating the advantages of analog in-memory computing. Moreover, existing high-speed self-terminating analog conductance state writing schemes can only program a very limited number of conductance states, typically 1-2 bits, which cannot meet the requirements of analog in-memory computing. Therefore, a high-speed, analog-to-digital conversion-free, self-determining and terminating analog conductance state writing scheme urgently needs to be developed. Summary of the Invention

[0004] To address the problems existing in the prior art, this invention proposes a high-speed writing method and application for the analog conductivity state of resistive memory. Based on a feedback loop in the continuous time domain consisting of an electrical signal connection input module, a reference resistor module, a resistive memory module, and a feedback module, the resistive memory can be autonomously programmed to reach any analog conductivity state within a continuous time. Furthermore, the feedback signal enables intelligent judgment of whether the target conductivity state has been reached and terminates the writing process. No additional analog-to-digital converter and digital logic operation module are required, thereby accelerating the programming process of the device.

[0005] The technical solution of this invention is as follows:

[0006] A high-speed writing method for analog conductance states of resistive memory is disclosed. This method is based on an analog circuit consisting of an input module, a reference resistor module, a resistive memory module, and a feedback module. The modules are connected by electrical signals to form a feedback loop in the continuous time domain. The resistive memory is autonomously programmed to reach any analog conductance state within a continuous time period. The writing process is then intelligently terminated based on whether the target conductance state has been reached, using the feedback signal. The method includes the following steps:

[0007] 1) Reset the conductance state of the resistive memory

[0008] 2) Adjust the conductance value of the reference resistor module to select the target conductance state.

[0009] 3) The DC voltage supplied by the input power supply of the input module is used to adjust the initial voltage value applied to the resistive memory.

[0010] 4) Connecting a resistive memory forms a complete feedback loop in the continuous time domain.

[0011] 5) Program the resistive memory's conductance state to be near the target conductance state, and implement the function of autonomously determining and terminating the process.

[0012] In the feedback loop within the continuous time domain, the feedback module detects the conductance difference between the memory and the reference resistor and feeds this conductance difference back to the input module as a voltage signal through the output terminal, completing the electrical signal feedback process. During this process, the resistive memory is programmed, and its conductance value continuously increases, approaching the target conductance value. The difference between the two becomes smaller and smaller. The input module will autonomously and continuously adjust the voltage value applied to the resistive memory according to the feedback voltage signal until the conductance value of the resistive memory reaches near the target conductance value. At this time, the voltage applied to the resistive memory will drop to 0, and the writing process will terminate. The entire judgment process and the termination of the writing process are autonomously realized by the feedback loop through the feedback of electrical signals. Therefore, the resistive memory can be written to any simulated conductance state.

[0013] Furthermore, the resistive memory module in step 1) is composed of a single resistive memory, which is connected to the common node of the feedback module and the reference resistor module through switch S1; a reset voltage pulse is applied to the resistive memory, that is, the top electrode TE terminal of the resistive memory is grounded and a positive voltage pulse is applied to the bottom electrode BE terminal. After the reset operation is completed, the resistive memory is adjusted to a low conductivity state.

[0014] Furthermore, the reference resistor module in step 2) includes a series of analog switches and fixed resistors. The conductance value of the reference resistor module is adjusted by turning off different switches or different numbers of switches. This conductance value serves as the target conductance state to be written into the resistive memory.

[0015] Furthermore, the input module in step 3) includes an input power supply, a fixed resistor, and an analog inverter constructed from an operational amplifier OA. The input power supply is directly connected to the inverting input resistor of the analog inverter through a fixed resistor, and the output of the analog inverter is connected to a resistive memory. The input power supply provides a DC voltage of a certain amplitude to the circuit to adjust the output voltage of the analog inverter, i.e., the initial voltage value applied to the resistive memory.

[0016] Furthermore, the feedback module in the feedback loop of step 4) consists of another operational amplifier OA and a fixed resistor on the output feedback loop. The OA operates as a positive feedback amplifier. The input of the feedback module is directly connected to the output of the reference resistor module and connected to the resistive memory module through switch S1. The fixed resistor on the output feedback loop is directly connected to the inverting input resistor of the analog inverter in the input module. The feedback module is used to sense the difference in conductance between the memory and the reference resistor and to provide direct feedback of the voltage signal, forming a feedback loop. Closing switch S1 allows the input module, the reference resistor module, the resistive memory module, and the feedback module to form a complete feedback loop in the continuous time domain.

[0017] Furthermore, the resistive memory is a resistive switching memory, a phase-change memory, a magnetic memory, or a ferroelectric memory.

[0018] This invention also applies the above method to resistive memory arrays, using the resistive memory as a selected single device within the array, and applying the feedback loop in the continuous time domain to the analog conductivity state write circuit of the resistive memory array. The analog conductivity state write circuit of the resistive memory array includes an N×N 1T1R array, a control circuit, the feedback loop in the continuous time domain, drive circuits for word lines WL / bit lines BL / source lines SL, and a multiplexer. The multiplexers near the bit lines BL and SL are equivalent to the external interfaces of the resistive memory modules. The word line WL corresponding to the resistive memory determines whether the resistive memory module is connected to the feedback loop. The control circuit is responsible for sending control signals to each module, controlling the connectivity and operating mode between modules. Applying the above method to the resistive memory array includes the following steps:

[0019] S1) Perform a Reset operation on the selected resistive memory.

[0020] First, select a resistive memory in the 1T1R array. Then, input a control signal through the control circuit to open the corresponding WL of the resistive memory and connect the corresponding SL and BL to the multiplexer. Apply a Reset voltage pulse to reset the selected resistive memory to a low conductivity state.

[0021] S2) Adjust the reference resistor module and select the target conductance state.

[0022] The control circuit transmits control signals to the reference resistor module, and adjusts the conductance value of the reference resistor module by turning off different switches or different numbers of switches. This conductance value is used as the target conductance state to be written into the resistive memory.

[0023] S3) Controlled by the control circuit, it adjusts the input voltage of the input module in the feedback loop to provide the initial voltage drop of the memory;

[0024] S4) The SL and BL corresponding to the memory are connected to the feedback loop through the multiplexer, the corresponding WL is opened, and the selected resistive memory is connected to the feedback loop to form a complete feedback loop in the continuous time domain.

[0025] S5) The selected resistive memory will be written to the vicinity of the target conductance state in a continuous time, and will realize the function of autonomous judgment and termination.

[0026] By repeating the above steps S1) to S5), all the memories in the array can be written to the set analog conductance state.

[0027] The beneficial effects of this invention are as follows:

[0028] This invention discloses a high-speed method and application for writing analog conductance states to resistive memory. Based on a feedback loop in the continuous-time domain consisting of an electrical signal connection input module, a reference resistor module, a resistive memory module, and a feedback module, the method can program the resistive memory to near any analog conductance value at extremely high write speeds within a continuous time period. Compared to traditional write-verify methods, this circuit eliminates the need for iterative processes, read-verify processes, and multiple write rounds, saving the overhead of analog-to-digital converters, logic judgment circuits, and other circuits, significantly improving the device programming speed and energy efficiency. Compared to other self-terminating one-step write schemes, this method greatly increases the programmable conductance states of the device, helping to further leverage the advantages of analog in-memory computing technology. Attached Figure Description

[0029] Figure 1 This is a flowchart of the high-speed writing method for the analog conductivity state of the resistive memory of the present invention;

[0030] Figure 2 This is a block diagram of the feedback loop used in the high-speed writing method of the analog conductivity state of the resistive memory of the present invention.

[0031] Figure 3 This is the feedback loop circuit diagram and its principle used in the high-speed writing method of the analog conductivity state of the resistive memory of the present invention;

[0032] Figure 4 This is a schematic diagram of the high-speed writing method of the analog conductivity state of the resistive memory of the present invention applied to a variable resistor array. Detailed Implementation

[0033] To more clearly illustrate the objectives, technical solutions, and advantages of this invention, a further detailed description is provided below in conjunction with the accompanying drawings. The description herein is merely illustrative and not intended to limit the scope of the invention.

[0034] This invention discloses a high-speed writing method for a resistive memory in an analog conductance state. The method is based on an analog circuit consisting of an input module, a reference resistor module, a resistive memory module, and a feedback module. The modules are connected by electrical signals to form a feedback loop in the continuous time domain. The resistive memory is autonomously programmed to reach any analog conductance state within a continuous time period. The writing process is then intelligently determined by the feedback signal to determine whether the target conductance state has been reached and to terminate the writing process. The flowchart is shown below. Figure 1 As shown, it includes the following steps:

[0035] 1) Reset the conductance state of the resistive memory in the Reset resistive memory module.

[0036] The resistive memory module consists of a single resistive memory, which is connected to the common node of the feedback module and the reference resistor module via switch S1. A Reset voltage pulse is applied to the resistive memory, that is, the top electrode TE terminal of the resistive memory is grounded and a positive voltage pulse is applied to the bottom electrode BE terminal. After the Reset operation is completed, the resistive memory is adjusted to a low conductivity state.

[0037] 2) Adjust the conductance value of the reference resistor module to select the target conductance state.

[0038] The reference resistor module includes a series of analog switches and fixed resistors. The conductance value of the reference resistor module is adjusted by turning off different switches or different numbers of switches. This conductance value serves as the target conductance state to be written into the resistive memory.

[0039] 3) The DC voltage supplied by the input power supply of the input module is used to adjust the initial voltage value applied to the resistive memory.

[0040] The input module includes an input power supply, a fixed resistor, and an analog inverter constructed from an operational amplifier OA. The input power supply is directly connected to the inverting input resistor of the analog inverter through a fixed resistor. The output of the analog inverter is connected to a resistive memory. The input power supply provides a DC voltage of a certain amplitude to the circuit to adjust the output voltage of the analog inverter, i.e., the initial voltage value applied to the resistive memory.

[0041] 4) Connecting a resistive memory forms a complete feedback loop in the continuous time domain.

[0042] The feedback module consists of another operational amplifier OA and a fixed resistor on the output feedback loop. The OA operates as a positive feedback amplifier. The input of the feedback module is directly connected to the output of the reference resistor module and to the resistive memory module through switch S1. The fixed resistor on the output feedback loop is directly connected to the inverting input resistor of the analog inverter in the input module. The feedback module is used to sense the difference in conductance between the memory and the reference resistor and to provide direct feedback of the electrical signal (voltage signal), forming a feedback loop. Closing switch S1 makes the input module, the reference resistor module, the resistive memory module, and the feedback module form a complete feedback loop in the continuous time domain.

[0043] 5) Program the resistive memory's conductance state to be near the target conductance state, and implement the function of autonomous judgment and termination;

[0044] In the feedback loop, the feedback module detects the conductance difference between the memory and the reference resistor and feeds this conductance difference back to the input module as a voltage signal through the output terminal, completing the electrical signal feedback process. During this process, the resistive memory is programmed, and its conductance value continuously increases, approaching the target conductance value. The difference between the two becomes smaller and smaller. The input module will autonomously and continuously adjust the voltage value applied to the resistive memory according to the feedback voltage signal until the conductance value of the resistive memory reaches near the target conductance value. At this time, the voltage applied to the resistive memory will drop to 0, which means that the writing process is terminated. The entire judgment process and the termination of the writing process are autonomously realized by the feedback loop through the feedback of electrical signals. Therefore, the resistive memory can be written to any simulated conductance state.

[0045] The resistive memory can be a resistive switching memory, a phase-change memory, a magnetic memory, or a ferroelectric memory.

[0046] Figure 2 This is a block diagram of the feedback loop used in the analog conductance state high-speed write method of resistive memory; this loop is... Figure 1The aforementioned feedback loop consists of an input module (S201), a reference resistor module (S202), a resistive memory module (S203), and a feedback module (S204). S201 defines the initial conditions applied to the resistive memory by the method, preventing it from being programmed to an extremely high conductance state within a picosecond timescale under high voltage conditions, ensuring that the time constant of the scheme matches the time constant of the resistive memory. Module S202 selects the conductance of the reference resistor; this conductance state serves as the target conductance state for the memory, meaning that the resistive memory in module S203 will be programmed to be near this conductance state. Module S204 provides direct feedback of the electrical signal (voltage signal), forming a feedback loop to determine the difference between the current conductance value and the target conductance value of the resistive memory. Once the target conductance value is reached, this module reduces or even removes the electrical stimulation applied to module S203, achieving autonomous judgment and termination. Notably, this programming process is implemented in a single step within a continuous timeframe, thus significantly improving the programming speed.

[0047] Figure 3 This invention presents the feedback loop circuit diagram and its principle for the high-speed writing method of the analog conductance state of the resistive memory. The reference resistor module consists of a series of analog switches and a fixed resistor. By selecting and closing different switches or different numbers of switches, the conductance value of the reference resistor can be adjusted. The input module consists of an input voltage, a fixed resistor, and an analog inverter constructed from an OA (Automatic Amplifier). The feedback module consists of an OA and a fixed resistor. The OA operates as a positive feedback amplifier (PFA). These modules are connected to form the feedback loop. In this invention, the target conductance value is denoted as G. target The conductance of the resistive memory is G. mem The conductance of the other fixed resistors is G0. Let V1, V2, V3, and V4 represent the input terminals of the inverter, the non-inverting input terminal of the PFA, the inverting input terminal of the inverter OA, and the output terminal, respectively. in V is the input voltage. out This is the output node of the PFA. The input module is connected to the feedback module and reference resistor module via node V1, and to the resistive memory module via node V4. The resistive memory module is connected to the common node V2 of the feedback module and reference resistor module via switch S1. When the loop is operating, if both OA are in the linear amplification region, according to the op-amp's "virtual short, virtual open" characteristics, V2 is 0, V1 = -V4, and the current flowing from node V2 to the amplifier input is 0. This means that the voltage drop across the reference resistor and the memory is the same, and the current flowing through them is also the same. Therefore, they should have the same or similar conductance values. Before the circuit reaches stability, i.e., before the memory's conductance approaches the target conductance value, the PFA operates in the saturation region. The circuit's dynamic equations should be analyzed, and G should be defined as...target / G mem =c,G target / G0=d, according to Kirchhoff's laws and Ohm's law, we can obtain:

[0048] (V in -V1)G0=(V1-V out )G0+(V1-V3)G0+(V1-V2)G target (V2-V4)G mem =(V1-V2)G target

[0049] (V1-V3)G0=(V3-V4)G0

[0050] The physical meanings of the above three equations are as follows: Input voltage V in The input current supplied to node V1 is equal to the sum of the currents flowing out of that node. The current flowing through the reference resistor and the memory is the same, and the current flowing through the two fixed resistors in the analog inverter is the same. Combining this with the open-loop transfer function equation of OA: V out =L(s)V2, V4 = -L(s)V3, where L0 is the DC gain factor, and ω0 is the 3-dB bandwidth. Combining the above five equations and omitting minor terms, we can obtain the dynamic equation of this circuit:

[0051]

[0052] This is equation (1) in the attached diagram. When c = 1, the coefficients of the last two terms in equation (1) become 0. At this time, the contribution of the preceding subterms to the second-order frequency domain equation cannot be ignored. Therefore, the dynamic equation of the circuit becomes:

[0053]

[0054] This is equation (2) in the attached diagram. According to the dynamic equation of the circuit, when c≤1, that is, when the conductance of the memory is greater than the target conductance value, the poles of the circuit are distributed on the left side of the complex plane. At this time, the voltage values ​​of nodes V2 and V4 will become 0, and both OA are in the linear amplification region. This means that the circuit can automatically reduce the voltage difference between its two ends to 0 when the variable resistor reaches near the target conductance, thus playing the function of automatic judgment and termination. When c>1, that is, when the conductance of the variable resistor is less than the target conductance value, the poles of the circuit are distributed on the right side of the complex plane. At this time, the voltage of node V2 is not 0, which means that PFA is in the saturation region. out =V DD V DD This is the circuit's supply voltage. According to Kirchhoff's current law, the voltage drop across the variable resistor at this time can be calculated as follows: At this point, the circuit is being programmed to operate on the variable resistor. Therefore, before the variable resistor is connected to the circuit, the value of capacitor c is infinite, and the voltage difference between nodes V2 and V4 is... Therefore, V can be adjusted. in To adjust the initial voltage drop of the circuit to accommodate different types of resistive memory.

[0055] Figure 4 This diagram illustrates the application of the high-speed analog conductance state writing method for resistive memory of this invention to a variable resistor array. It comprises a widely used N×N 1-transistor-1-resistor (1T1R) resistive memory array, an input circuit, a control circuit, the feedback loop proposed in this invention, drive circuits for WL / BL / SL, and a multiplexer. The control circuit is responsible for sending control signals to each module, controlling the connectivity and operating modes between modules. During operation, the control circuit transmits the digital signal corresponding to the target conductance state to the feedback loop, adjusting the conductance state of the reference resistor module to the target conductance value. Then, the control circuit adjusts the input voltage of the input module in the feedback loop to provide the initial voltage drop of the memory. Next, the control circuit transmits the address signal to the multiplexer of WL / BL / SL, connecting the corresponding resistive memory to the feedback loop and writing the resistive memory to the target analog conductance state. Repeating this process sequentially writes all the memories in the array to the set analog conductance states. The input circuit and ADC module are used for voltage signal input and output signal readout in subsequent calculation mode, and do not participate in the programming process of the 1T1R array.

[0056] The embodiments described above are not intended to limit the present invention. Any person skilled in the art can make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, the scope of protection of the present invention is defined by the scope of the claims.

Claims

1. A method for high-speed writing of a resistive memory in simulated conductance state, characterized in that, This method is based on an input module, An analog circuit consisting of a reference resistor module, a resistive memory module, and a feedback module is used to connect the modules based on electrical signals, forming a feedback loop in the continuous time domain. This allows the resistive memory to be autonomously programmed to reach any simulated conductance state within a continuous time period. The feedback signal intelligently determines whether the target conductance state has been reached and terminates the writing process accordingly. The circuit includes the following steps: 1) Reset the conductivity state of the resistive memory; The resistive memory module consists of a single resistive memory, which is connected to the common node of the feedback module and the reference resistor module via switch S1. A reset voltage pulse is applied to the resistive memory, that is, the top electrode TE terminal of the resistive memory is grounded and a positive voltage pulse is applied to the bottom electrode BE terminal. After the reset operation is completed, the resistive memory is adjusted to a low conductivity state. 2) Adjust the conductance value of the reference resistor module to select the target conductance state; 3) The DC voltage provided by the input power supply of the input module is used to adjust the initial voltage value applied to the resistive memory; The input module includes an input power supply, a fixed resistor, and an analog inverter constructed from an operational amplifier OA. The input power supply is directly connected to the inverting input resistor of the analog inverter through a fixed resistor. The output of the analog inverter is connected to a resistive memory. The input power supply provides a DC voltage of a certain amplitude to the circuit to adjust the output voltage of the analog inverter, i.e., the initial voltage value applied to the resistive memory. 4) Connecting a resistive memory forms a complete feedback loop in the continuous time domain; 5) Program the resistive memory's conductance state to be near the target conductance state, and implement the function of autonomous judgment and termination; In the feedback loop within the continuous time domain, the feedback module detects the conductance difference between the memory and the reference resistor and feeds this conductance difference back to the input module as a voltage signal through the output terminal, completing the electrical signal feedback process. During this process, the resistive memory is programmed, and its conductance value continuously increases, approaching the target conductance value. The difference between the two becomes smaller and smaller. The input module will autonomously and continuously adjust the voltage value applied to the resistive memory according to the feedback voltage signal until the conductance value of the resistive memory reaches near the target conductance value. At this time, the voltage applied to the resistive memory will drop to 0V, and the writing process will terminate. The entire judgment process and the termination of the writing process are autonomously realized by the feedback loop through the feedback of electrical signals. Therefore, the resistive memory can be written to any simulated conductance state.

2. The high-speed writing method for analog conductivity state of resistive memory as described in claim 1, characterized in that, The reference resistor module in step 2) includes a series of analog switches and fixed resistors. The conductance value of the reference resistor module is adjusted by turning off different switches or different numbers of switches. This conductance value serves as the target conductance state to be written into the resistive memory.

3. The high-speed writing method for analog conductivity state of resistive memory as described in claim 1, characterized in that, The feedback module in the feedback loop of step 4) consists of another operational amplifier OA and a fixed resistor on the output feedback loop. The OA operates as a positive feedback amplifier. The input of the feedback module is directly connected to the output of the reference resistor module and connected to the resistive memory module through switch S1. The fixed resistor on the output feedback loop is directly connected to the inverting input resistor of the analog inverter in the input module. The feedback module is used to sense the difference in conductance between the memory and the reference resistor and to provide direct feedback of the voltage signal, forming a feedback loop. Closing switch S1 causes the input module, the reference resistor module, the resistive memory module, and the feedback module to form a complete feedback loop in the continuous time domain.

4. The high-speed writing method for analog conductivity state of resistive memory as described in claim 1, characterized in that, The resistive memory is a resistive switching memory, a phase-change memory, a magnetic memory, or a ferroelectric memory.

5. A high-speed writing method for a resistive memory array in simulated conductance state, characterized in that, Using the resistive memory as described in claim 1 as a selected single device in a resistive memory array, the feedback loop in the continuous-time domain as described in claim 1 is applied to the analog conductivity state write circuit of the resistive memory array; the analog conductivity state write circuit of the resistive memory array includes a The 1T1R array, control circuit, feedback loop in the continuous time domain, drive circuit for word line WL / bit line BL / source line SL, and multiplexer, the multiplexer near bit line BL and source line SL is equivalent to the external interface of the resistive memory module, the word line WL corresponding to the resistive memory determines whether the resistive memory module is connected to the feedback loop, the control circuit is responsible for sending control signals to each module, controlling the connection and working mode between modules; applying the analog conductivity state high-speed writing method of the resistive memory as described in claim 1 to the resistive memory array includes the following steps: S1) Perform a Reset operation on the selected resistive memory; First, select a resistive memory in the 1T1R array. Then, input a control signal through the control circuit to open the corresponding WL of the resistive memory and connect the corresponding SL and BL to the multiplexer. Apply a Reset voltage pulse to reset the selected resistive memory to a low conductivity state. S2) Adjust the reference resistor module to select the target conductance state; The control circuit transmits control signals to the reference resistor module, and adjusts the conductance value of the reference resistor module by turning off different switches or different numbers of switches. This conductance value is used as the target conductance state to be written into the resistive memory. S3) Controlled by the control circuit, the input voltage of the input module in the feedback loop is adjusted to provide the initial voltage drop of the memory; S4) The SL and BL corresponding to the memory are connected to the feedback loop through the multiplexer, the corresponding WL is opened, and the selected resistive memory is connected to the feedback loop to form a complete feedback loop in the continuous time domain. S5) The selected resistive memory will be written to the vicinity of the target conductance state in a continuous time, and the function of autonomous judgment and termination will be realized. By repeating the above steps S1) to S5), all the memories in the array can be written to the set analog conductance state.

Images