US20220058470A1 - Synapse array of a neuromorphic device including a synapse array having a plurality of ferroelectricity field effect transistors - Google Patents
Synapse array of a neuromorphic device including a synapse array having a plurality of ferroelectricity field effect transistors Download PDFInfo
- Publication number
- US20220058470A1 US20220058470A1 US17/519,163 US202117519163A US2022058470A1 US 20220058470 A1 US20220058470 A1 US 20220058470A1 US 202117519163 A US202117519163 A US 202117519163A US 2022058470 A1 US2022058470 A1 US 2022058470A1
- Authority
- US
- United States
- Prior art keywords
- synapse
- field effect
- effect transistor
- ferroelectric field
- present disclosure
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 210000000225 synapse Anatomy 0.000 title claims abstract description 149
- 230000005669 field effect Effects 0.000 title claims abstract description 41
- 230000005621 ferroelectricity Effects 0.000 title 1
- 210000002364 input neuron Anatomy 0.000 claims abstract description 47
- 210000004205 output neuron Anatomy 0.000 claims abstract description 33
- 238000010586 diagram Methods 0.000 description 20
- 239000010410 layer Substances 0.000 description 20
- 230000010287 polarization Effects 0.000 description 11
- 229910052751 metal Inorganic materials 0.000 description 10
- 239000002184 metal Substances 0.000 description 10
- 238000007667 floating Methods 0.000 description 9
- 239000011229 interlayer Substances 0.000 description 9
- 230000002401 inhibitory effect Effects 0.000 description 8
- 239000000758 substrate Substances 0.000 description 8
- 229910052771 Terbium Inorganic materials 0.000 description 7
- 230000002964 excitative effect Effects 0.000 description 7
- 229910052715 tantalum Inorganic materials 0.000 description 7
- 230000000946 synaptic effect Effects 0.000 description 6
- 230000004888 barrier function Effects 0.000 description 5
- 238000004891 communication Methods 0.000 description 5
- 238000002955 isolation Methods 0.000 description 5
- 238000003909 pattern recognition Methods 0.000 description 5
- VYPSYNLAJGMNEJ-UHFFFAOYSA-N Silicium dioxide Chemical compound O=[Si]=O VYPSYNLAJGMNEJ-UHFFFAOYSA-N 0.000 description 4
- 229910052814 silicon oxide Inorganic materials 0.000 description 4
- 239000004020 conductor Substances 0.000 description 3
- 238000000034 method Methods 0.000 description 3
- 238000003491 array Methods 0.000 description 2
- 239000010949 copper Substances 0.000 description 2
- 229910021419 crystalline silicon Inorganic materials 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 229910000449 hafnium oxide Inorganic materials 0.000 description 2
- 238000003384 imaging method Methods 0.000 description 2
- 239000011810 insulating material Substances 0.000 description 2
- 230000010354 integration Effects 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 210000002569 neuron Anatomy 0.000 description 2
- 229910021420 polycrystalline silicon Inorganic materials 0.000 description 2
- 230000001242 postsynaptic effect Effects 0.000 description 2
- 210000005215 presynaptic neuron Anatomy 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- RYGMFSIKBFXOCR-UHFFFAOYSA-N Copper Chemical compound [Cu] RYGMFSIKBFXOCR-UHFFFAOYSA-N 0.000 description 1
- OAICVXFJPJFONN-UHFFFAOYSA-N Phosphorus Chemical compound [P] OAICVXFJPJFONN-UHFFFAOYSA-N 0.000 description 1
- XUIMIQQOPSSXEZ-UHFFFAOYSA-N Silicon Chemical compound [Si] XUIMIQQOPSSXEZ-UHFFFAOYSA-N 0.000 description 1
- NRTOMJZYCJJWKI-UHFFFAOYSA-N Titanium nitride Chemical compound [Ti]#N NRTOMJZYCJJWKI-UHFFFAOYSA-N 0.000 description 1
- 229910052782 aluminium Inorganic materials 0.000 description 1
- XAGFODPZIPBFFR-UHFFFAOYSA-N aluminium Chemical compound [Al] XAGFODPZIPBFFR-UHFFFAOYSA-N 0.000 description 1
- 229910052785 arsenic Inorganic materials 0.000 description 1
- RQNWIZPPADIBDY-UHFFFAOYSA-N arsenic atom Chemical compound [As] RQNWIZPPADIBDY-UHFFFAOYSA-N 0.000 description 1
- 210000004556 brain Anatomy 0.000 description 1
- 229910052802 copper Inorganic materials 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- WIHZLLGSGQNAGK-UHFFFAOYSA-N hafnium(4+);oxygen(2-) Chemical compound [O-2].[O-2].[Hf+4] WIHZLLGSGQNAGK-UHFFFAOYSA-N 0.000 description 1
- 150000002500 ions Chemical class 0.000 description 1
- 239000007769 metal material Substances 0.000 description 1
- 150000002739 metals Chemical class 0.000 description 1
- 230000003278 mimic effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 229910052698 phosphorus Inorganic materials 0.000 description 1
- 239000011574 phosphorus Substances 0.000 description 1
- 230000003389 potentiating effect Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 229910021332 silicide Inorganic materials 0.000 description 1
- FVBUAEGBCNSCDD-UHFFFAOYSA-N silicide(4-) Chemical compound [Si-4] FVBUAEGBCNSCDD-UHFFFAOYSA-N 0.000 description 1
- 229910052710 silicon Inorganic materials 0.000 description 1
- 239000010703 silicon Substances 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- MZLGASXMSKOWSE-UHFFFAOYSA-N tantalum nitride Chemical compound [Ta]#N MZLGASXMSKOWSE-UHFFFAOYSA-N 0.000 description 1
- WFKWXMTUELFFGS-UHFFFAOYSA-N tungsten Chemical compound [W] WFKWXMTUELFFGS-UHFFFAOYSA-N 0.000 description 1
- 229910052721 tungsten Inorganic materials 0.000 description 1
- 239000010937 tungsten Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/06—Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons
- G06N3/063—Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons using electronic means
-
- H—ELECTRICITY
- H10—SEMICONDUCTOR DEVICES; ELECTRIC SOLID-STATE DEVICES NOT OTHERWISE PROVIDED FOR
- H10B—ELECTRONIC MEMORY DEVICES
- H10B51/00—Ferroelectric RAM [FeRAM] devices comprising ferroelectric memory transistors
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- H01L27/11521—
-
- H01L27/1156—
-
- H01L27/11587—
-
- H01L27/1159—
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
- H01L29/00—Semiconductor devices specially adapted for rectifying, amplifying, oscillating or switching and having potential barriers; Capacitors or resistors having potential barriers, e.g. a PN-junction depletion layer or carrier concentration layer; Details of semiconductor bodies or of electrodes thereof ; Multistep manufacturing processes therefor
- H01L29/40—Electrodes ; Multistep manufacturing processes therefor
- H01L29/401—Multistep manufacturing processes
- H01L29/4011—Multistep manufacturing processes for data storage electrodes
- H01L29/40111—Multistep manufacturing processes for data storage electrodes the electrodes comprising a layer which is used for its ferroelectric properties
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
- H01L29/00—Semiconductor devices specially adapted for rectifying, amplifying, oscillating or switching and having potential barriers; Capacitors or resistors having potential barriers, e.g. a PN-junction depletion layer or carrier concentration layer; Details of semiconductor bodies or of electrodes thereof ; Multistep manufacturing processes therefor
- H01L29/40—Electrodes ; Multistep manufacturing processes therefor
- H01L29/43—Electrodes ; Multistep manufacturing processes therefor characterised by the materials of which they are formed
- H01L29/49—Metal-insulator-semiconductor electrodes, e.g. gates of MOSFET
- H01L29/51—Insulating materials associated therewith
- H01L29/516—Insulating materials associated therewith with at least one ferroelectric layer
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
- H01L29/00—Semiconductor devices specially adapted for rectifying, amplifying, oscillating or switching and having potential barriers; Capacitors or resistors having potential barriers, e.g. a PN-junction depletion layer or carrier concentration layer; Details of semiconductor bodies or of electrodes thereof ; Multistep manufacturing processes therefor
- H01L29/66—Types of semiconductor device ; Multistep manufacturing processes therefor
- H01L29/66007—Multistep manufacturing processes
- H01L29/66075—Multistep manufacturing processes of devices having semiconductor bodies comprising group 14 or group 13/15 materials
- H01L29/66227—Multistep manufacturing processes of devices having semiconductor bodies comprising group 14 or group 13/15 materials the devices being controllable only by the electric current supplied or the electric potential applied, to an electrode which does not carry the current to be rectified, amplified or switched, e.g. three-terminal devices
- H01L29/66409—Unipolar field-effect transistors
- H01L29/66477—Unipolar field-effect transistors with an insulated gate, i.e. MISFET
- H01L29/6684—Unipolar field-effect transistors with an insulated gate, i.e. MISFET with a ferroelectric gate insulator
-
- H—ELECTRICITY
- H10—SEMICONDUCTOR DEVICES; ELECTRIC SOLID-STATE DEVICES NOT OTHERWISE PROVIDED FOR
- H10B—ELECTRONIC MEMORY DEVICES
- H10B41/00—Electrically erasable-and-programmable ROM [EEPROM] devices comprising floating gates
- H10B41/30—Electrically erasable-and-programmable ROM [EEPROM] devices comprising floating gates characterised by the memory core region
-
- H—ELECTRICITY
- H10—SEMICONDUCTOR DEVICES; ELECTRIC SOLID-STATE DEVICES NOT OTHERWISE PROVIDED FOR
- H10B—ELECTRONIC MEMORY DEVICES
- H10B41/00—Electrically erasable-and-programmable ROM [EEPROM] devices comprising floating gates
- H10B41/70—Electrically erasable-and-programmable ROM [EEPROM] devices comprising floating gates the floating gate being an electrode shared by two or more components
-
- H—ELECTRICITY
- H10—SEMICONDUCTOR DEVICES; ELECTRIC SOLID-STATE DEVICES NOT OTHERWISE PROVIDED FOR
- H10B—ELECTRONIC MEMORY DEVICES
- H10B51/00—Ferroelectric RAM [FeRAM] devices comprising ferroelectric memory transistors
- H10B51/10—Ferroelectric RAM [FeRAM] devices comprising ferroelectric memory transistors characterised by the top-view layout
-
- H—ELECTRICITY
- H10—SEMICONDUCTOR DEVICES; ELECTRIC SOLID-STATE DEVICES NOT OTHERWISE PROVIDED FOR
- H10B—ELECTRONIC MEMORY DEVICES
- H10B51/00—Ferroelectric RAM [FeRAM] devices comprising ferroelectric memory transistors
- H10B51/20—Ferroelectric RAM [FeRAM] devices comprising ferroelectric memory transistors characterised by the three-dimensional arrangements, e.g. with cells on different height levels
-
- H—ELECTRICITY
- H10—SEMICONDUCTOR DEVICES; ELECTRIC SOLID-STATE DEVICES NOT OTHERWISE PROVIDED FOR
- H10B—ELECTRONIC MEMORY DEVICES
- H10B51/00—Ferroelectric RAM [FeRAM] devices comprising ferroelectric memory transistors
- H10B51/30—Ferroelectric RAM [FeRAM] devices comprising ferroelectric memory transistors characterised by the memory core region
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/049—Temporal neural networks, e.g. delay elements, oscillating neurons or pulsed inputs
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
- H01L29/00—Semiconductor devices specially adapted for rectifying, amplifying, oscillating or switching and having potential barriers; Capacitors or resistors having potential barriers, e.g. a PN-junction depletion layer or carrier concentration layer; Details of semiconductor bodies or of electrodes thereof ; Multistep manufacturing processes therefor
- H01L29/40—Electrodes ; Multistep manufacturing processes therefor
- H01L29/41—Electrodes ; Multistep manufacturing processes therefor characterised by their shape, relative sizes or dispositions
- H01L29/423—Electrodes ; Multistep manufacturing processes therefor characterised by their shape, relative sizes or dispositions not carrying the current to be rectified, amplified or switched
- H01L29/42312—Gate electrodes for field effect devices
- H01L29/42316—Gate electrodes for field effect devices for field-effect transistors
- H01L29/4232—Gate electrodes for field effect devices for field-effect transistors with insulated gate
- H01L29/42364—Gate electrodes for field effect devices for field-effect transistors with insulated gate characterised by the insulating layer, e.g. thickness or uniformity
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
- H01L29/00—Semiconductor devices specially adapted for rectifying, amplifying, oscillating or switching and having potential barriers; Capacitors or resistors having potential barriers, e.g. a PN-junction depletion layer or carrier concentration layer; Details of semiconductor bodies or of electrodes thereof ; Multistep manufacturing processes therefor
- H01L29/40—Electrodes ; Multistep manufacturing processes therefor
- H01L29/41—Electrodes ; Multistep manufacturing processes therefor characterised by their shape, relative sizes or dispositions
- H01L29/423—Electrodes ; Multistep manufacturing processes therefor characterised by their shape, relative sizes or dispositions not carrying the current to be rectified, amplified or switched
- H01L29/42312—Gate electrodes for field effect devices
- H01L29/42316—Gate electrodes for field effect devices for field-effect transistors
- H01L29/4232—Gate electrodes for field effect devices for field-effect transistors with insulated gate
- H01L29/42372—Gate electrodes for field effect devices for field-effect transistors with insulated gate characterised by the conducting layer, e.g. the length, the sectional shape or the lay-out
- H01L29/4238—Gate electrodes for field effect devices for field-effect transistors with insulated gate characterised by the conducting layer, e.g. the length, the sectional shape or the lay-out characterised by the surface lay-out
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
- H01L29/00—Semiconductor devices specially adapted for rectifying, amplifying, oscillating or switching and having potential barriers; Capacitors or resistors having potential barriers, e.g. a PN-junction depletion layer or carrier concentration layer; Details of semiconductor bodies or of electrodes thereof ; Multistep manufacturing processes therefor
- H01L29/66—Types of semiconductor device ; Multistep manufacturing processes therefor
- H01L29/68—Types of semiconductor device ; Multistep manufacturing processes therefor controllable by only the electric current supplied, or only the electric potential applied, to an electrode which does not carry the current to be rectified, amplified or switched
- H01L29/76—Unipolar devices, e.g. field effect transistors
- H01L29/772—Field effect transistors
- H01L29/78—Field effect transistors with field effect produced by an insulated gate
- H01L29/78391—Field effect transistors with field effect produced by an insulated gate the gate comprising a layer which is used for its ferroelectric properties
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Definitions
- the present disclosure relates to a neuromorphic device and a synapse array of the neuromorphic device having ferroelectric field effect transistors.
- a neuromorphic device based on the neuromorphic technology includes a plurality of pre-synaptic neurons, a plurality of post-synaptic neurons, and a plurality of synapses.
- the neuromorphic device outputs pulses or spikes having various levels, amplitude, and/or times, according to a learning state of the neuromorphic device.
- Embodiments of the present disclosure include a synapse array of a neuromorphic device including a synapse having ferroelectric field effect transistors.
- inventions of the present disclosure include a synapse array of a neuromorphic device having a high degree of integration using ferroelectric field effect transistors.
- a synapse array of the neuromorphic device may include an input neuron; an output neuron; and a synapse.
- the synapse may include a plurality of ferroelectric field effect transistors electrically connected to each other in parallel.
- a synapse array of a neuromorphic device may include a first input neuron; a second input neuron; an output neuron; and a synapse.
- the synapse may include a plurality of pairs of ferroelectric field effect transistors electrically connected to each other in parallel.
- Each of the plurality of pairs of ferroelectric field effect transistors may include a first ferroelectric field effect transistor and a second ferroelectric field effect transistor.
- the first ferroelectric field effect transistor and the second field effect transistor may be electrically connected to each other in series.
- a synapse array of a neuromorphic device may include an input neuron; an output neuron; a plurality of gating controllers; and a synapse.
- the synapse may include a ferroelectric field effect transistor having a common gate insulating layer, a common floating gate electrode, a plurality of individual ferroelectric films, and a plurality of individual control gate electrodes.
- FIG. 1A is a diagram schematically illustrating a synapse array of a neuromorphic device in accordance with an embodiment of the present disclosure.
- FIG. 1B is a diagram illustrating a synapse in accordance with an embodiment of the present disclosure.
- FIG. 1C is a layout schematically illustrating the synapse array of the neuromorphic device shown in FIG. 1A .
- FIG. 1D schematically illustrates a longitudinal cross-sectional view of the synapse array of the neuromorphic device in accordance with an embodiment of the present disclosure taken along a line I-I′ in FIG. 1C .
- FIG. 2A is a diagram schematically illustrating a synapse array of a neuromorphic device in accordance with an embodiment of the present disclosure
- FIG. 2B is a diagram illustrating a synapse in accordance with the embodiment of the present disclosure.
- FIG. 2C is a layout schematically illustrating the synapse array shown in FIG. 2A .
- FIGS. 2D and 2E schematically illustrate longitudinal cross-sectional views taken along a line II-II′ in FIG. 2C .
- FIGS. 2F and 2G are cross-sectional views taken along a line III-III′ in FIG. 2C .
- FIG. 3A is a diagram schematically illustrating a synapse array in accordance with an embodiment of the present disclosure.
- FIG. 3B is a diagram illustrating a synapse in accordance with an embodiment of the present disclosure.
- FIG. 3C is a schematic diagram showing a synaptic operation of a pair of the synapse transistors of a synapse in accordance with an embodiment of the present disclosure.
- FIG. 3D is a layout schematically illustrating the synapse array shown in FIG. 3A .
- FIGS. 3E and 3F schematically illustrate longitudinal cross-sectional views of a synapse array in accordance with embodiments of the present disclosure taken along a line IV-IV′ in FIG. 3D .
- FIG. 4A is a diagram schematically illustrating a synapse array in accordance with an embodiment of the present disclosure.
- FIG. 4B is a schematic diagram illustrating a synapse in accordance with the embodiment of the present disclosure.
- FIG. 5 is a diagram schematically illustrating a pattern recognition system in accordance with an embodiment of the present disclosure.
- FIG. 1A is a diagram schematically illustrating a synapse array of a neuromorphic device in accordance with an embodiment of the present disclosure
- FIG. 1B is a diagram illustrating a synapse in accordance with an embodiment of the present disclosure.
- a synapse array 100 a of a neuromorphic device in accordance with an embodiment of the present disclosure may include a plurality of input neurons 10 , a plurality of output neurons 20 , a plurality of gating controllers 40 , and a plurality of synapses 30 a.
- Each of the synapses 30 a may include a transistor having a common floating gate electrode Gf, a plurality of individual ferroelectric films Ff 1 to Ffn, and a plurality of control gate electrodes Gc 1 to Gcn.
- the input neuron 10 may be electrically connected to a drain electrode D of the synapse 30 a through an input neuron line 11 to provide a first reference voltage to the synapse 30 a .
- the first reference voltage may be a power supply voltage.
- the input neuron 10 may include a pre-synaptic neuron. That is, the input neuron 10 may provide a voltage or current to the synapse 30 a . In another embodiment of the present disclosure, the input neuron 10 may provide a pulse-shaped electrical signal to the synapse 30 a.
- the output neuron 20 may be electrically connected to the source electrode S of synapse 30 a through the output neuron line 21 to provide a second reference voltage to the synapse 30 a .
- the second reference voltage may be a ground voltage.
- the output neuron 20 may include a post-synaptic neuron. A voltage or current may be supplied from the synapse 30 a to the output neuron 20 .
- the output neuron 20 may receive and output the synaptic weight of the synapse 30 a .
- the output neuron 20 may provide a voltage or current to the synapse 30 a.
- the plurality of gating controllers 40 may be electrically connected to the plurality of control gate electrodes Gc 1 to Gcn of synapse 30 a through a plurality of gating control lines 41 , respectively.
- the gating controllers 40 may partially and independently turn on or turn off transistors in synapse 30 a .
- the plurality of gating controllers 40 may provide the same voltage, e.g., the read voltage Vrd, to the plurality of control gate electrodes Gc 1 to Gcn at the same time.
- the synapse 30 a may have various polarization voltages Vp according to the remanent polarization of the ferroelectric films Ff 1 to Ffn. That is, the synapse 30 a may have various synaptic weighting levels, partially or overall.
- the common floating gate electrode Gf may affect the conductivity or the resistance value of the synapse 30 a according to the various polarization voltages Vp of the ferroelectric films Ff 1 to Ffn.
- the polarization voltages Vp of the ferroelectric films Ff 1 to Ffn may differ from each other.
- the common floating gate electrode Gf may adjust a channel of the synapse 30 a by averaging, or totaling, the polarization voltages Vp of the ferroelectric films Ff 1 to Ffn.
- the synapse 30 a of the synapse array 100 a in accordance with an embodiment of the present disclosure can have various contributions to synapse weight through a plurality of separate individual ferroelectric films Ff 1 to Ffn, and can also output the total synapse weights (e.g., current) that is averaged over, or summed through, the common floating gate electrode (Gf).
- Gf common floating gate electrode
- FIG. 1C is a layout schematically illustrating the synapse array of the neuromorphic device shown in FIG. 1A .
- the synapse array 100 a of the neuromorphic device in accordance with an embodiment of the present disclosure may include a plurality of active regions Act, a plurality of gate lines GL 1 to GLn, a plurality of input neuron via plugs 10 V, a plurality of output neuron via plugs 20 V, and a plurality of gating via plugs 40 V.
- the plurality of active regions Act may extend in a first direction in parallel with each other.
- the plurality of gate lines GL 1 to GLn may extend in a second direction.
- the plurality of input neuron via plugs 10 V may be disposed on drain regions D of the plurality of active regions Act.
- the plurality of output neuron via plugs 20 V may be disposed on source regions S of the plurality of active regions Act.
- the plurality of gating via plugs 40 V may be disposed on end portions of the plurality of gate lines GL 1 to GLn.
- the first direction and the second direction may be perpendicular to each other. That is, the plurality of active regions Act and the plurality of gate lines GL 1 to GLn may extend in directions perpendicular to each other.
- the plurality of gate lines GL 1 to GLn may correspond to the plurality of control gate electrodes Gc 1 to Gcn or the plurality of gating control lines 41 shown in FIGS. 1A or 1B .
- FIG. 1D schematically illustrates a longitudinal cross-sectional view of the synapse array of the neuromorphic device in accordance with an embodiment of the present disclosure taken along line I-I′ in FIG. 1C .
- the synapse array 100 a of the neuromorphic device in accordance with an embodiment of the present disclosure may include the synapse 30 a disposed on the active region Act of a substrate Sub.
- the synapse array 100 a may include an interlayer insulating layer 50 , an input neuron via plug 10 V and an output neuron via plug 20 V, and an input neuron line 11 and an output neuron line 21 .
- the interlayer insulating layer 50 may cover the synapse 30 a.
- the input neuron via plug 10 V and the output neuron via plug 20 V may each vertically penetrate the interlayer insulating layer 50 and connect to the active region Act of the substrate Sub.
- the input neuron line 11 and the output neuron line 21 may be disposed on the interlayer insulating layer 50 and electrically connected to the input neuron via plug 10 V and the output neuron via plug 20 V, respectively.
- the substrate Sub may include, for example, a bulk single-crystalline silicon layer or an epitaxially grown single-crystalline silicon layer.
- the interlayer insulating layer 50 may include an insulating material such as silicon oxide (SiO 2 ).
- the input neuron via plug 10 V and the output neuron via plug 20 V may include at least one of a doped poly-crystalline silicon, a metal silicide, a barrier metal layer, or a conductor such as a metal.
- the input neuron line 11 and the output neuron line 21 may include a barrier metal layer or a conductor such as a metal.
- the synapse 30 a may include a common gate insulating film 31 , a common floating gate electrode Gf, individual ferroelectric films Ff 1 to Ffn, and individual control gate electrodes Gc 1 to Gcn.
- the individual ferroelectric films Ff 1 to Ffn and the individual control gate electrodes Gc 1 to Gcn may share the common gate insulating film 31 and the common floating gate electrode Gf.
- the active region Act, the common gate insulating film 31 , and the common floating gate electrode Gf may extend in the same direction, e.g., the first direction, and the individual ferroelectric films Ff 1 to Ffn and the individual control gate electrodes Gc 1 to Gcn may extend in the same direction, e.g., the second direction.
- the first direction and the second direction may be perpendicular to each other in a top or plan view.
- the common gate insulating film 31 may include an insulating material such as, for example, silicon oxide (SiO 2 ) or silicon oxynitride (SiON).
- the common floating gate electrode Gf may include a poly-crystalline silicon layer doped with n-type ions such as phosphorus (P) or arsenic (As), as non-limiting examples.
- the individual ferroelectric films Ff 1 to Ffn may include a ferroelectric material such as hafnium-oxide (HfOx), as non-limiting examples.
- the individual control gate electrodes Gc 1 to Gcn may include a conductor such as a barrier metal layer and/or a metal layer, as non-limiting examples.
- the barrier metal layer may include titanium-nitride (TiN), tantalum-nitride (TaN), or at least one of the other barrier materials, as non-limiting examples.
- the metal layer may include a metallic material such as tungsten (W), aluminum (Al), copper (Cu), or the other metals, as non-limiting examples.
- FIG. 2A is a diagram schematically illustrating a synapse array of a neuromorphic device in accordance with an embodiment of the present disclosure
- FIG. 2B is a diagram illustrating a synapse in accordance with the embodiment of the present disclosure.
- a synapse array 100 b in accordance with an embodiment of the present disclosure may include a plurality of input neurons 10 , a plurality of output neurons 20 , a plurality of gating controllers 40 , and a plurality of synapses 30 b.
- a synapse 30 b may include a plurality of synapse transistors T 1 to Tn connected in parallel.
- the synapse transistors T 1 to Tn may include ferroelectric field effect transistors.
- Drain electrodes D 1 to Dn of synapse transistors T 1 to Tn may be electrically connected to a common input neuron 10 and source electrodes S 1 to Sn may be electrically connected to a common output neuron 20 .
- Gate electrodes G 1 to Gn of the synapse transistors T 1 to Tn may include ferroelectric films Ff 1 to Ffn, respectively.
- the synapse transistors T 1 to Tn may independently have various levels of synapse weights according to the polarization voltages of the ferroelectric films Ff 1 to Ffn, respectively.
- Each of the gate electrodes G 1 to Gn of the synapse transistors T 1 to Tn may be electrically connected to different gating controllers 40 , respectively. That is, the synapse transistors T 1 to Tn may have independent and separate turn-on states and turn-off states. Therefore, a sum of the synapse weights of the synapse 30 b may be determined in accordance with the turn-on states and the turn-off states of the synapse transistors T 1 to Tn.
- FIG. 2C is a layout schematically illustrating the synapse array shown in FIG. 2A .
- the synapse array 100 b of the neuromorphic device in accordance with an embodiment of the present disclosure may include a plurality of active regions Act and a plurality of gate lines GL 1 to GLn.
- the plurality of active regions Act may extend in parallel with each other in a first direction.
- the plurality of gate lines GL 1 to GLn may extend in parallel with each other in a second direction.
- the plurality of active regions Act may include drain regions D, channel regions C, and source regions S extending in the first direction, respectively.
- the drain regions D and the source regions S of the active regions Act may be electrically connected to each other as a whole, in both the first and second directions, respectively.
- the drain regions D and the source regions S may be electrically connected in a direction in which the gate lines GL 1 to GLn extend, that is, in the second direction.
- FIGS. 2D and 2E schematically illustrate longitudinal cross-sectional views taken along a line II-II′ in FIG. 2C
- FIGS. 2F and 2G schematically illustrate cross-sectional views taken along a line III-III′ in FIG. 2C .
- synapse arrays 100 b in accordance with embodiments of the present disclosure may include a plurality of isolation regions Iso used to define a plurality of active regions Act, a plurality of gate lines GL 1 to GLn, an interlayer insulating layer 50 , a plurality of gating via plugs 40 V, and a plurality of gating control lines 41 .
- the isolation regions Iso and the active regions Act may be disposed in a substrate Sub and the gate lines GL 1 to GLn, the interlayer insulating layer 50 , the gating via plugs 40 V, and the gating control lines 41 may be disposed on the substrate Sub.
- each of the gate lines GL 1 to GLn of the synapse transistors T 1 to Tn may include a gate insulating film 31 , ferroelectric insulating layers Ff 1 to Ffn, and gate electrodes G 1 to Gn, respectively, disposed on the substrate Sub.
- the gate lines GL 1 to GLn of the synapse transistors T 1 to Tn of the synapse array 100 b may include a common gate insulating film 31 , a common ferroelectric film Ff, and individual gate electrodes G 1 to Gn.
- a gate insulating film 31 , ferroelectric films Ff 1 to Ffn, and gate electrodes G 1 to Gn may extend in parallel with each other.
- the gate insulating film 31 and the ferroelectric films Ff 1 to Ffn may have plate shapes, and the gate electrodes G 1 to Gn may have line shapes.
- the gate insulating film 31 may extend in the same direction as the active region Act.
- a ferroelectric film Ff and gate electrodes G 1 to Gn may have line shapes that extend in directions perpendicular to each other.
- the ferroelectric film Ff may be independently disposed so as to be separated from each other, such as in an island type.
- a gate insulating film 31 may have a plate shape, and ferroelectric films Ff 1 to Ffn and gate electrodes G 1 to Gn may have line shapes. That is, the ferroelectric films Ff 1 to Ffn and the gate electrodes G 1 to Gn may extend in the same direction.
- a gate insulating film 31 and a ferroelectric film Ff may have a plate shape, and gate electrodes G 1 to Gn may have line shapes.
- the ferroelectric film Ff may have a line shape extending in the first direction, and the gate electrodes G 1 to Gn may have line shapes extending in the second direction.
- FIG. 3A is a diagram schematically illustrating a synapse array in accordance with an embodiment of the present disclosure
- FIG. 3B is a diagram illustrating a synapse in accordance with an embodiment of the present disclosure.
- a synapse array 100 c in accordance with an embodiment of the present disclosure may include a plurality of first input neurons 10 a , a plurality of second input neurons 10 b , a plurality of output neurons 20 , a plurality of gating controllers 40 , and a plurality of synapses 30 c.
- Each of the plurality of synapses 30 c may include a plurality of pairs of synapse transistors TP 1 to TPn connected in parallel with each other.
- the pairs of the synapse transistors TP 1 to TPn may include a plurality of first transistors T 1 a to Tna and a plurality of second transistors T 1 b to Tnb connected in series, respectively. Referring to FIG.
- drain electrodes D 1 a to Dna of the first transistors T 1 a to Tna may be connected to a common first input neuron 10 a
- source electrodes S 1 b to Snb of the second transistors T 1 b to Tnb may be connected to a common second input neuron 10 b
- source electrodes S 1 a to Sna of the first transistors T 1 a to Tna and drain electrodes D 1 b to Dnb of the second transistors T 1 b to Tnb may be directly electrically connected to each other and connected to a common output neuron 20 , respectively.
- gate electrodes G 1 a to Gna of the first transistors T 1 a to Tna and gate electrodes G 1 b to Gnb of the second transistors T 1 b to Tnb are directly electrically connected to each other and connected to the gating controllers 40 , respectively.
- the first input neurons 10 a may provide a first reference voltage to the drain electrodes D 1 a to Dna of the first transistors T 1 a to Tna of synapse 30 c .
- the first input neuron 10 a may be connected to a power supply voltage node, and the first reference voltage may be a power supply voltage.
- the second input neurons 10 b may provide a second reference voltage to the source electrodes S 1 b to Snb of the second transistors T 1 b to Tnb of the synapse 30 c .
- the second input neurons 10 b may be coupled to a ground voltage node, and the second reference voltage may be a ground voltage.
- the first reference voltage may be relatively higher or greater than the second reference voltage.
- FIG. 3C is a schematic diagram showing a synaptic operation of a pair of the synapse transistors of a synapse in accordance with an embodiment of the present disclosure.
- a pair of the synapse transistors Ta and Tb of a synapse 30 c in accordance with an embodiment of the present disclosure can execute or operate an excitatory synapse operation and/or an inhibitory synapse operation according to polarization voltages Vpa and Vpb of gate electrodes Ga and Gb of the transistors Ta and Tb.
- the first transistor Ta allows an excitatory synapse current Ie to conduct from the first input neuron 10 a to the output neuron 20
- the second transistor Tb allows an inhibitory synapse current Ii to conduct from the output neuron 20 to the second input neuron 10 b.
- the excitatory synapse current Ie may be an excitatory synapse operation that increases current to the output neuron 20 to potentiate the synaptic weight
- the inhibitory synapse current Ii may be an inhibitory synapse operation that decreases current from the output neuron 20 to depress the synaptic weight.
- the inhibitory synapse current Ii may be greater than the excitatory synapse current Ie so that the pair of the synapse transistors Ta and Tb can perform an inhibitory synapse operation as a whole.
- the pair of the synapse transistors Ta and Tb can selectively perform either an excitatory synapse operation or an inhibitory synapse operation depending on the polarization voltages Vpa and Vpb of the gate electrodes Ga and Gb of the transistors Ta and Tb.
- FIG. 3D is a layout schematically illustrating the synapse array shown in FIG. 3A .
- the synapse array 100 c of the neuromorphic device in accordance with the embodiment of the present disclosure may include a plurality of active regions Act and a plurality of gate lines GL 1 to GLn.
- the plurality of active regions Act may extend in parallel with each other in a first direction.
- the plurality of gate lines GL 1 to GLn may extend in parallel with each other in a second direction.
- Each of the active regions Act may include a first drain region Da, a first channel region Ca, a first source region Sa, a second drain region Db, a second channel region Cb, and a second source region Sb.
- the first source region Sa and the second drain region Db may be one shared region in common.
- the first drain regions Da, the first channel regions Ca, the first source region Sa, the second drain region Db, the second channel regions Cb, and the second source regions Sb may overlap and intersect with the gate lines GL 1 to GLn.
- the first drain region Da, the first channel region Ca, and the first source region Sa may constitute the first transistors T 1 a to Tna shown in FIGS. 3A to 3C .
- the second drain region Db, the second channel region Cb, and the second source region Sb may constitute the second transistors T 1 b to Tnb shown in FIGS. 3A to 3C .
- FIGS. 3E and 3F schematically illustrate longitudinal cross-sectional views of a synapse array in accordance with embodiments of the present disclosure taken along a line IV-IV′ in FIG. 3D .
- a synapse array 100 c in accordance with an embodiment of the present disclosure may include isolation regions Iso, a gate line GL, an interlayer insulating layer 50 , a gating via plug 40 V, and a gating control line 41 .
- the isolation regions Iso may define active regions Act.
- the isolation regions Iso and the active regions Act may be disposed in a substrate Sub.
- the gate line GL, the interlayer insulating layer 50 , the gating via plug 40 V, and the gating control line 41 may be disposed on the substrate Sub.
- the gate line GL may include a gate insulating film 31 , a ferroelectric film Ff, and a gate electrode G.
- the gate insulating film 31 , the ferroelectric film Ff, and the gate electrode G may be a line shape crossing the active regions Act.
- the gate insulating film 31 and the ferroelectric film Ff may be a plate shape, and the gate electrode G may have a line shape.
- the ferroelectric film Ff and the gate electrode G may be line shapes extending in directions perpendicular to each other.
- the ferroelectric film Ff may be independently disposed so as to be separated from each other, such as in an island type.
- the gate insulating film 31 may be a plate shape, and the ferroelectric film Ff and the gate electrode G may be line shapes. That is, the ferroelectric film Ff and the gate electrode G may extend in directions perpendicular to each other.
- FIG. 4A is a diagram schematically illustrating a synapse array in accordance with an embodiment of the present disclosure
- FIG. 4B is a schematic diagram illustrating a synapse in accordance with the embodiment of the present disclosure.
- a synapse array 100 d in accordance with an embodiment of the present disclosure may include a plurality of first input neurons 10 a , a plurality of second input neurons 10 b , a plurality of output neurons 20 , a plurality of gating controllers 40 , and a plurality of synapses 30 d.
- Each of the synapses 30 d may include a plurality of pairs of synapse transistors TP 1 to TPn connected in parallel.
- the pairs of synapse transistors TP 1 to TPn may include first transistors T 1 a to Tna and second transistors T 1 b to Tnb connected in series, respectively. Referring to FIG.
- drain electrodes D 1 a to Dna of the first transistors T 1 a to Tna may be connected to a common first input neuron 10 a
- source electrodes S 1 b to Snb of the second transistors T 1 b to Tnb may be connected to a common second input neuron 10 b
- source electrodes S 1 a to Sna of the first transistors T 1 a to Tna and the drain electrodes D 1 b to Dnb of the second transistors T 1 b to Tnb may be directly electrically connected to each other and connected to a common output neuron 20 , respectively.
- the gate electrodes G 1 a to Gna of the first transistors T 1 a to Tna and the gate electrodes G 1 b to Gnb of the second transistors T 1 b to Tnb may be directly electrically connected to a plurality of gating controllers 40 , respectively.
- the first input neuron 10 a may provide a first reference voltage to the drain electrodes D 1 a to Dna of the first transistors T 1 a to Tna of the synapse 30 d .
- the first input neuron 10 a may be coupled to the power supply voltage node, and the first reference voltage may be the power supply voltage.
- the second input neuron 10 b may provide the second reference voltage to the source electrodes S 1 b to Snb of the second transistors T 1 b to Tnb of the synapse 30 d .
- the second input neuron 10 b may be coupled to a ground voltage node, and the second reference voltage may be a ground voltage.
- the first reference voltage may be higher or greater than the second reference voltage.
- FIG. 5 is a diagram schematically illustrating a pattern recognition system in accordance with an embodiment of the present disclosure.
- a pattern recognition system 900 may include a speech recognition system, an imaging recognition system, a code recognition system, a signal recognition system, and one or more systems for recognizing various patterns.
- the pattern recognition system 900 in accordance with the embodiment of the present disclosure may include a Central Processing Unit (CPU) 910 , a memory unit 920 , a communication control unit 930 , a network 940 , an output unit 950 , an input unit 960 , an Analog-Digital Converter (ADC) 970 , a neuromorphic unit 980 , and/or a bus 990 .
- the CPU 910 may generate and transmit various signals for a learning process of the neuromorphic unit 980 , and perform various processes and functions for recognizing patterns according to an output from the neuromorphic unit 980 .
- the CPU 910 may perform processes and functions for recognizing speech and imaging patterns based on an output from the neuromorphic unit 980 .
- the CPU 910 may be connected with the memory unit 920 , the communication control unit 930 , the output unit 950 , the ADC 970 , and the neuromorphic unit 980 through the bus 990 .
- the memory unit 920 may store various pieces of information, which are required to be stored in the pattern recognition system 900 .
- the memory unit 920 may include one or more of a volatile memory device, such as DRAM or SRAM, a nonvolatile memory, such as PRAM, MRAM, ReRAM or NAND flash memory, and various memory units, such as Hard Disk Drive (HDD) and Solid State Drive (SSD).
- a volatile memory device such as DRAM or SRAM
- a nonvolatile memory such as PRAM, MRAM, ReRAM or NAND flash memory
- various memory units such as Hard Disk Drive (HDD) and Solid State Drive (SSD).
- HDD Hard Disk Drive
- SSD Solid State Drive
- the communication control unit 930 may transmit and/or receive data to and/or from a communication control unit of another system through the network 940 .
- the communication control unit 930 may transmit speech and/or image recognition data through the network 940 .
- the output unit 950 may output data in various manners.
- the output unit 950 may include a speaker, a printer, a monitor, a display panel, a beam projector, a hologrammer, or other various output devices.
- the output unit 950 may output, for example, speech and/or image recognition data.
- the input unit 960 may include any of a microphone, a camera, a scanner, a touch pad, a keyboard, a mouse, a mouse pen, and one or more of various sensors.
- the ADC 970 may convert analog data inputted from the input unit 960 into digital data.
- the neuromorphic unit 980 may perform learning or recognition using the data outputted from the ADC 970 , and output data corresponding to recognized patterns.
- the neuromorphic unit 980 may include one or more of the neuromorphic devices in accordance with the various embodiments described above.
- the synapse arrays of a neuromorphic device including ferroelectric field effect transistors can have a higher degree of integration, a lower power consumption, and multiple resistance levels.
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Power Engineering (AREA)
- Microelectronics & Electronic Packaging (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computer Hardware Design (AREA)
- Condensed Matter Physics & Semiconductors (AREA)
- Ceramic Engineering (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Manufacturing & Machinery (AREA)
- Neurology (AREA)
- Semiconductor Memories (AREA)
- Thin Film Transistor (AREA)
- Non-Volatile Memory (AREA)
Abstract
A synapse array of a neuromorphic device comprises a first input neuron, a second input neuron, an output neuron, and a synapse. The synapse comprises a plurality of pairs of ferroelectric field effect transistors electrically connected to each other in parallel, each of the plurality of pairs of ferroelectric field effect transistors comprises a first ferroelectric field effect transistor and a second ferroelectric field effect transistor, and the first ferroelectric field effect transistor and the second ferroelectric field effect transistor are electrically connected to each other in series.
Description
- The present patent application a divisional application of a U.S. patent application Ser. No. 15/850,145, filed on Dec. 21, 2017, which claims priority under 35 U.S.C. § 119(a) to Korean Patent Application No. 10-2017-0069563, filed on Jun. 5, 2017, which is incorporated herein by reference in its entirety.
- The present disclosure relates to a neuromorphic device and a synapse array of the neuromorphic device having ferroelectric field effect transistors.
- Recently, much attention has been paid to devices in the field of neuromorphic technology, which use chips that mimic the human brain. A neuromorphic device based on the neuromorphic technology includes a plurality of pre-synaptic neurons, a plurality of post-synaptic neurons, and a plurality of synapses. The neuromorphic device outputs pulses or spikes having various levels, amplitude, and/or times, according to a learning state of the neuromorphic device.
- Embodiments of the present disclosure include a synapse array of a neuromorphic device including a synapse having ferroelectric field effect transistors.
- Other embodiments of the present disclosure include a synapse array of a neuromorphic device having a high degree of integration using ferroelectric field effect transistors.
- Further embodiments of the present disclosure include a synapse array of a neuromorphic device performing an excitatory synapse operation and an inhibitory synapse operation using ferroelectric field effect transistors.
- In accordance with an embodiment of the present disclosure, a synapse array of the neuromorphic device may include an input neuron; an output neuron; and a synapse. The synapse may include a plurality of ferroelectric field effect transistors electrically connected to each other in parallel.
- In accordance with the embodiment of the present disclosure, a synapse array of a neuromorphic device may include a first input neuron; a second input neuron; an output neuron; and a synapse. The synapse may include a plurality of pairs of ferroelectric field effect transistors electrically connected to each other in parallel. Each of the plurality of pairs of ferroelectric field effect transistors may include a first ferroelectric field effect transistor and a second ferroelectric field effect transistor. The first ferroelectric field effect transistor and the second field effect transistor may be electrically connected to each other in series.
- In accordance with an embodiment of the present disclosure, a synapse array of a neuromorphic device may include an input neuron; an output neuron; a plurality of gating controllers; and a synapse. The synapse may include a ferroelectric field effect transistor having a common gate insulating layer, a common floating gate electrode, a plurality of individual ferroelectric films, and a plurality of individual control gate electrodes.
-
FIG. 1A is a diagram schematically illustrating a synapse array of a neuromorphic device in accordance with an embodiment of the present disclosure. -
FIG. 1B is a diagram illustrating a synapse in accordance with an embodiment of the present disclosure. -
FIG. 1C is a layout schematically illustrating the synapse array of the neuromorphic device shown inFIG. 1A . -
FIG. 1D schematically illustrates a longitudinal cross-sectional view of the synapse array of the neuromorphic device in accordance with an embodiment of the present disclosure taken along a line I-I′ inFIG. 1C . -
FIG. 2A is a diagram schematically illustrating a synapse array of a neuromorphic device in accordance with an embodiment of the present disclosure, andFIG. 2B is a diagram illustrating a synapse in accordance with the embodiment of the present disclosure. -
FIG. 2C is a layout schematically illustrating the synapse array shown inFIG. 2A . -
FIGS. 2D and 2E schematically illustrate longitudinal cross-sectional views taken along a line II-II′ inFIG. 2C . -
FIGS. 2F and 2G are cross-sectional views taken along a line III-III′ inFIG. 2C . -
FIG. 3A is a diagram schematically illustrating a synapse array in accordance with an embodiment of the present disclosure. -
FIG. 3B is a diagram illustrating a synapse in accordance with an embodiment of the present disclosure. -
FIG. 3C is a schematic diagram showing a synaptic operation of a pair of the synapse transistors of a synapse in accordance with an embodiment of the present disclosure. -
FIG. 3D is a layout schematically illustrating the synapse array shown inFIG. 3A . -
FIGS. 3E and 3F schematically illustrate longitudinal cross-sectional views of a synapse array in accordance with embodiments of the present disclosure taken along a line IV-IV′ inFIG. 3D . -
FIG. 4A is a diagram schematically illustrating a synapse array in accordance with an embodiment of the present disclosure. -
FIG. 4B is a schematic diagram illustrating a synapse in accordance with the embodiment of the present disclosure. -
FIG. 5 is a diagram schematically illustrating a pattern recognition system in accordance with an embodiment of the present disclosure. - Various embodiments will be described below in more detail with reference to the accompanying drawings. Embodiments of the present disclosure may, however, have different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the claims to those skilled in the art.
- Throughout the specification, like reference numerals refer to the same elements. Therefore, although the same or similar reference numerals are not mentioned or described in the corresponding drawing, the reference numerals may be described with reference to other drawings. Furthermore, although elements are not represented by reference numerals, the elements may be described with reference to other drawings.
-
FIG. 1A is a diagram schematically illustrating a synapse array of a neuromorphic device in accordance with an embodiment of the present disclosure, andFIG. 1B is a diagram illustrating a synapse in accordance with an embodiment of the present disclosure. - Referring to
FIGS. 1A and 1B , asynapse array 100 a of a neuromorphic device in accordance with an embodiment of the present disclosure may include a plurality ofinput neurons 10, a plurality ofoutput neurons 20, a plurality ofgating controllers 40, and a plurality ofsynapses 30 a. - Each of the
synapses 30 a may include a transistor having a common floating gate electrode Gf, a plurality of individual ferroelectric films Ff1 to Ffn, and a plurality of control gate electrodes Gc1 to Gcn. - The
input neuron 10 may be electrically connected to a drain electrode D of thesynapse 30 a through aninput neuron line 11 to provide a first reference voltage to thesynapse 30 a. For example, the first reference voltage may be a power supply voltage. In an embodiment of the present disclosure, theinput neuron 10 may include a pre-synaptic neuron. That is, theinput neuron 10 may provide a voltage or current to thesynapse 30 a. In another embodiment of the present disclosure, theinput neuron 10 may provide a pulse-shaped electrical signal to thesynapse 30 a. - The
output neuron 20 may be electrically connected to the source electrode S ofsynapse 30 a through theoutput neuron line 21 to provide a second reference voltage to thesynapse 30 a. For example, the second reference voltage may be a ground voltage. In an embodiment of the present disclosure, theoutput neuron 20 may include a post-synaptic neuron. A voltage or current may be supplied from thesynapse 30 a to theoutput neuron 20. Thus, theoutput neuron 20 may receive and output the synaptic weight of thesynapse 30 a. In another embodiment of the present disclosure, theoutput neuron 20 may provide a voltage or current to thesynapse 30 a. - Referring again to
FIG. 1B , the plurality ofgating controllers 40 may be electrically connected to the plurality of control gate electrodes Gc1 to Gcn ofsynapse 30 a through a plurality ofgating control lines 41, respectively. Thegating controllers 40 may partially and independently turn on or turn off transistors insynapse 30 a. In the read-out mode, for example, the plurality ofgating controllers 40 may provide the same voltage, e.g., the read voltage Vrd, to the plurality of control gate electrodes Gc1 to Gcn at the same time. - The
synapse 30 a may have various polarization voltages Vp according to the remanent polarization of the ferroelectric films Ff1 to Ffn. That is, thesynapse 30 a may have various synaptic weighting levels, partially or overall. - The common floating gate electrode Gf may affect the conductivity or the resistance value of the
synapse 30 a according to the various polarization voltages Vp of the ferroelectric films Ff1 to Ffn. For example, the polarization voltages Vp of the ferroelectric films Ff1 to Ffn may differ from each other. The common floating gate electrode Gf may adjust a channel of thesynapse 30 a by averaging, or totaling, the polarization voltages Vp of the ferroelectric films Ff1 to Ffn. That is, thesynapse 30 a of thesynapse array 100 a in accordance with an embodiment of the present disclosure can have various contributions to synapse weight through a plurality of separate individual ferroelectric films Ff1 to Ffn, and can also output the total synapse weights (e.g., current) that is averaged over, or summed through, the common floating gate electrode (Gf). -
FIG. 1C is a layout schematically illustrating the synapse array of the neuromorphic device shown inFIG. 1A . Referring toFIG. 1C , thesynapse array 100 a of the neuromorphic device in accordance with an embodiment of the present disclosure may include a plurality of active regions Act, a plurality of gate lines GL1 to GLn, a plurality of input neuron viaplugs 10V, a plurality of output neuron viaplugs 20V, and a plurality of gating viaplugs 40V. - The plurality of active regions Act may extend in a first direction in parallel with each other. The plurality of gate lines GL1 to GLn may extend in a second direction. The plurality of input neuron via
plugs 10V may be disposed on drain regions D of the plurality of active regions Act. The plurality of output neuron viaplugs 20V may be disposed on source regions S of the plurality of active regions Act. The plurality of gating viaplugs 40V may be disposed on end portions of the plurality of gate lines GL1 to GLn. The first direction and the second direction may be perpendicular to each other. That is, the plurality of active regions Act and the plurality of gate lines GL1 to GLn may extend in directions perpendicular to each other. In an example, the plurality of gate lines GL1 to GLn may correspond to the plurality of control gate electrodes Gc1 to Gcn or the plurality ofgating control lines 41 shown inFIGS. 1A or 1B . -
FIG. 1D schematically illustrates a longitudinal cross-sectional view of the synapse array of the neuromorphic device in accordance with an embodiment of the present disclosure taken along line I-I′ inFIG. 1C . Referring toFIG. 1D , thesynapse array 100 a of the neuromorphic device in accordance with an embodiment of the present disclosure may include thesynapse 30 a disposed on the active region Act of a substrate Sub. - The
synapse array 100 a may include aninterlayer insulating layer 50, an input neuron viaplug 10V and an output neuron viaplug 20V, and aninput neuron line 11 and anoutput neuron line 21. The interlayer insulatinglayer 50 may cover thesynapse 30 a. - The input neuron via
plug 10V and the output neuron viaplug 20V may each vertically penetrate the interlayer insulatinglayer 50 and connect to the active region Act of the substrate Sub. Theinput neuron line 11 and theoutput neuron line 21 may be disposed on theinterlayer insulating layer 50 and electrically connected to the input neuron viaplug 10V and the output neuron viaplug 20V, respectively. - The substrate Sub may include, for example, a bulk single-crystalline silicon layer or an epitaxially grown single-crystalline silicon layer. The interlayer insulating
layer 50 may include an insulating material such as silicon oxide (SiO2). The input neuron viaplug 10V and the output neuron viaplug 20V may include at least one of a doped poly-crystalline silicon, a metal silicide, a barrier metal layer, or a conductor such as a metal. Theinput neuron line 11 and theoutput neuron line 21 may include a barrier metal layer or a conductor such as a metal. - The
synapse 30 a may include a commongate insulating film 31, a common floating gate electrode Gf, individual ferroelectric films Ff1 to Ffn, and individual control gate electrodes Gc1 to Gcn. For example, the individual ferroelectric films Ff1 to Ffn and the individual control gate electrodes Gc1 to Gcn may share the commongate insulating film 31 and the common floating gate electrode Gf. The active region Act, the commongate insulating film 31, and the common floating gate electrode Gf may extend in the same direction, e.g., the first direction, and the individual ferroelectric films Ff1 to Ffn and the individual control gate electrodes Gc1 to Gcn may extend in the same direction, e.g., the second direction. The first direction and the second direction may be perpendicular to each other in a top or plan view. - The common
gate insulating film 31 may include an insulating material such as, for example, silicon oxide (SiO2) or silicon oxynitride (SiON). The common floating gate electrode Gf may include a poly-crystalline silicon layer doped with n-type ions such as phosphorus (P) or arsenic (As), as non-limiting examples. The individual ferroelectric films Ff1 to Ffn may include a ferroelectric material such as hafnium-oxide (HfOx), as non-limiting examples. The individual control gate electrodes Gc1 to Gcn may include a conductor such as a barrier metal layer and/or a metal layer, as non-limiting examples. The barrier metal layer may include titanium-nitride (TiN), tantalum-nitride (TaN), or at least one of the other barrier materials, as non-limiting examples. The metal layer may include a metallic material such as tungsten (W), aluminum (Al), copper (Cu), or the other metals, as non-limiting examples. -
FIG. 2A is a diagram schematically illustrating a synapse array of a neuromorphic device in accordance with an embodiment of the present disclosure, andFIG. 2B is a diagram illustrating a synapse in accordance with the embodiment of the present disclosure. Referring toFIGS. 2A and 2B , asynapse array 100 b in accordance with an embodiment of the present disclosure may include a plurality ofinput neurons 10, a plurality ofoutput neurons 20, a plurality ofgating controllers 40, and a plurality ofsynapses 30 b. - Referring to
FIG. 2B , asynapse 30 b may include a plurality of synapse transistors T1 to Tn connected in parallel. The synapse transistors T1 to Tn may include ferroelectric field effect transistors. Drain electrodes D1 to Dn of synapse transistors T1 to Tn may be electrically connected to acommon input neuron 10 and source electrodes S1 to Sn may be electrically connected to acommon output neuron 20. Gate electrodes G1 to Gn of the synapse transistors T1 to Tn may include ferroelectric films Ff1 to Ffn, respectively. Accordingly, the synapse transistors T1 to Tn may independently have various levels of synapse weights according to the polarization voltages of the ferroelectric films Ff1 to Ffn, respectively. Each of the gate electrodes G1 to Gn of the synapse transistors T1 to Tn may be electrically connected todifferent gating controllers 40, respectively. That is, the synapse transistors T1 to Tn may have independent and separate turn-on states and turn-off states. Therefore, a sum of the synapse weights of thesynapse 30 b may be determined in accordance with the turn-on states and the turn-off states of the synapse transistors T1 to Tn. -
FIG. 2C is a layout schematically illustrating the synapse array shown inFIG. 2A . Referring toFIG. 2C , thesynapse array 100 b of the neuromorphic device in accordance with an embodiment of the present disclosure may include a plurality of active regions Act and a plurality of gate lines GL1 to GLn. - The plurality of active regions Act may extend in parallel with each other in a first direction. The plurality of gate lines GL1 to GLn may extend in parallel with each other in a second direction. The plurality of active regions Act may include drain regions D, channel regions C, and source regions S extending in the first direction, respectively. The drain regions D and the source regions S of the active regions Act may be electrically connected to each other as a whole, in both the first and second directions, respectively. For example, the drain regions D and the source regions S may be electrically connected in a direction in which the gate lines GL1 to GLn extend, that is, in the second direction.
-
FIGS. 2D and 2E schematically illustrate longitudinal cross-sectional views taken along a line II-II′ inFIG. 2C , andFIGS. 2F and 2G schematically illustrate cross-sectional views taken along a line III-III′ inFIG. 2C . - Referring to
FIGS. 2D, 2E, 2F and 2G ,synapse arrays 100 b in accordance with embodiments of the present disclosure may include a plurality of isolation regions Iso used to define a plurality of active regions Act, a plurality of gate lines GL1 to GLn, aninterlayer insulating layer 50, a plurality of gating viaplugs 40V, and a plurality of gating control lines 41. The isolation regions Iso and the active regions Act may be disposed in a substrate Sub and the gate lines GL1 to GLn, theinterlayer insulating layer 50, the gating viaplugs 40V, and thegating control lines 41 may be disposed on the substrate Sub. - Referring to
FIGS. 2A, 2B, 2C, 2D, 2E, and 2F , in embodiments, each of the gate lines GL1 to GLn of the synapse transistors T1 to Tn may include agate insulating film 31, ferroelectric insulating layers Ff1 to Ffn, and gate electrodes G1 to Gn, respectively, disposed on the substrate Sub. - Referring to
FIGS. 2A, 2B, 2C, 2D, 2E and 2G , in other embodiments, the gate lines GL1 to GLn of the synapse transistors T1 to Tn of thesynapse array 100 b may include a commongate insulating film 31, a common ferroelectric film Ff, and individual gate electrodes G1 to Gn. - Referring to
FIG. 2D , in an embodiment of the present disclosure, agate insulating film 31, ferroelectric films Ff1 to Ffn, and gate electrodes G1 to Gn may extend in parallel with each other. Alternatively, in another embodiment of the present disclosure, thegate insulating film 31 and the ferroelectric films Ff1 to Ffn may have plate shapes, and the gate electrodes G1 to Gn may have line shapes. In an embodiment of the present disclosure, thegate insulating film 31 may extend in the same direction as the active region Act. - Referring to
FIG. 2E , in an embodiment of the present disclosure, a ferroelectric film Ff and gate electrodes G1 to Gn may have line shapes that extend in directions perpendicular to each other. Alternatively, in another embodiment of the present disclosure, the ferroelectric film Ff may be independently disposed so as to be separated from each other, such as in an island type. - Referring to
FIG. 2F , in an embodiment of the present disclosure, agate insulating film 31 may have a plate shape, and ferroelectric films Ff1 to Ffn and gate electrodes G1 to Gn may have line shapes. That is, the ferroelectric films Ff1 to Ffn and the gate electrodes G1 to Gn may extend in the same direction. - Referring to
FIG. 2G , in an embodiment of the present disclosure, agate insulating film 31 and a ferroelectric film Ff may have a plate shape, and gate electrodes G1 to Gn may have line shapes. In another embodiment of the present disclosure, the ferroelectric film Ff may have a line shape extending in the first direction, and the gate electrodes G1 to Gn may have line shapes extending in the second direction. -
FIG. 3A is a diagram schematically illustrating a synapse array in accordance with an embodiment of the present disclosure, andFIG. 3B is a diagram illustrating a synapse in accordance with an embodiment of the present disclosure. - Referring to
FIGS. 3A and 3B , asynapse array 100 c in accordance with an embodiment of the present disclosure may include a plurality offirst input neurons 10 a, a plurality ofsecond input neurons 10 b, a plurality ofoutput neurons 20, a plurality ofgating controllers 40, and a plurality ofsynapses 30 c. - Each of the plurality of
synapses 30 c may include a plurality of pairs of synapse transistors TP1 to TPn connected in parallel with each other. The pairs of the synapse transistors TP1 to TPn may include a plurality of first transistors T1 a to Tna and a plurality of second transistors T1 b to Tnb connected in series, respectively. Referring toFIG. 3B , in thesynapse 30 c, drain electrodes D1 a to Dna of the first transistors T1 a to Tna may be connected to a commonfirst input neuron 10 a, and source electrodes S1 b to Snb of the second transistors T1 b to Tnb may be connected to a commonsecond input neuron 10 b, and source electrodes S1 a to Sna of the first transistors T1 a to Tna and drain electrodes D1 b to Dnb of the second transistors T1 b to Tnb may be directly electrically connected to each other and connected to acommon output neuron 20, respectively. - Furthermore, gate electrodes G1 a to Gna of the first transistors T1 a to Tna and gate electrodes G1 b to Gnb of the second transistors T1 b to Tnb are directly electrically connected to each other and connected to the
gating controllers 40, respectively. - The
first input neurons 10 a may provide a first reference voltage to the drain electrodes D1 a to Dna of the first transistors T1 a to Tna ofsynapse 30 c. For example, thefirst input neuron 10 a may be connected to a power supply voltage node, and the first reference voltage may be a power supply voltage. - The
second input neurons 10 b may provide a second reference voltage to the source electrodes S1 b to Snb of the second transistors T1 b to Tnb of thesynapse 30 c. For example, thesecond input neurons 10 b may be coupled to a ground voltage node, and the second reference voltage may be a ground voltage. The first reference voltage may be relatively higher or greater than the second reference voltage. -
FIG. 3C is a schematic diagram showing a synaptic operation of a pair of the synapse transistors of a synapse in accordance with an embodiment of the present disclosure. Referring toFIG. 3C , a pair of the synapse transistors Ta and Tb of asynapse 30 c in accordance with an embodiment of the present disclosure can execute or operate an excitatory synapse operation and/or an inhibitory synapse operation according to polarization voltages Vpa and Vpb of gate electrodes Ga and Gb of the transistors Ta and Tb. For example, when the read voltage Vrd is applied to the gate electrodes Ga and Gb of the transistors Ta and Tb from thegating controller 40, depending on the polarization voltages Vpa and Vb of the gate electrode Ga and Gb of the transistors Ta and Tb, the first transistor Ta allows an excitatory synapse current Ie to conduct from thefirst input neuron 10 a to theoutput neuron 20, and the second transistor Tb allows an inhibitory synapse current Ii to conduct from theoutput neuron 20 to thesecond input neuron 10 b. - The excitatory synapse current Ie may be an excitatory synapse operation that increases current to the
output neuron 20 to potentiate the synaptic weight, and the inhibitory synapse current Ii may be an inhibitory synapse operation that decreases current from theoutput neuron 20 to depress the synaptic weight. In another example, if the read voltage Vrd is zero volts (OV) and the polarization voltage Vpb of the second transistor Tb is greater than the polarization voltage Vpa of the first transistor Ta, the inhibitory synapse current Ii may be greater than the excitatory synapse current Ie so that the pair of the synapse transistors Ta and Tb can perform an inhibitory synapse operation as a whole. Thus, the pair of the synapse transistors Ta and Tb can selectively perform either an excitatory synapse operation or an inhibitory synapse operation depending on the polarization voltages Vpa and Vpb of the gate electrodes Ga and Gb of the transistors Ta and Tb. -
FIG. 3D is a layout schematically illustrating the synapse array shown inFIG. 3A . Referring toFIG. 3D , thesynapse array 100 c of the neuromorphic device in accordance with the embodiment of the present disclosure may include a plurality of active regions Act and a plurality of gate lines GL1 to GLn. - The plurality of active regions Act may extend in parallel with each other in a first direction. The plurality of gate lines GL1 to GLn may extend in parallel with each other in a second direction. Each of the active regions Act may include a first drain region Da, a first channel region Ca, a first source region Sa, a second drain region Db, a second channel region Cb, and a second source region Sb. The first source region Sa and the second drain region Db may be one shared region in common. The first drain regions Da, the first channel regions Ca, the first source region Sa, the second drain region Db, the second channel regions Cb, and the second source regions Sb may overlap and intersect with the gate lines GL1 to GLn. The first drain region Da, the first channel region Ca, and the first source region Sa may constitute the first transistors T1 a to Tna shown in
FIGS. 3A to 3C . The second drain region Db, the second channel region Cb, and the second source region Sb may constitute the second transistors T1 b to Tnb shown inFIGS. 3A to 3C . -
FIGS. 3E and 3F schematically illustrate longitudinal cross-sectional views of a synapse array in accordance with embodiments of the present disclosure taken along a line IV-IV′ inFIG. 3D . Referring toFIGS. 3E and 3F , asynapse array 100 c in accordance with an embodiment of the present disclosure may include isolation regions Iso, a gate line GL, aninterlayer insulating layer 50, a gating viaplug 40V, and agating control line 41. - The isolation regions Iso may define active regions Act. The isolation regions Iso and the active regions Act may be disposed in a substrate Sub. The gate line GL, the
interlayer insulating layer 50, the gating viaplug 40V, and thegating control line 41 may be disposed on the substrate Sub. The gate line GL may include agate insulating film 31, a ferroelectric film Ff, and a gate electrode G. - Referring to
FIG. 3E , in an embodiment of the present disclosure, thegate insulating film 31, the ferroelectric film Ff, and the gate electrode G may be a line shape crossing the active regions Act. Alternatively, in another embodiment of the present disclosure, thegate insulating film 31 and the ferroelectric film Ff may be a plate shape, and the gate electrode G may have a line shape. - Referring to
FIG. 3F , in an embodiment of the present disclosure, the ferroelectric film Ff and the gate electrode G may be line shapes extending in directions perpendicular to each other. Alternatively, in another embodiment of the present disclosure, the ferroelectric film Ff may be independently disposed so as to be separated from each other, such as in an island type. In an embodiment of the present disclosure, thegate insulating film 31 may be a plate shape, and the ferroelectric film Ff and the gate electrode G may be line shapes. That is, the ferroelectric film Ff and the gate electrode G may extend in directions perpendicular to each other. -
FIG. 4A is a diagram schematically illustrating a synapse array in accordance with an embodiment of the present disclosure, andFIG. 4B is a schematic diagram illustrating a synapse in accordance with the embodiment of the present disclosure. Referring toFIGS. 4A and 4B , asynapse array 100 d in accordance with an embodiment of the present disclosure may include a plurality offirst input neurons 10 a, a plurality ofsecond input neurons 10 b, a plurality ofoutput neurons 20, a plurality ofgating controllers 40, and a plurality ofsynapses 30 d. - Each of the
synapses 30 d may include a plurality of pairs of synapse transistors TP1 to TPn connected in parallel. The pairs of synapse transistors TP1 to TPn may include first transistors T1 a to Tna and second transistors T1 b to Tnb connected in series, respectively. Referring toFIG. 4B , in asynapse 30 d, drain electrodes D1 a to Dna of the first transistors T1 a to Tna may be connected to a commonfirst input neuron 10 a, source electrodes S1 b to Snb of the second transistors T1 b to Tnb may be connected to a commonsecond input neuron 10 b, and source electrodes S1 a to Sna of the first transistors T1 a to Tna and the drain electrodes D1 b to Dnb of the second transistors T1 b to Tnb may be directly electrically connected to each other and connected to acommon output neuron 20, respectively. - The gate electrodes G1 a to Gna of the first transistors T1 a to Tna and the gate electrodes G1 b to Gnb of the second transistors T1 b to Tnb may be directly electrically connected to a plurality of
gating controllers 40, respectively. - The
first input neuron 10 a may provide a first reference voltage to the drain electrodes D1 a to Dna of the first transistors T1 a to Tna of thesynapse 30 d. For example, thefirst input neuron 10 a may be coupled to the power supply voltage node, and the first reference voltage may be the power supply voltage. - The
second input neuron 10 b may provide the second reference voltage to the source electrodes S1 b to Snb of the second transistors T1 b to Tnb of thesynapse 30 d. For example, thesecond input neuron 10 b may be coupled to a ground voltage node, and the second reference voltage may be a ground voltage. The first reference voltage may be higher or greater than the second reference voltage. -
FIG. 5 is a diagram schematically illustrating a pattern recognition system in accordance with an embodiment of the present disclosure. For example, apattern recognition system 900 may include a speech recognition system, an imaging recognition system, a code recognition system, a signal recognition system, and one or more systems for recognizing various patterns. - Referring to
FIG. 5 , thepattern recognition system 900 in accordance with the embodiment of the present disclosure may include a Central Processing Unit (CPU) 910, amemory unit 920, acommunication control unit 930, anetwork 940, anoutput unit 950, aninput unit 960, an Analog-Digital Converter (ADC) 970, aneuromorphic unit 980, and/or abus 990. TheCPU 910 may generate and transmit various signals for a learning process of theneuromorphic unit 980, and perform various processes and functions for recognizing patterns according to an output from theneuromorphic unit 980. For example, theCPU 910 may perform processes and functions for recognizing speech and imaging patterns based on an output from theneuromorphic unit 980. - The
CPU 910 may be connected with thememory unit 920, thecommunication control unit 930, theoutput unit 950, theADC 970, and theneuromorphic unit 980 through thebus 990. - The
memory unit 920 may store various pieces of information, which are required to be stored in thepattern recognition system 900. Thememory unit 920 may include one or more of a volatile memory device, such as DRAM or SRAM, a nonvolatile memory, such as PRAM, MRAM, ReRAM or NAND flash memory, and various memory units, such as Hard Disk Drive (HDD) and Solid State Drive (SSD). - The
communication control unit 930 may transmit and/or receive data to and/or from a communication control unit of another system through thenetwork 940. For example, thecommunication control unit 930 may transmit speech and/or image recognition data through thenetwork 940. - The
output unit 950 may output data in various manners. For example, theoutput unit 950 may include a speaker, a printer, a monitor, a display panel, a beam projector, a hologrammer, or other various output devices. Theoutput unit 950 may output, for example, speech and/or image recognition data. - The
input unit 960 may include any of a microphone, a camera, a scanner, a touch pad, a keyboard, a mouse, a mouse pen, and one or more of various sensors. - The
ADC 970 may convert analog data inputted from theinput unit 960 into digital data. - The
neuromorphic unit 980 may perform learning or recognition using the data outputted from theADC 970, and output data corresponding to recognized patterns. Theneuromorphic unit 980 may include one or more of the neuromorphic devices in accordance with the various embodiments described above. - In accordance with embodiments of the present disclosure, the synapse arrays of a neuromorphic device including ferroelectric field effect transistors can have a higher degree of integration, a lower power consumption, and multiple resistance levels.
- Although various embodiments have been described for illustrative purposes, it will be apparent to those skilled in the art that various changes and modifications may be made without departing from the spirit and scope of the disclosure, as defined in the following claims.
Claims (10)
1. A synapse array of a neuromorphic device comprising:
a first input neuron;
a second input neuron;
an output neuron; and
a synapse,
wherein the synapse comprises a plurality of pairs of ferroelectric field effect transistors electrically connected to each other in parallel,
wherein each of the plurality of pairs of ferroelectric field effect transistors comprises a first ferroelectric field effect transistor and a second ferroelectric field effect transistor, and
wherein the first ferroelectric field effect transistor and the second ferroelectric field effect transistor are electrically connected to each other in series.
2. The synapse array of the neuromorphic device of claim 1 ,
wherein the first ferroelectric field effect transistor includes a drain electrode electrically connected to the first input neuron.
3. The synapse array of the neuromorphic device of claim 2 ,
wherein the first ferroelectric field effect transistor includes a source electrode electrically connected to the output neuron.
4. The synapse array of the neuromorphic device of claim 3 ,
wherein the first ferroelectric field effect transistor includes a body electrically connected to the source electrode of the first ferroelectric field effect transistor.
5. The synapse array of the neuromorphic device of claim 1 ,
wherein the second ferroelectric field effect transistor includes a source electrode electrically connected to the second input neuron.
6. The synapse array of the neuromorphic device of claim 5 ,
wherein the second ferroelectric field effect transistor includes a drain electrode electrically connected to the output neuron.
7. The synapse array of the neuromorphic device of claim 6 ,
wherein the second ferroelectric field effect transistor includes a body electrically connected to the drain electrode of the second ferroelectric field effect transistor.
8. The synapse array of the neuromorphic device of claim 1 ,
wherein the first ferroelectric field effect transistor includes a first gate electrode and the second field effect transistor includes a second gate electrode, and
wherein the first gate electrode and the second gate electrode are electrically connected to a common gating controller.
9. The synapse array of the neuromorphic device of claim 1 ,
wherein the first ferroelectric field effect transistor comprises an n-type ferroelectric field effect transistor, and
wherein the second ferroelectric field effect transistor comprises a p-type ferroelectric field effect transistor.
10. The synapse array of the neuromorphic device of claim 1 ,
wherein the first input neuron provides a first reference voltage to the synapse, and the second input neuron provides a second reference voltage to the synapse, and
wherein the first reference voltage is greater than the second reference voltage.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US17/519,163 US20220058470A1 (en) | 2017-06-05 | 2021-11-04 | Synapse array of a neuromorphic device including a synapse array having a plurality of ferroelectricity field effect transistors |
Applications Claiming Priority (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
KR10-2017-0069563 | 2017-06-05 | ||
KR1020170069563A KR20180133073A (en) | 2017-06-05 | 2017-06-05 | Synapse Array of a Neuromorphic Device Including a Synapse Array Having a Plurality of Ferro-electricity Field Effect Transistors |
US15/850,145 US11195087B2 (en) | 2017-06-05 | 2017-12-21 | Synapse array of a neuromorphic device including a synapse array having a plurality of ferroelectricity field effect transistors |
US17/519,163 US20220058470A1 (en) | 2017-06-05 | 2021-11-04 | Synapse array of a neuromorphic device including a synapse array having a plurality of ferroelectricity field effect transistors |
Related Parent Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US15/850,145 Division US11195087B2 (en) | 2017-06-05 | 2017-12-21 | Synapse array of a neuromorphic device including a synapse array having a plurality of ferroelectricity field effect transistors |
Publications (1)
Publication Number | Publication Date |
---|---|
US20220058470A1 true US20220058470A1 (en) | 2022-02-24 |
Family
ID=64460505
Family Applications (2)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US15/850,145 Active 2040-10-07 US11195087B2 (en) | 2017-06-05 | 2017-12-21 | Synapse array of a neuromorphic device including a synapse array having a plurality of ferroelectricity field effect transistors |
US17/519,163 Pending US20220058470A1 (en) | 2017-06-05 | 2021-11-04 | Synapse array of a neuromorphic device including a synapse array having a plurality of ferroelectricity field effect transistors |
Family Applications Before (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US15/850,145 Active 2040-10-07 US11195087B2 (en) | 2017-06-05 | 2017-12-21 | Synapse array of a neuromorphic device including a synapse array having a plurality of ferroelectricity field effect transistors |
Country Status (3)
Country | Link |
---|---|
US (2) | US11195087B2 (en) |
KR (1) | KR20180133073A (en) |
CN (1) | CN108987409B (en) |
Families Citing this family (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US12069870B2 (en) * | 2018-06-11 | 2024-08-20 | Cyberswarm, Inc. | Synapse array |
EP3654250B1 (en) * | 2018-11-13 | 2023-04-12 | IMEC vzw | Machine learning accelerator |
CN111493817B (en) * | 2019-01-31 | 2023-10-10 | 周冠谦 | Ductile flexible sensing device |
CN110232440B (en) * | 2019-06-11 | 2021-06-04 | 北京大学 | Pulse neuron circuit based on ferroelectric transistor |
US20200394502A1 (en) * | 2019-06-14 | 2020-12-17 | Macronix International Co., Ltd. | Neuromorphic computing device |
TWI717178B (en) * | 2019-12-30 | 2021-01-21 | 財團法人工業技術研究院 | A microelectromechanical apparatus having hermitic chamber |
US11437092B2 (en) * | 2020-05-27 | 2022-09-06 | Taiwan Semiconductor Manufacturing Company, Ltd. | Systems and methods to store multi-level data |
TWI784515B (en) | 2020-05-27 | 2022-11-21 | 台灣積體電路製造股份有限公司 | Memory system and method of operating memory system |
CN111834371B (en) * | 2020-07-30 | 2022-08-02 | 中国电子科技集团公司第五十八研究所 | Floating gate type FLASH synapse device structure and preparation method thereof |
CN112434802B (en) | 2020-11-06 | 2022-05-20 | 北京大学 | Method for realizing self-adaptive random pulse neuron based on ferroelectric transistor |
KR102619356B1 (en) * | 2021-10-25 | 2023-12-28 | 서울대학교산학협력단 | capacitor device for synapse, synapse and synaptic array based on capacitor |
Family Cites Families (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US4874963A (en) * | 1988-02-11 | 1989-10-17 | Bell Communications Research, Inc. | Neuromorphic learning networks |
US5371835A (en) * | 1990-02-02 | 1994-12-06 | Kabushikikaisha Wacom | Inductively coupled neural network |
JPH0731705B2 (en) * | 1992-08-24 | 1995-04-10 | 東京工業大学長 | Self-learning multiply-accumulate operation circuit element and circuit |
US5350953A (en) * | 1993-05-28 | 1994-09-27 | Swenson Richard M | Digitally weighted neuron for artificial neural network |
JPH09245110A (en) * | 1996-03-13 | 1997-09-19 | Tadahiro Omi | Feedback circuit |
US6847071B2 (en) | 2001-06-06 | 2005-01-25 | Matsushita Electric Industrial Co., Ltd. | Semiconductor device |
JP3833174B2 (en) * | 2001-06-22 | 2006-10-11 | 松下電器産業株式会社 | Semiconductor device and driving method thereof |
JP2013037737A (en) * | 2011-08-05 | 2013-02-21 | Tokai Univ | Neuron cmos circuit and electronic circuit provided with neuron cmos circuit |
EP2811453B1 (en) * | 2012-01-27 | 2019-05-01 | Korea Advanced Institute Of Science And Technology | Visual cortical circuit apparatus |
JP5858020B2 (en) * | 2013-10-03 | 2016-02-10 | 株式会社デンソー | Group information memory recognition device |
US9734880B1 (en) * | 2016-04-01 | 2017-08-15 | Intel Corporation | Apparatuses, methods, and systems for stochastic memory circuits using magnetic tunnel junctions |
-
2017
- 2017-06-05 KR KR1020170069563A patent/KR20180133073A/en unknown
- 2017-12-21 US US15/850,145 patent/US11195087B2/en active Active
-
2018
- 2018-03-29 CN CN201810270320.7A patent/CN108987409B/en active Active
-
2021
- 2021-11-04 US US17/519,163 patent/US20220058470A1/en active Pending
Also Published As
Publication number | Publication date |
---|---|
KR20180133073A (en) | 2018-12-13 |
CN108987409B (en) | 2023-10-10 |
US11195087B2 (en) | 2021-12-07 |
CN108987409A (en) | 2018-12-11 |
US20180349762A1 (en) | 2018-12-06 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US20220058470A1 (en) | Synapse array of a neuromorphic device including a synapse array having a plurality of ferroelectricity field effect transistors | |
US11341403B2 (en) | Synapse system of a neuromorphic device including a ferroelectric transistor | |
US11205117B2 (en) | Neuromorphic device having a plurality of synapses blocks | |
CN109214510B (en) | Nerve morphology multi-bit digital weight unit | |
US11157803B2 (en) | Neuromorphic device including a synapse having a variable resistor and a transistor connected in parallel with each other | |
US10833126B2 (en) | Three dimensional stacked semiconductor memory device | |
US11727977B2 (en) | Non-volatile analog resistive memory cells implementing ferroelectric select transistors | |
KR20160056816A (en) | Neuromorphic devices with excitatory and inhibitory functionality | |
US11443172B2 (en) | Synapse array of neuromorphic device including synapses having ferro-electric field effect transistors and operation method of the same | |
US10651316B2 (en) | Synaptic semiconductor device and neural networks using the same | |
US11586882B2 (en) | Synapse memory | |
KR20210151737A (en) | Synapse string and synapse string array for neural networks | |
Strukov et al. | Memory technologies for neural networks | |
US20180240846A1 (en) | Neuromorphic device including a synapse having carbon nano-tubes | |
KR101954254B1 (en) | Reconfigurable devices, device array for neuromorphic | |
US20230125501A1 (en) | Capacitor device for unit synapse, unit synapse and synapse array based on capacitor | |
US20180287056A1 (en) | Synapse and synapse array | |
KR20170080473A (en) | Neuromorphic device | |
US20180300612A1 (en) | Neuromorphic device and a synapse network including a post-synaptic neuron having a subtracting circuit | |
US20240069869A1 (en) | Multiply-accumulate operation device and neural network | |
US10565495B2 (en) | Synapse and neuromorphic device including the same | |
KR20200063555A (en) | Multilevel weighting device and neural network using the same |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
STPP | Information on status: patent application and granting procedure in general |
Free format text: DOCKETED NEW CASE - READY FOR EXAMINATION |