Semiconductor device and electronic apparatus

By optimizing the design of the computing unit and current generation circuit, the power consumption and area problems in the analog product sum calculation circuit are solved, and efficient continuous computing is achieved.

CN119032347APending Publication Date: 2024-11-26SEMICON ENERGY LAB CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202380034197.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Priority Date
2022-04-22
Filing Date
2023-04-10
Publication Date
2024-11-26

AI Technical Summary

Technical Problem

When existing analog product sum operation circuits are used in artificial neural networks, power consumption increases due to the digital analog conversion process, and multiple digital analog conversion circuits and analog digital conversion circuits are required, resulting in a larger circuit area.

Method used

A semiconductor device is designed, including a computing unit, a switching circuit and a current generation circuit. By optimizing the structure of the current generation and switching circuit, continuous computing is realized, and the number of conversions is reduced, and power consumption is reduced.

Benefits of technology

It realizes the reduction of power consumption and circuit area, supports continuous computing, and improves computing efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119032347A_ABST
    Figure CN119032347A_ABST
Patent Text Reader

Abstract

Provided is a semiconductor device having a small circuit scale and reduced power consumption. A semiconductor device includes first to fourth cells, first and second circuits, and first to fourth current generating circuits. The first unit is electrically connected to the third unit through a first wiring and a first current generation circuit, and is electrically connected to the first circuit through a second wiring. The second unit is electrically connected to the fourth unit through a third wiring and a second current generation circuit, and is electrically connected to the second circuit through a fourth wiring. The third unit is electrically connected to the second unit through a third current generation circuit and a fourth wiring. The fourth unit is electrically connected to the first unit through a fourth current generation circuit and a second wiring. Note that the first and second current generation circuits are used as current mirror circuits, and the third and fourth current generation circuits are used as function-type arithmetic circuits. A product operation is performed in the first and second units, and the result of the operation is held in the third and fourth units.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] One embodiment of the present invention relates to a semiconductor device and an electronic device.

[0002] In addition, one embodiment of the present invention is not limited to the above-mentioned technical field. The technical field of the invention disclosed in this specification and the like relates to an object, a driving method or a manufacturing method. In addition, one embodiment of the present invention relates to a process, a machine, a product or a composition of matter. Therefore, specifically, as examples of the technical field of one embodiment of the present invention disclosed in this specification, semiconductor devices, display devices, liquid crystal display devices, light-emitting devices, power storage devices, camera devices, storage devices, signal processing devices, processors, electronic devices, systems, their driving methods, their manufacturing methods or their inspection methods can be cited. Background Art

[0003] Nowadays, the development of integrated circuits modeled on the structure of the human brain is becoming increasingly popular. The integrated circuit is assembled with electronic circuits that are the structure of the brain and include circuits that are equivalent to the "neurons" and "synapses" of the human brain. Therefore, such integrated circuits are sometimes called "neuro-morphic", "brain-morphic" or "brain-inspired". The integrated circuit has a non-Neumann architecture, and compared with the Neumann architecture, which has higher power consumption as the processing speed increases, it can be expected to perform parallel processing with extremely low power consumption.

[0004] A data processing model that mimics a neural network and includes "neurons" and "synapses" is called an artificial neural network (ANN). For example, Non-Patent Documents 1 and 2 disclose a computing device that uses SRAM (Static Random Access Memory) to form an artificial neural network.

[0005] In addition, there are studies on using a computing device constituting an artificial neural network, for example, to correct an image displayed on a display device. For example, Patent Document 1 discloses a display device in which a computing circuit constituting an artificial neural network is used to adjust the brightness, color tone, etc. of a displayed image according to the preferences of a person viewing the image. [Prior technical literature] [Patent Document]

[0006] [Patent Document 1] Japanese Patent Application Publication No. 2018-36639 [Non-patent literature]

[0007] [Non-patent document 1] M. Kang et al., “IEEE Journal of Solid-State Circuits”, 2018, Volume 53, No. 2, p. 642-655. [Non-patent document 2] J. Zhang et al., “IEEE Journal of Solid-State Circuits”, 2017, Volume 52, No. 4, p. 915-924. Summary of the invention Technical problem to be solved by the invention

[0008] As an operation circuit constituting an artificial neural network, for example, an operation circuit that adds together analog currents corresponding to the products of weight coefficients and input data to perform a product-sum operation (hereinafter sometimes referred to as an analog product-sum operation circuit) can be cited. Since the operation circuit uses analog currents in the operation, the circuit scale can be reduced compared to an operation circuit constituted by a digital circuit, thereby reducing the circuit area. In addition, by designing the operation circuit in such a way that the analog current processed in the operation becomes smaller, the power consumption of the operation circuit can be reduced.

[0009] In addition, when using the analog product-sum operation circuit, it is necessary to convert the digital data input to the analog product-sum operation circuit into an analog current. In addition, since the operation result of the analog product-sum operation circuit is output as analog data, it is necessary to convert the analog data into digital data when processing the operation result in the digital circuit. In other words, in order to use the analog product-sum operation circuit, a digital-to-analog conversion circuit (digital voltage (signal)-analog current conversion circuit, IDAC) and an analog-to-digital conversion circuit (ADC) are required.

[0010] In particular, since multiple product-sum operations are sometimes performed in an artificial neural network, digital-to-analog conversion and analog-to-digital conversion are performed each time a product-sum operation is performed, and thus power consumption caused by these conversion processes is sometimes increased. In addition, when multiple product-sum operations are performed, it is preferred to use multiple analog product-sum operation circuits, but the same number of digital-to-analog conversion circuits and analog-to-digital conversion circuits as the analog product-sum operation circuits are required, respectively, so the circuit area is sometimes increased.

[0011] In addition, one of the purposes of one embodiment of the present invention is to provide a semiconductor device with reduced power consumption. In addition, one of the purposes of one embodiment of the present invention is to provide a semiconductor device with a small circuit area. In addition, one of the purposes of one embodiment of the present invention is to provide a semiconductor device that can perform operations continuously. In addition, one of the purposes of one embodiment of the present invention is to provide a novel semiconductor device. In addition, one of the purposes of one embodiment of the present invention is to provide an electronic device including the above-mentioned semiconductor device.

[0012] Note that the purpose of one embodiment of the present invention is not limited to the above purpose. The above purpose does not prevent the existence of other purposes. In addition, other purposes are purposes that are not mentioned above but will be described in the following description. Those skilled in the art can derive and appropriately extract the purposes not mentioned above from the description of the specification or drawings, etc. In addition, one embodiment of the present invention achieves at least one of the above purpose and other purposes. In addition, one embodiment of the present invention does not need to achieve all of the above purpose and other purposes. Solutions to technical problems (1) One embodiment of the present invention is a semiconductor device including a first unit, a second unit, a third unit, a fourth unit, a first circuit, a second circuit, a first current generating circuit, a second current generating circuit, a third current generating circuit, and a fourth current generating circuit.

[0014] The first unit is electrically connected to the first input terminal of the first current generating circuit through the first wiring, and the first unit is electrically connected to the first circuit and the fourth output terminal of the fourth current generating circuit through the second wiring. The second unit is electrically connected to the second input terminal of the second current generating circuit through the third wiring, and the second unit is electrically connected to the second circuit and the third output terminal of the third current generating circuit through the fourth wiring. The third unit is electrically connected to the first output terminal of the first current generating circuit and the third input terminal of the third current generating circuit. The fourth unit is electrically connected to the second output terminal of the second current generating circuit and the fourth input terminal of the fourth current generating circuit.

[0015] The first circuit has a function of generating a first current and outputting it to the second wiring. In addition, the second circuit has a function of generating a second current and outputting it to the fourth wiring. In addition, the first current generating circuit has a function of outputting a current from the first output terminal as a current mirror circuit, and the current amount corresponds to the current amount flowing through the first input terminal. In addition, the second current generating circuit has a function of outputting a current from the second output terminal as a current mirror circuit, and the current amount corresponds to the current amount flowing through the second input terminal. In addition, the third current generating circuit has a function of outputting a third current from the third output terminal as a function-type operation circuit, and the third current amount corresponds to the current amount flowing through the third input terminal. In addition, the fourth current generating circuit has a function of outputting a fourth current from the fourth output terminal as a function-type operation circuit, and the fourth current amount corresponds to the current amount flowing through the fourth input terminal.

[0016] The first unit has a function of maintaining a potential corresponding to the first data and a function of generating a fifth current and outputting it to the first wiring, the fifth current amount being equivalent to the product of the value of the first data and the value corresponding to the first current or the fourth current flowing through the second wiring. In addition, the second unit has a function of maintaining a potential corresponding to the second data and a function of generating a sixth current and outputting it to the third wiring, the sixth current amount being equivalent to the product of the value of the second data and the value corresponding to the second current or the third current flowing through the fourth wiring. In addition, the third unit has a function of maintaining a potential corresponding to the fifth current. In addition, the fourth unit has a function of maintaining a potential corresponding to the sixth current. (2) In the above (1), one embodiment of the present invention may include a first switching circuit, a second switching circuit, a third switching circuit, a fourth switching circuit, and a fifth switching circuit.

[0018] In particular, the first switching circuit preferably includes a first terminal, a second terminal, a third terminal and a fourth terminal. In addition, the second switching circuit preferably includes a fifth terminal, a sixth terminal and a seventh terminal. In addition, the third switching circuit preferably includes an eighth terminal, a ninth terminal and a tenth terminal. The fourth switching circuit preferably includes an eleventh terminal, a twelfth terminal and a thirteenth terminal. The fifth switching circuit preferably includes a fourteenth terminal, a fifteenth terminal and a sixteenth terminal.

[0019] Preferably, the first terminal of the first switching circuit is electrically connected to the first wiring, the second terminal of the first switching circuit is electrically connected to the first input terminal of the first current generating circuit, the third terminal of the first switching circuit is electrically connected to the third wiring, and the fourth terminal of the first switching circuit is electrically connected to the second input terminal of the second current generating circuit. In addition, preferably, the fifth terminal of the second switching circuit is electrically connected to the first output terminal of the first current generating circuit, the sixth terminal of the second switching circuit is electrically connected to the third unit, and the seventh terminal of the second switching circuit is electrically connected to the third input terminal of the third current generating circuit. In addition, preferably, the eighth terminal of the third switching circuit is electrically connected to the second output terminal of the second current generating circuit, the ninth terminal of the third switching circuit is electrically connected to the fourth unit, and the tenth terminal of the third switching circuit is electrically connected to the fourth input terminal of the fourth current generating circuit. In addition, preferably, the eleventh terminal of the fourth switching circuit is electrically connected to the first circuit, the twelfth terminal of the fourth switching circuit is electrically connected to the second wiring, and the thirteenth terminal of the fourth switching circuit is electrically connected to the fourth output terminal of the fourth current generating circuit. In addition, preferably, the fourteenth terminal of the fifth switching circuit is electrically connected to the second circuit, the fifteenth terminal of the fifth switching circuit is electrically connected to the fourth wiring, and the sixteenth terminal of the fifth switching circuit is electrically connected to the third output terminal of the third current generating circuit.

[0020] The first switching circuit preferably has a function of making the first terminal and the second terminal in a conductive state or a non-conductive state, a function of making the first terminal and the third terminal in a conductive state or a non-conductive state, and a function of making the third terminal and the fourth terminal in a conductive state or a non-conductive state. In addition, the second switching circuit preferably has a function of making the fifth terminal and the sixth terminal in a conductive state or a non-conductive state, and a function of making the sixth terminal and the seventh terminal in a conductive state or a non-conductive state. In addition, the third switching circuit preferably has a function of making the eighth terminal and the ninth terminal in a conductive state or a non-conductive state, and a function of making the ninth terminal and the tenth terminal in a conductive state or a non-conductive state. In addition, the fourth switching circuit preferably has a function of making the eleventh terminal and the twelfth terminal in a conductive state or a non-conductive state, and a function of making the twelfth terminal and the thirteenth terminal in a conductive state or a non-conductive state. In addition, the fifth switching circuit preferably has a function of making the fourteenth terminal and the fifteenth terminal in a conductive state or a non-conductive state, and a function of making the fifteenth terminal and the sixteenth terminal in a conductive state or a non-conductive state. (3) In addition, in the above (2), one embodiment of the present invention may also include a third circuit. In particular, the third circuit is preferably electrically connected to the first wiring. In addition, the third circuit preferably has a function of causing a seventh current corresponding to the first data to flow through the first unit and a function of causing an eighth current corresponding to the second data to flow through the second unit. (4) In the above (3), one embodiment of the present invention may include a fifth unit and a sixth unit. In particular, the first unit, the second unit, the fifth unit, and the sixth unit preferably include a first transistor, a second transistor, and a first capacitor, respectively.

[0023] Preferably, in each of the first unit, the second unit, the fifth unit and the sixth unit, one of the source and the drain of the first transistor is electrically connected to the gate of the second transistor, one of the source and the drain of the second transistor is electrically connected to one of the source and the drain of the first transistor, and one of the pair of terminals of the first capacitor is electrically connected to the gate of the second transistor.

[0024] In addition, in the first unit, it is preferred that the other of the source and the drain of the first transistor is electrically connected to the first wiring, and the other of the pair of terminals of the first capacitor is electrically connected to the second wiring. In addition, in the second unit, it is preferred that the other of the source and the drain of the first transistor is electrically connected to the third wiring, and the other of the pair of terminals of the first capacitor is electrically connected to the fourth wiring.

[0025] In addition, in the fifth unit, it is preferred that the other of the source and the drain of the first transistor is electrically connected to the second wiring, and the other of the pair of terminals of the first capacitor is electrically connected to the second wiring. In addition, in the sixth unit, it is preferred that the other of the source and the drain of the first transistor is electrically connected to the fourth wiring, and the other of the pair of terminals of the first capacitor is electrically connected to the fourth wiring.

[0026] In addition, the channel formation region of each of the first transistor and the second transistor preferably includes a first oxide semiconductor. In addition, the first oxide semiconductor preferably includes one or more selected from indium, zinc and element M. Element M is selected from gallium, aluminum, silicon, boron, yttrium, tin, copper, vanadium, beryllium, titanium, iron, nickel, germanium, zirconium, molybdenum, lanthanum, cerium, neodymium, hafnium, tantalum, tungsten, cobalt, magnesium and antimony. One or more. (5) In the above (4), one embodiment of the present invention may have a structure in which the third unit and the fourth unit each include a third transistor, a fourth transistor, a fifth transistor, and a second capacitor.

[0028] Preferably, in each of the third unit and the fourth unit, one of the source and the drain of the third transistor is electrically connected to the gate of the fourth transistor, one of the source and the drain of the fourth transistor is electrically connected to one of the source and the drain of the fifth transistor, the other of the source and the drain of the fifth transistor is electrically connected to the other of the source and the drain of the third transistor, and one of the pair of terminals of the first capacitor is electrically connected to the gate of the fourth transistor.

[0029] In the third unit, the other of the source and drain of the third transistor is preferably electrically connected to the sixth terminal of the second switching circuit. In addition, in the fourth unit, the other of the source and drain of the third transistor is preferably electrically connected to the ninth terminal of the third switching circuit.

[0030] In addition, the channel formation region of each of the third transistor and the fourth transistor preferably includes a second oxide semiconductor. In addition, the second oxide semiconductor preferably includes one or more selected from indium, zinc, and the element M described in (4). (6) In addition, in the above (5), one embodiment of the present invention may include a first layer, a second layer located above the first layer, and a third layer located above the second layer. In particular, the first layer preferably includes a first circuit, a second circuit, a third circuit, a first current generating circuit, a second current generating circuit, a third current generating circuit, and a fourth current generating circuit, the second layer preferably includes a third unit and a fourth unit, and the third layer preferably includes a first unit, a second unit, a fifth unit, and a sixth unit. (7) In addition, in the above (5), one embodiment of the present invention may include a first layer, a second layer located above the first layer, and a third layer located above the second layer, and include a fourth layer located above the third layer and a fifth layer located above the fourth layer. In particular, the first layer preferably includes a first circuit, a second circuit, a third circuit, a first current generating circuit, a second current generating circuit, a third current generating circuit, and a fourth current generating circuit, the second layer preferably includes a third unit, the third layer preferably includes a fourth unit, the fourth layer preferably includes a first unit and a fifth unit, and the fifth layer includes a second unit and a sixth unit. (8) Another embodiment of the present invention is an electronic device including any one of the semiconductor devices described in (1) to (7) above and a housing. Effects of the Invention

[0034] According to one embodiment of the present invention, a semiconductor device with reduced power consumption can be provided. In addition, according to one embodiment of the present invention, a semiconductor device with a small circuit area can be provided. In addition, according to one embodiment of the present invention, a semiconductor device capable of performing operations continuously can be provided. In addition, according to one embodiment of the present invention, a novel semiconductor device can be provided. In addition, according to one embodiment of the present invention, an electronic device including the above-mentioned semiconductor device can be provided.

[0035] Note that the effects of one embodiment of the present invention are not limited to the above effects. The effects listed above do not prevent the existence of other effects. In addition, other effects are effects that are not mentioned above but will be described in the following description. Those skilled in the art can derive and appropriately extract the effects that are not mentioned above from the description of the specification or drawings, etc. In addition, one embodiment of the present invention has at least one of the above effects and other effects. Therefore, one embodiment of the present invention sometimes does not have the effects listed above depending on the circumstances. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 is a block diagram illustrating a structural example of a semiconductor device. Figure 2 A circuit diagram illustrating a configuration example of a circuit included in a semiconductor device. Figure 3 A circuit diagram illustrating a configuration example of a circuit included in a semiconductor device. Figure 4A and Figure 4B A circuit diagram illustrating a configuration example of a circuit included in a semiconductor device. FIG. 5A to FIG. 5C A circuit diagram illustrating a configuration example of a circuit included in a semiconductor device. FIG. 6A to FIG. 6D A circuit diagram illustrating a configuration example of a circuit included in a semiconductor device. 7A to 7C A circuit diagram illustrating a configuration example of a circuit included in a semiconductor device. Fig. 8A and Figure 8B A circuit diagram illustrating a configuration example of a circuit included in a semiconductor device. 9A to 9C A circuit diagram illustrating a configuration example of a circuit included in a semiconductor device. Fig.10 A circuit diagram illustrating a configuration example of a circuit included in a semiconductor device. Fig.11A and Fig. 11B A circuit diagram illustrating a configuration example of a circuit included in a semiconductor device. Fig.12 It is a timing chart for explaining an operation example of a semiconductor device. FIG. 13A to FIG. 13C This is a block diagram illustrating an operation example of a semiconductor device. FIG. 14A to FIG. 14C This is a block diagram illustrating an operation example of a semiconductor device. Fig.15A and Fig. 15B This is a block diagram illustrating an operation example of a semiconductor device. Fig.16This is a flowchart illustrating an example of the operation of a semiconductor device. Fig.17 is a block diagram illustrating a structural example of a semiconductor device. Fig.18A is a perspective schematic diagram showing a structural example of a semiconductor device, Fig.18B is a block diagram showing a structural example of a semiconductor device. Fig.19 is a schematic cross-sectional view showing a structural example of a semiconductor device. Fig. 20 is a schematic cross-sectional view showing a structural example of a semiconductor device. Fig.21 It is a perspective schematic diagram showing a structural example of a semiconductor device. Fig. 22 is a block diagram showing a structural example of a semiconductor device. Fig.23A and Fig. 23B It is a diagram showing an example of an electronic component. Fig.24A and Fig. 24B is a diagram showing an example of an electronic device, FIG. 24C to FIG. 24E This is a diagram showing an example of a large computer. Fig.25 is a diagram showing an example of space equipment. Fig.26 is a diagram showing an example of a storage system that can be used in a data center. Modes for Carrying Out the Invention

[0037] In this specification, etc., a semiconductor device refers to a device that utilizes semiconductor characteristics, a circuit including a semiconductor element (e.g., a transistor, a diode, a photodiode), and a device including the circuit. In addition, a semiconductor device refers to all devices that can function by utilizing semiconductor characteristics. For example, an integrated circuit, a chip having an integrated circuit, and an electronic component containing a chip in a package are all examples of a semiconductor device. In addition, for example, a storage device, a display device, a light-emitting device, a lighting device, and an electronic device are sometimes semiconductor devices themselves, or sometimes include a semiconductor device.

[0038] In addition, in this specification, when it is described that "X is connected to Y", it means that the following cases are disclosed in this specification, etc.: a case where X is electrically connected to Y; a case where X is functionally connected to Y; and a case where X is directly connected to Y. Therefore, it is not limited to the connection relationship shown in the drawings or text, and other connection relationships are also within the scope described in the drawings or text. X and Y are objects (for example, devices, elements, circuits, wirings, electrodes, terminals, conductive films, and layers).

[0039] As an example of a case where X and Y are electrically connected, one or more elements (such as switches, transistors, capacitors, inductors, resistors, diodes, display devices, light-emitting devices, loads, etc.) that can electrically connect X and Y may be connected between X and Y. In addition, the switch has a function of controlling on or off. In other words, whether current flows is controlled by making the switch in a conducting state (on state) or a non-conducting state (off state).

[0040] As an example of a case where X and Y are functionally connected, for example, one or more circuits capable of functionally connecting X and Y (for example, a logic circuit (for example, an inverter, a NAND circuit, a NOR circuit), a signal conversion circuit (for example, a digital-to-analog conversion circuit, an analog-to-digital conversion circuit, a gamma correction circuit), a potential level conversion circuit (for example, a power supply circuit such as a boost circuit and a buck circuit), a level shift circuit that changes the potential level of a signal), a voltage source, a current source, a switching circuit, an amplifier circuit (for example, a circuit capable of increasing signal amplitude or current amount, an operational amplifier, a differential amplifier circuit, a source follower circuit, a buffer circuit), a signal generation circuit, a storage circuit, a control circuit, etc.) may be connected between X and Y. Note that, for example, even if another circuit is interposed between X and Y, when a signal output from X is transmitted to Y, X and Y can be said to be functionally connected.

[0041] Furthermore, when it is clearly stated that “X is electrically connected to Y”, the following cases are included: a case where X is electrically connected to Y (in other words, a case where X and Y are connected with another element or another circuit interposed therebetween); and a case where X and Y are directly connected (in other words, a case where X and Y are connected without another element or another circuit interposed therebetween).

[0042] In addition, in this specification, a circuit structure in which a wiring (wiring for supplying a constant potential or wiring for transmitting a signal) is electrically connected to a plurality of elements is adopted. For example, in this specification, a case in which X is directly connected to a wiring and Y is directly electrically connected to the wiring is sometimes described as "X and Y are directly connected".

[0043] In addition, for example, it can be expressed as "X, Y, the source of the transistor (sometimes referred to as one of the first terminal and the second terminal) and the drain of the transistor (sometimes referred to as the other of the first terminal and the second terminal) are electrically connected to each other, and X, the source of the transistor, the drain of the transistor, and Y are electrically connected in sequence." Or, it can be expressed as "the source of the transistor is electrically connected to X, the drain of the transistor is electrically connected to Y, and X, the source of the transistor, the drain of the transistor, and Y are electrically connected in sequence." Or, it can be expressed as "X is electrically connected to Y through the source and drain of the transistor, and X, the source of the transistor, the drain of the transistor, and Y are connected to each other in sequence." By specifying the connection order in the circuit structure using the same expression method as these examples, the source and drain of the transistor can be distinguished to determine the technical scope. Note that this expression method is an example and is not limited to the above expression method. Here, X and Y are objects (for example, devices, elements, circuits, wiring, electrodes, terminals, conductive films or layers, etc.).

[0044] In addition, even if independent components are electrically connected to each other on a circuit diagram, one component may have the functions of multiple components. For example, when a portion of a wiring is used as an electrode, one conductive film has the functions of both wiring and electrode. Therefore, the category of "electrically connected" in this specification also includes the case where one conductive film has the functions of multiple components.

[0045] In this specification, etc., a "resistance element" may be, for example, a circuit element having a resistance value higher than 0Ω or a wiring having a resistance value higher than 0Ω. Therefore, in this specification, etc., a "resistance element" includes a wiring having a resistance value, a transistor, a diode, or a coil through which current flows between a source and a drain. Therefore, a "resistance element" may sometimes be referred to as a "resistance", "load", or "region having a resistance value". In contrast, "resistance", "load", or "region having a resistance value" etc. may sometimes be referred to as a "resistance element". As for the resistance value, for example, it is preferably 1 mΩ or more and 10 Ω or less, more preferably 5 mΩ or more and 5 Ω or less, and even more preferably 10 mΩ or more and 1 Ω or less. In addition, for example, it may be 1 Ω or more and 1×10 9 Ω or less.

[0046] In this specification, etc., a "capacitor element" may be, for example, a circuit element having an electrostatic capacitance value higher than 0F, a wiring area having an electrostatic capacitance value higher than 0F, a parasitic capacitance, or a gate capacitance of a transistor. In addition, "capacitor element", "parasitic capacitance" or "gate capacitance" etc. may sometimes be referred to as a "capacitor". In contrast, a "capacitor" may sometimes be referred to as a "capacitor element", "parasitic capacitance" or "gate capacitance". In addition, a "capacitor" (a "capacitor" including three or more terminals) includes an insulator and a pair of conductors clamping the insulator. Thus, the "pair of conductors" of the "capacitor" may be referred to as a "pair of electrodes", "a pair of conductive areas", "a pair of areas" or "a pair of terminals". In addition, "one of a pair of terminals" and "the other of a pair of terminals" are sometimes referred to as a first terminal and a second terminal, respectively. In addition, the electrostatic capacitance value may be, for example, greater than 0.05fF and less than 10pF. In addition, for example, it may also be greater than 1pF and less than 10μF.

[0047] In this specification, etc., a transistor includes three terminals: a gate, a source, and a drain. The gate is used as a control terminal to control the conduction state of the transistor. The two terminals used as the source or drain are the input and output terminals of the transistor. Depending on the conductivity type of the transistor (n-channel type or p-channel type) and the level of the potential applied to the three terminals of the transistor, one of the two input and output terminals is used as the source and the other is used as the drain. Therefore, in this specification, etc., the source and the drain can be interchanged. In this specification, etc., when describing the connection relationship of the transistor, the expressions "one of the source and the drain" (first electrode or first terminal) and "the other of the source and the drain" (second electrode or second terminal) are used. In addition, depending on the structure of the transistor, a back gate is sometimes included in addition to the above three terminals. In this case, in this specification, etc., one of the gate and the back gate of the transistor is sometimes referred to as the first gate, and the other of the gate and the back gate of the transistor is sometimes referred to as the second gate. In addition, in the same transistor, "gate" and "back gate" can sometimes be interchanged. In addition, when a transistor includes three or more gates, each gate may be referred to as a first gate, a second gate, a third gate, etc. in this specification and the like.

[0048] For example, in this specification, a multi-gate structure transistor having more than two gate electrodes can be used as an example of a transistor. When a multi-gate structure is used, since the channel forming regions are connected in series, a structure in which multiple transistors are connected in series is formed. Therefore, by adopting a multi-gate structure, the off-state current can be reduced, and the voltage resistance of the transistor can be improved (improving reliability). Alternatively, by utilizing a multi-gate structure, when the transistor operates in a saturation region, even if the voltage between the drain and the source changes, the change in the current between the drain and the source is not too large, so that a voltage-current characteristic with a flat tilt angle can be obtained. When utilizing a voltage-current characteristic with a flat tilt angle, an ideal current source circuit or an active load with an extremely high resistance value can be realized. As a result, a differential circuit or a current mirror circuit with good characteristics can be realized.

[0049] In addition, a circuit diagram showing a circuit element sometimes includes a case where the circuit element has a plurality of circuit elements. For example, a circuit diagram showing a resistor includes a case where two or more resistors are connected in series. In addition, for example, a circuit diagram showing a capacitor includes a case where two or more capacitors are connected in parallel. In addition, for example, a circuit diagram showing a transistor includes a case where two or more transistors are connected in series and the gates of each transistor are electrically connected to each other. Similarly, for example, a circuit diagram showing a switch includes a case where the switch has two or more transistors, the two or more transistors are electrically connected in series or in parallel and the gates of each transistor are electrically connected to each other.

[0050] In this specification, etc., a node may be referred to as a terminal, wiring, electrode, conductive layer, conductor, impurity region, etc. depending on the circuit structure and device structure. In addition, a terminal, wiring, etc. may also be referred to as a node.

[0051] In addition, in this specification, "voltage" and "potential" can be appropriately interchanged. "Voltage" refers to the potential difference from the reference potential. For example, when the reference potential is the ground potential (ground potential), "voltage" can also be called "potential". The ground potential does not necessarily mean 0V. In addition, the potential is relative, and the potential supplied to the wiring, the potential applied to the circuit, the potential output from the circuit, etc., etc., also changes according to the change of the reference potential.

[0052] In addition, in this specification, "high-level potential" and "low-level potential" do not mean specific potentials. For example, when two wirings are both recorded as "wirings used to supply high-level potentials", the high-level potentials supplied by the two wirings may be different from each other. Similarly, when two wirings are both recorded as "wirings used to supply low-level potentials", the low-level potentials supplied by the two wirings may be different from each other.

[0053] In addition, "electric current" refers to the migration phenomenon (conduction) of electric charge. For example, the record of "conduction of a positively charged body" can be replaced by the record of "conduction of a negatively charged body in the opposite direction". Therefore, in this specification, etc., in the absence of special instructions, "electric current" refers to the migration phenomenon (conduction) of electric charge when carriers migrate. Here, electrons, holes, anions, cations, and complex ions can be cited as carriers, and the carriers are different according to the system (for example, semiconductors, metals, electrolytes, and vacuum) through which the current flows. In addition, the "direction of electric current" in wiring, etc. is the direction in which positively charged carriers migrate, and is recorded as a positive current amount. In other words, the direction in which negatively charged carriers migrate is opposite to the direction of electric current, and is recorded as a negative current amount. Therefore, in this specification, etc., in the absence of special instructions, regarding the positive and negative of the electric current (or the direction of the electric current), the record of "the electric current flows from element A to element B" can be replaced by the record of "the electric current flows from element B to element A". In addition, the record of "the electric current is input to element A" can be replaced by the record of "the electric current is output from element A".

[0054] In addition, in this specification, etc., ordinal numbers such as "first", "second", and "third" are added to avoid confusion between constituent elements. Therefore, the ordinal numbers do not limit the number of constituent elements. In addition, the ordinal numbers do not limit the order of the constituent elements. For example, in this specification, etc., the "first" constituent element in one embodiment may be referred to as the "second" constituent element in other embodiments or claims. In addition, for example, in this specification, etc., the constituent element referred to by "first" in one embodiment may be omitted in other embodiments or claims.

[0055] In this specification, etc., for the sake of convenience, words and phrases indicating configurations such as "upper" and "lower" are sometimes used to illustrate the positional relationship of constituent elements with reference to the drawings. In addition, the positional relationship of constituent elements is appropriately changed according to the direction in which each constituent element is described. Therefore, the words and phrases described in the specification, etc. are not limited, and the words and phrases can be appropriately replaced according to the situation. For example, if it is an expression of "an insulator located on the top surface of a conductor", by rotating the direction of the illustrated drawing by 180 degrees, it can be replaced with "an insulator located on the bottom surface of a conductor".

[0056] In addition, the terms "above" or "below" are not limited to the case where the positional relationship of the components is "directly above" or "directly below" and in direct contact. For example, if it is an expression "electrode B on insulating layer A", it is not necessarily necessary that electrode B is formed in direct contact with insulating layer A, and the case where other components are included between insulating layer A and electrode B may also be included. In addition, similarly, for example, if it is an expression "electrode B above insulating layer A", it is not necessarily necessary that electrode B is formed in direct contact with insulating layer A, and the case where other components are included between insulating layer A and electrode B may also be included. In addition, similarly, for example, if it is an expression "electrode B below insulating layer A", it is not necessarily necessary that electrode B is formed in direct contact with below insulating layer A, and the case where other components are included between insulating layer A and electrode B may also be included.

[0057] In addition, in this specification, etc., words such as "row" and "column" are sometimes used to describe the components configured in a matrix and their positional relationship. In addition, the positional relationship of the components is appropriately changed according to the direction in which each component is described. Therefore, it is not limited to the words described in the specification, etc., and the words can be appropriately replaced according to the situation. For example, by rotating the direction of the drawing by 90 degrees, the expression of "row direction" can sometimes be replaced by "column direction".

[0058] In addition, in this specification, etc., words such as "film" and "layer" can be interchanged depending on the situation. For example, "conductive layer" can sometimes be interchanged with "conductive film". In addition, "insulating film" can sometimes be interchanged with "insulating layer". In addition, other words can be used instead of words such as "film" and "layer" depending on the situation or condition. For example, "conductive layer" or "conductive film" can sometimes be interchanged with "conductive body". In addition, for example, "insulating layer" or "insulating film" can sometimes be interchanged with "insulator".

[0059] Note that in this specification, etc., the words "electrode", "wiring" and "terminal" do not functionally limit their constituent elements. For example, sometimes an "electrode" is used as a part of a "wiring", and vice versa. Furthermore, the words "electrode" or "wiring" also include the case where a plurality of "electrodes" or "wirings" are formed as one. In addition, for example, sometimes a "terminal" is used as a part of a "wiring" or "electrode", and vice versa. Furthermore, the word "terminal" also includes the case where one or more selected from "electrode", "wiring" and "terminal" are formed as one. Therefore, for example, an "electrode" can be a part of a "wiring" or a "terminal", and for example, a "terminal" can be a part of a "wiring" or an "electrode". In addition, the words "electrode", "wiring" or "terminal" are sometimes replaced with words such as "region" depending on the situation.

[0060] In this specification, etc., depending on the situation or conditions, the words "wiring", "signal line" or "power line" can be interchanged with each other. For example, "wiring" can sometimes be converted to "signal line". In addition, for example, "wiring" can sometimes be converted to "power line". Vice versa, "signal line" or "power line" can sometimes be converted to "wiring". "Power line" can sometimes be converted to "signal line". Vice versa, "signal line" can sometimes be converted to "power line". In addition, depending on the situation or conditions, "potential" applied to the wiring can sometimes be converted to "signal". Vice versa, "signal" can sometimes be converted to "potential".

[0061] In addition, in this specification, etc., sometimes the working method of the semiconductor device is described with reference to a timing diagram. In addition, the timing diagram used in this specification, etc. shows an ideal working example, and is not limited to the period, signal (for example, potential or current) and timing shown in the timing diagram unless otherwise specified. In the timing diagram of this specification, etc., the size and timing of the signal (for example, potential or current) input to each wiring (including nodes) in the timing diagram can be changed according to the situation. For example, even if two periods with equal intervals are shown in the timing diagram, the lengths of the two periods are sometimes different. In addition, for example, even if one of the two periods is shown to be long and the other is short, the lengths of the two periods may sometimes be the same, or sometimes one of the two periods may be short and the other may be long.

[0062] In this specification, etc., a flowchart is sometimes used to illustrate the working method of a semiconductor device. In addition, in this specification, etc., the processing shown in the flowchart is classified by work and shown as independent steps. However, in actual processing, it is difficult to classify the processing shown in the flowchart by work, and sometimes one step involves multiple steps or multiple steps involve one step. Therefore, the processing shown in the flowchart is not limited to the steps described in the specification, and can be appropriately replaced according to the situation. Specifically, the order of steps can be exchanged, and steps can be added or deleted according to the situation.

[0063] In this specification, etc., metal oxide refers to the oxide of metal in a broad sense. Metal oxides are classified into oxide insulators, oxide conductors (including transparent oxide conductors) and oxide semiconductors (Oxide Semiconductor, also referred to as OS), etc. For example, when the channel formation region of a transistor contains a metal oxide, the metal oxide is sometimes referred to as an oxide semiconductor. In other words, when the metal oxide can constitute the channel formation region of a transistor having at least one of an amplification function, a rectification function and a switching function, the metal oxide can be referred to as a metal oxide semiconductor. In addition, an OS transistor can be referred to as a transistor containing a metal oxide or an oxide semiconductor.

[0064] In addition, in this specification and the like, a metal oxide containing nitrogen may also be referred to as a metal oxide (metal oxide). In addition, a metal oxide containing nitrogen may also be referred to as a metal oxynitride (metal oxynitride).

[0065] In addition, in this specification, etc., impurities of a semiconductor refer to substances other than the main components constituting the semiconductor layer. For example, an element with a concentration of less than 0.1 at.% is an impurity. When impurities are contained, one or more selected from the group consisting of an increase in the defect state density in the semiconductor, a decrease in carrier mobility, and a decrease in crystallinity sometimes occur. When the semiconductor is an oxide semiconductor, as impurities that change the semiconductor properties, there are, for example, Group 1 elements, Group 2 elements, Group 13 elements, Group 14 elements, Group 15 elements, or transition metals other than the main components, in particular, for example, hydrogen (contained in water), lithium, sodium, silicon, boron, phosphorus, carbon, and nitrogen.

[0066] In this specification, etc., a switch refers to an element that has a function of controlling whether to allow current to flow by changing to a conducting state (on state) or a non-conducting state (off state). Alternatively, a switch refers to an element that has a function of selecting and switching a current path. Therefore, a switch sometimes includes two or more terminals for allowing current to flow in addition to a control terminal. As an example of a switch, an electric switch or a mechanical switch can be used. In other words, a switch is not limited to a specific element as long as it can control the current.

[0067] Examples of electrical switches include transistors (e.g., bipolar transistors or MOS transistors), diodes (e.g., PN diodes, PIN diodes, Schottky diodes, metal-insulator-metal (MIM: Metal Insulator Metal) diodes, MIS (Metal Insulator Semiconductor: Metal-Insulator-Semiconductor) diodes, or diode-connected transistors (transistors with gate and drain in the on state)), or logic circuits combining these elements. When a transistor is used as a switch, the "on state" of the transistor refers to, for example, a state in which the source electrode and the drain electrode of the transistor are electrically short-circuited or a state in which current can flow between the source electrode and the drain electrode. In addition, the "non-conducting state" of the transistor refers to a state in which the source electrode and the drain electrode of the transistor are electrically disconnected. When the transistor is used only as a switch, there is no particular restriction on the polarity (conductivity type) of the transistor.

[0068] An example of a mechanical switch is a switch using MEMS (micro-electromechanical system) technology. This switch has a mechanically movable electrode and operates by controlling conduction and non-conduction by moving the electrode.

[0069] In this specification, "parallel" refers to a state where the angle formed by two straight lines is greater than -10° and less than 10°. Therefore, it also includes a state where the angle is greater than -5° and less than 5°. "Approximately parallel" refers to a state where the angle formed by two straight lines is greater than -30° and less than 30°. In addition, "perpendicular" refers to a state where the angle formed by two straight lines is greater than 80° and less than 100°. Therefore, it also includes a state where the angle is greater than 85° and less than 95°. "Approximately perpendicular" refers to a state where the angle formed by two straight lines is greater than 60° and less than 120°.

[0070] In this specification, etc., the structure described in each embodiment can be appropriately combined with the structure described in other embodiments to constitute one mode of the present invention. In addition, when a plurality of structural examples are described in one embodiment, these structural examples can be appropriately combined.

[0071] In addition, the content (or part thereof) described in a certain embodiment may be applied / combined / replaced with other content (or part thereof) described in the embodiment and at least one of the content (or part thereof) described in one or more other embodiments.

[0072] Note that the contents described in the embodiments refer to the contents described in each embodiment using various drawings or the contents described in the text described in the specification.

[0073] In addition, more figures can be formed by combining a figure (or a part thereof) shown in a certain embodiment with other parts of the figure, other figures (or a part thereof) shown in the embodiment, and at least one of the figures (or a part thereof) shown in one or more other embodiments.

[0074] The embodiments described in this specification are described with reference to the accompanying drawings. However, a person skilled in the art can easily understand the fact that the embodiments can be implemented in a plurality of different forms, and the methods and details can be transformed into various forms without departing from the purpose and scope of the present invention. Therefore, the present invention should not be interpreted as being limited to the contents described in the embodiments. Note that in the structure of the invention in the embodiments, the same symbols are sometimes used in different drawings to represent the same parts or parts with the same functions, and repeated descriptions are omitted. In stereograms, etc., for the sake of clarity, the illustration of some constituent elements is sometimes omitted.

[0075] In this specification, etc., when the same symbol is used for multiple elements and it is necessary to distinguish them, a symbol such as "_1", "[n]", "[m, n]" is sometimes added to the symbol for identification. In addition, in the drawings, etc., when a symbol is added to a symbol for identification, such as "_1", "[n]", "[m, n]", etc., if it is not necessary to distinguish them in this specification, etc., a symbol such as "_1", "[n]", "[m, n]" is sometimes not added for identification.

[0076] In the drawings, the size, thickness of a layer or area is sometimes exaggerated for the sake of clarity. Therefore, the present invention is not limited to the dimensions in the drawings. In addition, in the drawings, ideal examples are schematically shown, and the shapes or values ​​shown in the drawings are not limited. For example, the unevenness of signals, voltages or currents caused by noise or timing deviations may be included.

[0077] (Implementation Method 1) In this embodiment, a calculation circuit of a semiconductor device which is one embodiment of the present invention is described.

[0078] <Configuration Example of Arithmetic Circuit> Figure 1 The arithmetic circuit CDV shown is a semiconductor device of one embodiment of the present invention, and is an arithmetic circuit capable of continuously performing a plurality of product-sum operations and function operations.

[0079] The arithmetic circuit CDV includes an arithmetic unit array MACA1, an arithmetic unit array MACA2, a memory cell array MEMA1, and a memory cell array MEMA2.

[0080] In addition, the operation circuit CDV includes a circuit XCS1 and a circuit WSD1 as a driving circuit of the operation unit array MACA1. In addition, the operation circuit CDV includes a circuit XCS2 and a circuit WSD2 as a driving circuit of the operation unit array MACA2. In addition, the operation circuit CDV includes a circuit WCS and a circuit ITRZ as a driving circuit of the operation unit array MACA1 and the operation unit array MACA2.

[0081] In addition, the arithmetic circuit CDV includes a circuit WWD1 and a circuit RWD1 as a driving circuit of the memory cell array MEMA1. In addition, the arithmetic circuit CDV includes a circuit WWD2 and a circuit RWD2 as a driving circuit of the memory cell array MEMA2.

[0082] In addition, the operation circuit CDV includes a current generating circuit CM1 , a current generating circuit CM2 , a current generating circuit RL1 , and a current generating circuit RL2 .

[0083] The current generating circuit CM1 includes terminals C1Ti[1] to C1Ti[s] and terminals C1To[1] to C1To[s]. The current generating circuit CM2 includes terminals C2Ti[1] to C2Ti[s] and terminals C2To[1] to C2To[s]. The current generating circuit RL1 includes terminals R1Ti[1] to R1Ti[s] and terminals R1To[1] to R1To[s]. The current generating circuit RL2 includes terminals R2Ti[1] to R2Ti[s] and terminals R2To[1] to R2To[s].

[0084] The operation circuit CDV includes a switching circuit SWC1 , a switching circuit SWC2 , a switching circuit SWC3 , a switching circuit SWC5 , and a switching circuit SWC6 .

[0085] The switching circuit SWC1 includes terminals T1a[1] to T1a[s], terminals T1b[1] to T1b[s], terminals T1c[1] to T1c[s], and terminals T1d[1] to T1d[s]. The switching circuit SWC2 includes terminals T2a[1] to T2a[s], terminals T2b[1] to T2b[s], and terminals T2c[1] to T2c[s]. The switching circuit SWC3 includes terminals T3a[1] to T3a[s], terminals T3b[1] to T3b[s], and terminals T3c[1] to T3c[s]. The switching circuit SWC5 includes terminals T5a[1] to T5a[s], terminals T5b[1] to T5b[s], and terminals T5c[1] to T5c[s]. The switching circuit SWC6 includes terminals T6a[1] to T6a[s], terminals T6b[1] to T6b[s], and terminals T6c[1] to T6c[s].

[0086] As an example, each of the arithmetic unit array MACA1 and the arithmetic unit array MACA2 includes a plurality of arithmetic units arranged in a matrix of s rows and s columns (s is an integer greater than or equal to 1).

[0087] In the row direction of the arithmetic cell array MACA1, wirings XCL1[1] to XCL1[s] and wirings WSL1[1] to WSL1[s] are extended. In the column direction of the arithmetic cell array MACA1, wirings WCL1[1] to WCL1[s] are extended.

[0088] In addition, wiring XCL2[1] to wiring XCL2[s] and wiring WSL2[1] to wiring WSL2[s] are extended in the row direction of the arithmetic unit array MACA2. In addition, wiring WCL2[1] to wiring WCL2[s] are extended in the column direction of the arithmetic unit array MACA2.

[0089] The circuit WCS is electrically connected to the wiring WCL1[1] to the wiring WCL1[s]. The circuit WSD1 is electrically connected to the wiring WSL1[1] to the wiring WSL1[s]. The terminal T5a[1] to the terminal T5a[s] of the switching circuit SWC5 is electrically connected to the circuit XCS1.

[0090] The terminal T5b[k] (k is an integer greater than or equal to 1 and less than or equal to s) of the switching circuit SWC5 is electrically connected to the wiring XCL1[k] (note that Figure 1 Terminal T5b[k] and wiring XCL1[k] are not shown in the figure. Note that Figure 1The structure in which the terminal T5b[1] of the switching circuit SWC5 is electrically connected to the wiring XCL1[1] and the terminal T5b[s] of the switching circuit SWC5 is electrically connected to the wiring XCL1[s]. In addition, the terminal T5c[k] of the switching circuit SWC5 is electrically connected to the terminal R2To[k] of the current generating circuit RL2 (note that in Figure 1 Terminal T5c[k] and terminal R2To[k] are not shown in the figure. Figure 1 , a structure in which a terminal T5c[1] of the switching circuit SWC5 is electrically connected to a terminal R2To[1] of the current generating circuit RL2, and a terminal T5c[s] of the switching circuit SWC5 is electrically connected to a terminal R2To[s] of the current generating circuit RL2 is shown.

[0091] Note that in this embodiment, the first to s-th components are sometimes replaced with the k-th component for description. For example, the terminals T5a[1] to T5a[s] of the switching circuit SWC5 are sometimes replaced with the terminal T5a[k] of the switching circuit SWC5 for description. Also, for example, the terminals R2To[1] to R2To[s] of the current generating circuit RL2 are sometimes replaced with the terminal R2To[k] of the current generating circuit RL2 for description.

[0092] The terminal T1a[k] of the switching circuit SWC1 is electrically connected to the wiring WCL1[k] (note that Figure 1 The terminal T1a[k] and the wiring WCL1[k] are not shown in the figure. In addition, the terminal T1b[k] of the switching circuit SWC1 is electrically connected to the terminal C1Ti[k] of the current generating circuit CM1 (note that Figure 1 The terminal T1b[k] and the terminal C1Ti[k] are not shown in the figure.) The terminal T1c[k] of the switching circuit SWC1 is electrically connected to the wiring WCL2[k] (note that Figure 1 The terminal T1c[k] and the wiring WCL2[k] are not shown in the figure. In addition, the terminal T1d[k] of the switching circuit SWC1 is electrically connected to the terminal C2Ti[k] of the current generating circuit CM2 (note that Figure 1 Terminal T1d[k] and terminal C2Ti[k] are not shown in the figure.

[0093] The circuit WSD2 is electrically connected to the wiring WSL2[1] to the wiring WSL2[s]. In addition, the terminal T6a[1] to the terminal T6a[s] of the switching circuit SWC6 are electrically connected to the circuit XCS2. The terminal T6b[k] of the switching circuit SWC6 is electrically connected to the wiring XCL2[k] (note that in Figure 1Terminal T6b[1] and wiring XCL2[k] are not shown in the figure. In addition, terminal T6c[k] of switching circuit SWC6 is electrically connected to terminal R1To[k] of current generating circuit RL1 (note that Figure 1 Terminal T6c[k] and terminal R1To[k] are not shown in the figure).

[0094] The memory cell array MEMA1 and the memory cell array MEMA2 each include a plurality of calculation units arranged in a matrix of s rows and s columns, for example.

[0095] Furthermore, wirings WWL1[1] to WWL1[s] and wirings RWL1[1] to RWL1[s] extend in the row direction of the memory cell array MEMA1. Furthermore, wirings BL1[1] to BL1[s] extend in the column direction of the arithmetic cell array MEMA1.

[0096] In addition, wirings WWL2[1] to WWL2[s] and wirings RWL2[1] to RWL2[s] extend in the row direction of the memory cell array MEMA2. In addition, wirings BL2[1] to BL2[s] extend in the column direction of the arithmetic cell array MEMA2.

[0097] The circuit WWD1 is electrically connected to the wiring WWL1 [ 1 ] to the wiring WWL1 [s]. In addition, the circuit RWD1 is electrically connected to the wiring RWL1 [ 1 ] to the wiring RWL1 [s].

[0098] The terminal T2b[k] of the switching circuit SWC2 is electrically connected to the wiring BL1[k] (note that Figure 1 The terminal T2b[k] and the wiring BL1[k] are not shown in the figure. In addition, the terminal T2a[k] of the switching circuit SWC2 is electrically connected to the terminal C1To[k] of the current generating circuit CM1 (note that Figure 1 Terminal T2a[k] and terminal C1To[k] are not shown in the figure. In addition, terminal T2c[k] of switching circuit SWC2 is electrically connected to terminal R1Ti[k] of current generating circuit RL1 (note that Figure 1 Terminal T2c[k] and terminal R1Ti[k] are not shown in the figure.

[0099] The circuit WWD2 is electrically connected to the wiring WWL2[1] to the wiring WWL2[s]. In addition, the circuit RWD2 is electrically connected to the wiring RWL2[1] to the wiring RWL2[s].

[0100] The terminal T3b[k] of the switching circuit SWC3 is electrically connected to the wiring BL2[k] (note that Figure 1The terminal T3b[k] and the wiring BL2[k] are not shown in the figure). The terminal T3a[k] ​​of the switching circuit SWC3 is electrically connected to the terminal C2To[k] of the current generating circuit CM2 (note that Figure 1 Terminal T3a[k] ​​and terminal C2To[k] are not shown in the figure. In addition, terminal T3c[k] of switching circuit SWC3 is electrically connected to terminal C2Ti[k] of current generating circuit RL2 (note that Figure 1 Terminal T3c[k] and terminal R2Ti[k] are not shown in the figure.

[0101] <Configuration Example of Each Circuit Included in the Arithmetic Circuit> Next, explain Figure 1 The arithmetic circuit CDV shown includes various components.

[0102] Figure 2 is selectively shown Figure 1 FIG. 1 is a diagram of the operation unit array MACA1, the circuit WSD1, the circuit WCS, the circuit XCS1 and the switching circuit SWC5 in the operation circuit CDV of FIG. Figure 2 The circuit structures of the arithmetic unit array MACA1, the circuit WCS, and the circuit XCS1 are also shown.

[0103] in addition, Figure 3 is selectively shown Figure 1 FIG. 1 is a diagram of the operation unit array MACA2, the circuit WSD2, the circuit XCS2, the circuit ITRZ and the switching circuit SWC6 in the operation circuit CDV. In particular, Figure 3 The circuit structures of the arithmetic unit array MACA2, the circuit XCS2, and the circuit ITRZ are also shown.

[0104] in addition, Figure 4A is selectively shown Figure 1 FIG. 1 is a diagram of a memory cell array MEMA1, a circuit WWD1, and a circuit RWD1 in a computing circuit CDV of FIG. Figure 4A The circuit structure of the memory cell array MEMA1 is also shown.

[0105] in addition, Figure 4B is selectively shown Figure 1 FIG. 1 is a diagram of a memory cell array MEMA2, a circuit WWD2, and a circuit RWD2 in a computing circuit CDV of FIG. 1 . In particular, Figure 4B The circuit structure of the memory cell array MEMA2 is also shown.

[0106] [Calculation unit array MACA1 and calculation unit array MACA2] The calculation unit array MACA1 and the calculation unit array MACA2 both have the function of performing product-sum calculations of a plurality of first data and a plurality of second data, for example. Note that in this embodiment, each first data and each second data is a positive value or a value of "0".

[0107] The computing unit array MACA1 includes, for example, units IM1[1,1] to IM1[s,s]. Furthermore, the computing unit array MACA1 includes units IMD1[1] to IMD1[s]. Furthermore, the computing unit array MACA2 includes, for example, units IM2[1,1] to IM2[s,s]. Furthermore, the computing unit array MACA2 includes units IMD2[1] to IMD2[s].

[0108] Note that the structure of the arithmetic unit array MACA2 is the same as that of the arithmetic unit array MACA1. That is, the structures of the units IM2[1,1] to IM2[s,s] are the same as the structures of the units IM1[1,1] to IM1[s,s], and the structures of the units IMD2[1] to IMD2[s] are the same as the structures of the units IMD1[1] to IMD1[s]. Therefore, with regard to the structures of the units IM2[1,1] to IM2[s,s] and the units IMD2[1] to IMD2[s], refer to the description of the structures of the units IM1[1,1] to IM1[s,s] and the units IMD1[1] to IMD1[s] described below. In addition, at this time, in the description of the structure of the following IM1[1, 1] to unit IM1[s, s] and unit IMD1[1] to unit IMD1[s], the operation unit array MACA1 is replaced by the operation unit array MACA2, the wiring WSL1[1] to wiring WSL1[s] is replaced by the wiring WSL2[1] to wiring WSL2[s], the wiring WCL1[1] to wiring WCL1[s] is replaced by the wiring WCL2[1] to wiring WCL2[s], and the wiring XCL1[1] to wiring XCL1[s] is replaced by the wiring XCL2[1] to wiring XCL2[s], to describe the structure of unit IM2[1, 1] to unit IM2[s, s] and unit IMD2[1] to unit IMD2[s].

[0109] The units IM1[1,1] to IM1[s,s] are used as operation units, for example. In addition, as an example, the units IMD1[1] to IMD1[s] have a function of holding a potential corresponding to reference data for operation in the units IM1[1,1] to IM1[s,s]. Note that the reference data will be described in detail later.

[0110] Each of the units IM1[1,1] to IM1[s,s] includes, for example, a transistor F1, a transistor F2, a transistor F5, and a capacitor C5. Each of the units IMD1[1] to IMD1[s] includes, for example, a transistor F1d, a transistor F2d, a transistor F5d, and a capacitor C5d.

[0111] In particular, the structures (including channel length and channel width) of the transistors F1 included in each of the units IM1[1,1] to IM1[s,s] are preferably identical to each other, the structures of the transistors F2 included in each of the units IM1[1,1] to IM1[s,s] are preferably identical to each other, and the structures of the transistors F5 included in each of the units IM1[1,1] to IM1[s,s] are preferably identical to each other. In addition, the structures of the transistors F1d included in each of the units IMD1[1] to IMD1[s] are preferably identical to each other, the structures of the transistors F2d included in each of the units IMD1[1] to IMD1[s] are preferably identical to each other, and the structures of the transistors F5d included in each of the units IMD1[1] to IMD1[s] are preferably identical to each other. In addition, the structures of the transistors F1 and F1d are preferably identical to each other, the structures of the transistors F2 and F2d are preferably identical to each other, and the structures of the transistors F5 and F5d are preferably identical to each other.

[0112] When the structures of the transistors are identical to each other, the electrical characteristics of each transistor can be made substantially equal. Therefore, when the structures of the transistors F1 included in each of the units IM1[1,1] to IM1[s,s] are identical, the structures of the transistors F2 included in each of the units IM1[1,1] to IM1[s,s] are identical, and the structures of the transistors F5 included in each of the units IM1[1,1] to IM1[s,s] are identical, the units IM1[1,1] to IM1[s,s] can perform almost the same operation under the same conditions. Here, the same conditions refer to, for example, the potentials of the source, drain, and gate of the transistor F1, the potentials of the source, drain, and gate of the transistor F2, the potentials of the source, drain, and gate of the transistor F5, and the voltages input to the units IM1[1,1] to IM1[s,s]. Similarly, when the transistors F1d included in the units IMD1[1] to IMD1[s] are identical in structure, the transistors F2d included in the units IMD1[1] to IMD1[s] are identical in structure, and the transistors F5d included in the units IMD1[1] to IMD1[s] are identical in structure, the units IMD1[1] to IMD1[s] can perform almost identical operations under identical conditions. Here, identical conditions refer to, for example, the potentials of the source, drain, and gate of the transistor F1d, the potentials of the source, drain, and gate of the transistor F2d, the potentials of the source, drain, and gate of the transistor F5d, and the voltages input to the units IMD1[1] to IMD1[s].

[0113] In addition, unless otherwise stated, the case where the transistor F1 and the transistor F1d work as a switching element is included. That is, the following case is included: the gate voltage, source voltage, and drain voltage of each of the above transistors are appropriately biased so that the transistor works as a switching element. Note that one embodiment of the present invention is not limited to this. For example, the transistor F1 and the transistor F1d can work in the linear region or the saturation region when they are in the on state, and can also work in the linear region sometimes and in the saturation region sometimes.

[0114] In addition, unless otherwise specified, the case where the transistor F2 and the transistor F2d operate in the subthreshold region is included (in other words, in the transistor F2 or the transistor F2d, the gate-source voltage is lower than the threshold voltage, and more preferably, the drain current increases exponentially relative to the gate-source voltage). In other words, the following cases are included: the gate voltage, source voltage and drain voltage of the above-mentioned transistors are appropriately biased so that the transistor operates in the subthreshold region. Therefore, the case where the transistor F2 and the transistor F2d operate in a manner where the off-state current flows between the source and the drain is included.

[0115] In addition, the transistor F5 and the transistor F5d are used as clamp transistors (sometimes referred to as clamp FETs), for example. Therefore, it is preferable that a constant potential is supplied to each gate of the transistor F5 and the transistor F5d. In addition, by providing the transistor F5 (transistor F5d), it is possible to prevent drain-induced barrier lowering (DIBL) in the transistor F2 (transistor F2d), and the details thereof will be described later.

[0116] On the other hand, when the drain induced barrier lowering (DIBL) in the transistor F2 (transistor F2d) can be ignored, the transistor F5 (transistor F5d) may not be provided in the cell IM1 (cell IMD1).

[0117] One or both of the transistor F1 and the transistor F1d are, for example, preferably OS transistors. In particular, as metal oxides contained in the channel formation region of the OS transistor, for example, indium oxide, gallium oxide and zinc oxide can be cited. In addition, the metal oxide preferably contains one or more selected from indium, element M and zinc. Note that element M is one or more selected from gallium, aluminum, silicon, boron, yttrium, tin, copper, vanadium, beryllium, titanium, iron, nickel, germanium, zirconium, molybdenum, lanthanum, cerium, neodymium, hafnium, tantalum, tungsten, cobalt, magnesium and antimony. In particular, element M is preferably one or more selected from aluminum, gallium, yttrium and tin. Note that metal oxides containing indium, element M and zinc are sometimes referred to as In-M-Zn oxides.

[0118] In particular, as a metal oxide for the semiconductor layer, an oxide containing indium (In), gallium (Ga) and zinc (Zn) (also referred to as IGZO) is preferably used. Alternatively, an oxide containing indium, tin and zinc (also referred to as ITZO (registered trademark)) is preferably used. In addition, an oxide containing indium (In), gallium (Ga), tin (Sn) and zinc (Zn) is preferably used. Alternatively, an oxide containing indium (In), aluminum (Al) and zinc (Zn) (also referred to as IAZO) is preferably used. Alternatively, an oxide containing indium (In), aluminum (Al), gallium (Ga) and zinc (Zn) (also referred to as IAGZO) is preferably used.

[0119] In addition, the metal oxide contained in the channel formation region of the OS transistor preferably has a stacked structure of multiple oxide layers with different chemical compositions. For example, consider an oxide layer with a two-layer structure having a first layer and a second layer located directly above the first layer. The atomic number ratio of the element M relative to the metal element of the main component in the metal oxide used for the first layer is preferably greater than the atomic number ratio of the element M relative to the metal element of the main component in the metal oxide used for the second layer. In addition, the atomic number ratio of the element M relative to In in the metal oxide used for the first layer is preferably greater than the atomic number ratio of the element M relative to In in the metal oxide used for the second layer. By adopting this structure, it is possible to suppress the diffusion of impurities and oxygen from the structure formed below the first layer to the second layer.

[0120] Furthermore, the atomic number ratio of In to element M in the metal oxide used in the second layer is preferably greater than the atomic number ratio of In to element M in the metal oxide used in the first layer. With this structure, the OS transistor can obtain large on-state current and high frequency characteristics.

[0121] Specifically, for example, as the metal oxide for the first layer, a metal oxide having a composition of In:M:Zn=1:3:2 [atomic ratio] or a composition close thereto, In:M:Zn=1:3:4 [atomic ratio] or a composition close thereto, or In:M:Zn=1:1:0.5 [atomic ratio] or a composition close thereto may be used. Furthermore, as the metal oxide for the second layer, a metal oxide having a composition of In:M:Zn=1:1:1 [atomic ratio] or a composition close thereto, In:M:Zn=1:1:1.2 [atomic ratio] or a composition close thereto, In:M:Zn=1:1:2 [atomic ratio] or a composition close thereto, or In:M:Zn=4:2:3 [atomic ratio] or a composition close thereto may be used. Note that the nearby composition includes a range of ±30% of the desired atomic ratio.

[0122] In addition, as one or both of the transistors F1 and F1d, in addition to the OS transistor, a transistor containing silicon in the channel formation region (hereinafter referred to as a Si transistor) can be used. As silicon, for example, amorphous silicon (sometimes referred to as hydrogenated amorphous silicon), microcrystalline silicon, polycrystalline silicon, or single crystal silicon can be used.

[0123] In addition, as one or both of the transistor F1 and the transistor F1d, in addition to the OS transistor and the Si transistor, for example, a transistor containing germanium or the like in the channel formation region, a transistor containing a compound semiconductor such as zinc selenide, cadmium sulfide, gallium arsenide, indium phosphide, gallium nitride or silicon germanium in the channel formation region, a transistor containing carbon nanotubes in the channel formation region, or a transistor containing an organic semiconductor in the channel formation region can be used.

[0124] By using an OS transistor as one or both of the transistor F1 and the transistor F1d, the leakage current of one or both of the transistor F1 and the transistor F1d can be suppressed, thereby reducing the power consumption of the operation circuit. Specifically, when one or both of the transistor F1 and the transistor F1d are in a non-conducting state, the leakage current from the holding node (for example, the node N or the node Nd described later) to the write word line can be made very small, so the refresh work of the potential of the holding node can be reduced. In addition, the power consumption of the product-sum operation circuit can be reduced by reducing the refresh work. In addition, by making the leakage current from the holding node to any of the wiring WCL and the wiring XCL very small, the unit can maintain the potential of the holding node for a long time, thereby improving the calculation accuracy of the operation circuit.

[0125] Furthermore, by using an OS transistor as one or both of the transistor F2 and the transistor F2d, it is possible to operate the transistors in a wider current range in the subthreshold region, thereby reducing current consumption.

[0126] In addition, by using OS transistors as transistors F2, transistor F2d, transistor F5, and transistor F5d, transistors F2, transistor F2d, transistor F5, and transistor F5d can be manufactured simultaneously with transistor F1 and transistor F1d, thereby sometimes shortening the manufacturing process of the operation circuit. In addition, Si transistors can be used as transistors F2, transistor F2d, transistor F5, and transistor F5d in addition to OS transistors.

[0127] In addition, when semiconductor devices are highly integrated on a chip, heat is sometimes generated in the chip due to circuit driving. Due to this heat, the temperature of the transistor increases, so the characteristics of the transistor change, which may lead to changes in field effect mobility, a decrease in operating frequency, etc. The heat resistance of OS transistors is higher than that of Si transistors, so changes in field effect mobility caused by temperature changes are less likely to occur, and the operating frequency is less likely to decrease. Moreover, in OS transistors, even if the temperature increases, it is easy to maintain the characteristic that the drain current increases exponentially with respect to the gate-source voltage. Therefore, by using OS transistors, it is easy to perform calculations, processing, etc. even in a high temperature environment. When a semiconductor device with high resistance to heat caused by driving is constructed, an OS transistor is preferably used as a transistor.

[0128] In the units IM1[1,1] to IM1[s,s], the first terminal of the transistor F1 is electrically connected to the gate of the transistor F2. The first terminal of the transistor F2 is electrically connected to the wiring VE0. The first terminal of the capacitor C5 is electrically connected to the gate of the transistor F2. The second terminal of the transistor F2 is electrically connected to the first terminal of the transistor F5. In addition, the second terminal of the transistor F5 is electrically connected to the second terminal of the transistor F1, and the gate of the transistor F5 is electrically connected to the wiring VE1.

[0129] In addition, in the units IM1[1,1] to IM1[s,s], the second terminal of the transistor F2 is electrically connected in series with the wiring WCL1 via the first terminal and the second terminal of the transistor F5, thereby preventing the wiring WCL1 from directly applying a high-level potential to the second terminal of the transistor F2. Thus, the drain-induced barrier lowering in the transistor F2 can be prevented.

[0130] In the cells IM1[1,1] to IM1[s,s], if the second terminal of the transistor F2 is directly electrically connected to the wiring WCL1 (i.e., the transistor F5 is not provided), a high-level potential is sometimes directly applied from the wiring WCL1 to the second terminal of the transistor F2, and drain-induced barrier lowering occurs in the transistor F2. When drain-induced barrier lowering occurs in the transistor F2, the threshold voltage of the transistor F2 decreases, and thus the voltage range of the subthreshold region of the transistor F2 sometimes changes. Therefore, when the transistor F5 is not provided in the cells IM1[1,1] to IM1[s,s], the current flowing through the subthreshold region of the transistor F2 sometimes fluctuates.

[0131] In the units IMD1[1] to IMD1[s], the first terminal of the transistor F1d is electrically connected to the gate of the transistor F2d. The first terminal of the transistor F2d is electrically connected to the wiring VE0. The first terminal of the capacitor C5d is electrically connected to the gate of the transistor F2d. The second terminal of the transistor F2d is electrically connected to the first terminal of the transistor F5d. In addition, the second terminal of the transistor F5d is electrically connected to the second terminal of the transistor F1d, and the gate of the transistor F5d is electrically connected to the wiring VE1.

[0132] Like the transistor F5 in each of the cells IM1 [1, 1] to IM1 [s, s], the transistor F5d in the cells IMD1 [1] to IMD1 [s] also has a function of preventing the drain-induced barrier lowering in the transistor F2d.

[0133] Figure 2The transistor F1, transistor F2, transistor F5, transistor F1d, transistor F2d and transistor F5d shown include a back gate. Although the connection structure of the back gate is not shown, the electrical connection object of the back gate can be determined when designing. For example, in a transistor including a back gate, in order to increase the on-state current of the transistor, the gate and the back gate can be electrically connected. That is, for example, the gate of transistor F1 can be electrically connected to the back gate, and the gate of transistor F1d can be electrically connected to the back gate. In addition, for example, in a transistor including a back gate, in order to change the threshold voltage of the transistor or reduce the off-state current of the transistor, a wiring for electrically connecting the back gate of the transistor to an external circuit can also be provided to supply a potential to the back gate of the transistor through the external circuit.

[0134] In addition, although Figure 2 The transistor F1, transistor F2, and transistor F5 shown include back gates, but the semiconductor device of one embodiment of the present invention is not limited thereto. Figure 2 The transistors F1, F2 and F5 shown may also be transistors without a back gate, that is, single-gate structures. Alternatively, some transistors may include a back gate and other transistors may not include a back gate.

[0135] In addition, although Figure 2 Although the transistor F1 , the transistor F2 , and the transistor F5 are shown as n-channel transistors, the semiconductor device of one embodiment of the present invention is not limited thereto. For example, an n-channel transistor may be used as the transistor F1 , and p-channel transistors may be used as the transistor F2 and the transistor F5 .

[0136] Note that the above-mentioned examples of changing the structure and polarity of the transistors are not limited to the transistors F1, F2, and F5. For example, the same applies to the transistors F1d, F2d, and F5d, transistors described elsewhere in the specification, and transistors shown in other drawings.

[0137] Wiring VE0 is used as wiring for allowing current to flow between the first terminal and the second terminal of each transistor F2 of unit IM1[1,1] to unit IM1[s,s]. In addition, wiring VE0 is used as wiring for allowing current to flow between the first terminal and the second terminal of transistor F2d of unit IMD1[1] to unit IMD1[s]. As an example, wiring VE0 is used as wiring for supplying a constant potential. The constant potential may be, for example, a low level potential or a ground potential.

[0138] Wiring VE1 is used as wiring for applying a potential to the gates of transistors F5 of cells IM1[1,1] to IM1[s,s] and to the gates of transistors F5d of cells IMD1[1] to IMD1[s]. Note that this potential is preferably within a range where transistors F5 and F5d are used as clamp transistors.

[0139] In the cell IM1[1,1], the second terminal of the transistor F1 and the second terminal of the transistor F5 are electrically connected to the wiring WCL1[1], and the gate of the transistor F1 is electrically connected to the wiring WSL1[1]. The second terminal of the capacitor C5 is electrically connected to the wiring XCL1[1]. Figure 2 In the illustrated unit IM1 [1, 1], a connection portion between the first terminal of the transistor F1, the gate of the transistor F2, and the first terminal of the capacitor C5 is represented as a node N [1, 1].

[0140] In the cell IM1[s, 1], the second terminal of the transistor F1 and the second terminal of the transistor F5 are electrically connected to the wiring WCL1[1], and the gate of the transistor F1 is electrically connected to the wiring WSL1[s]. The second terminal of the capacitor C5 is electrically connected to the wiring XCL1[s]. Figure 2 In the illustrated unit IM1 [s, 1], a connection portion between the first terminal of the transistor F1 , the gate of the transistor F2 , and the first terminal of the capacitor C5 is represented as a node N [s, 1].

[0141] In the cell IM1[1, s], the second terminal of the transistor F1 and the second terminal of the transistor F5 are electrically connected to the wiring WCL1[s], and the gate of the transistor F1 is electrically connected to the wiring WSL1[1]. The second terminal of the capacitor C5 is electrically connected to the wiring XCL1[1]. Figure 2 In the illustrated unit IM1 [1, s], a connection portion between the first terminal of the transistor F1, the gate of the transistor F2, and the first terminal of the capacitor C5 is represented as a node N [1, s].

[0142] In the cell IM1[s, s], the second terminal of the transistor F1 and the second terminal of the transistor F5 are electrically connected to the wiring WCL1[s], and the gate of the transistor F1 is electrically connected to the wiring WSL1[s]. The second terminal of the capacitor C5 is electrically connected to the wiring XCL1[s]. Figure 2 In the illustrated unit IM1 [s, s], a connection portion between the first terminal of the transistor F1 , the gate of the transistor F2 , and the first terminal of the capacitor C5 is represented as a node N [s, s].

[0143] In the cell IMD1[1], the second terminal of the transistor F1d and the second terminal of the transistor F5d are electrically connected to the wiring XCL1[1], and the gate of the transistor F1d is electrically connected to the wiring WSL1[1]. The second terminal of the capacitor C5d is electrically connected to the wiring XCL1[1]. Figure 2 In the illustrated unit IMD1 [ 1 ], a connection portion between the first terminal of the transistor F1 d , the gate of the transistor F2 d , and the first terminal of the capacitor C5 d is represented as a node Nd [ 1 ].

[0144] In the cell IMD1[s], the second terminal of the transistor F1d and the second terminal of the transistor F5d are electrically connected to the wiring XCL1[s], and the gate of the transistor F1d is electrically connected to the wiring WSL1[s]. The second terminal of the capacitor C5d is electrically connected to the wiring XCL1[s]. Figure 2 In the illustrated unit IMD1[s], a connection portion between the first terminal of the transistor F1d, the gate of the transistor F2d, and the first terminal of the capacitor C5d is represented as a node Nd[s].

[0145] The node N[1,1], the node N[1,s], the node N[s,1], the node N[s,s], the node Nd[1], and the node Nd[s] are used as holding nodes of each unit.

[0146] In the units IM1[1,1] to IM1[s,s], for example, when the transistors F1 and F5 are in the on state, the gate and the second terminal of the transistor F2 are in the on state. When the constant potential supplied by the wiring VE0 is set to the ground potential (GND), the transistor F1 is in the on state, and a current of the current amount I flows from the wiring WCL1 to the second terminal of the transistor F2, the potential of the gate (node ​​N) of the transistor F2 is determined according to the current amount I. Note that since the transistor F1 is in the on state, ideally, the potential of the second terminal of the transistor F2 is equal to the potential of the gate (node ​​N) of the transistor F2. Here, by turning off the transistor F1, the potential of the gate (node ​​N) of the transistor F2 is maintained by the capacitor C5. Thus, the transistor F2 can allow a current of the current amount I corresponding to the ground potential of the first terminal of the transistor F2 and the potential of the gate (node ​​N) of the transistor F2 to flow between the source and the drain of the transistor F2. In this specification and the like, such an operation is referred to as “setting (programming) the amount of current flowing between the source and drain of the transistor F2 of the unit IM1 to 1” and the like.

[0147] Note that in the above description, by replacing transistor F1 with transistor F1d, replacing transistor F2 with transistor F2d, and replacing node N with node Nd, the amount of current flowing between the source and drain of transistor F2d can also be set in the same manner in units IMD1[1] to IMD1[s].

[0148] [Circuit WSD1 and Circuit WSD2] For example, the circuit WSD1 has a function of supplying a predetermined signal to the wiring WSL1 [k] when writing first data to each arithmetic cell included in the arithmetic cell array MACA1 to thereby select a row of the arithmetic cell array MACA1 to which the first data is to be written.

[0149] Similarly, the circuit WSD2 has a function of supplying a predetermined signal to the wiring WSL2[k] when writing first data to each arithmetic cell included in the arithmetic cell array MACA2, thereby selecting a row of the arithmetic cell array MACA2 to which the first data is to be written.

[0150] For example, in Figure 2 In the circuit WSD1, a high-level potential is supplied to the wiring WSL1[1] and a low-level potential is supplied to the wiring WSL1[2] (not shown) to the wiring WSL1[s], thereby turning on the transistor F1 and the transistor F1d including the gate electrically connected to the wiring WSL1[1] and turning off the transistor F1 and the transistor F1d including the gate electrically connected to each of the wirings WSL1[2] to the wiring WSL1[s].

[0151] In addition, the structures of the circuit WSD1 and the circuit WSD2 may be the same as each other.

[0152] [Circuit WCS] The circuit WCS has the following functions, for example: obtaining first data as digital data from outside the operation circuit CDV, converting the first data into analog data (current), and supplying the first data converted into analog data to the operation cells included in the operation cell array MACA1 or the operation cells included in the operation cell array MACA2. For example, when the circuit WCS writes the first data to the operation cells in the k-th column of the operation cell array MACA1, the circuit WCS supplies the first data to the operation cells in the k-th column of the operation cell array MACA1 through the wiring WCL1[k]. In addition, when the circuit WCS writes the first data to the operation cells in the k-th column of the operation cell array MACA2, the circuit WCS supplies the first data to the operation cells in the k-th column of the operation cell array MACA2 through the wiring WCL1[k] and the wiring WCL2[k].

[0153] The circuit WCS includes, for example, the circuit SWCA and the circuits WCSa[1] to WCSa[s].

[0154] The circuit SWCA has a function of establishing a conductive state or a non-conductive state between the wiring WCL1[k] and the circuit WCSa[k].

[0155] The circuit SWCA includes, for example, switches SA[1] to SA[s].

[0156] A first terminal of the switch SA[k] is electrically connected to the wiring WCL1[k], a second terminal of the switch SA[k] is electrically connected to the circuit WCSa[k], and a control terminal of the switch SA[k] is electrically connected to the wiring SWLA.

[0157] As the switch SA[k], for example, an analog switch or an electric switch such as a transistor can be used. In particular, it is preferable to use the above-mentioned transistor as the electric switch as the switch SA[k], and it is more preferable to use an OS transistor. Note that when an electric switch is used as the switch SA[k], in addition to the OS transistor, for example, a Si transistor can be used as the electric switch. In addition, the switch SA[k] can also use a mechanical switch, for example.

[0158] Note that in this specification, etc. Figure 2 The switch SA[k] shown is in an open state when a high-level potential is supplied to the control terminal and is in a closed state when a low-level potential is supplied to the control terminal.

[0159] As an example, the wiring SWLA is used as a wiring for switching the on state and the off state of the switch SA[k]. Therefore, the wiring SWLA is supplied with a high-level potential or a low-level potential.

[0160] As described above, the circuit SWCA is used as a circuit for making the circuit WCS and the wiring WCL1[k] conductive or non-conductive. In other words, the circuit SWCA switches the conductive and non-conductive state between the circuit WCS and the wiring WCL1[k] by using the switch SA[k].

[0161] In addition, each of the circuit WCSa[ 1 ] to the circuit WCSa[s] is electrically connected to each of the wiring IWL[ 1 ] to the wiring IWL[s] on a one-to-one basis.

[0162] Each of the wiring IWL[ 1 ] to the wiring IWL[s] is used as a wiring that transmits first data, which is digital data, from the outside of the operation circuit CDV to each of the circuit WCSa[ 1 ] to the circuit WCSa[s].

[0163] The circuit WCSa[k] has a function of, for example, obtaining first data from the wiring IWL[k] and supplying a signal corresponding to the first data to the wiring WCL1[k]. Specifically, the circuit WCSa[k] supplies the first data to be stored in each cell included in the arithmetic unit array MACA1 or the arithmetic unit array MACA2 when the switch SA[k] is in the on state. Figure 2In the operation unit array MACA1 and the operation unit array MACA2, the signal is preferably analog data (current).

[0164] For example, the circuit WCSa[k] may have Figure 5A The structure shown. Note that Figure 5A In order to show the electrical connection between the circuit WCSa[k] and its surrounding circuits, the circuit SWCA, the switch SA[k], the wiring SWLA and the wiring WCL1[k] are also shown.

[0165] therefore, Figure 5A The switch SA[k] shown can be Figure 2 Similarly, the wiring WCL1[k] may be extended to be arranged at any one of the switches SA[1] to SA[s] included in the circuit SWCA. Figure 2 Any one of the wirings WCL1[1] to WCL1[s] in the arithmetic unit array MACA1.

[0166] Therefore, the wiring WCL1[k] is electrically connected to the circuit WCSa[k] via the switch SA[k].

[0167] Figure 5A The circuit WCSa[k] shown includes, for example, a switch SWW. A first terminal of the switch SWW is electrically connected to a second terminal of the switch SA[k], and a second terminal of the switch SWW is electrically connected to a wiring VINIL1. The wiring VINIL1 is used as a wiring for supplying a potential for initialization to the wiring WCL, and the potential for initialization may be a ground potential (GND), a low-level potential, or a high-level potential. Note that the switch SWW is in an on state only when a potential for initialization is supplied to the wiring WCL1[k], and is in an off state otherwise.

[0168] As the switch SWW, for example, an analog switch or an electric switch such as a transistor can be used. In addition, when a transistor is used as the switch SWW, for example, the transistor can be a transistor having the same structure as the transistor F1 or the transistor F2. In addition, in addition to the electric switch, a mechanical switch can also be used.

[0169] Also, as an example, Figure 5A The circuit WCSa[k] includes a plurality of current sources CS. Specifically, the circuit WCSa[k] has a configuration of converting M bits (2 M The first data output of the circuit WCSa[k] is a function of the current value (where M is an integer greater than 1). M -1 current source CS. The circuit WCSa[k] includes, for example, one current source CS that outputs information corresponding to the value of the first bit as a current, two current sources CS that output information corresponding to the value of the second bit as a current, and 2M-1 A current source CS outputs information corresponding to the value of the Mth bit as a current.

[0170] exist Figure 5A In the embodiment, each current source CS includes a terminal U1 and a terminal U2. The terminal U1 of each current source CS is electrically connected to the second terminal of the switch SA included in the circuit SWCA. In addition, the terminal U2 of one current source CS is electrically connected to the wiring DW[1], and the terminals U2 of two current sources CS are electrically connected to the wiring DW[2], respectively. K-1 The terminals U2 of the current sources CS are electrically connected to the wirings DW[M], respectively.

[0171] The multiple current sources CS in the circuit WCSa[k] have the function of converting the same constant current I Wut Note that, in fact, in the manufacturing stage of the operation circuit CDV, errors may occur due to fluctuations in the electrical characteristics of the transistors in each current source CS. Therefore, the constant current I outputted from each terminal U1 of the plurality of current sources CS is Wut The error of is preferably within 10%, more preferably within 5%, and further preferably within 1%. Note that in this embodiment, it is assumed that the constant current I output from the terminal U1 of the plurality of current sources CS in the circuit WCSa[k] Wut There is no error between them.

[0172] The wiring DW[1] to the wiring DW[M] correspond to the above-mentioned wiring IWL[k] and are used as wiring for obtaining the first data as digital data from the outside. Specifically, the wiring DW[1] to the wiring DW[M] are used to transmit the constant current I output from the current source CS electrically connected to them. Wut For example, when the wiring DW[1] is supplied with a high-level potential, the current source CS electrically connected to the wiring DW[1] makes I Wut The current flows through the second terminal of the switch SA[k], and when the wiring DW[1] is supplied with a low-level potential, the current source CS electrically connected to the wiring DW[1] does not output I Wut For example, when the wiring DW[2] is supplied with a high-level potential, the two current sources CS electrically connected to the wiring DW[2] make a total of 2I Wut A constant current flows through the second terminal of the switch SA[k], and when the wiring DW[2] is supplied with a low-level potential, the current source CS electrically connected to the wiring DW[2] does not output a total of 2I Wut For example, when the wiring DW[M] is supplied with a high-level potential, the 2 M-1 The current source CS makes a total of 2 M-1 I WutA constant current flows through the second terminal of the switch SA[k], and when the wiring DW[M] is supplied with a low-level potential, the current source CS electrically connected to the wiring DW[M] does not output a total of 2 M-1 I Wut of constant current.

[0173] The current supplied by one current source CS electrically connected to the wiring DW[1] corresponds to the value of the first bit, the current supplied by two current sources CS electrically connected to the wiring DW[2] corresponds to the value of the second bit, and the current supplied by two current sources CS electrically connected to the wiring DW[M] corresponds to the value of the second bit. M-1 The current supplied by the current source CS is equivalent to the value of the Mth bit. Here, consider the circuit WCSa when M is 2. For example, when the value of the first bit is "1" and the value of the second bit is "0", the wiring DW[1] is supplied with a high level potential and the wiring DW[2] is supplied with a low level potential. At this time, as a constant current, I Wut Flows from the circuit WCSa to the second terminal of the switch SA[k] of the circuit SWCA. In addition, for example, when the value of the first bit is "0" and the value of the second bit is "1", the wiring DW[1] is supplied with a low level potential and the wiring DW[2] is supplied with a high level potential. At this time, as a constant current, 2I Wut Flows from the circuit WCSa to the second terminal of the switch SA[k] of the circuit SWCA. For example, when the value of the first bit is "1" and the value of the second bit is "1", the wiring DW[1] and the wiring DW[2] are supplied with a high level potential. At this time, 3I Wut The constant current flows from the circuit WCSa to the second terminal of the switch SA[k] of the circuit SWCA. In addition, for example, when the value of the first bit is "0" and the value of the second bit is "0", the wiring DW[1] and the wiring DW[2] are supplied with a low-level potential. At this time, the constant current does not flow from the circuit WCSa to the second terminal of the switch SA[k] of the circuit SWCA.

[0174] Notice, Figure 5A The circuit WCSa[k] is shown when M is an integer greater than 3, but when M is 1, Figure 3 The circuit WCSa of A may be configured without providing a current source CS electrically connected to the wiring DW[2] to the wiring DW[M]. In addition, when M is 2, Figure 3 The circuit WCSa of A may be configured without providing a current source CS electrically connected from the wiring DW[3] (not shown) to the wiring DW[M].

[0175] Next, a specific configuration example of the current source CS will be described.

[0176] Fig. 6A The current source CS1 shown is applicable to Figure 5AIn the circuit WCSa, the current source CS1 includes a transistor Tr1 and a transistor Tr2.

[0177] The first terminal of transistor Tr1 is electrically connected to wiring VDDL, and the second terminal of transistor Tr1 is electrically connected to the gate of transistor Tr1, the back gate of transistor Tr1, and the first terminal of transistor Tr2. The second terminal of transistor Tr2 is electrically connected to terminal U1, and the gate of transistor Tr2 is electrically connected to terminal U2. In addition, terminal U2 is electrically connected to wiring DW.

[0178] Wiring DW is Figure 5A Any one of the wiring DW[1] to the wiring DW[M].

[0179] The wiring VDDL is used as a wiring for supplying a constant potential. The constant potential may be, for example, a high-level potential.

[0180] When the constant potential supplied by the wiring VDDL is set to a high-level potential, a high-level potential is input to the first terminal of the transistor Tr1. In addition, the potential of the second terminal of the transistor Tr1 is set to a potential lower than the high-level potential. At this time, the first terminal of the transistor Tr1 is used as a drain, and the second terminal of the transistor Tr1 is used as a source. In addition, the gate of the transistor Tr1 and the second terminal of the transistor Tr1 are electrically connected, so the gate-source voltage of the transistor Tr1 is 0V. Therefore, when the threshold voltage of the transistor Tr1 is within an appropriate range, a current (drain current) in the current range of the subthreshold region flows between the first terminal and the second terminal of the transistor Tr1. In the case where the transistor Tr1 is an OS transistor, the amount of current is preferably, for example, 1.0×10 -8 A or less, more preferably 1.0×10 -12 A is less than or equal to 1.0×10 -15 A or less. In addition, for example, the current is more preferably in a range that increases exponentially with respect to the gate-source voltage. That is, the transistor Tr1 is used as a current source that allows a current in the current range when operating in the subthreshold region to flow. This current is equivalent to the above-mentioned I Wut Or the following I Xut .

[0181] Transistor Tr2 is used as a switching element. In addition, when the potential of the first terminal of transistor Tr2 is higher than the potential of the second terminal of transistor Tr2, the first terminal of transistor Tr2 is used as a drain, and the second terminal of transistor Tr2 is used as a source. In addition, the back gate of transistor Tr2 and the second terminal of transistor Tr2 are electrically connected, so the voltage between the back gate and the source is 0V. Therefore, when the threshold voltage of transistor Tr2 is within an appropriate range, transistor Tr2 is in an open state when a high level potential is input to the gate of transistor Tr2, and transistor Tr2 is in a closed state when a low level potential is input to the gate of transistor Tr2. Specifically, when transistor Tr2 is in an open state, the current in the current range of the above-mentioned subthreshold region flows from the second terminal of transistor Tr1 to terminal U1, and when transistor Tr2 is in a closed state, the current does not flow from the second terminal of transistor Tr1 to terminal U1.

[0182] Note that it can be used Figure 5A The circuit of the current source CS in the circuit WCSa[k] is not limited to Fig. 6A For example, the current source CS1 has a structure in which the back gate of the transistor Tr2 and the second terminal of the transistor Tr2 are electrically connected, but a structure in which the back gate of the transistor Tr2 is electrically connected to other wiring may also be adopted. Figure 6B An example of such a structure is shown. Figure 6B The current source CS2 shown has a structure in which the back gate of the transistor Tr2 is electrically connected to the wiring VTHL. In the current source CS2, since the wiring VTHL is electrically connected to an external circuit, etc., the external circuit can be used to supply a predetermined potential to the wiring VTHL to supply the predetermined potential to the back gate of the transistor Tr2. As a result, the threshold voltage of the transistor Tr2 can be changed. In particular, by increasing the threshold voltage of the transistor Tr2, the off-state current of the transistor Tr2 can be reduced.

[0183] In addition, for example, the current source CS1 has a structure in which the back gate of the transistor Tr1 and the second terminal of the transistor Tr1 are electrically connected, but a structure in which the voltage between the back gate and the second terminal of the transistor Tr2 is held by a capacitor may be adopted. Figure 6C An example of such a structure is shown. Figure 6CThe current source CS3 shown includes a transistor Tr3 and a capacitor C7 in addition to the transistor Tr1 and the transistor Tr2. The current source CS3 is different from the current source CS1 in that the second terminal of the transistor Tr1 is electrically connected to the back gate of the transistor Tr1 through the capacitor C7; and the back gate of the transistor Tr1 is electrically connected to the first terminal of the transistor Tr3. In addition, the current source CS3 has a structure in which the second terminal of the transistor Tr3 is electrically connected to the wiring VTL and the gate of the transistor Tr3 is electrically connected to the wiring VWL. In the current source CS3, by supplying a high level potential to the wiring VWL to turn on the transistor Tr3, the wiring VTL and the back gate of the transistor Tr1 can be turned on. At this time, a predetermined potential can be input from the wiring VTL to the back gate of the transistor Tr1. And, by supplying a low level potential to the wiring VWL to turn off the transistor Tr3, the voltage between the second terminal of the transistor Tr1 and the back gate of the transistor Tr1 can be maintained by the capacitor C7. That is, by determining the voltage supplied to the back gate of the transistor Tr1 by the wiring VTL, the threshold voltage of the transistor Tr1 can be changed, and the threshold voltage of the transistor Tr1 can be fixed by the transistor Tr3 and the capacitor C7.

[0184] In addition, for example, it can be used Figure 5A The circuit structure of the current source CS in the circuit WCSa[k] can also be Fig.6D The current source CS4 is shown. The current source CS4 has a Figure 6C The back gate of the transistor Tr2 in the current source CS3 is electrically connected to the wiring VTHL but not electrically connected to the second terminal of the transistor Tr2. Figure 6B Similarly to the current source CS2, the current source CS4 can change the threshold voltage of the transistor Tr2 according to the potential supplied by the wiring VTHL.

[0185] In the current source CS4, when a large current flows between the first terminal and the second terminal of the transistor Tr1, in order to make the current flow from the terminal U1 to the outside of the current source CS4, it is necessary to increase the on-state current of the transistor Tr2. In this case, in the current source CS4, by supplying a high-level potential to the wiring VTHL, the threshold voltage of the transistor Tr2 is reduced and the on-state current of the transistor Tr2 is increased, so that the large current flowing between the first terminal and the second terminal of the transistor Tr1 can flow from the terminal U1 to the outside of the current source CS4.

[0186] By Figure 5A The current source CS in the circuit WCSa[k] is used as FIG. 6A to FIG. 6DThe circuit WCSa can output a current corresponding to the M-bit first data by any of the current sources CS1 to CS4 shown. In addition, the current amount can be, for example, the current amount flowing between the first terminal and the second terminal when the transistor F1 operates in the subthreshold region.

[0187] In addition, as Figure 5A The circuit WCSa[k] can also be used Fig. 6A The current source CS1 is shown. Figure 5B The circuit WCSa[k] has wiring DW[1] to wiring DW[M], each of which is connected to a Fig. 6A In addition, when the channel width of transistor Tr1[1] is w[1], the channel width of transistor Tr1[2] is w[2], and the channel width of transistor Tr1[M] is w[M], the ratio of the channel widths is w[1]:w[2]:w[M]=1:2:2 M-1 The source-drain current flowing through a transistor operating in the subthreshold region is proportional to the channel width and is therefore Figure 5A Similarly, the circuit WCSa[k] Figure 5B The circuit WCSa[k] shown can output a current corresponding to M bits of first data.

[0188] Note that transistors that can be used for transistor F1 or transistor F2 can be used as transistor Tr1 (including transistor Tr1[1] to transistor Tr1[M]), transistor Tr2 (including transistor Tr2[1] to transistor Tr2[M]), and transistor Tr3, for example. In particular, it is preferable to use OS transistors as transistor Tr1 (including transistor Tr1[1] to transistor Tr1[M]), transistor Tr2 (including transistor Tr2[1] to transistor Tr2[M]), and transistor Tr3.

[0189] [Circuit XCS1 and Circuit XCS2] The circuit XCS1 has the following functions, for example: obtaining second data as digital data from outside the operation circuit CDV, converting the second data into analog data (current), and supplying the second data to the operation cells included in the operation cell array MACA1. For example, when the circuit XCS1 supplies the second data to the operation cells in the k-th row in the operation cell array MACA1, the circuit XCS1 supplies the second data to the operation cells in the k-th row of the operation cell array MACA1 through the wiring XCL1[k].

[0190] Similarly, the circuit XCS2 has a function of supplying second data to the arithmetic units included in the arithmetic unit array MACA2. For example, when the circuit XCS2 supplies the second data to the arithmetic units in the k-th row in the arithmetic unit array MACA2, the circuit XCS2 supplies the second data to the arithmetic units in the k-th row of the arithmetic unit array MACA2 through the wiring XCL2[k].

[0191] Note that the circuit structure of the circuit XCS2 is the same as the circuit structure of the circuit XCS1. Therefore, with respect to the circuit structure of the circuit XCS2, refer to the description of the circuit structure of the circuit XCS1 described below. In addition, at this time, in the description of the circuit structure of the circuit XCS1 described below, the circuit structure of the circuit XCS2 is described by replacing the arithmetic unit array MACA1 with the arithmetic unit array MACA2, replacing the wiring XCL1[1] to the wiring XCL1[s] with the wiring XCL2[1] to the wiring XCL2[s], and replacing the terminal T5a[1] to the terminal T5a[s] of the switching circuit SWC5 with the terminal T6a[1] to the terminal T6a[s] of the switching circuit SWC6.

[0192] The circuit XCS1 includes, for example, a circuit XCSa[1] to a circuit XCSa[s].

[0193] exist Figure 2 In the example, the circuit XCSa[1] is electrically connected to the terminal T5a[1] of the switching circuit SWC5. In addition, as an example, the circuit XCSa[s] is electrically connected to the terminal T5a[s] of the switching circuit SWC5.

[0194] In addition, each of the circuit XCSa[ 1 ] to the circuit XCSa[s] is electrically connected to each of the wiring IXL[ 1 ] to the wiring IXL[s] on a one-to-one basis.

[0195] Each of the wiring IXL[1] to the wiring IXL[s] is used as a wiring that transmits second data as digital data from the outside of the operation circuit CDV to each of the circuit XCSa[1] to the circuit XCSa[s].

[0196] Each of the circuits XCSa[1] to XCSa[s] has, for example, a function of obtaining reference data described later from each of the wirings IXL[1] to IXL[s] and supplying a signal corresponding to the reference data to the wirings XCL1[1] to XCL1[s]. Alternatively, as an example, each of the circuits XCSa[1] to XCSa[s] has a function of obtaining second data from each of the wirings IXL[1] to IXL[s] and supplying a signal corresponding to the second data. Note that in Figure 2 In the computing unit array MACA1, the above-mentioned signals are preferably analog data (current).

[0197] Figure 5C It is shown Figure 2 A block diagram of an example of a circuit XCSa[k] included in the circuit XCS1. Figure 5C , a circuit XCSa[k] corresponding to any one of the circuits XCSa[1] to XCSa[s] is selectively shown. Figure 5C In order to show the electrical connection between the circuit XCS and its peripheral circuits, the wiring XCL1[k], the selection circuit SWC5, the terminal T5a[k], the terminal T5b[k], and the terminal T5c[k] are also shown.

[0198] Therefore, the wiring XCL1[k] is electrically connected to the circuit XCSa[k]. In addition, since the switching circuit SWC5 is provided between the wiring XCL1[k] and the circuit XCSa[k], the switching circuit SWC5 can switch between the conductive state and the non-conductive state between the circuit XCSa[k] and the wiring WCL1[k].

[0199] Figure 5C The circuit XCSa[k] shown includes, for example, a switch SWX. A first terminal of the switch SWX is electrically connected to the wiring XCL1[k], and a second terminal of the switch SWX is electrically connected to the wiring VINIL2. The wiring VINIL2 is used as a wiring for supplying a potential for initialization to the wiring XCL1[k]. The potential for initialization may be a ground potential (GND), a low-level potential, or a high-level potential. In addition, the potential for initialization supplied by the wiring VINIL2 may be equal to the potential supplied by the wiring VINIL1. Note that the switch SWX is in an open state only when the potential for initialization is supplied to the wiring XCL1[k], and is in a closed state otherwise.

[0200] The switch SWX may be, for example, a switch that can be used for the switch SWW.

[0201] in addition, Figure 5C The circuit structure of the circuit XCSa[k] can be adopted with Figure 5A Specifically, the circuit XCSa[k] has the function of outputting reference data as a current and the function of outputting L bits (2 L The second data output of the circuit XCSa[k] is a function of the current amount. L -1 current source CS. The circuit XCSa[k] includes a current source CS that outputs information corresponding to the value of the first bit as a current, two current sources CS that output information corresponding to the value of the second bit as a current, and 2 L-1 A current source CS outputs information corresponding to the value of the Lth bit as a current.

[0202] In addition, as reference data that the circuit XCSa[k] outputs as a current, for example, information that the value of the first bit is "1" and the values ​​of the second bit and subsequent bits are "0" may be used.

[0203] exist Figure 5C In the example, the terminal U2 of one current source CS is electrically connected to the wiring DX[1], and the terminals U2 of the two current sources CS are electrically connected to the wiring DX[2], respectively. L-1 The terminals U2 of the current sources CS are electrically connected to the wiring DX[L], respectively.

[0204] The multiple current sources CS in the circuit XCSa[k] have the same constant current output I from terminal U1. Xut In addition, the wiring DX[1] to the wiring DX[L] are equivalent to the above-mentioned wiring IXL[k], and are used as wiring to obtain reference data or second data as digital data from the outside. Specifically, the wiring DX[1] to the wiring DX[L] are used to transmit the current source CS electrically connected to them to output I Xut That is, the circuit XCSa[k] has a function of causing a current corresponding to L bits of information transmitted from the wiring DX[1] to the wiring DX[L] to flow through the wiring XCL1[k].

[0205] Specifically, consider the circuit XCSa[k] when L is 2. For example, when the value of the first bit is "1" and the value of the second bit is "0", the wiring DX[1] is supplied with a high level potential and the wiring DX[2] is supplied with a low level potential. At this time, as a constant current, I Xut Flows from the circuit XCSa[k] to the wiring XCL. In addition, for example, when the value of the first bit is "0" and the value of the second bit is "1", the wiring DX[1] is supplied with a low level potential and the wiring DX[2] is supplied with a high level potential. At this time, as a constant current, 2I Xut From the circuit XCSa[k] to the wiring XCL1[k]. For example, when the value of the first bit is "1" and the value of the second bit is "1", the wiring DX[1] and the wiring DX[2] are supplied with a high level potential. At this time, 3I Xut Flows from the circuit XCSa[k] to the wiring XCL1[k]. In addition, for example, when the value of the first bit is "0" and the value of the second bit is "0", the wiring DX[1] and the wiring DX[2] are supplied with a low level potential. At this time, the constant current does not flow from the circuit XCSa[k] to the wiring XCL1[k]. Note that at this time, in this specification, etc., it is sometimes recorded that the current with a current amount of 0 flows from the circuit XCSa[k] to the wiring XCL[k]. In addition, the current amounts 0 and I output by the circuit XCSa[k] are Xut、2I Xut 、3I Xut The second data output by the circuit XCSa[k] may be, in particular, the current I output by the circuit XCSa[k]. Xut It can be the reference data output by the circuit XCSa[k].

[0206] When an error occurs due to fluctuations in the electrical characteristics of the transistors in each current source CS of the circuit XCSa[k], the constant current I outputted from each terminal U1 of the plurality of current sources CS is Xut The error of is preferably within 10%, more preferably within 5%, and further preferably within 1%. Note that in this embodiment, it is assumed that the constant current I output from the terminal U1 of the plurality of current sources CS in the circuit XCSa[k] Xut The description is given without any error.

[0207] In addition, similarly to the current source CS of the circuit WCSa[k], the current source CS of the circuit XCSa[k] can be used. FIG. 6A to FIG. 6D In this case, the wiring DX can be used instead of any one of the current sources CS1 to CS4. FIG. 6A to FIG. 6D The wiring DW shown. Thus, the circuit XCSa[k] can cause a current in a subthreshold current range to flow through the wiring XCL[k] as reference data or L-bit second data.

[0208] in addition, Figure 5C The circuit XCSa[k] can be used with Figure 5B The circuit structure is the same as the circuit WCSa[k] shown in the figure. In this case, the circuit XCSa[k] is used instead of Figure 5B The circuit WCSa[k] shown can be considered by replacing the wiring IWL[k] with the wiring IXL[k], the wiring DW[1] with the wiring DX[1], the wiring DW[2] with the wiring DX[2], the wiring DW[M] with the wiring DX[L], the switch SWW with the switch SWX, and the wiring VINIL1 with the wiring VINIL2.

[0209] [Memory Cell Arrays MEMA1 and MEMA2] The memory cell array MEMA1 has a function of storing the calculation results in the calculation cell array MACA1, for example. Specifically, the calculation results are stored in the memory cells included in the memory cell array MEMA1. In addition, the memory cell array MEMA1 may also store the calculation results in the calculation cell array MACA2, depending on the situation.

[0210] Similarly, the memory cell array MEMA2 has the function of storing the calculation results in the calculation cell array MACA2, for example. Specifically, the calculation results are stored in the storage cells included in the memory cell array MEMA2. In addition, depending on the situation, the memory cell array MEMA2 can also store the calculation results in the calculation cell array MACA1.

[0211] The memory cell array MEMA1 includes, for example, cells MC1 [1, 1] to MC1 [s, s]. Furthermore, the memory cell array MEMA2 includes, for example, cells MC2 [1, 1] to MC2 [s, s].

[0212] Note that the structure of the memory cell array MEMA2 is the same as that of the memory cell array MEMA1. That is, the structures of the cells MC2[1,1] to MC2[s,s] are the same as the structures of the cells MC1[1,1] to MC1[s,s]. Therefore, with respect to the structures of the cells MC2[1,1] to MC2[s,s], refer to the description of the structures of the cells MC1[1,1] to MC1[s,s] described below. In addition, at this time, in the description of the structure of the following unit MC1[1, 1] to the unit MC1[s, s], the memory cell array MEMA1 is replaced by the memory cell array MEMA2, the wiring WWL1[1] to the wiring WWL1[s] is replaced by the wiring WWL2[1] to the wiring WWL2[s], the wiring RWL1[1] to the wiring RWL1[s] is replaced by the wiring RWL2[1] to the wiring RWL2[s], and the wiring BL1[1] to the wiring BL1[s] is replaced by the wiring BL2[1] to the wiring BL2[s], to describe the structure of the unit MC2[1, 1] to the unit MC2[s, s].

[0213] The cell MC1 [1, 1] to the cell MC1 [s, s] are used as memory cells, for example.

[0214] For example, each of the cells MC1 [ 1 , 1 ] to MC1 [s, s] includes a transistor F7 , a transistor F8 , a transistor F9 , and a capacitor C6 .

[0215] In particular, the structures (including channel length and channel width) of the transistors F7 included in the cells MC1[1,1] to MC1[s,s] are preferably the same as each other, the structures of the transistors F8 included in the cells MC1[1,1] to MC1[s,s] are preferably the same as each other, and the structures of the transistors F9 included in the cells MC1[1,1] to MC1[s,s] are preferably the same as each other. For the advantages of the transistors having the same structure, refer to the description of the transistors F1, F2, and F5.

[0216] In addition, unless otherwise stated, the case where the transistor F7 and the transistor F9 work as a switching element is included. That is, the following case is included: the gate voltage, source voltage and drain voltage of each of the above transistors are appropriately biased so that the transistor works as a switching element. Note that one embodiment of the present invention is not limited to this. For example, the transistor F7 and the transistor F9 can work in the linear region or the saturation region when they are in the on state, and can also work in the linear region sometimes and in the saturation region sometimes.

[0217] In addition, unless otherwise specified, the case where the transistor F8 operates in the subthreshold region is included (in other words, in the transistor F8, the gate-source voltage is lower than the threshold voltage, and more preferably, the drain current increases exponentially relative to the gate-source voltage). In other words, the following cases are included: the gate voltage, source voltage, and drain voltage of the transistor F8 are appropriately biased so that the transistor operates in the subthreshold region. Therefore, the case where the transistor F8 operates in a manner where the off-state current flows between the source and the drain is included.

[0218] As each of the transistors F7 to F9 , for example, a transistor that can be used as the transistor F1 , the transistor F2 , or the transistor F5 can be used.

[0219] In the cells MC1[1,1] to MC1[s,s], the first terminal of the transistor F7 is electrically connected to the gate of the transistor F8. The first terminal of the transistor F8 is electrically connected to the wiring VE2. The first terminal of the capacitor C6 is electrically connected to the gate of the transistor F8. The second terminal of the transistor F8 is electrically connected to the first terminal of the transistor F9. In addition, the second terminal of the transistor F9 is electrically connected to the second terminal of the transistor F7, and the second terminal of the capacitor C6 is electrically connected to the wiring VE3.

[0220] Note, about Figure 4A or Figure 4B The transistors F7 to F9 in the circuit each have a back gate, and reference is made to the description of the back gate of each of the transistors F1, F2, F5, F1d, F2d, and F5d described above.

[0221] Wiring VE2 is used as wiring for passing current between the first terminal and the second terminal of each transistor F8 of cells MC1[1,1] to MC1[s,s]. In addition, as an example, wiring VE2 is used as wiring for supplying a constant potential. The constant potential may be, for example, a low level potential or a ground potential.

[0222] The wiring VE3 is used as a wiring for applying a potential to the second terminal of each capacitor C6 of the cells MC1 [1, 1] to MC1 [s, s]. This potential may be, for example, a low-level potential or a ground potential.

[0223] In cell MC1 [1, 1], the second terminal of transistor F7 and the second terminal of transistor F9 are electrically connected to wiring BL1 [1], the gate of transistor F7 is electrically connected to wiring WWL1 [1], and the gate of transistor F9 is electrically connected to wiring RWL1 [1].

[0224] In cell MC1[s, 1], the second terminal of transistor F7 and the second terminal of transistor F9 are electrically connected to wiring BL1[1], the gate of transistor F7 is electrically connected to wiring WWL1[s], and the gate of transistor F9 is electrically connected to wiring RWL1[s].

[0225] In cell MC1[1,s], the second terminal of transistor F7 and the second terminal of transistor F9 are electrically connected to wiring BL1[s], the gate of transistor F7 is electrically connected to wiring WWL1[1], and the gate of transistor F9 is electrically connected to wiring RWL1[1].

[0226] In cell MC1[s, s], the second terminal of transistor F7 and the second terminal of transistor F9 are electrically connected to wiring BL1[s], the gate of transistor F7 is electrically connected to wiring WWL1[s], and the gate of transistor F9 is electrically connected to wiring RWL1[s].

[0227] Note that the first terminal of each capacitor C6 of the unit IM1 [1, 1] to the unit IM1 [s, s] is used as a holding node of each unit.

[0228] Next, the writing operation to the unit MC1[1,1] to the unit MC1[s,s] is described. For example, when the transistor F7 and the transistor F9 are in the on state, the gate of the transistor F8 is in the on state with the second terminal. When the constant potential supplied by the wiring VE2 is set to the ground potential (GND), the transistor F7 is in the on state and the current of the current amount I flows from the wiring BL1 to the second terminal of the transistor F8, the potential of the gate (node ​​N) of the transistor F8 is determined according to the current amount I. Note that since the transistor F7 is in the on state, ideally, the potential of the second terminal of the transistor F8 is equal to the potential of the gate (node ​​N) of the transistor F8. Here, by turning off the transistor F7, the potential of the gate (node ​​N) of the transistor F8 is maintained by the capacitor C6. Then, by turning off the transistor F9, the writing operation ends.

[0229] Through the above-described writing operation, the amount of current flowing between the source and the drain of the transistor F8 of the cell MC1 is set to I. Note that when the transistor F9 is in the off state, current does not flow between the source and the drain of the transistor F8.

[0230] When the written data is read from the cell MC1 [1, 1] to the cell MC1 [s, s], the transistor F9 is turned on to allow a current of the current amount I set in the transistor F8 to flow through the wiring BL1.

[0231] [Switching circuit SWC1 to switching circuit SWC3, switching circuit SWC5, switching circuit SWC6] The switching circuit SWC1 has, for example, a function of placing the terminal T1a[k] and the terminal T1b[k] in a conductive state or a non-conductive state, a function of placing the terminal T1c[k] and the terminal T1d[k] in a conductive state or a non-conductive state, and a function of placing the terminal T1a[k] and the terminal T1c[k] in a conductive state or a non-conductive state.

[0232] Fig. 7A The configuration example of the switching circuit SWC1 is shown. As an example, Fig. 7A The switching circuit SWC1 includes switches S1a[1] to S1a[s], switches S1b[1] to S1b[s], and switches S1c[1] to S1c[s].

[0233] The first terminal of the switch S1a[k] is electrically connected to the terminal T1a[k], and the second terminal of the switch S1a[k] is electrically connected to the terminal T1b[k]. In addition, the first terminal of the switch S1b[k] is electrically connected to the terminal T1a[k], and the second terminal of the switch S1b[k] is electrically connected to the terminal T1c[k]. In addition, the first terminal of the switch S1c[k] is electrically connected to the terminal T1d[k], and the second terminal of the switch S1c[k] is electrically connected to the terminal T1c[k].

[0234] In addition, each control terminal of switch S1a[1] to switch S1a[s] is electrically connected to wiring SWL1a. In addition, each control terminal of switch S1b[1] to switch S1b[s] is electrically connected to wiring SWL1b. In addition, each control terminal of switch S1c[1] to switch S1c[s] is electrically connected to wiring SWL1c.

[0235] The switching circuit SWC2 has, for example, a function of making the connection between the terminal T2a[k] and the terminal T2b[k] conductive or non-conductive, and a function of making the connection between the terminal T2b[k] and the terminal T2c[k] conductive or non-conductive.

[0236] Figure 7B The configuration example of the switching circuit SWC2 is shown. As an example, Figure 7B The switching circuit SWC2 includes switches S2a[1] to S2a[s] and switches S2b[1] to S2b[s].

[0237] The first terminal of the switch S2a[k] is electrically connected to the terminal T2a[k], and the second terminal of the switch S2a[k] is electrically connected to the terminal T2b[k]. In addition, the first terminal of the switch S2b[k] is electrically connected to the terminal T2b[k], and the second terminal of the switch S2b[k] is electrically connected to the terminal T2c[k].

[0238] In addition, each control terminal of the switch S2a[1] to the switch S2a[s] is electrically connected to the wiring SWL2a. In addition, each control terminal of the switch S2b[1] to the switch S2b[s] is electrically connected to the wiring SWL2b.

[0239] The switching circuit SWC3 has, for example, a function of making the connection between the terminal T3a[k] ​​and the terminal T3b[k] conductive or non-conductive, and a function of making the connection between the terminal T3b[k] and the terminal T3c[k] conductive or non-conductive.

[0240] Figure 7C The configuration example of the switching circuit SWC3 is shown. As an example, Figure 7C The switching circuit SWC3 includes switches S3a[1] to S3a[s] and switches S3b[1] to S3b[s].

[0241] The first terminal of switch S3a[k] ​​is electrically connected to terminal T3a[k], and the second terminal of switch S3a[k] ​​is electrically connected to terminal T3b[k]. In addition, the first terminal of switch S3b[k] is electrically connected to terminal T3b[k], and the second terminal of switch S3b[k] is electrically connected to terminal T3c[k].

[0242] In addition, each control terminal of the switch S3a[1] to the switch S3a[s] is electrically connected to the wiring SWL3a. In addition, each control terminal of the switch S3b[1] to the switch S3b[s] is electrically connected to the wiring SWL3b.

[0243] The switching circuit SWC5 has, for example, a function of making the connection between the terminal T5a[k] and the terminal T5b[k] conductive or non-conductive, and a function of making the connection between the terminal T5b[k] and the terminal T5c[k] conductive or non-conductive.

[0244] Fig. 8A The configuration example of the switching circuit SWC5 is shown. As an example, Fig. 8A The switching circuit SWC5 includes switches S5a[1] to S5a[s] and switches S5b[1] to S5b[s].

[0245] The first terminal of switch S5a[k] is electrically connected to terminal T5a[k], and the second terminal of switch S5a[k] is electrically connected to terminal T5b[k]. In addition, the first terminal of switch S5b[k] is electrically connected to terminal T5b[k], and the second terminal of switch S5b[k] is electrically connected to terminal T5c[k].

[0246] In addition, each control terminal of the switch S5a[1] to the switch S5a[s] is electrically connected to the wiring SWL5a. In addition, each control terminal of the switch S5b[1] to the switch S5b[s] is electrically connected to the wiring SWL5b.

[0247] The switching circuit SWC6 has, for example, a function of making the connection between the terminal T6a[k] and the terminal T6b[k] conductive or non-conductive, and a function of making the connection between the terminal T6b[k] and the terminal T6c[k] conductive or non-conductive.

[0248] Figure 8B An example of the structure of the switching circuit SWC6 is shown. As an example, Figure 8B The switching circuit SWC6 includes switches S6a[1] to S6a[s] and switches S6b[1] to S6b[s].

[0249] The first terminal of switch S6a[k] is electrically connected to terminal T6a[k], and the second terminal of switch S6a[k] is electrically connected to terminal T6b[k]. In addition, the first terminal of switch S6b[k] is electrically connected to terminal T6b[k], and the second terminal of switch S6b[k] is electrically connected to terminal T6c[k].

[0250] In addition, each control terminal of the switch S6a[1] to the switch S6a[s] is electrically connected to the wiring SWL6a. In addition, each control terminal of the switch S6b[1] to the switch S6b[s] is electrically connected to the wiring SWL6b.

[0251] Note that the above-mentioned switches S1a[1] to S1a[s], switches S1b[1] to S1b[s], switches S1c[1] to S1c[s], switches S2a[1] to S2a[s], switches S2b[1] to S2b[s], switches S3a[1] to S3a[s], switches S3b[1] to S3b[s], switches S5a[1] to S5a[s], switches S5b[1] to S5b[s], switches S6a[1] to S6a[s], and switches S6b[1] to S6b[s], for example, can all use switches that can be used as switch SA[k].

[0252] In this specification and the like, each of the switches is turned on when a high-level potential is supplied to the control terminal, and is turned off when a low-level potential is supplied to the control terminal.

[0253] The wiring S1La, wiring S1Lb, wiring S1Lc, wiring S2La, wiring S2Lb, wiring S3La, wiring S3Lb, wiring S5La, wiring S5Lb, wiring S6La, and wiring S6Lb are used as wiring for switching the on state and the off state of the switch including the electrically connected control terminal. Therefore, for example, the wiring is supplied with a high level potential or a low level potential.

[0254] By using the switching circuits SWC1 to SWC3 , the switching circuit SWC5 , and the switching circuit SWC6 , it is possible to select an input terminal for inputting a signal into the inside of the switching circuit and an output terminal for outputting the signal to the outside.

[0255] [Current Generation Circuit CM1 and Current Generation Circuit CM2] The current generating circuit CM1 has, for example, a function of outputting a current to the terminal C1To[k], the current amount of which is equal to the current amount output to the terminal C1Ti[k]. In addition, the current generating circuit CM2 has, for example, a function of outputting a current to the terminal C2To[k], the current amount of which is equal to the current amount output to the terminal C2Ti[k], similarly to the current generating circuit CM1.

[0256] Fig.9A Specific circuit configuration examples of the current generating circuit CM1 and the current generating circuit CM2 are shown. Fig.9A The current generating circuit CM shown has a circuit structure that can be used for the current generating circuit CM1 and the current generating circuit CM2, including circuits CG[1] to CG[s].

[0257] The circuits CG[1] to CG[s] may have the same structure as one another. Note that in this embodiment mode, the description is given on the premise that the circuits CG[1] to CG[s] have the same structure as one another.

[0258] In addition, when Fig.9A The current generating circuit CM is used to Figure 1 When the current generating circuit CM1 of the operation circuit CDV is Fig.9A The current generating circuit CM includes terminals CTi[1] to CTi[s] which correspond to Figure 1 The current generating circuit CM1 of the terminal C1Ti[1] to the terminal C1Ti[s], Fig.9A The current generating circuit CM includes terminals CTo[1] to CTo[s] which correspond to Figure 1The current generating circuit CM1 of the terminal C1To[1] to the terminal C1To[s]. Fig.9A The current generating circuit CM is used to Figure 1 When the current generating circuit CM2 of the operation circuit CDV is Fig.9A The current generating circuit CM includes terminals CTi[1] to CTi[s] which are equivalent to Figure 1 The current generating circuit CM2 of the terminal C2Ti[1] to the terminal C2Ti[s], Fig.9A The current generating circuit CM includes terminals CTo[1] to CTo[s] which correspond to Figure 1 Terminal C2To[1] to terminal C2To[s] of the current generating circuit CM2.

[0259] The circuit CG[k] (not shown) includes a transistor Tr7, a transistor Tr7m, a transistor Tr8, and a transistor Tr8m. In Fig. 9, the transistor Tr7, the transistor Tr7m, the transistor Tr8, and the transistor Tr8m are all n-channel transistors.

[0260] In circuit CG[k], a first terminal of transistor Tr7 is electrically connected to wiring VDE, and a second terminal of transistor Tr7 is electrically connected to a gate of transistor Tr7, a first terminal of transistor Tr8, a gate of transistor Tr8, a gate of transistor Tr8m, and terminal CTi[k] (not shown). In addition, a second terminal of transistor Tr8 is electrically connected to wiring VSE.

[0261] The first terminal of transistor Tr7m is electrically connected to wiring VDE, and the second terminal of transistor Tr7m is electrically connected to the gate of transistor Tr7m, the first terminal of transistor Tr8m, and terminal CTo[k] (not shown). The second terminal of transistor Tr8m is electrically connected to wiring VSE.

[0262] The wiring VDE is used as a wiring for supplying a potential to the first terminal of the transistor Tr7 and the first terminal of the transistor Tr7m in each of the circuits CG[1] to CG[s]. The constant potential may be, for example, a high-level potential.

[0263] The wiring VSE is used as a wiring for supplying a potential to the second terminal of the transistor Tr8 and the second terminal of the transistor Tr8m in each of the circuits CG[1] to CG[s]. The constant potential may be, for example, a ground potential, a low-level potential, or a negative potential.

[0264] exist Fig.9AIn the embodiment, when the constant potential supplied by the wiring VDE is set to a high-level potential, a high-level potential is input to the first terminal of the transistor Tr7. In addition, the potential of the second terminal of the transistor Tr7 is set to a potential lower than the high-level potential. At this time, the first terminal of the transistor Tr7 is used as a drain, and the second terminal of the transistor Tr7 is used as a source. In addition, the gate of the transistor Tr7 and the second terminal of the transistor Tr7 are electrically connected, so the gate-source voltage of the transistor Tr7 is 0V. Therefore, when the threshold voltage of the transistor Tr7 is within an appropriate range, a current (drain current) in the current range of the subthreshold region flows between the first terminal and the second terminal of the transistor Tr7. In the case where the transistor Tr7 is an OS transistor, the amount of this current is preferably, for example, 1.0×10 -8 A or less, more preferably 1.0×10 -12 A is less than or equal to 1.0×10 -15 A or less. In addition, for example, the current is more preferably in a range that increases exponentially with respect to the gate-source voltage. That is, the transistor Tr7 is used as a current source that allows a current in the current range when operating in the subthreshold region to flow. Note that the above content can also be applied to the transistor Tr7m.

[0265] In addition, Fig.9A In the embodiment, the first terminal of transistor Tr8 is electrically connected to the gate of transistor Tr8 and the gate of transistor Tr8m, so the connection structure of transistor Tr8 and transistor Tr8m is a current mirror circuit. That is, ideally, the amount of current flowing between the source and drain of transistor Tr8 is equal to the amount of current flowing between the source and drain of transistor Tr8m.

[0266] When the current flowing between the source and drain of transistor Tr7 and transistor Tr7m is I SC And the current output from the circuit CG[k] to the terminal CTi[k] is I OP When the current flowing between the source and drain of transistor Tr8 is I SC -I OP Therefore, the current flowing between the source and drain of transistor Tr8m is also I SC -I OP Therefore, the amount of current output from the circuit CG[k] to the terminal CTo[k] is I SC -(I SC -I OP )=I OP Thus, the circuit CG[k] can output a current to the terminal CTo[k], the amount of which is equal to the amount of current output to the terminal CTi[k].

[0267] In addition, the structures of the circuits CG[1] to CG[s] of the semiconductor device according to one embodiment of the present invention are not limited to Fig.9A As one embodiment of the present invention, the structures of the circuits CG[1] to CG[s] of the semiconductor device may be changed according to the situation. Fig.9A The structure shown is formed by the structure.

[0268] For example, Fig.9A The circuits CG[1] to CG[s] shown can also be changed to Fig. 9B The structures of circuits CG[1] to CG[s] are shown. Fig. 9B The circuit CG[k] (not shown) has the following structure: Fig.9A The circuit CG[k] is further provided with a transistor Tr9 and a transistor Tr9m as n-channel transistors, the transistor Tr8 and the transistor Tr9 are connected in common source and gate, and the transistor Tr8m and the transistor Tr9m are connected in common source and gate. Fig. 9B As shown, by connecting the transistors included in the current mirror circuit in common source and common gate configurations, the operation of the current mirror circuit can be made more stable.

[0269] In addition, for example, as Figure 1 The structures of the current generating circuit CM1 and the current generating circuit CM2 shown in the figure can also be adopted Fig. 9C The structure of the current generating circuit CM is shown.

[0270] Fig. 9C The circuit CG[1] to the circuit CG[s] included in the current generating circuit CM include a transistor Tr10, a transistor Tr10m, a transistor Tr11, and a transistor Tr11m. Note that the transistor Tr10, the transistor Tr10m, the transistor Tr11, and the transistor Tr11m are all p-channel transistors.

[0271] exist Fig. 9C In the circuit CG[k], the first terminal of the transistor Tr11 is electrically connected to the wiring VDE, and the second terminal of the transistor Tr11 is electrically connected to the gate of the transistor Tr11, the gate of the transistor Tr11m, and the first terminal of the transistor Tr10. In addition, the second terminal of the transistor Tr10 is electrically connected to the gate of the transistor Tr10, the gate of the transistor Tr10m, and the terminal CTi[k]. In addition, the first terminal of the transistor Tr11m is electrically connected to the wiring VDE, and the second terminal of the transistor Tr11m is electrically connected to the first terminal of the transistor Tr10m. In addition, the second terminal of the transistor Tr10m is electrically connected to the terminal CTo[k].

[0272] Fig. 9CThe circuit CG[k] has a structure of a current mirror circuit, in which the transistor Tr10 is connected to the transistor Tr11 and the transistor Tr10m is connected to the transistor Tr11m. Thus, the circuit CG[k] can output a current to the terminal CTo[k], which is equal to the current output to the terminal CTi[k]. Note that Fig. 9C The circuit CG[k] may be, for example, a current mirror circuit without the transistor Tr11 and the transistor Tr11m (without forming a common source and common gate connection).

[0273] [Current Generation Circuit RL1 and Current Generation Circuit RL2] The current generating circuit RL1 includes, for example, a function-type operation circuit. Specifically, for example, the current generating circuit RL1 has the following functions: performing a function-type operation with a value corresponding to the current amount output to the terminal R1Ti[k] as an input value, and outputting a current corresponding to the result of the operation to the terminal R1To[k].

[0274] In addition, the current generating circuit RL2 may include, for example, a function-based operation circuit similar to that of the current generating circuit RL1 .

[0275] As the function-type operation circuit in the current generating circuit RL1 and the current generating circuit RL2, for example, an operation circuit of a sigmoid function, a tanh function, a softmax function, a ReLU function, or a threshold function can be used. In addition, the current generating circuit RL1 and the current generating circuit RL2 may include a circuit for performing pooling processing instead of a function-type operation circuit.

[0276] Here, as an example, a circuit configuration in which the current generating circuit RL1 and the current generating circuit RL2 include a calculation circuit of the ReLU function will be described.

[0277] Fig.10 The current generating circuit RL shown is an example of a circuit configuration of the current generating circuit RL1 or the current generating circuit RL2 including a calculation circuit of a ReLU function. Fig.10 The current generating circuit RL shown includes, for example, a circuit RCG[1] to a circuit RCG[s].

[0278] The circuits RCG[1] to RCG[s] may have the same structure as one another. Note that in this embodiment mode, the description is given on the premise that the circuits RCG[1] to RCG[s] have the same structure as one another.

[0279] In addition, when Fig.10 The current generating circuit RL is used Figure 1 When the current generating circuit RL1 of the operation circuit CDV, Fig.10The current generating circuit RL includes terminals RTi[1] to RTi[s] which are equivalent to Figure 1 The current generating circuit RL1 of the terminal R1Ti[1] to the terminal R1Ti[s], Fig.10 The current generating circuit RL includes terminals RTo[1] to RTo[s] which are equivalent to Figure 1 The current generating circuit RL1 of the terminal R1To[1] to the terminal R1To[s]. In addition, when Fig.10 The current generating circuit RL is used Figure 1 When the current generating circuit RL2 of the operation circuit CDV, Fig.10 The current generating circuit RL includes terminals RTi[1] to RTi[s] which are equivalent to Figure 1 The current generating circuit RL2 of the terminal R2Ti[1] to the terminal R2Ti[s], Fig.10 The current generating circuit RL includes terminals RTo[1] to RTo[s] which are equivalent to Figure 1 The current generating circuit RL2 has terminals R2To[1] to R2To[s].

[0280] The circuit RCG[k] (not shown) includes, for example, a transistor Tr12, a transistor Tr12m, a transistor Tr13, a transistor Tr13m, a transistor Tr14, a transistor Tr14m, a transistor Tr15, a transistor Tr15m, a transistor Tr16, a transistor Tr16m, a transistor Tr17, and a transistor Tr17m. Fig.10 In the embodiment, transistor Tr12, transistor Tr12m, transistor Tr13, transistor Tr13m, transistor Tr16, transistor Tr16m, transistor Tr17, and transistor Tr17m are all p-channel transistors. Fig.10 Among them, transistor Tr14, transistor Tr14m, transistor Tr15, and transistor Tr15m are all n-channel transistors.

[0281] In the circuit RCG[k], the first terminal of the transistor Tr13 is electrically connected to the wiring VDE, the second terminal of the transistor Tr13 is electrically connected to the first terminal of the transistor Tr12, and the gate of the transistor Tr13 is electrically connected to the gate of the transistor Tr13m, the second terminal of the transistor Tr12, and the terminal RTi[k]. In addition, the gate of the transistor Tr12 is electrically connected to the wiring RSWL1 and the gate of the transistor Tr12m. In addition, the first terminal of the transistor Tr13m is electrically connected to the wiring VDE, and the second terminal of the transistor Tr13m is electrically connected to the first terminal of the transistor Tr12m.

[0282] The second terminal of transistor Tr12m is electrically connected to the input terminal of constant current source CI, the first terminal of transistor Tr14, the gate of transistor Tr15, and the gate of transistor Tr15m. In addition, the second terminal of transistor Tr14 is electrically connected to the first terminal of transistor Tr15, and the gate of transistor Tr14 is electrically connected to wiring RSWL2 and the gate of transistor Tr14m. In addition, the second terminal of transistor Tr15 is electrically connected to wiring VSE.

[0283] An output terminal of the constant current source CI is electrically connected to the wiring VSE2.

[0284] The first terminal of transistor Tr17 is electrically connected to wiring VDE, and the second terminal of transistor Tr17 is electrically connected to the first terminal of transistor Tr16. In addition, the second terminal of transistor Tr16 is electrically connected to the gate of transistor Tr17, the gate of transistor Tr17m, and the first terminal of transistor Tr14m. In addition, the second terminal of transistor Tr14m is electrically connected to the first terminal of transistor Tr15m, and the second terminal of transistor Tr15m is electrically connected to wiring VSE.

[0285] In addition, the gate of transistor Tr16 is electrically connected to wiring RSWL3 and the gate of transistor Tr16m. In addition, the first terminal of transistor Tr17m is electrically connected to wiring VDE, and the second terminal of transistor Tr17m is electrically connected to the first terminal of transistor Tr16m. In addition, the second terminal of transistor Tr16m is electrically connected to terminal RTo[k].

[0286] Regarding the wiring VDE, refer to the details of the wiring VDE in the description of the current generating circuit CM. Regarding the wiring VSE, refer to the details of the wiring VSE in the description of the current generating circuit CM.

[0287] Wiring VSE2 is used as wiring for supplying a constant potential to the output terminal of constant current source CI. The constant potential may be, for example, a low level potential, a ground potential, or a negative potential. In addition, the constant potential supplied by wiring VSE2 may also be equal to the constant potential supplied by wiring VSE.

[0288] exist Fig.10In the embodiment, transistor Tr12, transistor Tr12m, transistor Tr14, transistor Tr14m, transistor Tr16, and transistor Tr16m are used as cascode connection transistors, for example. Therefore, as an example, wiring RSWL1 is used as wiring for supplying bias potential to each gate of transistor Tr12 and transistor Tr12m. In addition, as an example, wiring RSWL2 is used as wiring for supplying bias potential to each gate of transistor Tr14 and transistor Tr14m. In addition, as an example, wiring RSWL3 is used as wiring for supplying bias potential to each gate of transistor Tr16 and transistor Tr16m.

[0289] In addition, the transistor Tr12, the transistor Tr12m, the transistor Tr14, the transistor Tr14m, the transistor Tr16, and the transistor Tr16m can also be used as switch transistors. For example, by supplying a high level potential to the wiring RSWL1, the transistor Tr12 and the transistor Tr12m can be turned off. In addition, by supplying a low level potential to the wiring RSWL2, the transistor Tr15 and the transistor Tr15m can be turned off. In addition, by supplying a high level potential to the wiring RSWL3, the transistor Tr16 and the transistor Tr16m can be turned off.

[0290] exist Fig.10 In the embodiment, the gate of transistor Tr13 is electrically connected to terminal RTi[k] and the gate of transistor Tr13m, so the connection structure of transistor Tr13 and transistor Tr13m is a current mirror circuit. That is, ideally, the amount of current flowing between the source and drain of transistor Tr13 is equal to the amount of current flowing between the source and drain of transistor Tr13m.

[0291] Similarly, in Fig.10 In the embodiment, the first terminal of transistor Tr15 is electrically connected to the gate of transistor Tr15 and the gate of transistor Tr15m through transistor Tr14, so the connection structure of transistor Tr15 and transistor Tr15m is a current mirror circuit. That is to say, ideally, the amount of current flowing through the source and drain of transistor Tr15 is equal to the amount of current flowing through the source and drain of transistor Tr15m. In addition, similarly, the gate of transistor Tr17 is electrically connected to the gate of transistor Tr17m and the second terminal of transistor Tr16, so the connection structure of transistor Tr17 and transistor Tr17m is a current mirror circuit. That is to say, ideally, the amount of current flowing through the source and drain of transistor Tr17 is equal to the amount of current flowing through the source and drain of transistor Tr17m.

[0292] For example, the constant current source CI sets the current I STD The constant current is output to the wiring VSE2.STD Equivalent to the reference value of the ReLU function in the circuit RCG[k].

[0293] The output current I from terminal RTi[k] OP The current and I OP >I STD In the case of RCG[k], the circuit outputs a current I from the terminal RTo[k] OP -I STD In addition, when I OP ≤I STD In this case, no current is output from terminal RTo[k].

[0294] [Circuit ITRZ] The circuit ITRZ includes, for example, a function operation circuit and an analog-to-digital conversion circuit. In particular, the function operation circuit preferably has a function of performing a function operation with a value corresponding to the input current as an input value and outputting digital data (voltage) corresponding to the result of the operation.

[0295] Fig.11A A circuit configuration example of the circuit ITRZ is shown. Fig.11A The circuit ITRZ shown is available for Figure 1 and Figure 3 The circuit shown is an example of an ITRZ circuit. Note that Fig.11A In order to show the electrical connection between the circuit ITRZ and its surrounding circuits, the wiring WCL2[k] is also shown. Figure 1 and Figure 3 The switch SB[k] is any one of the wirings WCL2[1] to WCL2[n] included in the arithmetic circuit CDV. Figure 3 The circuit SWCB shown includes any one of the switches SB[1] to SB[s].

[0296] As the switch SB[k], for example, a switch that can be used as the switch SA[k] can be used. For example, as the switch SB[k], an electric switch such as an analog switch or a mechanical switch can be used.

[0297] Fig.11A The circuit ITRZ includes the circuit SWCB, the circuit ITRZa[1] to the circuit ITRZa[s]. Note that Fig.11A 1 and 2 selectively show a circuit ITRZa[k] as any one of the circuits ITRZa[1] to ITRZa[s]. In addition, the circuit ITRZa[k] includes a conversion circuit RL3[k] and an analog-to-digital conversion circuit ADC.

[0298] In addition, the conversion circuit RL3[k] includes a terminal R3Ti[k] and a terminal R3To[k].

[0299] The first terminal of the switch SB[k] is electrically connected to the wiring WCL2[k], and the second terminal of the switch SB[k] is electrically connected to the terminal R3Ti[k] of the conversion circuit RL3[k]. In addition, the terminal R3To[k] of the conversion circuit RL3[k] is electrically connected to the input terminal of the analog-to-digital conversion circuit ADC, and the output terminal of the analog-to-digital conversion circuit ADC is electrically connected to the wiring OL[k].

[0300] Wiring OL[k](in Figure 1 In the example, the wiring OL[1] to the wiring OL[s] are used as wiring for outputting the operation result in the operation circuit CDV as digital data to the outside of the operation circuit CDV.

[0301] The conversion circuit RL3[k] may be the above-mentioned function-type operation circuit. As the function-type operation circuit, for example, an operation circuit of a sigmoid function, a tanh function, a softmax function, a ReLU function, or a threshold function may be used. In addition, the conversion circuit RL3 may include a circuit for performing pooling processing instead of a function-type operation circuit. In addition, the conversion circuit RL3[k] preferably outputs a voltage from the terminal R3To[k].

[0302] In addition, the conversion circuit RL3[k] may also be a current-voltage conversion circuit.

[0303] When the conversion circuit RL3[k] is a current-voltage conversion circuit, it is preferred that, for example, the conversion circuit RL3[k] generates an analog voltage corresponding to the current input from the wiring WCL2[k] to the terminal R3Ti[k] of the conversion circuit RL3[k] through the switch SB[k], and outputs it to the terminal R3To[k] of the conversion circuit RL3[k].

[0304] In addition, the analog-to-digital conversion circuit ADC preferably converts the analog voltage supplied from the terminal R3To[k] of the conversion circuit RL3[k] into a digital signal and outputs it to the wiring OL[k].

[0305] in addition, Fig. 11B A configuration example of a circuit ITRZ when the conversion circuit RL3[k] is a current-voltage conversion circuit is shown. Fig. 11B The conversion circuit RL3[k] shown includes, for example, a load LE and an operational amplifier OP1.

[0306] The inverting input terminal of the operational amplifier OP1 is electrically connected to the first terminal of the load LE and the second terminal of the switch SB[k]. The non-inverting input terminal of the operational amplifier OP1 is electrically connected to the wiring VRL. The output terminal of the operational amplifier OP1 is electrically connected to the second terminal of the load LE and the terminal R3To[k].

[0307] The wiring VRL is used as a wiring for supplying a constant potential. The constant potential may be, for example, a ground potential (GND), a low-level potential, or the like.

[0308] In particular, when the constant potential supplied by the wiring VRL is the ground potential (GND), the inverting input terminal of the operational amplifier OP1 is virtually grounded, so the analog voltage output to the wiring OL[k] can be a voltage based on the ground potential (GND).

[0309] When the circuit ITRZ has Fig.11A When the structure is such that the value corresponding to the amount of current flowing from the wiring WCL2[k] through the switch SB[k] to the terminal R3Ti[k] of the conversion circuit RL3[k] can be output as an analog voltage to the terminal R3To[k]. In addition, the analog voltage can be converted into a digital signal by the analog-to-digital conversion circuit ADC and output to the wiring OL[k].

[0310] <Calculation work in the calculation unit array MACA1> Next, an example of the calculation operation of the calculation unit array MACA1 will be described. Note that although an example of the calculation operation of the calculation unit array MACA1 will be described below, the calculation unit array MACA2 can also perform the same calculation operation.

[0311] Fig.12 A timing chart showing an operation example of the arithmetic unit array MACA1, the circuit WCS, the circuit XCS1, and the switching circuit SWC1. Fig.12 The timing diagram shows the potential changes of the wiring SWLA, wiring SWL1a, wiring SWL1b, wiring SWL1c, wiring WSL[i] (here, i is an integer greater than 1 and less than s-1), wiring WSL1[i+1], wiring XCL1[i], wiring XCL1[i+1], node N[i, j] (here, j is an integer greater than 1 and less than s-1), node N[i+1, j], node Nd[i], and node Nd[i+1] during the period from time T11 to time T23 and in the vicinity thereof. And, Fig.12 The timing diagram also shows the current I flowing between the first terminal and the second terminal of the transistor F2 in the unit IM1[i,j]. F2 [i, j], the current I flowing between the first terminal and the second terminal of the transistor F2d in the unit IMD1[i] F2d[i], the current I flowing between the first terminal and the second terminal of the transistor F2 in the unit IM1[i+1,j] F2 [i+1, j], the current I flowing between the first terminal and the second terminal of the transistor F2d in the unit IMD1[i+1] F2d [i+1] respective changes.

[0312] Use as circuit WCS Figure 5A Circuit WCS is used as circuit XCS1 Figure 5C Circuit XCS1.

[0313] Note that in this working example, the potential of wiring VE0 is the ground potential GND. In addition, before time T11, the potentials of nodes N[i, j], N[i+1, j], Nd[i], and Nd[i+1] are set to the ground potential GND as an initial setting. Specifically, for example, by Figure 5A The potential of the wiring VINIL1 used for initialization is set to the ground potential GND, so that the switch SWW, the switch S3, the transistors F1 in the unit IM1[i, j] and the unit IM1[i+1, j] are turned on, so that the potential of the node N[i, j] and the node N[i+1, j] can be the ground potential GND. In addition, for example, by Figure 5C The potential of the wiring VINIL2 used for initialization is set to the ground potential GND, so that the switch SWX, the transistors F1d in the unit IMD1[i] and the unit IMD1[i+1] are in the on state, so that the potentials of the nodes Nd[i] and the node Nd[i+1] can be the ground potential GND.

[0314] In this working example, switches S5a[1] to S5a[s] of switching circuit SWC5 are always on, and switches S5b[1] to S5b[s] are always off. In other words, the current from circuit XCS1 flows through wiring XCL1[1] to wiring XCL1[s].

[0315] [Time T11 to Time T12] During the period from time T11 to time T12, a high-level potential is applied to the wiring SWLA (at Fig.12 A low level potential is applied to the wiring SWL1a (in Fig.12=In the switching circuit SWC1, a low level potential is applied to the wiring SWL1b, and a low level potential is applied to the wiring SWL1c. As a result, in the circuit WCS, a high level potential is applied to the control terminals of the switches SA[1] to SA[s], and the switches SA[1] to SA[s] are all in the on state. In addition, in the switching circuit SWC1, a low level potential is applied to the control terminals of the switches S1a[1] to S1a[s], the switches S1b[1] to S1b[s], and the switches S1c[1] to S1c[s], and the switches S1a[1] to S1a[s] are all in the off state.

[0316] In addition, during the period from time T11 to time T12, a low level potential is applied to the wiring WSL1[i] and the wiring WSL1[i+1]. As a result, a low level potential is applied to the gate of the transistor F1 in the unit IM1[i,1] to the unit IM1[i,s] of the i-th row of the calculation unit array MACA1 and the gate of the transistor F1d in the unit IMD1[i], and each transistor F1 and transistor F1d is turned off. In addition, a low level potential is applied to the gate of the transistor F1 in the unit IM1[i+1,1] to the unit IM1[i+1,s] of the i+1-th row of the calculation unit array MACA1 and the gate of the transistor F1d in the unit IMD1[i+1], and each transistor F1 and transistor F1d is turned off.

[0317] In addition, during the period from time T11 to time T12, the ground potential GND is applied to the wiring XCL1[i] and the wiring XCL1[i+1]. Figure 5C When the wiring XCL1[k] is the wiring XCL1[i] and the wiring XCL1[i+1], the potential of the wiring VINIL2 used for initialization is set to the ground potential GND and the switch SWX is turned on, so that the potential of the wiring XCL1[i] and the wiring XCL1[i+1] can be set to the ground potential GND.

[0318] In addition, during the period from time T11 to time T12, each of the wirings WCL1[1] to WCL1[s] electrically connected to each other through different switches SA is connected to the wirings WCL1[1] to WCL1[s]. Figure 5A In the circuits WCSa[1] to WCSa[s], the wirings DW[1] to DW[M] are not input with the first data. Figure 5AIn each of the circuits WCSa[1] to WCSa[s], a low-level potential is input to each of the wirings DW[1] to DW[M]. In addition, during the period from time T11 to time T12, a low-level potential is input to each of the wirings DW[1] to DW[M] electrically connected to the wirings XCL1[1] to XCL1[s]. Figure 5C In the circuit XCSa[1] to the circuit WCSa[s], the wiring DX[1] to the wiring DX[L] are not input with the second data. Figure 5C In each of the circuits XCSa[1] to WCSa[s], a low-level potential is input to the wirings DX[1] to DX[L].

[0319] In addition, during the period from time T11 to time T12, current does not flow through the wiring WCL1[j], the wiring XCL1[i], and the wiring XCL1[i+1]. Therefore, I F2 [i, j], I F2d [i]、I F2 [i+1, j], I F2d [i+1] is 0.

[0320] [Time T12 to Time T13] During the period from time T12 to time T13, a high level potential is applied to the wiring WSL1[i]. As a result, a high level potential is applied to the gate of the transistor F1 in the unit IM1[i,1] to the unit IM1[i,s] of the i-th row of the operation unit array MACA1 and the gate of the transistor F1d in the unit IMD1[i], and each transistor F1 and transistor F1d is in the on state. In addition, during the period from time T12 to time T13, a low level potential is applied to the wiring WSL1[1] to the wiring WSL1[s] other than the wiring WSL1[i], and the transistor F1 in the unit IM1[1,1] to the unit IM1[s,s] other than the i-th row of the operation unit array MACA1 and the transistor F1d in the unit IMD1[1] to the unit IMD1[s] other than the i-th row are in the off state.

[0321] Then, the ground potential GND is applied to the wiring XCL1 [ 1 ] to the wiring XCL1 [s] from before the time T12 .

[0322] [Time T13 to Time T14] During the period from time T13 to time T14, the current of the current amount I0[i,j] flows from the circuit WCSa[j] to the operation unit array MACA1 through the switch SA[j] as the first data. Figure 5AWhen the wiring WCL1[k] in the circuit is the wiring WCL1[j], by inputting a signal corresponding to the first data to each of the wirings DW[1] to DW[M], the current I0[i, j] flows from the circuit WCSa[j] to the second terminal of the switch SA[j]. That is, when the value of the M-bit signal input as the first data is α[i, j] (α[i, j] is 0 or more and 2 K -1 or less), I0[i,j]=α[i,j]×I Wut .

[0323] Note that when α[i, j] is 0, I0[i, j]=0 holds, so strictly speaking, current does not flow from circuit WCSa through switch SA[j] to the operation unit array MACA1, but in this specification, etc., it is sometimes recorded as "current I0[i, j]=0 flows" etc.

[0324] During the period from time T13 to time T14, the first terminal of the transistor F1 in the unit IM1[i, j] in the i-th row of the operation unit array MACA1 and the wiring WCL1[j] are in a conductive state, and the first terminal of the transistor F1 in the unit IM1[1, j] to the unit IM1[s, j] other than the i-th row of the operation unit array MACA1 and the wiring WCL1[j] are in a non-conductive state, so the current of the current amount I0[i, j] flows from the wiring WCL1[j] to the unit IM1[i, j].

[0325] In addition, by turning on the transistor F1 in the unit IM1[i, j], the gate and drain of the transistor F2 in the unit IM1[i, j] are in a conductive state (diode connection structure). Therefore, when the current flows from the wiring WCL1[j] to the unit IM1[i, j], the potential of the gate of the transistor F2 is approximately equal to the potential of the second terminal of the transistor F2. This potential depends on the amount of current flowing from the wiring WCL1[j] to the unit IM1[i, j] and the potential of the first terminal of the transistor F2 (here, GND). In the present working example, the current of the current amount I0[i, j] flows from the wiring WCL1[j] to the unit IM1[i, j], and the potential of the gate (node ​​N[i, j]) of the transistor F2 becomes V g [i, j]. That is, in transistor F2, the gate-source voltage becomes V g [i, j]-GND, the current of the current amount I0[i, j] is set as the current flowing between the first terminal and the second terminal of the transistor F2.

[0326] Here, the threshold voltage of transistor F2 is V th [i, j], the current I0[i, j] when the transistor F2 operates in the subthreshold region can be expressed by the following formula.

[0327] [Formula 1] I0[i,j]=I a exp{J(V g [i, j]-V th [i, j])} …(1.1)

[0328] Note, I a Indicates V g [i, j] is V th The drain current at [i, j], J represents the correction coefficient determined by temperature, device structure, etc.

[0329] In addition, during the period from time T13 to time T14, the current I ref0 The current flows from the circuit XCS1 to the wiring XCL1[i] as reference data. Specifically, Figure 6C When the wiring XCL1[k] in the example is the wiring XCL1[i], a high-level potential is input to the wiring DX[1], and a low-level potential is input to the wiring DX[2] to the wiring DX[L], the current I ref0 From the circuit XCSa[i] to the wiring XCL1[i]. That is, I ref0 =I Xut .

[0330] During the period from time T13 to time T14, the first terminal of the transistor F1d in the unit IMD1[i] is in conduction with the wiring XCL1[i], and thus the current I ref0 The current flows from the wiring XCL1[i] to the cell IMD1[i].

[0331] As with the cell IM1[i,j], by turning on the transistor F1d in the cell IMD1[i], the gate and drain of the transistor F2d in the cell IMD1[i] are in a conducting state (diode connection structure). Therefore, when current flows from the wiring XCL1[i] to the cell IMD1[i], the potential of the gate of the transistor F2d and the second terminal of the transistor F2d are substantially equal. This potential depends on the amount of current flowing from the wiring XCL1[i] to the cell IMD1[i] and the potential of the first terminal of the transistor F2d (here, GND). In this working example, the current I ref0 The current flows from the wiring XCL1[i] to the unit IMD1[i], and the gate of the transistor F2d (node ​​Nd[i]) has V gm [i], the potential of the wiring XCL1[i] at this time is also V gm [i] That is, in transistor F2d, the gate-source voltage is V gm [i]-GND, the current Iref0 The current is set to be the current flowing between the first terminal and the second terminal of the transistor F2d.

[0332] Here, the threshold voltage of transistor F2d is V thm [i], the current I when transistor F2d operates in the subthreshold region ref0 It can be expressed by the following formula.

[0333] [Formula 2] I ref0 =I a exp{J(V gm [i]-V thm [i])} …(1.2)

[0334] Note that the correction coefficient J is the same as that of the transistor F2 in the unit IM1[i, j]. For example, the device structure (including the channel length and the channel width) of the transistor is the same. In addition, although there is a non-uniformity in the correction coefficient J of each transistor due to the non-uniformity during manufacturing, the non-uniformity is suppressed to a degree that is sufficient for the discussion described later in actual use.

[0335] Here, the weight coefficient w[i, j] as the first data is defined as follows.

[0336] [Formula 3] w[i, j] = exp{J(V g [i, j]-V th [i,]-V gm [i]+V thm [i])} …(1.3)

[0337] Therefore, when using equation (1.3), equation (1.4), I0[i, j] = α[i, j] × I Wut and I ref0 =I Xut When , equation (1.1) can be rewritten as follows.

[0338] [Formula 4]

[0339] Note that Figure 5A The current I output by the current source CS of the circuit WCSa[j] Wut and Figure 5C The current I output by the current source CS of the circuit XCSa[i] Xut When they are equal, w[i, j] = α[i, j]. That is, when I Wut and I Xut When they are equal, α[i, j] is equivalent to the value of the first data, so IWut and I Xut Preferably equal. In this working example, it is assumed that I Wut with I Xut Equal to explain.

[0340] [Time T14 to Time T15] During the period from time T14 to time T15, a low level potential is applied to the wiring WSL1[i]. As a result, a low level potential is applied to the gate of the transistor F1 in the unit IM1[i,1] to the unit IM1[i,s] in the i-th row of the arithmetic unit array MACA1 and the gate of the transistor F1d in the unit IMD1[i], and each transistor F1 and transistor F1d is turned off.

[0341] By turning off the transistor F1 in the cell IM1[i,j], the capacitor C5 holds the difference between the potential of the gate of the transistor F2 (node ​​N[i,j]) and the potential of the wiring XCL1[i]. g [i, j]-V gm [i] In addition, by turning off the transistor F1 in the unit IMD1[i], the capacitor C5d maintains the difference between the potential of the gate of the transistor F2d (node ​​Nd[i]) and the potential of the wiring XCL1[i] at 0. Note that the voltage held by the capacitor C5d may become a voltage other than 0 (here, for example, V ds ). At this time, the potential of the node Nd[i] is regarded as the potential of the wiring XCL1[i] plus V ds The potential of .

[0342] [Time T15 to Time T16] During the period from time T15 to time T16, GND is applied to the wiring XCL1[i]. Figure 5C When the wiring XCL1[k] in the circuit is the wiring XCL1[i], the potential of the wiring VINIL2 for initialization is set to the ground potential GND and the switch SWX is turned on, so that the potential of the wiring XCL1[i] can be set to the ground potential GND.

[0343] Therefore, the potential of the node N[i,1] to the node N[i,n] changes due to the capacitive coupling through the capacitor C5 in each of the units IM1[i,1] to the unit IM1[i,s] in the i-th row of the operation unit array MACA1, and the potential of the node Nd[i] changes due to the capacitive coupling through the capacitor C5d in the unit IMD1[i].

[0344] The amount of change in potential from node N[i,1] to node N[i,s] is a potential obtained by multiplying the amount of change in potential of wiring XCL1[i] by a capacitance coupling coefficient depending on the structure of each unit IM1[i,1] to unit IM1[i,s] in the arithmetic unit array MACA1. The capacitance coupling coefficient is calculated based on the capacitance of capacitor C5, the gate capacitance of transistor F2, parasitic capacitance, etc. In each of units IM1[i,1] to unit IM1[i,s], when the capacitance coupling coefficient through capacitor C5 is p, the potential of node N[i,j] of unit IM1[i,j] decreases by p (V gm [i]-GND).

[0345] Similarly, when the potential of the wiring XCL1[i] changes, the potential of the node Nd[i] also changes due to the capacitive coupling through the capacitor C5d in the unit IMD1[i]. When the capacitive coupling coefficient through the capacitor C5d is p, which is the same as that of the capacitor C5, the potential of the node Nd[i] of the unit IMD1[i] decreases by p (V gm [i]-GND).

[0346] exist Fig.12 In the timing chart of , p=1 is taken as an example. Therefore, the potential of the node Nd[i] during the period from time T15 to time T16 is GND.

[0347] As a result, the potential of the node N[i,j] of the unit IM1[i,j] decreases, and the transistor F2 is turned off. Similarly, the potential of the node Nd[i] of the unit IMD1[i] decreases, and the transistor F2d is also turned off. F2 [i, j] and I F2d [i] are all 0.

[0348] [Time T16 to Time T17] During the period from time T16 to time T17, a high level potential is applied to the wiring WSL1[i+1]. As a result, a high level potential is applied to the gate of the transistor F1 in the unit IM1[i+1,1] to the unit IM1[i+1,s] of the i+1th row of the operation unit array MACA1 and the gate of the transistor F1d in the unit IMD1[i+1], and each transistor F1 and transistor F1d is in the on state. In addition, during the period from time T16 to time T17, a low level potential is applied to the wiring WSL1[1] to the wiring WSL1[s] other than the wiring WSL1[i+1], and the transistor F1 in the unit IM1[1,1] to the unit IM1[s,s] other than the i+1th row of the operation unit array MACA1 and the transistor F1d in the unit IMD1[1] to the unit IMD1[s] other than the i+1th row are in the off state.

[0349] Then, the ground potential GND is applied to the wiring XCL1 [ 1 ] to the wiring XCL1 [ s ] from before the time T16 .

[0350] [Time T17 to Time T18] During the period from time T17 to time T18, the current of the current amount I0[i+1,j] flows from the circuit WCSa[j] to the operation unit array MACA1 through the switch SA[j] as the first data. Figure 5A When the wiring WCL1[k] in the circuit is the wiring WCL1[j+1], the current I0[i+1, j] flows from the circuit WCSa[j] to the second terminal of the switch SA[j] by inputting a signal corresponding to the first data to each of the wirings DW[1] to DW[M]. That is, when the value of the M-bit signal input as the first data is α[i+1, j] (α[i+1, j] is 0 or more and 2 K -1 or less), I0[i+1,j]=α[i+1,j]×I Wut .

[0351] Note that when α[i+1, j] is 0, I0[i+1, j]=0 holds, so strictly speaking, current does not flow from circuit WCSa[j] through switch SA[j] to the operation unit array MACA1, but in this specification, etc., it is sometimes recorded as "current I0[i+1, j]=0 flows" in the same way as in the case of I0[i, j]=0.

[0352] At this time, the first terminal of the transistor F1 in the unit IM1[i+1, j] in the i+1th row of the operation unit array MACA1 and the wiring WCL1[j] are in a conductive state, and the first terminal of the transistor F1 in the unit IM1[1, j] to the unit IM1[m, j] other than the i+1th row of the operation unit array MACA1 and the wiring WCL1[j] are in a non-conductive state, so the current of the current amount I0[i+1, j] flows from the wiring WCL1[j] to the unit IM1[i+1, j].

[0353] In addition, by turning on the transistor F1 in the unit IM1[i+1, j], the gate and drain of the transistor F2 in the unit IM1[i+1, j] are in a conductive state (diode connection structure). Therefore, when the current flows from the wiring WCL1[j] to the unit IM1[i+1, j], the potential of the gate of the transistor F2 is approximately equal to the potential of the second terminal of the transistor F2. This potential depends on the amount of current flowing from the wiring WCL1[j] to the unit IM1[i+1, j] and the potential of the first terminal of the transistor F2 (here, GND). In the present working example, the current of the current amount I0[i+1, j] flows from the wiring WCL1[j] to the unit IM1[i+1, j], and the potential of the gate (node ​​N[i+1, j]) of the transistor F2 becomes V g [i+1, j]. That is, in transistor F2, the gate-source voltage becomes V g [i+1, j]-GND, the current of the current amount I0[i+1, j] is set as the current flowing between the first terminal and the second terminal of the transistor F2.

[0354] Here, the threshold voltage of transistor F2 is V th When the transistor F2 is in the subthreshold region, the current I0[i+1,j] can be expressed by the following equation: Note that the correction coefficient is J, as is the case with the transistor F2 in the unit IM1[i,j] and the transistor F2d in the unit IMD1[i].

[0355] [Formula 5] I0[i+1,j]=I a exp{J(V g [i+1, j]-V th [i+1, j])}

[0356] In addition, during the period from time T17 to time T18, the current I ref0 The current flows from the circuit XCSa[i+1] to the wiring XCL1[i+1] as reference data. Specifically, as in the period from time T13 to time T14, Figure 5CWhen the wiring XCL1[k] in the circuit is the wiring XCL1[i+1], the wiring DX[1] is input with a high level potential, and the wirings DX[2] to DX[L] are input with a low level potential, and the current I ref0 =I Xut The flow is from the circuit XCSa[i+1] to the wiring XCL1[i+1].

[0357] During the period from time T17 to time T18, the first terminal of the transistor F1d in the unit IMD1[i+1] is in conduction with the wiring XCL1[i+1], and thus the current I ref0 The current flows from the wiring XCL1[i+1] to the cell IMD1[i+1].

[0358] As with the cell IM1[i+1, j], by turning on the transistor F1d in the cell IMD1[i+1], the gate and drain of the transistor F2d in the cell IMD1[i+1] are in a conducting state (diode connection). Therefore, when current flows from the wiring XCL1[i+1] to the cell IMD1[i+1], the potential of the gate of the transistor F2d and the second terminal of the transistor F2d are substantially equal. This potential depends on the amount of current flowing from the wiring XCL1[i+1] to the cell IMD1[i+1] and the potential of the first terminal of the transistor F2d (here, GND). In this working example, the gate of the transistor F2d is connected to the drain of the transistor F2d by the current I ref0 The current flows from the wiring XCL1[i+1] to the unit IMD1[i+1], and the gate of the transistor F2 (node ​​Nd[i+1]) has V gm [i+1], the potential of the wiring XCL1[i+1] is also V gm [i+1] That is, in transistor F2d, the gate-source voltage is V gm [i+1]-GND, the current I ref0 The current is set to flow between the first terminal and the second terminal of the transistor F2d.

[0359] Here, the threshold voltage of transistor F2d is V thm When [i+1, j], the current I when transistor F2d operates in the subthreshold region ref0 It can be expressed by the following equation: Note that the correction coefficient J is the same as that of the transistor F2 in the unit IM1[i+1,j].

[0360] [Formula 6] Iref0=I a exp{J(V gm [i+1]-V thm [t+1])} …(1.6)

[0361] Here, the weight coefficient w[i+1, j] as the first data is defined as follows.

[0362] [Formula 7]

[0363] Therefore, when using equation (1.3), equation (1.6), I 0r [i, j] = α[i, j] × I Wut and I ref0 =I Xut When , equation (1.5) can be rewritten as follows.

[0364] [Formula 8]

[0365] Note that, as mentioned above, in this working example, it is assumed that I Wut with I Xut Equal to explain.

[0366] [Time T18 to Time T19] During the period from time T18 to time T19, a low level potential is applied to the wiring WSL1[i+1]. As a result, a low level potential is applied to the gate of the transistor F1 in the cells IM1[i+1,1] to IM1[i+1,n] in the i+1th row of the arithmetic cell array MACA1 and the gate of the transistor F1d in the cell IMD1[i+1], and each transistor F1 and transistor F1d is turned off.

[0367] By turning off the transistor F1 in the cell IM1[i+1,j], the capacitor C5 holds the difference between the potential of the gate of the transistor F2 (node ​​N[i+1,j]) and the potential of the wiring XCL1[i+1]. g [i+1, j]-V gm [i+1]. In addition, by turning off the transistor F1 in the unit IMD1[i+1], the capacitor C5d maintains the difference between the potential of the gate of the transistor F2d (node ​​Nd[i+1]) and the potential of the wiring XCL1[i+1] at 0. Note that the voltage held by the capacitor C5d may become a non-zero voltage (here, for example, V ds ). At this time, the potential of the node Nd[i+1] is regarded as the potential of the wiring XCL1[i+1] plus V ds The potential of .

[0368] [Time T19 to Time T20] During the period from time T19 to time T20, the ground potential GND is applied to the wiring XCL1[i+1]. Figure 5C When the wiring XCL1[k] in is the wiring XCL[i+1], the potential of the wiring VINIL2 for initialization is set to the ground potential GND and the switch SWX is turned on, so that the potential of the wiring XCL1[i+1] can be set to the ground potential GND.

[0369] Therefore, the potential of the node N[i,1] to the node N[i+1,s] changes due to the capacitive coupling through the capacitor C5 in each of the units IM1[i+1,1] to the unit IM1[i+1,n] in the i+1th row of the operation unit array MACA1, and the potential of the node Nd[i+1] changes due to the capacitive coupling through the capacitor C5d in the unit IMD1[i+1].

[0370] The amount of change in potential from the node N[i+1,1] to the node N[i+1,n] is a potential obtained by multiplying the amount of change in potential of the wiring XCL1[i+1] by a capacitance coupling coefficient depending on the structure of each unit IM1[i+1,1] to the unit IM1[i+1,s] in the arithmetic unit array MACA1. The capacitance coupling coefficient is calculated based on the capacitance of the capacitor C5, the gate capacitance of the transistor F2, the parasitic capacitance, etc. In each of the units IM1[i+1,1] to the unit IM1[i+1,s], when the capacitance coupling coefficient through the capacitor C5 is p, which is the same as the capacitance coupling coefficient through the capacitor C5 of each unit IM1[i,1] to the unit IM1[i,s], the potential of the node N[i+1,j] of the unit IM1[i+1,j] decreases by p (V gm [i+1]-GND).

[0371] Similarly, when the potential of the wiring XCL1[i+1] changes, the potential of the node Nd[i+1] also changes due to the capacitive coupling through the capacitor C5d in the unit IMD1[i+1]. When the capacitive coupling coefficient through the capacitor C5d is p, which is the same as that of the capacitor C5, the potential of the node Nd[i+1] of the unit IMD1[i+1] decreases by p (V gm [i+1]-GND).

[0372] exist Fig.12 In the timing chart of , p=1 is taken as an example. Therefore, the potential of the node Nd[i+1] during the period from time T20 to time T21 is GND.

[0373] As a result, the potential of the node N[i+1,j] of the unit IM1[i+1,j] decreases, and the transistor F2 is turned off. Similarly, the potential of the node Nd[i+1] of the unit IMD1[i+1] decreases, and the transistor F2d is also turned off. Therefore, during the period from time T19 to time T20, I F2 [i+1, j], I F2d [i+1] is always 0.

[0374] [Time T20 to Time T21] During the period from time T20 to time T21, a low-level potential is applied to the wiring SWLA, thereby applying a low-level potential to each control terminal of the switches SA[1] to SA[s], and the switches SA[1] to SA[s] are all in the off state.

[0375] [Time T21 to Time T22] During time T21 to time T22, a high level potential is applied to wiring SWL1a, thereby applying a high level potential to the control terminal of each of switches S1a[1] to S1a[s] of switching circuit SWC1, and switches S1a[1] to S1a[s] are all turned on.

[0376] [Time T22 to Time T23] During the period from time T22 to time T23, the current I ref0 x[i] times x[i]I ref0 The current of flows from the circuit XCS1 to the wiring XCL1[i] as the second data. Specifically, for example, Figure 5C When the wiring XCL1[k] in the example is the wiring XCL1[i], a high level potential or a low level potential is input to each of the wirings DX[1] to DX[L] according to the value of x[i], and the current x[i]I ref0 =x[i]I Xut From the circuit XCSa[i] to the wiring XCL1[i]. Note that in this working example, x[i] corresponds to the value of the second data. At this time, the potential of the wiring XCL1[i] changes from 0 to V gm [i]+ΔV[i].

[0377] When the potential of the wiring XCL1[i] changes, the potentials of the nodes N[i,1] to N[i,s] also change due to the capacitive coupling of the capacitors C5 in the cells IM1[i,1] to IM1[i,s] in the i-th row of the arithmetic cell array MACA1. Therefore, the potential of the node N[i,j] of the cell IM1[i,j] becomes V g [i, j]+pΔV[i].

[0378] Similarly, when the potential of the wiring XCL1[i] changes, the potential of the node Nd[i] also changes due to capacitive coupling through the capacitor C5d in the cell IMD1[i]. Therefore, the potential of the node Nd[i] of the cell IMD1[i] becomes V gm [i]+pΔV[i].

[0379] Thus, during the period from time T22 to time T23, the current I1[i,j] flowing between the first terminal and the second terminal of the transistor F2 and the current I1[i,j] flowing between the first terminal and the second terminal of the transistor F2d are equal to or greater than the current I1[i,j] flowing between the first terminal and the second terminal of the transistor F2d. ref1 [i, j] can be expressed as follows.

[0380] [Formula 9] I1[i, j]=I a exp{J(V g [i, j]+pΔV[i]-V th [ii, j])} =I0[i,j]exp(JpΔV[i])…(1.9)

[0381] [Formula 10] I ref1 [i]=I a exp{J(V gm [i]+pΔV[i]-V thm [i])} =x[i]I ref0 …(1.10)

[0382] According to equations (1.9) and (1.10), x[i] can be expressed as follows.

[0383] [Formula 11] x[i]=exp(JpΔV[i])…(1.11)

[0384] Therefore, equation (1.9) can be rewritten as follows.

[0385] [Formula 12] I1[i,j]=x[i]w[i,j]I ref0 …(1.12)

[0386] That is, the amount of current flowing between the first terminal and the second terminal of the transistor F2 in the unit IM1 [i, j] is proportional to the product of the first data w [i, j] and the second data x [i].

[0387] In addition, during the period from time T22 to time T23, the current I ref0 x[i+1] times x[i+1]Iref0 The current of flows from the circuit XCS1 to the wiring XCL1[i+1] as the second data. Specifically, for example, Figure 5C When the wiring XCL1[k] in the example is the wiring XCL1[i+1], a high level potential or a low level potential is input to each of the wirings DX[1] to DX[L] according to the value of x[i+1], and the current x[i+1]I ref0 =x[i+1]I Xut From the circuit XCSa[i+1] to the wiring XCL1[i+1]. Note that in this working example, x[i+1] corresponds to the value of the second data. At this time, the potential of the wiring XCL1[i+1] changes from 0 to V gm [i+1]+ΔV[i+1].

[0388] When the potential of the wiring XCL1[i+1] changes, the potentials of the nodes N[i+1,1] to N[i+1,s] also change due to the capacitive coupling of the capacitors C5 in the cells IM1[i+1,1] to IM1[i+1,s] in the i+1th row of the arithmetic cell array MACA1. Therefore, the potential of the node N[i+1,j] of the cell IM1[i+1,j] becomes V g [i+1, j]+pΔV[i+1].

[0389] Similarly, when the potential of the wiring XCL1[i+1] changes, the potential of the node Nd[i+1] also changes due to capacitive coupling through the capacitor C5d in the cell IMD1[i+1]. Therefore, the potential of the node Nd[i+1] of the cell IMD1[i+1] becomes V gm [i+1]+pΔV[i+1].

[0390] Thus, during the period from time T22 to time T23, the current I1[i+1, j] flowing between the first terminal and the second terminal of the transistor F2 and the current I1[i+1, j] flowing between the first terminal and the second terminal of the transistor F2d are equal to or less than the current I1[i+1, j] flowing between the first terminal and the second terminal of the transistor F2d. ref1 [i+1, j] can be expressed as follows.

[0391] [Formula 13] I1[i+1,j] =I a exp{J(V g [i+1,j]+pΔV[i+1]-V th [i+1,j])} =I0[i+1,j]exp(JpΔV[i+1])…(1.13)

[0392] [Formula 14] Iref1[i+1] =I a exp{J(V gm [i+1]+pΔV[i+1]-V thm [i+1])} =x[i+1]Iref0…(1.14)

[0393] According to equations (1.13) and (1.14), x[i+1] can be expressed as follows.

[0394] [Formula 15] x[i+1]=exp(JpΔV[i+1])…(1.15)

[0395] Therefore, equation (1.13) can be rewritten as follows.

[0396] [Formula 16] I1[i+1,j]=x[i+1]w[i+1,j]I ref0 …(1.15)

[0397] That is, the amount of current flowing between the first terminal and the second terminal of the transistor F2 in the unit IM1[i+1,j] is proportional to the product of w[i+1,j] as the first data and x[i+1] as the second data.

[0398] In addition, during the period from time T22 to time T23, the switches S1a[1] to S1a[s] of the switching circuit SWC1 are in the on state, and the switches S1b[1] to S1b[s] are in the off state, so that Figure 1 The terminal C1Ti[j] (not shown) of the current generating circuit CM1 supplies the current I1[i, j] (Formula (1.12)) flowing from the wiring WCL1[j] to the unit IM1[i, j] and the current I1[i+1, j] (Formula (1.16)) flowing from the wiring WCL1[j] to the unit IM1[i+1, j]. At this time, the total amount of current flowing from the terminal C1Ti[j] of the current generating circuit CM1 to the unit IM1[i, j] and the unit IM1[i+1, j] through the switching circuit SWC1 is I SUM [j], according to equations (1.12) and (1.16), I SUM [j] can be expressed by the following formula.

[0399] [Formula 17] l SUM [j]=I1[i,j]+I1[i+1,j] =I ref0(x[i]w[i,j]+x[+1]w[i+1,j])…(1.17)

[0400] Therefore, the amount of current output from the terminal C1Ti[j] of the current generating circuit CM1 is a current amount proportional to the product of w[i, j] and w[i+1, j] as the first data and x[i] and x[i+1] as the second data.

[0401] Note that in the above working example, although the sum of the currents flowing through the unit IM1[i, j] and the unit IM1[i+1, j] is described, the sum of the currents flowing through each of the unit IM1[1, j] to the unit IM1[s, j] as a plurality of units may be described. In this case, the equation (1.17) can be rewritten as follows.

[0402] [Formula 18]

[0403] Therefore, the arithmetic circuit MAC1 including the arithmetic unit array MACA1 of more than three rows and multiple columns can also perform the product-sum operation as described above. In this case, the arithmetic unit array MACA1 uses one of the multiple columns as a storage unit. ref0 and ref0 As the current amount unit (the above-mentioned unit IMD1), the product-sum operation processing of the remaining number of columns in the plurality of columns can be performed simultaneously. In other words, by increasing the number of columns of the memory cell array, a semiconductor device capable of high-speed product-sum operation processing can be provided.

[0404] <Operation Method Example 1 of Arithmetic Circuit CDV> Next, explain Figure 1 An example of the working method of the operation circuit CDV.

[0405] FIG. 13A to FIG. 15B As Figure 1 An example of the operation method of the operation circuit CDV is a block diagram illustrating the flow of data input to the operation circuit CDV, data output from the operation circuit CDV, and data in the operation circuit CDV. Note that FIG. 13A to FIG. 15B In the figure, shaded arrows represent digital data and white arrows represent analog data.

[0406] Fig.13A The following operation is shown: the first data W as digital data (1) The first data W is input from the outside of the operation circuit CDV to the circuit WCS, and the circuit WCS converts the first data W (1) The first data W converted into analog data (current) is converted into analog data (current). (1)Note that the first data is written to each of the units IM1[1,1] to IM1[s,s] of the operation unit array MACA1, so the first data W (1) The first data W is processed into s rows and s columns (details will be described later). (1) It is one of the multipliers and multiplicands processed in the first product-sum operation.

[0407] Fig. 13B The following operation is shown: the second data X as digital data (1) The second data X is input from the outside of the operation circuit CDV to the circuit XCS1, and the circuit XCS1 converts the second data X (1) Convert the second data X into analog data (current) (1) Input to the operation unit array MACA1. Note that the second data X (1) is a row and column with s groups of second data (a row and a column with s rows and s columns), wherein one group includes s second data (details will be described later). Therefore, with s second data as a group, the second data X of the first group to the sth group are (1) The second data XCS1 is sequentially input to the operation unit array MACA1 through the wiring XCL1[1] to the wiring XCL1[s]. (1) It is the other of the multiplier and multiplicand processed in the second product-sum operation.

[0408] in addition, Fig. 13B The following operation is shown: the operation unit array MACA1 outputs Fig.13A The first data W is written into the computing unit array MACA1 in the step (1) With the second data X being input (1) The result of the first product-sum operation X (1) W (1) , X (1) W (1) Write to the memory cell array MEMA1. Note that X (1) W (1) A row or column with s rows and s columns.

[0409] Fig. 13C The following operation is shown: the first data W as digital data (2) The first data W is input from the outside of the operation circuit CDV to the circuit WCS, and the circuit WCS converts the first data W (2) The first data W converted into analog data (current) is converted into analog data (current). (2) Write to the arithmetic unit array MACA2. Note that with W (1) Similarly, the first data W(2) A row or column with s rows and s columns.

[0410] Fig.14A The following operation is shown: the memory cell array MEMA1 outputs Fig. 13B The step is written in X (1) W (1) , the current generating circuit RL1 obtains X (1) W (1) , output F(X (1) W (1) Note that F(x) is a function that is operated in the current generating circuit RL1, and is specifically defined as F(X (1) W (1) )=X (2) . X (2) is processed as the second data for the second product-sum operation. (2) Sent to the computing unit array MACA2.

[0411] in addition, Fig.14A The following operation is shown: the output is Fig. 13C The first data W is written into the computing unit array MACA2 in the step (2) The X input to the arithmetic unit array MACA2 (2) The result of the second product-sum operation X (2) W (2) , X (2) W (2) is written to the memory cell array MEMA2. Note that with X (1) W (1) Likewise, X (2) W (2) A row or column with s rows and s columns.

[0412] Fig. 14B The following operation is shown: the first data W as digital data (3) The first data W is input from the outside of the operation circuit CDV to the circuit WCS, and the circuit WCS converts the first data W (3) The first data W converted into analog data (current) is converted into analog data (current). (3) Write to the arithmetic unit array MACA1. Note that with W (1) and W (2) Similarly, the first data W (3) A row or column with s rows and s columns.

[0413] Fig. 14C The following operation is shown: the storage cell array MEMA2 outputs Fig.14A The step is written in X (2) W (2) , the current generating circuit RL2 obtains X(2) W (2) , output F(X (2) W (2) Note that, similar to the current generating circuit RL1, the current generating circuit RL2 also performs the operation of the function F(x). (2) W (2) )=X (3) , X (3) is processed as the second data for the third product-sum operation. (3) Sent to the computing unit array MACA1.

[0414] in addition, Fig. 14C The following operation is shown: the output is Fig. 14B The first data W is written into the computing unit array MACA1 in the step (3) The X input to the arithmetic unit array MACA1 (3) The result of the third product-sum operation is X (3) W (3) , X (3) W (3) is written to the memory cell array MEMA2. Note that with X (1) W (1) and X (2) W (2) Likewise, X (3) W (3) A row or column with s rows and s columns.

[0415] Later, by repeatedly FIG. 13C to FIG. 14C It can perform multiple product and sum operations and function operations in succession.

[0416] Fig.15A and Fig. 15B Note that the operation of performing the final product-sum operation in the operation unit array MACA1 is as follows: Fig.15A The work is performed when the final product and sum operation is performed in the operation unit array MACA2 Fig. 15B work.

[0417] First, explain Fig.15A .exist Fig.15A In the step, the final product-sum operation is performed in the operation unit array MACA1 to obtain the result X (N) W (N) (Here, N is an odd number) is held in the memory cell array MEMA1. Fig.15A In the memory cell array MEMA1, the output X (N) W (N) , circuit ITRZ obtains X (N) W (N), output data T of the operation circuit CDV. Note that the circuit ITRZ also performs the operation of the function F(x) similarly to the current generating circuits RL1 and RL2. Therefore, it can be expressed as T = F(x) (N) W (N) ).

[0418] Next, explain Fig. 15B .exist Fig. 15B In the step, the final product-sum operation is performed in the operation unit array MACA2 to obtain the result X (N) W (N) (Here, N is an even number) is held in the memory cell array MEMA2. Fig. 15B In the memory cell array MEMA2, the output X (N) W (N) , circuit ITRZ obtains X (N) W (N) , output data T of the operation circuit CDV. For T, refer to Fig.15A Description.

[0419] Reference above FIG. 13A to FIG. 15B The operation method of the operation circuit CDV is described below. FIG. 13A to FIG. 15B Detailed operation of the circuit included in the arithmetic circuit CDV.

[0420] <Operation method example 2 of calculation circuit CDV> Fig.16 It is shown Figure 1 FIG. 1 is a flowchart of an example of the working method of the operation circuit CDV. Fig.16 As shown, Figure 1 The working method of the operation circuit CDV includes steps ST1 to ST16, step ST3A, and steps BR1 to BR4.

[0421] [Step ST1] Step ST1 includes a preliminary operation of writing a plurality of first data to the operation unit array MACA1 in order to perform the Nth (N in step ST1 is an odd number greater than 1) product-sum operation. Note that when step ST1 is performed for the first time, N is 1. In addition, when step ST1 is performed for the second time, N is 3.

[0422] For example, step ST1 includes an operation in which the switching circuit SWC1 causes the terminal T1a[k] to be in a non-conductive state with the terminal T1b[k] and causes the terminal T1a[k] to be in a non-conductive state with the terminal T1c[k].

[0423] Specifically, in Fig. 7AIn the switching circuit SWC1, a low-level potential is supplied from the wiring SWL1a to the control terminal of the switch S1a[k] to turn off the switch S1a[k], and a low-level potential is supplied from the wiring SWL1b to the control terminal of the switch S1b[k] to turn off the switch S1b[k]. Note that this operation is equivalent to Fig.12 The timing diagram shows the operation of the switching circuit SWC1 from time T11 to time T12.

[0424] In addition, for example, step ST1 includes an operation in which the switching circuit SWC5 brings the terminal T5a[k] and the terminal T5b[k] into conduction and brings the terminal T5b[k] and the terminal T5c[k] into non-conduction.

[0425] Specifically, in Fig. 8A In the switching circuit SWC5, a high-level potential is supplied from the wiring SWL5a to the control terminal of the switch S5a[k] to turn on the switch S5a[k], and a low-level potential is supplied from the wiring SWL5b to the control terminal of the switch S5b[k] to turn off the switch S5b[k].

[0426] In addition, for example, step ST1 includes an operation of bringing the circuit WCSa[k] and the wiring WCL1[k] into conduction in the circuit WCS.

[0427] Specifically, in Figure 2 In the circuit SWCA, a high-level potential is supplied from the wiring SWLA to the control terminal of the switch SA[k] to turn on the switch SA[k].

[0428] Thus, preparations for writing a plurality of first data into the arithmetic unit array MACA1 are completed.

[0429] [Step ST2] In step ST2, a plurality of first data are written into the operation unit array MACA1 to perform the Nth (when step ST2 is performed for the first time, N=1) product-sum operation.

[0430] For example, referring to the above Fig.12 The first data is written into the units IM1[1,1] to IM1[s,s] of the operation unit array MACA1 from time T11 to time T20. In addition, the first data written into the unit IM1[i,j] of the operation unit array MACA1 is denoted as W (N) [i, j]. Note that W (N) N in [i, j] represents the number of product-sum operations, for example, W (1) [i, j] means the first data used when performing the first product-sum operation.

[0431] At this time, the first data written into the operation unit array MACA1 is recorded as W in rows and columns. (1) In addition, W (N) It can be expressed as follows.

[0432] [Formula 19]

[0433] Note that the operations performed in step ST1 and step ST2 are equivalent to Fig.13A In addition, when N is other than 1, it is equivalent to Fig. 14B .

[0434] [Step BR1] In step BR1, the operation of the operation circuit CDV determines whether N is 1. In the operation of the operation circuit CDV, if N is 1, the process proceeds to step ST3, and if N is not 1, the process proceeds to step ST3A.

[0435] [Step ST3] Step ST3 includes preliminary operations for inputting a plurality of second data to the operation unit array MACA1 when the first product-sum operation is performed (when N is 1).

[0436] For example, step ST3 includes an operation in which the switching circuit SWC5 causes the terminal T5a[k] and the terminal T5b[k] to be in a conductive state, and causes the terminal T5b[k] and the terminal T5c[k] to be in a non-conductive state.

[0437] Specifically, in Fig. 8A In the switching circuit SWC5, a high-level potential is supplied from the wiring SWL5a to the control terminal of the switch S5a[k] to turn on the switch S5a[k], and a low-level potential is supplied from the wiring SWL5b to the control terminal of the switch S5b[k] to turn off the switch S5b[k].

[0438] [Step ST3A] Step ST3A includes preliminary operations to input a plurality of second data into the operation unit array MACA1 when performing the Nth (here, N is an odd number greater than or equal to 3) product-sum operation.

[0439] For example, step ST3A includes an operation in which the switching circuit SWC5 causes the terminal T5a[k] and the terminal T5b[k] to be in a non-conductive state and causes the terminal T5b[k] and the terminal T5c[k] to be in a conductive state.

[0440] Specifically, in Fig. 8AIn the switching circuit SWC5, a low-level potential is supplied from the wiring SWL5a to the control terminal of the switch S5a[k] to turn off the switch S5a[k], and a high-level potential is supplied from the wiring SWL5b to the control terminal of the switch S5b[k] to turn on the switch S5b[k].

[0441] [Step BR2] In step BR2, the operation of the operation circuit CDV determines whether the processing is completed with the Nth (when step BR2 is performed for the first time, N is 1) product-sum operation. In the operation of the operation circuit CDV, if the processing is not completed with the Nth product-sum operation, the process proceeds to step ST4, and if the processing is completed with the Nth product-sum operation, the process proceeds to step ST13.

[0442] [Step BR3] In step BR3, it is determined in the operation of the operation circuit CDV whether N is 1. Note that in the case where step BR3 is performed for the first time, N is 1. In the operation of the operation circuit CDV, if N is 1, the process proceeds to step ST4, and if N is not 1, the process proceeds to step ST11.

[0443] [Step ST4] Step ST4 includes preliminary work for performing the first product-sum operation.

[0444] For example, step ST4 includes an operation in which the switching circuit SWC1 causes the terminal T1a[k] and the terminal T1b[k] to be in a conductive state, and causes the terminal T1a[k] and the terminal T1c[k] to be in a non-conductive state.

[0445] Specifically, in Fig. 7A In the switching circuit SWC1, a high level potential is supplied from the wiring SWL1a to the control terminal of the switch S1a[k] to turn on the switch S1a[k], and a low level potential is supplied from the wiring SWL1b to the control terminal of the switch S1b[k] to turn off the switch S1b[k]. Note that this operation is equivalent to Fig.12 The timing diagram shows the operation of the switching circuit SWC1 from time T21 to time T22.

[0446] In addition, for example, step ST4 includes an operation in which the switching circuit SWC2 causes the terminal T2a[k] and the terminal T2b[k] to be in a conductive state, and causes the terminal T2b[k] and the terminal T2c[k] to be in a non-conductive state.

[0447] Specifically, in Figure 7BIn the switching circuit SWC2, a high-level potential is supplied from the wiring SWL2a to the control terminal of the switch S2a[k] to turn on the switch S2a[k], and a low-level potential is supplied from the wiring SWL2b to the control terminal of the switch S2b[k] to turn off the switch S2b[k].

[0448] In addition, for example, step ST4 includes an operation of making the circuit WCSa[k] and the wiring WCL1[k] in the circuit WCS non-conductive.

[0449] Specifically, in Figure 2 In the circuit SWCA, a low-level potential is supplied from the wiring SWLA to the control terminal of the switch SA[k] to turn the switch SA[k] off.

[0450] Thus, in the operation unit array MACA1, preparations for performing the first product-sum operation are completed.

[0451] [Step ST5] Step ST5 includes performing the first product-sum operation in the operation unit array MACA1. (1) With the second data X (1) Note that X will be explained later. (1) .

[0452] In step ST5, referring to the above Fig.12 The timing diagram from time T22 to time T23 shows that the second data is sent to each of the units IMD1[1] to IMD1[s] and the units IM1[1,1] to IM1[s,s] of the arithmetic unit array MACA1.

[0453] In step ST5, circuit XCS1 sequentially generates currents corresponding to the second data of the first to sth groups, and sends the second data of each group to units IMD1[1] to IMD1[s] and units IM1[1,1] to IM1[s,s] of the arithmetic unit array MACA1.

[0454] For example, the circuit XCS1 generates the second data of the first group whose amount is X (1) (1) [1]×I Xut To X (1) (1) [s]×I Xut The current is sent to the arithmetic unit array MACA1. Note that X (N) (h) [i] N represents the number of product-sum operations, X (1) (h)[i] means the second data used in the first product-sum operation. (N) (h) The h in [i] means that the second data is included in the hth group (h is an integer greater than or equal to 1 and less than or equal to s). (N) (h) The i in [i] means that the second data is sent to the i-th row of the arithmetic unit array MACA1, that is, to the wiring XCL1[i]. (N) (h) [1] is the second data transmitted to the wiring XCL1[1].

[0455] In addition, for example, the circuit XCS1 generates a quantity X as the second data of the sth group. (1) (s) [1]×I Xut To X (1) (s) [s]×I Xut The current is sent to the computing unit array MACA1.

[0456] Here, in step ST5, the second data input to the operation unit array MACA1 is represented as rows and columns as X (1) Note that the X included in the hth group and input to the i-th row of the operation unit array MACA1 (1) (h) [i] Recorded in X (1) h rows and i columns. At this time, X (1) It can be represented by the following formula.

[0457] [Formula 20]

[0458] Here, consider the case where the second data of the hth group is input to the operation unit array MACA1. At this time, the amount of current flowing from the terminal C1Ti[j] of the current generating circuit CM1 to the wiring WCL1[j] extending in the operation unit array MACA1 is I SUM (1) (h) [j]. Note that I SUM (N) (h) [j] N represents the number of product-sum operations, I SUM (1) (h) [j] means the current amount corresponding to the result of the first product-sum operation. SUM (1) (h) The h of [j] means using the second data included in the h-th group.

[0459] In addition, in step ST5, the switch S1a[j] of the switching circuit SWC1 is in the open state, and the switch S1b[j] of the switching circuit SWC1 is in the closed state, so I SUM (1) (h) The current of [j] flows from the terminal C1Ti[j] of the current generating circuit CM1 to the unit IM1[1,j] to the unit IM1[s,j] in the j-th column of the arithmetic unit array MACA1 through the wiring WCL1[j].

[0460] At this time, the amount of current flowing from each of the terminals C1Ti[1] to C1Ti[s] of the current generating circuit CM1 to the wiring WCL1[1] to the wiring WCL1[s] extending in the arithmetic unit array MACA1 can be expressed as follows using (Formula 1.17).

[0461] [Formula 21]

[0462] In addition, since I is output from the terminal C1Ti[j] of the current generating circuit CM1 SUM (1) (h) [j], so the terminal C1To[j] of the current generating circuit CM1 also outputs I SUM (1) (h) [j] current.

[0463] In step ST5, the switch S2a[j] of the switching circuit SWC2 is turned on, and the switch S2b[j] of the switching circuit SWC2 is turned off. Therefore, the I from the terminal C1To[j] of the current generating circuit CM1 is SUM (1) (h) The current [j] flows to the wiring BL1 [j] of the j-th column of the memory cell array MEMA1.

[0464] Here, in Figure 1 and Figure 4AIn the memory cell array MEMA1, the circuit WWD1 supplies a high level potential to the wiring WWL1[h], and the circuit RWD1 supplies a high level potential to the wiring RWL1[h]. As a result, a high level potential is applied to the gate of the transistor F7 and the gate of the transistor F9 in the cells MC1[h,1] to the cells MC1[h,s] of the h-th row of the memory cell array MEMA1, and each transistor F7 and transistor F9 is in an on state. At this time, a low level potential is supplied from the circuit WWD1 to the wiring WWL1[1] to the wiring WWL1[s] other than the wiring WWL1[h], and a low level potential is supplied from the circuit RWD1 to the wiring RWL1[1] to the wiring RWL1[s] other than the wiring RWL1[h]. Therefore, the transistors F7 and transistors F9 in the cells MC1[1,1] to the cells MC1[s,s] other than the h-th row of the memory cell array MEMA1 are in an off state.

[0465] At this time, the first terminal of the transistor F8 in the cell MC1[h, j] of the h-th row of the memory cell array MEMA1 is in a conductive state with the wiring BL1[j], and the first terminal of the transistor F8 in the cells MC1[1, j] to MC1[s, j] other than the h-th row of the memory cell array MEMA1 is in a non-conductive state with the wiring BL1[j], so the current I SUM (1) (h) The current of [j] flows from the wiring BL1[j] to the cell MC1[h,j].

[0466] In addition, by turning on the transistor F7 in the cell MC1[h, j], the gate and drain of the transistor F8 in the cell MC1[h, j] are in a conducting state (diode connection structure). Therefore, when the current flows from the wiring BL1[j] to the cell MC1[h, j], the potential of the gate of the transistor F8 and the second terminal of the transistor F8 are substantially equal. This potential depends on the amount of current flowing from the wiring BL1[j] to the cell MC1[h, j] and the potential of the first terminal of the transistor F8. This potential is the potential supplied by the wiring VE2, and in this working example, it is, for example, the ground potential GND. In addition, in this working example, the current I SUM (1) (h) The current of [j] flows from the wiring BL1[j] to the cell MC1[h,j], and the potential of the gate of the transistor F8 becomes, for example, V MEM [h, j]. That is, in transistor F8, the gate-source voltage is V MEM [h, j]-GND, as the current flowing between the first terminal and the second terminal of the transistor F8, the current amount I is set SUM (1)(h) [j].

[0467] Then, in Figure 1 and Figure 4A In the memory cell array MEMA1, a low level potential is supplied from the circuit WWD1 to the wiring WWL1[h], and a low level potential is supplied from the circuit RWD1 to the wiring RWL1[h]. As a result, a low level potential is applied to the gate of the transistor F7 and the gate of the transistor F9 in the cells MC1[h,1] to MC1[h,s] of the h-th row of the memory cell array MEMA1, and the respective transistors F7 and F9 are in the off state.

[0468] By turning off the transistor F7 in the cell MC1[h,j], the difference V between the potential of the gate of the transistor F8 and the potential of the wiring VE3 (here, for example, the ground potential GND) is held in the capacitor C6. MEM [h, j] - GND. As a result, the current I flowing between the source and drain of the transistor F8 of the unit MC1 [h, j] can be maintained. SUM (1) (h) [j]. In addition, by turning off the transistor F9 in the cell MC1 [h, j], the current I in the cell MC1 [h, j] is set. SUM (1) (h) The current of [j] does not flow from the wiring BL1[j] to the cell MC1[h,j].

[0469] As described above, corresponding to W as the first data (1) The j-th column of W (1) [1, j] to W (1) [s, j] and X as the second data of the hth group (1) (h) [1] to X (1) (h) The value of the product-sum operation of [s] is the current I SUM (1) (h) [j] is set in MC1[h, j] of the memory cell array MEMA1.

[0470] Specifically, for example, corresponding to W as the first data (1) The first column of W (1) [1, 1] to W (1) [s, 1] and X as the second data of the first group (1) (1) [1] to X (1) (1) The value of the product-sum operation of [s] is the current I SUM(1) (1) [1] is set in the cell MC1[1,1] of the memory cell array MEMA1. In addition, for example, corresponding to W as the first data (1) W of the sth column (1) [1, s] to W (1) [s, s] and X as the second data of the sth group (1) (s) [1] to X (1) (s) The value of the product-sum operation of [s] is the current I SUM (1) (s) [s] is set in MC1[s, s] of the memory cell array MEMA1.

[0471] In the operation cell array MACA1, the second data of the first group to the sth group are sequentially input from the circuit XCS1. Each time the second data is input from the circuit XCS1, the operation result is sequentially written from the first row of the memory cell array MEMA1. The current amount of each of the cells MC1[1,1] to MC1[s,s] set in the memory cell array MEMA1 is recorded as I SUM (1) Note that, corresponding to W as the first data (1) The j-th column of W (1) [1, j] to W (1) [s, j] and X as the second data of the hth group (1) (h) [1] to X (1) (h) The value of the product-sum operation of [s] is the current I SUM (1) (h) [j] Recorded in I SUM (1) h rows and j columns. At this time, I SUM (1) It can be represented by the following formula.

[0472] [Formula 22]

[0473] The first product-sum operation is completed at step ST5. Note that when the process goes from step ST5 to step ST6, the second product-sum operation is performed after step ST6.

[0474] Note that the work performed in steps ST3 to ST5 is equivalent to Fig. 13B The block diagram shown in the figure shows the operation. At this time, I SUM (1) Equivalent to X(1) W (1) .

[0475] [Step ST6] Step ST6 includes preliminary work of writing a plurality of first data to the operation unit array MACA2 for performing the N+1th (N+1 in step ST6 is an even number greater than 2) product-sum operation. Note that N+1 is 2 when performing step ST6 for the first time.

[0476] For example, step ST6 includes the operation of switching circuit SWC1 to make the terminals T1a[k] and T1b[k] non-conductive, to make the terminals T1a[k] and T1c[k] conductive, and to make the terminals T1c[k] and T1d[k] non-conductive.

[0477] Specifically, in Fig. 7A In the switching circuit SWC1, a low-level potential is supplied from the wiring SWL1a to the control terminal of the switch S1a[k] to turn off the switch S1a[k], a high-level potential is supplied from the wiring SWL1b to the control terminal of the switch S1b[k] to turn on the switch S1b[k], and a low-level potential is supplied from the wiring SWL1c to the control terminal of the switch S1c[k] to turn off the switch S1c[k].

[0478] In addition, for example, step ST6 includes an operation in which the switching circuit SWC6 causes the terminal T6a[k] and the terminal T6b[k] to be in a conductive state, and causes the terminal T6b[k] and the terminal T6c[k] to be in a non-conductive state.

[0479] Specifically, in Figure 8B In the switching circuit SWC6, a high-level potential is supplied from the wiring SWL6a to the control terminal of the switch S6a[k] to turn on the switch S6a[k], and a low-level potential is supplied from the wiring SWL6b to the control terminal of the switch S6b[k] to turn off the switch S6b[k].

[0480] In addition, for example, step ST6 includes an operation of bringing the circuit WCSa[k] and the wiring WCL1[k] into conduction in the circuit WCS.

[0481] Specifically, in Figure 2 In the circuit SWCA, a high-level potential is supplied from the wiring SWLA to the control terminal of the switch SA[k] to turn on the switch SA[k].

[0482] In addition, for example, step ST6 includes an operation of making the circuit ITRZa[k] and the wiring WCL2[k] in the circuit ITRZ non-conductive.

[0483] Specifically, in the circuit SWCB of FIG. 11 , a low-level potential is supplied from the wiring SWLB to the control terminal of the switch SB[k] to turn the switch SB[k] off.

[0484] That is, through the above operation, the circuit WCSa[k], the wiring WCL1[k], and the wiring WCL2[k] inside the circuit WCS are in a conductive state, thereby completing the preparation for writing a plurality of first data into the arithmetic unit array MACA2.

[0485] [Step ST7] Step ST7 includes a process of writing a plurality of first data into the operation unit array MACA2 in order to perform the N+1th (in the case of performing step ST7 for the first time, N+1=2) product-sum operation.

[0486] For example, referring to the above Fig.12 The first data is written into the units IM2[1,1] to IM2[s,s] of the operation unit array MACA2 from time T11 to time T20. In addition, the first data written into the unit IM2[i,j] of the operation unit array MACA2 is denoted as W (N+1) [i, j].

[0487] At this time, the first data written into the operation unit array MACA2 is recorded as W in rows and columns. (N+1) In addition, W (N+1) It can be expressed as follows.

[0488] [Formula 23]

[0489] Note that the operations performed in steps ST6 and ST7 are equivalent to the operations performed in the case of N=2. Fig. 13C The block diagram shown works.

[0490] [Step ST8] Step ST8 includes preliminary operations for inputting a plurality of second data into the operation unit array MACA2 when performing the N+1th (in the case of performing step ST8 for the first time, N+1=2) product-sum operation.

[0491] For example, step ST8 includes an operation in which the switching circuit SWC6 causes the terminal T6a[k] and the terminal T6b[k] to be in a non-conductive state and causes the terminal T6b[k] and the terminal T6c[k] to be in a conductive state.

[0492] Specifically, in Figure 8BIn the switching circuit SWC6, a low-level potential is supplied from the wiring SWL6a to the control terminal of the switch S6a[k] to turn off the switch S6a[k], and a high-level potential is supplied from the wiring SWL6b to the control terminal of the switch S6b[k] to turn on the switch S6b[k].

[0493] [Step BR4] In step BR4, it is determined whether the operation circuit CDV has completed the processing with the N+1th (when step BR4 is performed for the first time, N+1=2) product-sum operation. In the operation circuit CDV, if the processing has not been completed with the N+1th product-sum operation, the process proceeds to step ST9, and if the processing has been completed with the N+1th product-sum operation, the process proceeds to step ST15.

[0494] [Step ST9] Step ST9 includes preliminary work for performing the N+1th (in the case of performing step ST7 for the first time, N+1=2) product-sum operation.

[0495] For example, step ST9 includes an operation in which the switching circuit SWC1 causes the terminal T1a[k] and the terminal T1c[k] to be in a non-conductive state and causes the terminal T1c[k] and the terminal T1d[k] to be in a conductive state.

[0496] Specifically, in Fig. 7A In the switching circuit SWC1, a low-level potential is supplied from the wiring SWL1b to the control terminal of the switch S1b[k] to turn off the switch S1b[k], and a high-level potential is supplied from the wiring SWL1c to the control terminal of the switch S1c[k] to turn on the switch S1c[k].

[0497] In addition, for example, step ST9 includes an operation in which the switching circuit SWC2 causes the terminal T2a[k] and the terminal T2b[k] to be in a non-conductive state and causes the terminal T2b[k] and the terminal T2c[k] to be in a conductive state.

[0498] Specifically, in Figure 7B In the switching circuit SWC2, a low-level potential is supplied from the wiring SWL2a to the control terminal of the switch S2a[k] to turn off the switch S2a[k], and a high-level potential is supplied from the wiring SWL2b to the control terminal of the switch S2b[k] to turn on the switch S2b[k].

[0499] In addition, for example, step ST9 includes an operation in which the switching circuit SWC3 causes the terminal T3a[k] ​​and the terminal T3b[k] to be in a conductive state, and causes the terminal T3b[k] and the terminal T3c[k] to be in a non-conductive state.

[0500] Specifically, in Figure 7C In the switching circuit SWC3, a high-level potential is supplied from the wiring SWL3a to the control terminal of the switch S3a[k] ​​to turn on the switch S3a[k], and a low-level potential is supplied from the wiring SWL3b to the control terminal of the switch S3b[k] to turn off the switch S3b[k].

[0501] In addition, for example, step ST9 includes an operation of making the circuit ITRZa[k] and the wiring WCL2[k] in the circuit ITRZ non-conductive.

[0502] Specifically, in the circuit SWCB of FIG. 11 , a low-level potential is supplied from the wiring SWLB to the control terminal of the switch SB[k] to turn the switch SB[k] off.

[0503] Thus, in the operation unit array MACA2, preparations for performing the N+1th product-sum operation are completed.

[0504] [Step ST10] Step ST10 includes performing the N+1th product-sum operation of the plurality of first data and the plurality of second data in the operation unit array MACA2. Note that in the case of performing the first step ST10, N+1=2nd time. (N+1) With the second data X (N+1) Note that X will be explained later. (N+1) .

[0505] In step ST10, the results of the first product-sum operation of the first row to the s-th row are sequentially read out from the memory cell array MEMA1.

[0506] Here, consider the case where the first operation result is read from the cells MC1[h, 1] to the cells MC1[h, s] of the h-th row of the memory cell array MEMA1. Specifically, for example, in the memory cell array MEMA1, the circuit RWD1 supplies a high-level potential to the wiring RWL1[h]. As a result, a high-level potential is applied to the gate of the transistor F9 in the cells MC1[h, 1] to the cells MC1[h, s] of the h-th row of the memory cell array MEMA1, and each transistor F9 is in an on state. At this time, a low-level potential is supplied from the circuit RWD1 to the wiring RWL1[1] to the wiring RWL1[s] other than the wiring RWL1[h]. Therefore, the transistor F9 in the cells MC1[1, 1] to the cells MC1[s, s] other than the h-th row of the memory cell array MEMA1 is in an off state.

[0507] At this time, the first terminal of the transistor F8 in the cell MC1[h, j] of the h-th row of the memory cell array MEMA1 is in a conducting state with the wiring BL1[j], and the first terminal of the transistor F8 in the cells MC1[1, j] to MC1[s, j] other than the h-th row of the memory cell array MEMA1 is in a non-conducting state with the wiring BL1[j]. Furthermore, due to the switching circuit SWC2, the terminal R1Ti[j] of the current generating circuit RL1 is in a conducting state with the wiring BL1[j], so the current I SUM (N) (h) The current of [j] flows from the terminal R1Ti[j] of the current generating circuit RL1 to the cell MC1 [h, j] through the wiring BL1 [j].

[0508] The current generating circuit RL1 generates a current corresponding to the amount of current I input to the terminal R1Ti[j]. SUM (N) (h) [j] is the input value and a function operation is performed, and a current is output to the terminal R1To[j], and the current corresponds to the result of the operation. SUM (N) (h) The result of the function when [j] is the input value is defined as F(I SUM (N) (h) [j]) = X (N+1) (h) [j]×I Xut That is, the current generating circuit RL1 receives a current I through the terminal R1Ti[j]. SUM (N) (h) The current [j] is output from terminal R1To[j], and the current X (N+1) (h) [j]×I Xut In addition, X (N+1) (h) [j] is the second data in the N+1th product-sum operation.

[0509] Here, the second data output from the terminal R1To[1] to the terminal R1To[s] of the current generating circuit RL1 is represented as X in rows and columns. (N+1) Note that since the current I flows from the terminal R1Ti[j] of the current generating circuit RL1 to the cell MC1[h, j] of the h-th row of the memory cell array MEMA1 SUM (N) (h) The current of [j], the second data X output from the terminal R1To[j] (N+1)(h) [j] Recorded in X (N+1) The h-th row and j-th column of .

[0510] [Formula 24]

[0511] In addition, X (N+1) The h line of X (N+1) (h) [1] to X (N+1) (h) [s] is the second data of the hth group of the N+1th operation circuit.

[0512] In addition, due to the switching circuit SWC6, the terminal R1To[i] of the current generating circuit RL1 and the wiring XCL2[i] are in a conductive state, so the current I SUM (N+1) (h) The current of [i] flows from the terminal R1To[i] of the current generating circuit RL1 to the cell IMD2[i], the cell IM2[i,1] to the cell IM2[i,s] through the wiring XCL2[i].

[0513] Here, by reading the results of the N-th product-sum operation from the first row to the s-th row from the memory cell array MEMA1 row by row, currents corresponding to the second data of the first group to the s-th group are sequentially output from the terminal R1To[1] to the terminal R1To[s] of the current generating circuit RL1. Thus, the second data of a group set in each row is sent to the units IMD2[1] to the units IMD2[s] and the units IM2[1,1] to the units IM2[s,s] of the arithmetic unit array MACA2 in the order from the first row to the s-th row of the memory cell array MEMA1.

[0514] Specifically, for example, the current generating circuit RL1 converts its quantity to X (N+1) (1) [1]×I Xut To X (N+1) (1) [s]×I Xut The current of the first group is output to the operation unit array MACA1 as the second data of the first group. In addition, for example, the current generating circuit RL1 outputs its quantity as X (N+1) (s) [1]×I Xut To X (N+1) (s) [s]×I Xut The current is output as the second data of the sth group to the operation unit array MACA1.

[0515] Note that, regarding the product-sum operation in the operation unit array MACA2, reference is made to Fig.12 The timing diagram from time T22 to time T23 or step ST5 is described.

[0516] Here, consider the case where the second data of the hth group is input to the operation unit array MACA1. At this time, the amount of current flowing from the terminal C2Ti[j] of the current generating circuit CM2 to the wiring WCL2[j] extending in the operation unit array MACA2 is I SUM (N+1) (h) [j].

[0517] In addition, in step ST10, the switch S1b[j] of the switching circuit SWC1 is in the closed state, and the switch S1c[j] of the switching circuit SWC1 is in the open state, so I SUM (N+1) (h) The current of [j] flows from the terminal C2Ti[j] of the current generating circuit CM2 to the unit IM2[1,j] to the unit IM2[s,j] in the j-th column of the arithmetic unit array MACA2 through the wiring WCL2[j].

[0518] At this time, the amount of current flowing from each of the terminals C2Ti[1] to C2Ti[s] of the current generating circuit CM2 to the wiring WCL2[1] to the wiring WCL2[s] extending in the operation unit array MACA2 can be expressed as follows using (Formula 1.17).

[0519] [Formula 25]

[0520] In addition, since I is output from the terminal C2Ti[j] of the current generating circuit CM2 SUM (N+1) (h) [j], so the terminal C2To[j] of the current generating circuit CM2 also outputs I SUM (N+1) (h) [j] current.

[0521] In step ST10, the switch S2a[j] of the switching circuit SWC2 is turned on, and the switch S2b[j] of the switching circuit SWC2 is turned off. Therefore, the I2 from the terminal C2To[j] of the current generating circuit CM2 is SUM (N+1) (h) The current [j] flows to the wiring BL2[j] of the j-th column of the memory cell array MEMA2.

[0522] Here, in Figure 1 and Figure 4BIn the memory cell array MEMA2, a high level potential is supplied from the circuit WWD2 to the wiring WWL2[h], and a high level potential is supplied from the circuit RWD2 to the wiring RWL2[h]. As a result, a high level potential is applied to the gate of the transistor F7 and the gate of the transistor F9 in the cells MC2[h,1] to the cells MC2[h,s] of the h-th row of the memory cell array MEMA2, and each transistor F7 and transistor F9 is in an on state. At this time, a low level potential is supplied from the circuit WWD2 to the wiring WWL2[1] to the wiring WWL2[s] other than the wiring WWL2[h], and a low level potential is supplied from the circuit RWD2 to the wiring RWL2[1] to the wiring RWL2[s] other than the wiring RWL2[h]. Therefore, the transistors F7 and transistors F9 in the cells MC2[1,1] to the cells MC2[s,s] other than the h-th row of the memory cell array MEMA2 are in an off state.

[0523] At this time, the first terminal of the transistor F8 in the cell MC1[h, j] of the h-th row of the memory cell array MEMA2 is in a conductive state with the wiring BL2[j], and the first terminal of the transistor F8 in the cells MC2[1, j] to MC2[s, j] other than the h-th row of the memory cell array MEMA2 is in a non-conductive state with the wiring BL2[j], so the current I SUM (2) (h) The current of [j] flows from the wiring BL2[j] to the cell MC2[h,j].

[0524] In addition, by turning on the transistor F7 in the cell MC2[h, j], the gate and drain of the transistor F8 in the cell MC2[h, j] are in a conducting state (diode connection). Therefore, when the current flows from the wiring BL2[j] to the cell MC2[h, j], the potential of the gate of the transistor F8 and the second terminal of the transistor F8 are substantially equal. This potential depends on the amount of current flowing from the wiring BL2[j] to the cell MC2[h, j] and the potential of the first terminal of the transistor F8. This potential is the potential supplied by the wiring VE2, and in the present working example, it is, for example, the ground potential GND. In addition, in the present working example, the current I SUM (N+1) (h) The current of [j] flows from the wiring BL2[j] to the cell MC2[h,j], and the potential of the gate of the transistor F8 becomes, for example, V MEM [h, j]. That is, in transistor F8, the gate-source voltage is V MEM [h, j]-GND, as the current flowing between the first terminal and the second terminal of the transistor F8, the current amount I is set SUM (N+1)(h) [j].

[0525] Then, in Figure 1 and Figure 4A In the memory cell array MEMA2, a low level potential is supplied from the circuit WWD2 to the wiring WWL2[h], and a low level potential is supplied from the circuit RWD2 to the wiring RWL2[h]. As a result, a low level potential is applied to the gate of the transistor F7 and the gate of the transistor F9 in the cells MC2[h,1] to the cells MC2[h,s] of the h-th row of the memory cell array MEMA2, and the respective transistors F7 and F9 are in the off state.

[0526] By turning off the transistor F7 in the cell MC2 [h, j], the difference V between the potential of the gate of the transistor F8 and the potential of the wiring VE3 (here, for example, the ground potential GND) is held in the capacitor C6. MEM [h, j] - GND. Thus, the current I flowing between the source and drain of the transistor F8 of the unit MC2 [h, j] can be maintained. SUM (N+1) (h) [j]. In addition, by turning off the transistor F9 in the cell MC2 [h, j], the current I in the cell MC2 [h, j] is set. SUM (N+1) (h) The current of [j] does not flow from the wiring BL2[j] to the cell MC2[h,j].

[0527] As described above, corresponding to W as the first data (N+1) The j-th column of W (N+1) [1, j] to W (N+1) [s, j] and X as the second data of the hth group (N+1) (h) [1] to X (N+1) (h) The value of the product-sum operation of [s] is the current I SUM (N+1) (h) [j] is set in MC2[h, j] of the memory cell array MEMA2.

[0528] Specifically, for example, corresponding to W as the first data (N+1) The first column of W (N+1) [1, 1] to W (N+1) [s, 1] and X as the second data of the first group (N+1) (1) [1] to X (N+1) (1) The value of the product-sum operation of [s] is the current I SUM(N +1) (1) [1] is set in the cell MC2[1,1] of the memory cell array MEMA2. In addition, for example, corresponding to W as the first data (N +1) W of the sth column (N+1) [1, s] to W (N+1) [s, s] and X as the second data of the sth group (N+1) (s) [1] to X (N+1) (s) The value of the product-sum operation of [s] is the current I SUM (N+1) (s) [s] is set in MC2[s, s] of the memory cell array MEMA2.

[0529] In the operation cell array MACA2, the second data of the first group to the sth group are sequentially input from the memory cell array MEMA1. Each time the second data is input from the circuit XCS1, the operation result is sequentially written from the first row of the memory cell array MEMA2. The current amount of each of the cells MC2[1,1] to MC2[1,1] set in the memory cell array MEMA2 is recorded as I SUM (N+1) Note that, corresponding to W as the first data (N+1) The j-th column of W (N+1) [1, j] to W (N+1) [s, j] and X as the second data of the hth group (N+1) (h) [1] to X (N+1) (h) The value of the product-sum operation of [s] is the current I SUM (N +1) (h) [j] Recorded in I SUM (N+1) h rows and j columns. At this time, I SUM (N+1) It can be represented by the following formula.

[0530] [Formula 26]

[0531] The N+1th product-sum operation is completed by step ST10. After step ST10 is completed, the value of N is increased by 2, and the process proceeds to step ST1. For example, when N=1 in step ST10 (when the N+1=2th product-sum operation is performed in step ST10), N=3 in the next step ST1.

[0532] Note that the operations performed in steps ST8 to ST10 are equivalent to the operations performed in the case where N=2. Fig.14A The block diagram shown in the figure shows the operation. At this time, I SUM (2) Equivalent to X (2) W (2) .

[0533] [Step ST11] Step ST11 includes preliminary operations for performing the Nth (in step ST11, N is an odd number greater than or equal to 3) product-sum operation. Note that N is 3 when step ST11 is performed for the first time.

[0534] For example, step ST11 includes an operation in which the switching circuit SWC1 causes the terminal T1a[k] to be in a conductive state with the terminal T1b[k] and causes the terminal T1a[k] to be in a non-conductive state with the terminal T1b[k].

[0535] Specifically, in Fig. 7A In the switching circuit SWC1, a high level potential is supplied from the wiring SWL1a to the control terminal of the switch S1a[k] to turn on the switch S1a[k], and a low level potential is supplied from the wiring SWL1b to the control terminal of the switch S1b[k] to turn off the switch S1b[k]. Note that this operation is equivalent to Fig.12 The timing diagram shows the operation of the switching circuit SWC1 from time T21 to time T22.

[0536] In addition, for example, step ST11 includes an operation in which the switching circuit SWC2 causes the terminal T2a[k] and the terminal T2b[k] to be in a conductive state, and causes the terminal T2b[k] and the terminal T2c[k] to be in a non-conductive state.

[0537] Specifically, in Figure 7B In the switching circuit SWC2, a high-level potential is supplied from the wiring SWL2a to the control terminal of the switch S2a[k] to turn on the switch S2a[k], and a low-level potential is supplied from the wiring SWL2b to the control terminal of the switch S2b[k] to turn off the switch S2b[k].

[0538] In addition, for example, step ST11 includes an operation in which the switching circuit SWC3 puts the terminal T3a[k] ​​and the terminal T3b[k] in a non-conductive state and puts the terminal T3b[k] and the terminal T3c[k] in a conductive state.

[0539] Specifically, in Figure 7CIn the switching circuit SWC3, a low-level potential is supplied from the wiring SWL3a to the control terminal of the switch S3a[k] ​​to turn off the switch S3a[k], and a high-level potential is supplied from the wiring SWL3b to the control terminal of the switch S3b[k] to turn on the switch S3b[k].

[0540] In addition, for example, step ST11 includes an operation of making the circuit WCSa[k] and the wiring WCL1[k] in the circuit WCS non-conductive.

[0541] Specifically, in Figure 2 In the circuit SWCA, a low-level potential is supplied from the wiring SWLA to the control terminal of the switch SA[k] to turn the switch SA[k] off.

[0542] Thus, in the operation unit array MACA1, preparation for performing the N-th product-sum operation of the plurality of first data and the plurality of second data is completed.

[0543] [Step ST12] Step ST12 includes performing the Nth (in step ST12, N is an odd number greater than 3) product-sum operation in the operation unit array MACA1. (N) With the second data X (N) Note that X will be explained later. (N) .

[0544] In step ST12, the results of the N-1th product-sum operation in the 1st row to the sth row are sequentially read out from the memory cell array MEMA2.

[0545] Here, consider the case where the N-1th operation result is read from the cell MC2[h, 1] to the cell MC2[h, s] of the h-th row of the memory cell array MEMA2. Specifically, for example, in the memory cell array MEMA2, the circuit RWD2 supplies a high-level potential to the wiring RWL2[h]. As a result, a high-level potential is applied to the gate of the transistor F9 in the cell MC2[h, 1] to the cell MC2[h, s] of the h-th row of the memory cell array MEMA2, and each transistor F9 is in an on state. At this time, a low-level potential is supplied from the circuit RWD2 to the wiring RWL2[1] to the wiring RWL2[s] other than the wiring RWL2[h]. Therefore, the transistor F9 in the cell MC2[1, 1] to the cell MC2[s, s] other than the h-th row of the memory cell array MEMA2 is in an off state.

[0546] At this time, the first terminal of the transistor F8 in the cell MC2[h,j] of the h-th row of the memory cell array MEMA2 is in a conductive state with the wiring BL2[j], and the first terminal of the transistor F8 in the cells MC2[1,j] to MC2[s,j] other than the h-th row of the memory cell array MEMA2 is in a non-conductive state with the wiring BL2[j]. Furthermore, due to the switching circuit SWC3, the terminal R2Ti[j] of the current generating circuit RL2 is in a conductive state with the wiring BL2[j], so the current I SUM (N-1) (h) The current of [j] flows from the terminal R2Ti[j] of the current generating circuit RL2 to the cell MC1 [h, j] through the wiring BL2 [j].

[0547] Similar to the current generating circuit RL1, the current generating circuit RL2 generates a current corresponding to the current amount I input to the terminal R2Ti[j]. SUM (N-1) (h) [j] is the input value and a function operation is performed, and a current is output to the terminal R2To[j], and the current corresponds to the result of the operation. SUM (N-1) (h) The result of the function when [j] is the input value is defined as F(I SUM (N-1) (h) [j]) = X (N) (h) [j]×I Xut That is, the current generating circuit RL2 receives a current I through the terminal R2Ti[j]. SUM (N-1) (h) The current [j] is output from terminal R2To[j], and the current X (N) (h) [j]×I Xut In addition, X (N) (h) [j] is the second data in the Nth product-sum operation.

[0548] Here, the second data output from the terminal R2To[1] to the terminal R2To[s] of the current generating circuit RL2 is represented as X in rows and columns. (N) Note that since the current I flows from the terminal R2Ti[j] of the current generating circuit RL2 to the cell MC2[h, j] of the h-th row of the memory cell array MEMA2 SUM (N-1) (h) The current of [j], the second data X output from the terminal R2To[j](N) (h) [j] Recorded in X (N) The h-th row and j-th column of .

[0549] [Formula 27]

[0550] In addition, X (N) The h line of X (N) (h) [1] to X (N) (h) [s] is the second data of the hth group of the Nth operation circuit.

[0551] In addition, due to the switching circuit SWC5, the terminal R2To[i] of the current generating circuit RL2 and the wiring XCL1[i] are in a conductive state, so the current I SUM (N) (h) The current of [i] flows from the terminal R2To[i] of the current generating circuit RL2 to the cell IMD1[i], the cell IM1[i,1] to the cell IM1[i,s] through the wiring XCL1[i].

[0552] Here, by reading the results of the N-1th product-sum operation from the first row to the sth row from the memory cell array MEMA2 row by row, currents corresponding to the second data of the first group to the sth group are sequentially output from the terminal R2To[1] to the terminal R2To[s] of the current generating circuit RL2. Thus, the second data of a group set in each row is sent to the units IMD1[1] to the units IMD1[s] and the units IM1[1,1] to the units IM1[s,s] of the operation unit array MACA1 in the order from the first row to the sth row of the memory cell array MEMA2.

[0553] Then, similarly to step ST5, a product-sum operation is performed in the operation cell array MACA1, and the operation result is written into the memory cell array MEMA1. At this time, the current amount of each of the cells MC1[1,1] to MC1[s,s] set in the memory cell array MEMA1 is recorded as I SUM (N) In addition, corresponding to W as the first data (N) The j-th column of W (N) [1, j] to W (N) [s, j] and X as the second data of the hth group (N) (h) [1] to X (N) (h) The value of the product-sum operation of [s] is the current I SUM (N)(h) [j] Recorded in I SUM (N) h rows and j columns. At this time, I SUM (N) It can be expressed by the following formula.

[0554] [Formula 28]

[0555] The Nth product-sum operation is completed at step ST12. Note that when the process goes from step ST12 to step ST6, the N+1th product-sum operation is performed after step ST6.

[0556] Note that the operations performed in step ST11 and step ST12 are equivalent to the operations performed in step ST11 and step ST12 in the case of N=3. Fig. 14C The block diagram shown in the figure shows the operation. At this time, I SUM (3) Equivalent to X (3) W (3) .

[0557] [Step ST13] Step ST13 includes preliminary work for inputting a plurality of second data into the operation unit array MACA1 and performing the final product-sum operation.

[0558] For example, step ST13 includes the operation of switching circuit SWC1 to make terminals T1a[k] and T1b[k] non-conductive, terminals T1a[k] and T1c[k] conductive, and terminals T1c[k] and T1d[k] non-conductive.

[0559] Specifically, in Fig. 7A In the switching circuit SWC1, a low-level potential is supplied from the wiring SWL1a to the control terminal of the switch S1a[k] to turn off the switch S1a[k], a high-level potential is supplied from the wiring SWL1b to the control terminal of the switch S1b[k] to turn on the switch S1b[k], and a low-level potential is supplied from the wiring SWL3a to the control terminal of the switch S1c[k] to turn off the switch S1c[k].

[0560] In addition, for example, step ST13 includes an operation in which the switching circuit SWC6 brings the terminal T6a[k] and the terminal T6b[k] into conduction and brings the terminal T6b[k] and the terminal T6c[k] into non-conduction.

[0561] Specifically, in Figure 8BIn the switching circuit SWC6, a high-level potential is supplied from the wiring SWL6a to the control terminal of the switch S6a[k] to turn on the switch S6a[k], and a low-level potential is supplied from the wiring SWL6b to the control terminal of the switch S6b[k] to turn off the switch S6b[k].

[0562] In addition, for example, step ST13 includes an operation of making the circuit WCSa[k] and the wiring WCL1[k] in the circuit WCS non-conductive.

[0563] Specifically, in Figure 2 In the circuit SWCA, a low-level potential is supplied from the wiring SWLA to the control terminal of the switch SA[k] to turn the switch SA[k] off.

[0564] In addition, for example, step ST13 includes an operation of making the circuit ITRZa[k] and the wiring WCL2[k] in the circuit ITRZ non-conductive.

[0565] Specifically, in the circuit SWCB of FIG. 11 , a high-level potential is supplied from the wiring SWLB to the control terminal of the switch SB[k] to turn on the switch SB[k].

[0566] Thus, in the operation unit array MACA1, preparations for the final sum-of-products operation are completed.

[0567] [Step ST14] Step ST14 includes performing the final product-sum operation in the operation unit array MACA1. (N) With the second data X (N) The product-sum operation of (where N is an odd number greater than 1).

[0568] Specifically, for example, in step ST14, similarly to step ST5 or step ST10, second data is sent from wiring XCL1[1] to wiring XCL1[s] to each of cells IMD1[1] to IMD1[s] and cells IM1[1,1] to IM1[s,s] of the arithmetic cell array MACA1.

[0569] Here, in step ST14, the second data input to the operation unit array MACA1 is represented as rows and columns as X (N) (Here, N is an odd number greater than 1.) (N) Same as formula (1.27).

[0570] In addition, in step ST14, similar to step ST5 or step ST10, the second data of the first group to the sth group are sequentially input into each of the units IMD1[1] to IMD1[s] and the units IM1[1,1] to IM1[s,s] of the operation unit array MACA1, thereby flowing a current through each of the wirings WCL1[1] to WCL1[s], and the amount of the current corresponds to the result of the product-sum operation of the first data and the second data.

[0571] In step ST14, the circuit XCS2 supplies the ground potential GND to each of the wirings XCL2[1] to XCL2[s]. Thus, in the j-th column of the arithmetic cell array MAC2, the current does not flow from the wiring WCL2[j] to the cells IM2[1,j] to IM2[s,j] of the arithmetic cell array MAC2.

[0572] In step ST14, the switch S1a[j] of the switching circuit SWC1 is in the closed state, the switch S1b[j] of the switching circuit SWC1 is in the open state, and the switch S1c[j] of the switching circuit SWC1 is in the closed state. Fig.11A The switch SB[j] (not shown) of the circuit SWCB is in the open state.

[0573] Therefore, I SUM (N) (h) The current [j] flows from the circuit ITRZa[j] (not shown) of the circuit ITRZ of FIG. 11 through the wiring WCL1[j] and the wiring WCL2[j] to the cell IM1[1,j] to the cell IM1[s,j] in the j-th column of the arithmetic unit array MACA1.

[0574] Here, consider the case where the second data of the hth group is input to the operation unit array MACA1. At this time, the amount of current flowing from the circuit ITRZa[j] to the wiring WCL1[j] extending in the operation unit array MACA1 is I SUM (N) (h) [j]. Note that I SUM (N) (h) [j] N represents the number of product-sum operations, I SUM (N) (h) [j] means the current amount corresponding to the result of the Nth product-sum operation. SUM (N) (h) The h of [i] means using the second data included in the h-th group.

[0575] At this time, the amount of current flowing from each of the circuit ITRZa[1] to the circuit ITRZa[s] to the wiring WCL1[1] to the wiring WCL1[s] extending in the arithmetic unit array MACA1 can be expressed as follows using (Formula 1.17).

[0576] [Formula 29]

[0577] In addition, in each of the circuits ITRZa[1] to ITRZa[s], the switching circuit RL3[j] is switched to correspond to the amount of current I input to the terminal R3Ti[j]. SUM (N) (h) The value of [j] is used as the input value to perform a function operation, and a current is output to the terminal R3To[j], and the current corresponds to the result of the operation. SUM (N) (h) The result of the function when [j] is the input value is defined as F(I SUM (N) (h) [j]) = T (h) [j]×I Tut . Note that I Tut For T (h) The reference current that flows when [j] is 1. The conversion circuit RL3 is set to SUM (N) (h) The current of [j] is input to terminal R3Ti[j], and the current T is output from terminal R3To[j]. (h) [j]×I Tut In addition, T (h) [j] is the calculation result obtained using the calculation circuit CDV.

[0578] Here, the calculation results output from each terminal R3To[1] to terminal R3To[s] of the conversion circuit RL3[1] to the conversion circuit RL3[s] are recorded as T. Note that the current I SUM (N) (h) The current of [j] flows from the terminal R3Ti[j] of the switching circuit RL3[j] to the cell MC1[h, j] of the h-th row of the memory cell array MEMA1, and the operation result T is output from the terminal R3To[j]. (h) [j] is recorded in the hth row and jth column of T.

[0579] [Formula 30]

[0580] Note that the hth row of T (h) [1] to T (h) [s] is a result obtained by using the second data of the hth group through the Nth product-sum operation and output from each terminal R3To[1] to the terminal R3To[s] of the conversion circuit RL3[1] to the conversion circuit RL3[s].

[0581] In the operation unit array MACA1, by sequentially inputting the second data of the first to s-th groups, the operation results of the first to s-th groups are sequentially output from the terminals R3To[1] to R3To[s] of the conversion circuit RL3.

[0582] Then, in the circuit ITRZa[j], the operation results T are sequentially outputted from the terminal R3To[j] of the conversion circuit RL3. (1) [j] to T (s) [j] is converted into a digital signal by the analog-to-digital conversion circuit ADC and output from the wiring OL[j].

[0583] Note that the operations performed in steps ST13 and ST14 are equivalent to Fig.15A The block diagram shows the operation.

[0584] [Step ST15] Step ST15 includes preliminary work for inputting a plurality of second data into the operation unit array MACA2 and performing the final product-sum operation.

[0585] For example, step ST15 includes the operation of switching circuit SWC1 to make terminals T1a[k] and T1b[k] non-conductive, terminals T1a[k] and T1c[k] conductive, and terminals T1c[k] and T1d[k] non-conductive.

[0586] Specifically, in Fig. 7A In the switching circuit SWC1, a low-level potential is supplied from the wiring SWL1a to the control terminal of the switch S1a[k] to turn off the switch S1a[k], a high-level potential is supplied from the wiring SWL1b to the control terminal of the switch S1b[k] to turn on the switch S1b[k], and a low-level potential is supplied from the wiring SWL3a to the control terminal of the switch S1c[k] to turn off the switch S1c[k].

[0587] In addition, for example, step ST15 includes an operation in which the switching circuit SWC5 brings the terminal T5a[k] and the terminal T5b[k] into conduction and brings the terminal T5b[k] and the terminal T5c[k] into non-conduction.

[0588] Specifically, in Fig. 8A In the switching circuit SWC5, a high-level potential is supplied from the wiring SWL5a to the control terminal of the switch S5a[k] to turn on the switch S5a[k], and a low-level potential is supplied from the wiring SWL5b to the control terminal of the switch S5b[k] to turn off the switch S5b[k].

[0589] In addition, for example, step ST15 includes an operation of making the circuit WCSa[k] and the wiring WCL1[k] in the circuit WCS non-conductive.

[0590] Specifically, in Figure 2 In the circuit SWCA, a low-level potential is supplied from the wiring SWLA to the control terminal of the switch SA[k] to turn the switch SA[k] off.

[0591] In addition, for example, step ST15 includes an operation of making the circuit ITRZa[k] and the wiring WCL2[k] in the circuit ITRZ non-conductive.

[0592] Specifically, in the circuit SWCB of FIG. 11 , a high-level potential is supplied from the wiring SWLB to the control terminal of the switch SB[k] to turn on the switch SB[k].

[0593] Thus, in the arithmetic unit array MACA2, preparations for the final sum-of-products operation are completed.

[0594] [Step ST16] Step ST16 includes performing the product-sum operation of the last plurality of first data and the plurality of second data in the operation unit array MACA2. (N+1) With the second data X (N+1) The product-sum operation of (here, N+1 is an even number greater than 2).

[0595] Specifically, for example, in step ST16, similarly to step ST5 or step ST10, second data is sent from wiring XCL2[1] to wiring XCL2[s] to each of cells IMD2[1] to IMD2[s] and cells IM2[1,1] to IM2[s,s] of the arithmetic unit array MACA2.

[0596] Here, in step ST16, the second data input to the operation unit array MACA2 is represented as rows and columns as X (N+1) (Here, N+1 is an even number greater than 2.) (N) Same as formula (1.27).

[0597] In addition, in step ST16, similar to step ST5 or step ST10, the second data of the first group to the sth group are sequentially input into each of the units IMD2[1] to IMD2[s] and the units IM2[1,1] to IM2[s,s] of the operation unit array MACA2, thereby flowing a current through each of the wirings WCL2[1] to WCL2[s], and the amount of the current corresponds to the result of the product-sum operation of the first data and the second data.

[0598] In step ST16, the circuit XCS1 supplies the ground potential GND to each of the wirings XCL1[1] to XCL1[s]. Thus, in the j-th column of the arithmetic cell array MAC1, the current does not flow from the wiring WCL1[j] to the cells IM1[1,j] to IM1[s,j] of the arithmetic cell array MAC1.

[0599] In addition, in step ST16, the switch S1a[j] of the switching circuit SWC1 is in the closed state, the switch S1b[j] of the switching circuit SWC1 is in the open state, and the switch S1c[j] of the switching circuit SWC1 is in the closed state. Fig.11A The switch SB[j] (not shown) of the circuit SWCB is in the open state.

[0600] Therefore, I SUM (N+1) (h) The current [j] flows from the circuit ITRZa[j] (not shown) of the circuit ITRZ of FIG. 11 through the wiring WCL2[j] to the cell IM2[1,j] to the cell IM2[s,j] in the j-th column of the arithmetic unit array MACA2.

[0601] Here, consider the case where the second data of the hth group is input to the operation unit array MACA2. At this time, the amount of current flowing from the circuit ITRZa[j] to the wiring WCL2[j] extending in the operation unit array MACA2 is I SUM (N+1) (h) [j]. Note that I SUM (N) (h) [j] N represents the number of product-sum operations, I SUM (N+1) (h) [j] means the current amount corresponding to the result of the N+1th product-sum operation. SUM (N) (h) The h of [i] means using the second data included in the h-th group.

[0602] At this time, the amount of current flowing from each of the circuits ITRZa[1] to ITRZa[s] to the wirings WCL1[1] to WCL1[s] extending in the arithmetic unit array MACA1 can be expressed using (Formula 1.17) in the same manner as Formula (1.25).

[0603] In addition, in each of the circuits ITRZa[1] to ITRZa[s], the switching circuit RL3[j] is switched to correspond to the amount of current I input to the terminal R3Ti[j]. SUM (N+1) (h) The value of [j] is used as an input value to perform a function operation, and a current is output to the terminal R3To[j], and the current amount corresponds to the result of the operation. As in step ST14, by defining the function as F(x), the conversion circuit RL3 converts the current amount I SUM (N+1) (h) The current of [j] is input to terminal R3Ti[j], and the current T is output from terminal R3To[j]. (h) [j]×I Tut Thus, similarly to step ST14, T is obtained as the calculation result of the circuit CVD. (h) [j].

[0604] In addition, the calculation results output from each terminal R3To[1] to terminal R3To[s] of the conversion circuit RL3[1] to the conversion circuit RL3[s] are the same matrix T as the formula (1.30).

[0605] Note that the hth row of T (h) [1] to T (h) [s] is a result obtained by using the second data of the hth group through the Nth product-sum operation and output from each terminal R3To[1] to the terminal R3To[s] of the conversion circuit RL3[1] to the conversion circuit RL3[s].

[0606] In the operation unit array MACA2, by sequentially inputting the second data of the first to s-th groups, the operation results of the first to s-th groups are sequentially output from the terminals R3To[1] to R3To[s] of the conversion circuit RL3.

[0607] Then, in the circuit ITRZa[j], the operation results T are sequentially outputted from the terminal R3To[j] of the conversion circuit RL3. (1) [j] to T (s) [j] is converted into a digital signal by the analog-to-digital conversion circuit ADC and output from the wiring OL[j].

[0608] Note that the operations performed in steps ST15 and ST16 are equivalent to Fig. 15B The block diagram shows the operation.

[0609] By using the operation circuit CDV described in this embodiment, one or more product-sum operations can be performed. In addition, when multiple product-sum operations are performed in the existing operation circuit, digital-to-analog conversion and analog-to-digital conversion are performed before and after each operation. However, by using the operation circuit CDV, as digital-to-analog conversion, it is only necessary to generate analog current once by the circuit WCS, and as analog-to-digital conversion, it is only necessary to process once by the analog-to-digital conversion circuit ADC in the circuit ITRZ. That is, by using the operation circuit CDV, the number of digital-to-analog conversion circuits and analog-to-digital conversion circuits can be reduced, so the circuit area of ​​the operation circuit CDV is smaller than that of the existing structure. In addition, compared with the existing structure, the power consumption of the operation circuit CDV can be reduced.

[0610] Note that in this embodiment, an example is described in which the plurality of arithmetic units in the arithmetic unit array MACA1, the plurality of arithmetic units in the arithmetic unit array MACA2, the plurality of storage units in the storage unit array MEMA1, and the plurality of storage units in the storage unit array MEMA2 are all configured in a matrix shape of s rows and s columns, but the number of rows of the arithmetic unit array MACA1, the arithmetic unit array MACA2, the storage unit array MEMA1, and the storage unit array MEMA2 may be completely or partially different. In addition, the number of columns of the arithmetic unit array MACA1, the arithmetic unit array MACA2, the storage unit array MEMA1, and the storage unit array MEMA2 may be completely or partially different.

[0611] The semiconductor device of one embodiment of the present invention is not limited to the structure of the arithmetic circuit CDV described in this embodiment. The semiconductor device of one embodiment of the present invention may have a structure in which the arithmetic circuit CDV described in this embodiment is appropriately modified.

[0612] For example, you can also Figure 1 The operation circuit CDV shown is changed to Fig.17 Arithmetic circuit CDVA is a modified example of the structure of the arithmetic circuit CDV, and is different from the arithmetic circuit CDV in that the current generating circuit CM1 is replaced with the current generating circuit RL1 and the current generating circuit CM2 is replaced with the current generating circuit RL2.

[0613] Specifically, terminal T1b[k] is electrically connected to terminal R1Ti[k], terminal R1To[k] is electrically connected to terminal T2a[k], terminal T1d[k] is electrically connected to terminal R2Ti[k], terminal R2To[k] is electrically connected to terminal T3a[k], terminal T2c[k] is electrically connected to terminal C1Ti[k], terminal C1To[k] is electrically connected to terminal T6c[k], terminal T3c[k] is electrically connected to terminal C2Ti[k], and terminal C2To[k] is electrically connected to terminal T5c[k].

[0614] Therefore, in the operation circuit CDVA, the potential corresponding to the amount of current output from the current generating circuit RL1 is written to the memory cell array MEMA1, and the potential corresponding to the amount of current output from the current generating circuit RL2 is written to the memory cell array MEMA2. Even with this structure, the operation circuit CDVA can perform multiple product-sum operations and function-type operations in the same way as the operation circuit CDV.

[0615] Note that this embodiment mode can be appropriately combined with other embodiment modes described in this specification. For example, the configuration, structure, method, and the like described in this embodiment mode can be appropriately combined with the configuration, structure, method, and the like described in other embodiment modes.

[0616] (Implementation Method 2) In this embodiment, a configuration example of the arithmetic circuit CDV described in the above embodiment will be described.

[0617] Fig.18A It is a perspective view schematically showing a calculation circuit CDV of a semiconductor device according to one embodiment of the present invention. Fig.18A The computing circuit CDV shown in the figure includes, for example, a circuit layer PHRL, a storage layer OMEL, and a computing layer OMAL. In addition, the circuit layer PHRL is located below the storage layer OMEL, and the computing layer OMAL is located above the storage layer OMEL. In other words, Fig.18A The operation circuit CDV has a structure in which a circuit layer PHRL, a memory layer OMEL, and an operation layer OMAL are stacked in this order from the bottom.

[0618] Fig.18B It is shown Fig.18A The block diagram shows structural examples of the circuit layer PHRL, the storage layer OMEL and the operation layer OMAL.

[0619] exist Fig.18B In the example, the circuit layer PHRL includes Figure 1 The circuit WCS, the circuit XCS1, the circuit XCS2, the circuit ITRZ, the current generating circuit CM1, the current generating circuit CM2, the current generating circuit RL1 and the current generating circuit RL2 are shown. In addition, the storage layer OMEL includes, for example Figure 1 The memory cell arrays MEMA1 and MEMA2, the circuits WWD1 and WWD2, the circuits RWD1 and RWD2 are shown. In addition, the operation layer OMAL includes, for example, the operation cell arrays MACA1 and MACA2, the circuits WSD1 and WSD2.

[0620] Notice, Figure 1 The switching circuit SWC1 , the switching circuit SWC2 , the switching circuit SWC3 , the switching circuit SWC5 , and the switching circuit SWC6 shown may be included in the circuit layer PHRL, may be included in the storage layer OMEL, and may also be included in the operation layer OMAL.

[0621] The circuit layer PHRL can be formed, for example, by providing circuit elements such as transistors and capacitors on a substrate. In addition, as the substrate, a semiconductor substrate (for example, a single crystal substrate made of silicon or germanium) can be used. As a substrate other than a semiconductor substrate, for example, an SOI (Silicon On Insulator) substrate, a glass substrate, a quartz substrate, a plastic substrate, a sapphire glass substrate, a metal substrate, a stainless steel substrate, a substrate containing a stainless steel foil, a tungsten substrate, a substrate containing a tungsten foil, a flexible substrate, a laminated film, a paper containing a fibrous material, or a substrate film can be used. As an example of a glass substrate, barium borosilicate glass, aluminoborosilicate glass, or soda-lime glass can be cited. As an example of a flexible substrate, a laminated film, or a substrate film, plastics represented by polyethylene terephthalate (PET), polyethylene naphthalate (PEN), polyether sulfone (PES), and polytetrafluoroethylene (PTFE) can be cited. Alternatively, as another example, a synthetic resin such as an acrylic resin can be cited. Alternatively, as another example, polypropylene, polyester, polyvinyl fluoride, or polyvinyl chloride can be cited. Other examples include polyamide, polyimide, aramid, epoxy resin, inorganic vapor-deposited film, or paper. When the manufacturing process of the calculation circuit CDV includes heat treatment, it is preferable to select a material with high heat resistance as the substrate.

[0622] Note that in this embodiment mode, description is given on the assumption that a substrate included in the circuit layer PHRL is a semiconductor substrate including silicon.

[0623] By using a semiconductor substrate made of silicon as a substrate included in the circuit layer PHRL, for example, a transistor included in each of the circuits WCS, XCS1, XCS2, ITRZ, current generating circuit CM1, CM2, RL1, and RL2 can be formed in the semiconductor substrate. In this case, the transistor is a Si transistor. Since the Si transistor has a high field effect mobility, a large on-state current can flow. As a result, the driving speed of each of the above circuits can be increased, and the width of the signal range can be expanded.

[0624] In addition, by directly forming the storage layer OMEL on the upper part of the circuit layer PHRL, a stacked structure of the circuit layer PHRL and the storage layer OMEL can be manufactured. Alternatively, the storage layer OMEL can also have a structure in which circuit elements such as transistors and capacitors are arranged on a substrate and the substrate is installed on the upper part of the circuit layer PHRL to manufacture the stacked structure.

[0625] When the storage layer OMEL is directly formed on the upper part of the circuit layer PHRL, the storage layer OMEL preferably includes an OS transistor. Since the OS transistor can be formed not only on a semiconductor substrate but also on an insulator substrate, a conductor substrate, a conductive film, an insulating film, or a semiconductor film, it is easy to be provided on a semiconductor substrate (on the circuit layer PHRL) on which a Si transistor is formed.

[0626] In addition, when a circuit element such as a transistor or a capacitor is formed on a substrate as a storage layer OMEL and the substrate is mounted on the circuit layer PHRL, a flip-chip bonding method or a wire bonding method can be used. Alternatively, a first bonding layer can be provided on one side of the circuit layer PHRL, a second bonding layer can be provided on the substrate of the storage layer OMEL, and the first bonding layer and the second bonding layer can be bonded together using one or both of a surface activated bonding method and a hydrophilic bonding method, thereby mounting the storage layer OMEL on the circuit layer PHRL. In particular, copper (Cu) is used for the conductors included in each of the first bonding layer and the second bonding layer, and bonding copper (Cu) to each other is called Cu-Cu bonding.

[0627] <Cross-sectional structure example 1> Next, explain Fig.18A and Fig.18B A specific structural example of the operation circuit CDV is shown. Fig.19 yes Fig.18A and Fig.18B The schematic cross-sectional view of an example of the arithmetic circuit CDV shown.

[0628] Fig.19 FIG. 4 is a schematic cross-sectional view of the circuit layer PHRL, the storage layer OMEL and the operation layer OMAL. Fig.19 In the computing circuit CDV, the storage layer OMEL is directly formed on the circuit layer PHRL, and the computing layer OMAL is directly formed on the storage layer OMEL.

[0629] Fig.19 The transistor 400 included in the circuit layer PHRL is shown. The transistor 400 is provided on the substrate 311 and includes: a conductor 316 used as a gate; an insulator 315 and an insulator 317 used as a gate insulator; a semiconductor region 313 including a portion of the substrate 311; and a low resistance region 314a and a low resistance region 314b used as a source region or a drain region. The transistor 400 can be a p-channel transistor or an n-channel transistor. The substrate 311 can use a single crystal silicon substrate, for example.

[0630] Here, in Fig.19 In the transistor 400 shown, the semiconductor region 313 (a part of the substrate 311) forming the channel has a convex shape. In addition, a conductor 316 is provided in a manner that covers the side and top surfaces of the semiconductor region 313 via an insulator 315. In addition, the conductor 316 can also use a material that adjusts the work function. Because the convex portion of the semiconductor substrate is utilized, this transistor 400 is also referred to as a FIN-type transistor. In addition, an insulator for forming a mask for the convex portion can also be included in a manner that contacts the top of the convex portion. In addition, although a portion of the semiconductor substrate is processed to form a convex portion is shown here, the SOI substrate can also be processed to form a semiconductor film having a convex shape.

[0631] Notice, Fig.19 The structure of the transistor 400 shown is only an example and is not limited to the above structure. An appropriate transistor may be used according to the circuit structure or driving method.

[0632] A wiring layer including an interlayer film, wiring and a plug may also be provided between each structure. In addition, the wiring layer may be provided in multiple layers according to the design. In addition, in this specification, wiring and a plug electrically connected to the wiring may also be a component. That is, a part of the conductor is sometimes used as wiring, and a part of the conductor is sometimes used as a plug.

[0633] For example, an insulator 320, an insulator 324, and an insulator 326 are sequentially stacked as interlayer films on the transistor 400. In addition, a conductor 328 and the like are embedded in the insulator 320. In addition, a conductor 330 and the like are embedded in the insulator 324 and the insulator 326. In addition, the conductor 328 and the conductor 330 are used as contact plugs or wirings.

[0634] In addition, the insulator used as an interlayer film can also be used as a planarization film to cover the concavo-convex shape thereunder. For example, in order to improve the flatness of the top surface of the insulator 320, the top surface can also be planarized by a planarization process using a chemical mechanical polishing (CMP) method.

[0635] A wiring layer may also be provided on the insulator 326 and the conductor 330. Fig.19 In the embodiment, an insulator 350, an insulator 357, an insulator 352, and an insulator 354 are sequentially stacked on the insulator 326 and the conductor 330. A conductor 356 is formed between the insulator 350, the insulator 357, and the insulator 352. The conductor 356 is used as a contact plug or wiring.

[0636] An insulator 354 is provided on the insulator 352 and the conductor 356. The insulator 354 may also have embedded therein contact plugs or wirings for electrically connecting to an upper circuit (eg, a circuit in the memory layer OMEL or a circuit in the computing layer OMAL).

[0637] in addition, Fig.19 Schematic diagram showing the cell MC included in the storage layer OMEL. Specifically, Fig.19 The transistor F7, the transistor F8, the transistor F9 and the capacitor C6 included in the cell MC are shown. Note that the cell MC may be any of the cells MC1[1,1] to MC1[s,s] and MC2[1,1] to MC2[s,s] described in the above embodiment.

[0638] exist Fig.19 In the storage layer OMEL of the operation circuit CDV, the transistor F7 and the capacitor C6 are located above the transistor F8 and the transistor F9.

[0639] exist Fig.19 In the memory layer OMEL, transistors F8 and F9 are provided in a manner of using one island-shaped semiconductor layer in common. Specifically, a gate insulating film and a gate electrode of transistor F8 are formed in one of two regions of one island-shaped semiconductor layer, and a gate insulating film and a gate electrode of transistor F9 are formed in the other of two regions of one island-shaped semiconductor layer.

[0640] In addition, Fig.19 In the storage layer OMEL, both transistors F8 and transistor F9 use transistors including back gates. In particular, the back gate of transistor F8 is located in a region overlapping with the gate insulating film and gate electrode of transistor F8 below the above-mentioned one island-shaped semiconductor layer, and the back gate of transistor F9 is located in a region overlapping with the gate insulating film and gate electrode of transistor F9 below the above-mentioned one island-shaped semiconductor layer.

[0641] In addition, one of the source electrode and the drain electrode of the transistor F8 is electrically connected to a conductor corresponding to the wiring VE2. In addition, one of the source electrode and the drain electrode of the transistor F9 is electrically connected to a conductor corresponding to the wiring BL. The wiring BL may be any one of the wirings BL1[1] to BL1[s] and the wirings BL2[1] to BL2[s] described in the above embodiment. In addition, as an example, the wiring VE2 and the wiring BL extend in the channel width direction of the transistor F8 or the transistor F9.

[0642] The conductor serving as the gate electrode of the transistor F9 is extended in the channel width direction. The conductor corresponds to the wiring RWL. The wiring RWL may be any of the wiring RWL1[1] to the wiring RWL1[s] and the wiring RWL2[1] to the wiring RWL2[s] described in the above embodiment.

[0643] An insulator serving as an interlayer film is formed between the transistors F8 and F9 and the transistor F7. In addition, the insulator has openings in the region overlapping with the gate electrode of the transistor F8 and the region overlapping with the wiring BL, and the openings are filled with conductors. One conductor is electrically connected to one of the source electrode and the drain electrode of the transistor F7, and the other conductor is electrically connected to the other of the source electrode and the drain electrode of the transistor F7.

[0644] As described above, transistor F7 is located above transistor F8 and transistor F9. In addition, a dielectric of capacitor C6 is formed to cover the end of the island-shaped semiconductor layer of transistor F7, and a conductor corresponding to the second terminal of capacitor C6 is formed on the dielectric. The conductor corresponds to wiring VE3.

[0645] In addition, the potentials supplied to the wiring VE2 and the wiring VE3 may be equal to each other as described in Embodiment 1. When the potentials supplied to the wiring VE2 and the wiring VE3 are equal to each other, the wiring VE2 and the wiring VE3 may be electrically connected to each other (not shown).

[0646] The gate insulating film and gate electrode of the transistor F7 are formed in the region of the island-shaped semiconductor layer of the transistor F7. In particular, the conductor of the gate electrode of the transistor F7 is provided so as to extend in the channel width direction. In addition, the conductor corresponds to the wiring WWL. The wiring WWL may be any one of the wiring WWL1[1] to the wiring WWL1[s] and the wiring WWL2[1] to the wiring WWL2[s] described in the above embodiment.

[0647] In addition, the transistor F7 is a transistor including a back gate similarly to the transistor F8 and the transistor F9. In particular, the back gate of the transistor F7 is located in a region overlapping with the gate insulating film and the gate electrode of the transistor F7 under the island-shaped semiconductor layer.

[0648] As described above, in transistor F7, transistor F8 and transistor F9, the gate and the back gate are configured in a manner that clamps the channel formation region of the semiconductor. The gate and the back gate are formed by a conductor. The back gate can have the same function as the gate. In addition, by changing the potential of the back gate, the threshold voltage of the transistor can be changed. The potential of the back gate can be equal to the potential of the gate, or it can be a ground potential or an arbitrary potential.

[0649] In addition, since the gate and back gate are formed of a conductor, they also have the function of preventing the electric field generated outside the transistor from affecting the semiconductor forming the channel (especially the electrostatic shielding function). That is, it is possible to prevent the electrical characteristics of the transistor from changing due to the influence of external electric fields such as static electricity. In addition, by providing a back gate, the change in the threshold voltage of the transistor before and after the bias thermal stress test (sometimes referred to as the BT test) can be reduced.

[0650] For example, by using a transistor including a back gate as the transistor F7, the influence of the external electric field can be reduced and the closed state can be stably maintained. Therefore, the data written to the first terminal of the capacitor C6 can be stably maintained. By providing a back gate, the operation of the cell MC is stable, and the reliability of the storage layer OMEL including the cell MC can be improved.

[0651] As the semiconductor layer forming the channels of the transistors F7, F8, and F9, one or more of a single crystal semiconductor, a polycrystalline semiconductor, a microcrystalline semiconductor, an amorphous semiconductor, etc. can be used. As the semiconductor material, for example, silicon, germanium, etc. can be used as described in Embodiment 1. In addition, compound semiconductors such as silicon germanium, silicon carbide, gallium arsenide, oxide semiconductors, and nitride semiconductors can also be used.

[0652] In addition, transistors F7, F8, and F9 are preferably transistors (OS transistors) using an oxide semiconductor that is one of metal oxides in a semiconductor layer that forms a channel. Since the band gap of the oxide semiconductor is greater than 2 eV, the off-state current is significantly small. Therefore, the power consumption of the unit MC can be reduced. Therefore, the power consumption of the operation circuit CDV including the unit MC can be reduced.

[0653] Furthermore, a memory cell including an OS transistor may be referred to as an “OS memory.” Furthermore, an operation circuit CDV including the memory cell may also be referred to as an “OS memory.”

[0654] In addition, the OS transistor operates stably even in a high temperature environment, and the characteristics vary little. For example, even in a high temperature environment, its off-state current hardly increases. Specifically, even in an ambient temperature above room temperature and below 200°C, its off-state current hardly increases. In addition, even in a high temperature environment, the on-state current is not easy to decrease. Therefore, the OS memory operates stably even in a high temperature environment and has high reliability.

[0655] in addition, Fig.19 The unit IM included in the operation layer OMAL is shown. Specifically, Fig.19 The transistor F1, transistor F2, transistor F5 and capacitor C5 included in the unit IM are shown. Note that the unit IM may be any of the units IM1[1,1] to IM1[s,s] and IM2[1,1] to IM2[s,s] described in the above embodiments.

[0656] In addition, if Fig.19 As shown, the structure of the unit IM in the operation layer OMAL can be the same as the structure of the unit MC in the storage layer OMEL. Therefore, the structure of the unit IM in the operation layer OMAL refers to the description of the unit MC in the storage layer OMEL. Note that in the description of the unit MC in the storage layer OMEL, the transistor F7 is replaced by the transistor F1, the transistor F8 is replaced by the transistor F2, the transistor F9 is replaced by the transistor F5, the capacitor C6 is replaced by the capacitor C5, the wiring BL is replaced by the wiring WCL, the wiring VE2 is replaced by the wiring VE0, the wiring RWL is replaced by the wiring VE1, the wiring VE3 is replaced by the wiring XCL, and the wiring WWL is replaced by the wiring WSL to describe the unit IM in the operation layer OMAL.

[0657] The wiring WCL may be any one of the wiring WCL1[1] to the wiring WCL1[s] and the wiring WCL2[1] to the wiring WCL2[s] described in the above embodiment. In addition, the wiring WSL may be any one of the wiring WSL1[1] to the wiring WSL1[s] and the wiring WSL2[1] to the wiring WSL2[s] described in the above embodiment. In addition, the wiring XCL may be any one of the wiring XCL1[1] to the wiring XCL1[s] and the wiring XCL2[1] to the wiring XCL2[s] described in the above embodiment.

[0658] <Cross-sectional structure example 2> Fig. 20 is with Fig.19 Different Fig.18A and Fig.18B The schematic cross-sectional view of an example of the arithmetic circuit CDV shown.

[0659] Fig. 20 The storage layer OMEL and the operation layer OMAL in the computing circuit CDV of the embodiment of the present invention both include a substrate. Fig.19 The operation circuit CDV is different.

[0660] Fig. 20 The storage layer OMEL of the operation circuit CDV includes a substrate BS1. In addition, transistors F7, F8, F9 and capacitor C6 are formed on the substrate BS1. Note that the structure of the cell MC formed on the substrate BS1 is the same as that of the cell MC formed on the substrate BS1. Fig.19 The unit MC of the operation circuit CDV is the same, but it can also be changed according to the situation Fig. 20 The structure of the unit MC.

[0661] in addition, Fig. 20 The operation layer OMAL of the operation circuit CDV includes a substrate BS2. In addition, a transistor F1, a transistor F2, a transistor F5 and a capacitor C5 are formed on the substrate BS2. Note that the structure of the unit IM formed on the substrate BS2 is similar to that of the unit IM formed on the substrate BS2. Fig.19 The unit IM of the operation circuit CDV is the same, but it can also be changed according to the situation Fig. 20 The structure of the unit IM.

[0662] That is to say, Fig. 20 The arithmetic circuit CDV has a structure in which a substrate BS1 in a memory layer OMEL and a substrate BS2 in an arithmetic layer OMAL are mounted on a substrate 311 on which a circuit XCS1 , a circuit WCS and a circuit ITRZ are formed.

[0663] Note that the substrate BS1 and the substrate BS2 may be a substrate that can be used in the circuit layer PHRL (e.g., the substrate 311). For example, by using a semiconductor substrate made of silicon as both the substrate BS1 and the substrate BS2, the transistors F1, F2, F5, and F7 to F9 may all be Si transistors.

[0664] In addition, as a method of mounting the substrate BS1 on the substrate 311 and a method of mounting the substrate BS2 on the substrate BS1, as described above, a flip chip bonding method or a wire bonding method may be used. Alternatively, a bonding layer may be provided between the bonded substrates to use one or both of a surface activated bonding method and a hydrophilic bonding method.

[0665] Note that the semiconductor device of one embodiment of the present invention is not limited to Fig.18A , Fig.18B , Fig.19 and Fig. 20 The semiconductor device of one embodiment of the present invention may also have a structure appropriately changed. Fig.18A , Fig.18B , Fig.19 and Fig. 20 The structure formed.

[0666] Fig.21 Show Fig.18A An example of modification of the arithmetic circuit CDV is shown. Fig.21 The operation circuit CDV shown includes a storage layer OMEL1 and a storage layer OMEL2 instead of the storage layer OMEL, and includes an operation layer OMAL1 and an operation layer OMAL2 instead of the operation layer OMAL. Fig.18A The operation circuit CDV is different. In other words, Fig.21 The computing circuit CDV includes a circuit layer PHRL, a storage layer OMEL1, a storage layer OMEL2, a computing layer OMAL1 and a computing layer OMAL2.

[0667] Fig. 22 It is shown Fig.21 The block diagram shows structural examples of the circuit layer PHRL, the storage layer OMEL1, the storage layer OMEL2, the operation layer OMAL1 and the operation layer OMAL2.

[0668] exist Fig. 22 In, with Fig.18B The circuit layer PHRL shown in the figure is similar, and the circuit layer PHRL includes, for example Figure 1 The circuit WCS, the circuit XCS1, the circuit XCS2, the circuit ITRZ, the current generating circuit CM1, the current generating circuit CM2, the current generating circuit RL1 and the current generating circuit RL2 are shown. In addition, the storage layer OMEL1 includes, for example Figure 1 The memory cell array MEMA1, the circuit WWD1 and the circuit RWD1 are shown. In addition, the memory layer OMEL2 includes, for example Figure 1 The memory cell array MEMA2, the circuit WWD2 and the circuit RWD2 are shown. In addition, the operation layer OMAL1 includes, for example, the operation cell array MACA1 and the circuit WSD1. In addition, the operation layer OMAL2 includes, for example, the operation cell array MACA2 and the circuit WSD2.

[0669] As described above, two or four storage layers OMEL and operation layers OMAL may be arranged above the circuit layer PHRL in the operation circuit CDV. Note that the number of storage layers OMEL and operation layers OMAL arranged above the circuit layer PHRL may be 3 or 5 or more in total.

[0670] This embodiment mode can be appropriately combined with other embodiment modes described in this specification. For example, the configuration, structure, method, etc. described in this embodiment mode can be appropriately combined with the configuration, structure, method, etc. described in other embodiment modes.

[0671] (Implementation method 3) In this embodiment, a transistor including an oxide semiconductor in a channel formation region (OS transistor) is described. In the description of the OS transistor, a comparison with a transistor including silicon in a channel formation region (also referred to as a Si transistor) is briefly described.

[0672] [OS Transistor] It is preferable to use an oxide semiconductor with a low carrier concentration for the OS transistor. For example, the carrier concentration of the channel formation region of the oxide semiconductor is 1×10 18 cm -3 Below, preferably below 1×10 17 cm -3 , more preferably less than 1×10 16 cm -3 , and more preferably less than 1×10 13 cm -3 , and further preferably less than 1×10 10 cm -3 , and is 1×10 -9 cm -3 In the case of reducing the carrier concentration of the oxide semiconductor film, the impurity concentration in the oxide semiconductor film can be reduced to reduce the defect state density. In this specification, etc., a state in which the impurity concentration is low and the defect state density is low is referred to as high-purity intrinsic or substantially high-purity intrinsic. In addition, an oxide semiconductor with a low carrier concentration is sometimes referred to as a high-purity intrinsic or substantially high-purity intrinsic oxide semiconductor.

[0673] Since high-purity intrinsic or substantially high-purity intrinsic oxide semiconductors have a low defect state density, they sometimes have a low trap state density. In addition, it takes a long time for the charges captured by the trap states of the oxide semiconductor to disappear, and sometimes they act like fixed charges. Therefore, sometimes the electrical characteristics of a transistor in which a channel formation region is formed in an oxide semiconductor with a high trap state density are unstable.

[0674] Therefore, in order to stabilize the electrical characteristics of the transistor, it is effective to reduce the impurity concentration in the oxide semiconductor. In order to reduce the impurity concentration in the oxide semiconductor, it is preferable to also reduce the impurity concentration in the nearby film. As impurities, hydrogen, nitrogen, etc. can be cited. Note that impurities in oxide semiconductors refer to, for example, elements other than the main components that constitute the oxide semiconductor. For example, an element with a concentration of less than 0.1 atomic % can be said to be an impurity.

[0675] In an OS transistor, when impurities and oxygen vacancies exist in the channel formation region of an oxide semiconductor, the electrical characteristics are easily changed and reliability may be reduced. In addition, in an OS transistor, hydrogen enters the oxygen vacancies in the oxide semiconductor to form defects (hereinafter sometimes referred to as V O H), electrons that become carriers may be generated. In addition, when V O When the voltage is 0.1V, the donor concentration in the channel formation region increases. As the donor concentration in the channel formation region increases, the threshold voltage is sometimes non-uniform. Therefore, when oxygen vacancies are included in the channel formation region of the oxide semiconductor, the transistor is in a normally-on state (a state in which a channel exists and current flows in the transistor even when no voltage is applied to the gate electrode). Therefore, in the channel formation region of the oxide semiconductor, it is preferable to minimize impurities, oxygen vacancies, and V O H.

[0676] In addition, the band gap of the oxide semiconductor is preferably larger than the band gap of silicon (typically 1.1 eV), preferably 2 eV or more, more preferably 2.5 eV or more, and more preferably 3.0 eV or more. By using an oxide semiconductor with a larger band gap than silicon, the off-state current (also called Ioff) of the transistor can be reduced.

[0677] For example, in Si transistors, as the miniaturization of transistors progresses, short channel effects (Short Channel Effect: also called SCE) appear. Therefore, miniaturization of Si transistors is difficult. One of the reasons for the short channel effect is that the band gap of silicon is small. On the other hand, in OS transistors, oxide semiconductors, which are semiconductor materials with large band gaps, are used, so the short channel effect can be suppressed. In other words, OS transistors are transistors with no short channel effect or very little short channel effect.

[0678] The short channel effect refers to the degradation of electrical characteristics that occurs with the miniaturization of transistors (reduction of channel length). Specific examples of the short channel effect include a redu...

Claims

1. A semiconductor device, include: Unit 1; Unit 2; Unit 3; Unit 4; First Circuit; Second Circuit; a first current generating circuit; a second current generating circuit; a third current generating circuit; as well as a fourth current generating circuit, wherein the first unit is electrically connected to the first input terminal of the first current generating circuit through a first wiring, The first unit is electrically connected to the first circuit and the fourth output terminal of the fourth current generating circuit through a second wiring. The second unit is electrically connected to the second input terminal of the second current generating circuit via a third wiring. The second unit is electrically connected to the second circuit and the third output terminal of the third current generating circuit through a fourth wiring. The third unit is electrically connected to the first output terminal of the first current generating circuit and the third input terminal of the third current generating circuit. The fourth unit is electrically connected to the second output terminal of the second current generating circuit and the fourth input terminal of the fourth current generating circuit. The first circuit has a function of generating a first current and outputting it to the second wiring, The second circuit has a function of generating a second current and outputting it to the fourth wiring, The first current generating circuit has a function of outputting a current from the first output terminal as a current mirror circuit, the amount of the current corresponding to the amount of the current flowing through the first input terminal, The second current generating circuit has a function of outputting a current from the second output terminal as a current mirror circuit, the amount of the current corresponding to the amount of the current flowing through the second input terminal, The third current generating circuit has a function of outputting a third current from the third output terminal as a function-type operation circuit, the third current amount corresponding to the current amount flowing through the third input terminal, The fourth current generating circuit has a function of outputting a fourth current from the fourth output terminal as a function-type operation circuit, the fourth current amount corresponding to the current amount flowing through the fourth input terminal, The first unit has a function of holding a potential corresponding to first data and a function of generating and outputting a fifth current to the first wiring, the fifth current amount being equivalent to the product of a value of the first data and a value corresponding to the first current or the fourth current flowing through the second wiring, The second unit has a function of holding a potential corresponding to second data and a function of generating and outputting a sixth current to the third wiring, the sixth current amount being equivalent to the product of a value of the second data and a value corresponding to the second current or the third current flowing through the fourth wiring, The third unit has a function of maintaining a potential corresponding to the fifth current, Furthermore, the fourth unit has a function of maintaining a potential corresponding to the sixth current.

2. The semiconductor device according to claim 1, further comprising: include: a first switching circuit; a second switching circuit; a third switching circuit; a fourth switching circuit; as well as a fifth switching circuit, The first switching circuit includes a first terminal, a second terminal, a third terminal and a fourth terminal, The second switching circuit includes a fifth terminal, a sixth terminal and a seventh terminal, The third switching circuit includes an eighth terminal, a ninth terminal and a tenth terminal, The fourth switching circuit includes an eleventh terminal, a twelfth terminal and a thirteenth terminal, The fifth switching circuit includes a fourteenth terminal, a fifteenth terminal and a sixteenth terminal, The first terminal of the first switching circuit is electrically connected to the first wiring. The second terminal of the first switching circuit is electrically connected to the first input terminal of the first current generating circuit, The third terminal of the first switching circuit is electrically connected to the third wiring. The fourth terminal of the first switching circuit is electrically connected to the second input terminal of the second current generating circuit, The fifth terminal of the second switching circuit is electrically connected to the first output terminal of the first current generating circuit, The sixth terminal of the second switching circuit is electrically connected to the third unit, The seventh terminal of the second switching circuit is electrically connected to the third input terminal of the third current generating circuit, The eighth terminal of the third switching circuit is electrically connected to the second output terminal of the second current generating circuit, The ninth terminal of the third switching circuit is electrically connected to the fourth unit, The tenth terminal of the third switching circuit is electrically connected to the fourth input terminal of the fourth current generating circuit, The eleventh terminal of the fourth switching circuit is electrically connected to the first circuit, The twelfth terminal of the fourth switching circuit is electrically connected to the second wiring. The thirteenth terminal of the fourth switching circuit is electrically connected to the fourth output terminal of the fourth current generating circuit, The fourteenth terminal of the fifth switching circuit is electrically connected to the second circuit, The fifteenth terminal of the fifth switching circuit is electrically connected to the fourth wiring, The sixteenth terminal of the fifth switching circuit is electrically connected to the third output terminal of the third current generating circuit, The first switching circuit has a function of making the first terminal and the second terminal conductive or non-conductive, a function of making the first terminal and the third terminal conductive or non-conductive, and a function of making the third terminal and the fourth terminal conductive or non-conductive, The second switching circuit has a function of making the fifth terminal and the sixth terminal conductive or non-conductive, and a function of making the sixth terminal and the seventh terminal conductive or non-conductive, The third switching circuit has a function of making the eighth terminal and the ninth terminal conductive or non-conductive, and a function of making the ninth terminal and the tenth terminal conductive or non-conductive. The fourth switching circuit has a function of making the eleventh terminal and the twelfth terminal conductive or non-conductive, and a function of making the twelfth terminal and the thirteenth terminal conductive or non-conductive, And the fifth switching circuit has a function of making the fourteenth terminal and the fifteenth terminal in a conductive state or a non-conductive state, and a function of making the fifteenth terminal and the sixteenth terminal in a conductive state or a non-conductive state.

3. The semiconductor device according to claim 2, further comprising: include: Third Circuit, wherein the third circuit is electrically connected to the first wiring, And the third circuit has a function of causing a seventh current corresponding to the first data to flow through the first unit and a function of causing an eighth current corresponding to the second data to flow through the second unit.

4. The semiconductor device according to claim 3, further comprising: include: Unit 5; as well as Unit 6, The first unit, the second unit, the fifth unit and the sixth unit each include a first transistor, a second transistor and a first capacitor, The channel formation region of each of the first transistor and the second transistor includes a first oxide semiconductor. The first oxide semiconductor includes one or more selected from indium, zinc and element M, The element M is one or more selected from gallium, aluminum, silicon, boron, yttrium, tin, copper, vanadium, beryllium, titanium, iron, nickel, germanium, zirconium, molybdenum, lanthanum, cerium, neodymium, hafnium, tantalum, tungsten, cobalt, magnesium and antimony, In each of the first unit, the second unit, the fifth unit, and the sixth unit, One of the source and the drain of the first transistor is electrically connected to the gate of the second transistor, One of the source and the drain of the second transistor is electrically connected to one of the source and the drain of the first transistor, One of a pair of terminals of the first capacitor is electrically connected to a gate of the second transistor, In the first unit, The other of the source and the drain of the first transistor is electrically connected to the first wiring, The other of the pair of terminals of the first capacitor is electrically connected to the second wiring, In the second unit, The other of the source and the drain of the first transistor is electrically connected to the third wiring, The other of the pair of terminals of the first capacitor is electrically connected to the fourth wiring, In the fifth unit, The other of the source and the drain of the first transistor is electrically connected to the second wiring, The other of the pair of terminals of the first capacitor is electrically connected to the second wiring, And in the sixth unit, The other of the source and the drain of the first transistor is electrically connected to the fourth wiring, The other of the pair of terminals of the first capacitor is electrically connected to the fourth wiring.

5. The semiconductor device according to claim 4, wherein the third unit and the fourth unit each include a third transistor, a fourth transistor, a fifth transistor and a second capacitor, The channel formation region of each of the third transistor and the fourth transistor includes a second oxide semiconductor, The second oxide semiconductor includes one or more selected from indium, zinc and the element M, In each of the third unit and the fourth unit, One of the source and the drain of the third transistor is electrically connected to the gate of the fourth transistor, One of the source and the drain of the fourth transistor is electrically connected to one of the source and the drain of the fifth transistor, The other of the source and the drain of the fifth transistor is electrically connected to the other of the source and the drain of the third transistor, One of a pair of terminals of the first capacitor is electrically connected to a gate of the fourth transistor, In the third unit, The other of the source and the drain of the third transistor is electrically connected to the sixth terminal of the second switching circuit, And in the fourth unit, The other of the source and the drain of the third transistor is electrically connected to a ninth terminal of the third switching circuit.

6. The semiconductor device according to claim 5, include: First floor; a second layer located above the first layer; as well as The third layer is located above the second layer. The first layer includes the first circuit, the second circuit, the third circuit, the first current generating circuit, the second current generating circuit, the third current generating circuit and the fourth current generating circuit, The second layer includes the third unit and the fourth unit, And the third layer includes the first unit, the second unit, the fifth unit and the sixth unit.

7. The semiconductor device according to claim 5, include: First floor; a second layer located above the first layer; a third layer located above the second layer; a fourth layer located above the third layer; as well as a fifth layer located above the fourth layer, The first layer includes the first circuit, the second circuit, the third circuit, the first current generating circuit, the second current generating circuit, the third current generating circuit and the fourth current generating circuit, The second layer includes the third unit, The third layer includes the fourth unit, The fourth layer includes the first unit and the fifth unit, The fifth layer includes the second unit and the sixth unit.

8. An electronic device, include: The semiconductor device according to any one of claims 1 to 7; as well as Frame.