CN106845634B - A kind of neuron circuit based on memory resistor - Google Patents
A kind of neuron circuit based on memory resistor Download PDFInfo
- Publication number
- CN106845634B CN106845634B CN201611235356.9A CN201611235356A CN106845634B CN 106845634 B CN106845634 B CN 106845634B CN 201611235356 A CN201611235356 A CN 201611235356A CN 106845634 B CN106845634 B CN 106845634B
- Authority
- CN
- China
- Prior art keywords
- cynapse
- memristor
- neuron
- resistance
- neuron circuit
- 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.)
- Active
Links
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
Abstract
The invention discloses a kind of neuron circuits based on memory resistor, in the present invention, the memristor of cynapse array selects part volatibility bipolarity electric resistance changing device, and the memristor for expressing membrane potential of neurons selects volatibility electric resistance changing device, neuron circuit is constructed, and there is cynapse basic unit.The neuron circuit can be realized the integration discharging function in biological neuron, give expression to local hierarchical current potential, cynapse has part volatibility, can express the relevant plasticity of activity schedule, has great similarity in terms of information is stored, transmitted with processing with biologically neuron and cynapse.The present invention can provide basic unit for hardware simulation cerebral nerve network structure, overcome Neural spike train time delay of the existing technology, it is difficult to realize the technical problems such as High Density Integration, the information processing system of class brain can be used to construct, can quickly handle bulk information parallel and calculate in network in the neurology for realizing brain to have very big application value.
Description
Technical field
The invention belongs to semiconductor message areas, and in particular to a kind of neuron circuit based on memory resistor.The circuit
For a kind of neuron for artificial neural network and cynapse basic unit, deposited with biologically neuron and cynapse in information
Storage, transmitting and processing aspect have great similarity, can be used to construct the neural computing network of class brain.
Background technique
The brain of people cognitive function, language understanding, in terms of be better than contemporary most computers, have simultaneously
There is the features such as small in size, low in energy consumption, high-efficient, fault tolerant concurrent operation.Traditional computer is based on von Karman structure
, information processing is performed separately with storage, and concurrent operation is indifferent.It is different from computer, in big intracerebral, the processing of information with
Storage carries out in the same time and place.Human brain is by about 1011A neuron passes through about 1015A cynapse is connected with each other, and forms one
Huge neural network can quickly handle bulk information parallel.
Neuron plays key effect in brain information treatment process, and the major function of neuron is to handle and transmit
Information, and complete the cell membrane that this function relies primarily on neuronal cell.In the brain, neuron receives from the emerging of dendron
Putting forth energy property or inhibitory synapse current potential generate the local hierarchical current potential with certain timeliness, and are integrated.Lipid bilayer
The current potential of cell membrane then can accordingly change, and when reaching certain value, neuron can then generate action potential, issue signal, and
Signal is transmitted to next neuron via cynapse by aixs cylinder.Integration electric discharge is most basic one of the function of neuron.Mind
Key effect is also played through cynapse (connecting portion of two neurons), plasticity, i.e., the form and function of cynapse are by the external world
The influence of stimulation and the characteristic to change, are the physiological Foundations of brain learning Yu memory, information processing and storage.Therefore,
The key that artificial neural network is constructed from hardware is to develop the artificial neuron with class nervous function, and have the function of cynapse
Energy.
The neuron realized using traditional cmos circuit needs to use complicated transistor and capacitor, and is difficult to and high
Density cynapse array is integrated, also, traditional silicon-based transistor neuron is mainly used for Digital Logical Circuits, pulse nerve at present
Function is simultaneously not implemented.And the special circuit for simulating a synaptic function just needs tens triodes, and the neural network of human brain
In there are about 1015A cynapse.Therefore, established on hardware based on traditional cmos circuit is with the comparable huge neural network of human brain
It is unpractical.
In recent years about memristor studies have shown that there is the memristor of simple sandwich structure, in the work of electric pulse
Conductive continuous enhancing may be implemented under and reduce, be used for imictron and cynapse basic function, thus receive extensively
Ground concern.Firstly, the plasticity of electric conductivity and biology cynapse that memristor gradually changes has great similarity.Memory resistor
Electric conductivity enhances the enhancing that can simulate biology synaptic connection strengths；The reduction of memory resistor electric conductivity can simulate Synaptic junction
The inhibition of intensity.And neuron handle and transmit electric signal when, cell membrane potential is also continuously to accumulate, i.e. integration process.
Thus memristor can be used as the variation of key element simulation film potential, in conjunction with other elements, realize the integration electric discharge of neuron
Function.
The Resistance states of memristor currently used for analog neuron cynapse be it is completely non-volatile, i.e., its electric conductivity is outside
Added electric field is maintained at certain numerical value after removing, do not change over time.However, the bonding strength of cynapse is in telecommunications in biology
Number effect after, can first enhance, gradually decay to a certain extent then as the time.This dynamic at any time of synaptic plasticity becomes
The process of change can be realized the time encoding to bursting activity, have to functions such as realization brain learning, memory, forgettings important
Meaning.
And the neuron circuit report based on memristor building is very few, is to be badly in need of overcoming the problems, such as.
Memristor structure is simple, is designed by cross spider, can High Density Integration.Thus, neuron is constructed by memristor
Circuit, and there is synaptic function, it is significant.Meanwhile the device for expressing synaptic plasticity uses part volatibility memristor, more
The nearly biology cynapse characteristic of adjunction, this neuron will have major application prospect in Artificial Intelligence Circuits.
Summary of the invention
The present invention proposes a kind of neuron circuit based on memory resistor building, to realize the plasticity of neuronal synapse
The simulation that flash-over characteristic is integrated with membrane potential of neurons overcomes Neural spike train time delay of the existing technology, it is difficult to real
The technical problems such as existing High Density Integration.
Neuron circuit includes cynapse array, dendron, pericaryon and aixs cylinder.Cynapse is used to receive and adjust upper level
The action potential that neuron circuit occurs, is transmitted to pericaryon through dendron, be transmitted to through aixs cylinder after integration electric discharge again under
Level-one neuron circuit.Dendron is realized by connecting wire；Aixs cylinder is realized by connecting wire；Cynapse part, by there is part easily
The memristor for the property lost is realized；Pericaryon, by the memristor of the expression membrane potential of neurons of complete volatibility or part volatibility
Device and other corresponding electronic device devices are realized.
Based on the above design, specific technical solution of the present invention is as follows:
A kind of neuron circuit based on memory resistor, including cynapse array, switching tube, adder, expression neuron membrane
Memristor, comparator, the Spike signal generator of current potential；Wherein:
The cynapse array is for receiving the action potential that upper level neuron circuit transmits comprising several are arranged side by side
Part volatibility memristor, (in the present invention, aixs cylinder is by leading for each aixs cylinder of each memristor one end and upper level neuron circuit
Line is realized) it is connected, other end tandem is all the way, to be connected through switching with adder input terminal；
The adder is used to integrate the action potential of each input terminal of cynapse array, for adjusting expression membrane potential of neurons
Memristor resistance, to realize simulation to membrane potential of neurons；
The memristor of the expression membrane potential of neurons is complete volatibility memristor, for imictron cell membrane
Local hierarchical current potential；One terminates the output end of the adder, and the other end is divided into two-way, connects divider resistance ground connection all the way；Separately
Comparator is connect all the way, is sent after action potential partial pressure to comparator input terminal after the integration that will acquire；
Another input of comparator terminates reference voltage VR, for comparing divider resistance voltage-to-ground and reference voltage VR
Size；When divider resistance voltage-to-ground is greater than VRWhen, conduction level is exported, cut-off level is otherwise exported；
The Spike signal generator input terminal is connected with comparator output terminal, and Spike signal generator exports three roads letter
Number, it is connected all the way with the control electrode of the switching tube, the movement for control switch pipe；Second road signal connects cynapse array
Tandem end, for adjusting the transmission efficiency of cynapse array；Third road signal connects next stage neuron circuit, as next stage nerve
First circuit input signal.Usually when comparator output cut-off level, Spike signal generator stops working, and switching tube is connected；
When comparator exports conduction level, Spike signal generator exports a cut-off signals, disconnects switching tube；
When work, it is added via the upper level neuron signal of cynapse array input by adder, it is real carries out electric signal
When integrate, after integration electric signal amplitude rise to threshold value (this value depend on it is used expression membrane potential of neurons memristor
Resistance switching performance, those skilled in the art know that electric resistance changing device has corresponding threshold value) when, make to express membrane potential of neurons
Memristor resistance value reduce so that divider resistance voltage-to-ground increase；When divider resistance voltage-to-ground is more than reference voltage VR
When, comparator exports conduction level, and the electric signal electric discharge of Spike signal generator simulation biology according to the pre-stored data is realized
The integration discharging function of neuron is completed in the sending of action potential；The reference voltage VRIt is according to expression membrane potential of neurons
The resistance value of memristor and the resistance value size of divider resistance and the amplitude of neuron action potential mutually weigh depending on be arranged
One constant pressure.
Further, each cynapse in the cynapse array is realized using part volatibility memristor.
Further, the operational amplifier that the adder uses.
Further, the memristor of the expression membrane potential of neurons selects part volatibility device or completely non-volatile
Device, Spike signal generator output end are connected with the output end of the memristor of expression membrane potential of neurons and adder, are used for
In neuron circuit electric discharge, the memristor resistance of resetting expression membrane potential of neurons makes it be restored to high-impedance state, realizes cell
The expression of film initial potential.
Further, adjusting of the discharge cell to cynapse Array transfer efficiency is according to neuron activity timing phase
The plasticity STDP principle of pass carries out each cynapse resistance respectively:
After discharge cell issues electric signal, discovery upper level neuron has also generated a movement electricity after a short time
Position, the cynapse resistance being attached thereto become larger, and transmission efficiency becomes smaller；After discharge cell issues electric signal, discovery upper level nerve
Member has also sent out an action potential before a bit of time, and the cynapse resistance being attached thereto becomes smaller, and transmission efficiency becomes larger.
Further, the reference voltage VRSize Criterion of Selecting is to ensure that when only a small number of input signal inputs, whole
Electric signal amplitude is less than reference voltage V after conjunctionR, and when compared with multiple input signals or higher synaptic efficacy, electric signal after integration
Amplitude is greater than reference voltage VR
In the present invention, the memristor of cynapse array selects part volatibility bipolarity electric resistance changing device, expresses neuron
The memristor of the memristor of film potential selects volatibility electric resistance changing device, and MOS transistor T selects p-type transistor or other pressures
Control switch, reverse phase summation operation device, phase inverter, resistance, comparator, Spike signal generator etc. be mature commercial device or
Equipment.Emulation completion is carried out by the neuron circuit to selected device and building, and there is cynapse basic unit.The neuron
Circuit can be realized the integration discharging function in biological neuron, give expression to local hierarchical current potential, and cynapse has part volatibility,
The relevant plasticity of activity schedule can be expressed.
The present invention can be realized the integration discharging function in biological neuron, give expression to local hierarchical current potential, to other minds
The electric signal come through member transmitting carries out space-time integration, meanwhile, the electric signal of the action potential of generation and the transmitting of other neurons can
To be adjusted by the relevant plasticity of activity schedule (Spike-timing-dependent plasticity, STDP) rule
The plasticity of cynapse realizes the time encoding to bursting activity.The cynapse has part volatibility, is more nearly biological true
Property.This neuron can provide basic unit for hardware simulation cerebral nerve network structure, overcome mind of the existing technology
Postpone through first discharge time, it is difficult to realize the technical problems such as High Density Integration.This neuron circuit is to construct class brain
Information processing system can quickly handle bulk information parallel, overcome traditional computer based on von Karman structure in image
Identification, the deficiency of the intelligence aspect such as self adaptive control, study, reasoning, decision.
Detailed description of the invention
Fig. 1 is neuron circuit proposed by the present invention；
Fig. 2 is the electrical property of memristor array of the present invention；
Fig. 3 is the electrical property of memristor 2 of the present invention；
Fig. 4 is a kind of embodiment of neuron circuit；
Fig. 5 is the adjusting of the three kinds of situations and the relevant plasticity of activity schedule of neuron circuit emulation；
Fig. 6 is the neuron circuit proposed by the present invention based on part volatibility bipolarity electric resistance changing device.
Specific embodiment
By example, property feature is described further for the essence of the present invention with reference to the accompanying drawing.It needs to illustrate herein
It is the explanation of these embodiments to be used to help to understand the present invention, but and do not constitute a limitation of the invention.
Embodiment:
In the present solution, cynapse array is memristor array, adder constitutes reverse phase by operational amplifier combination resistance and asks
And device, and then one phase inverter of connection will restore polarity of voltage, integrate the signal of input.The expression membrane potential of neurons
Memristor is memristor 2, is connected to inverter output.Neuron dendron, aixs cylinder are expressed with conducting wire in the present invention.Memristor 2
The other end divides two-way, meets divider resistance R all the wayc, another way connects comparator, and comparator output terminal connects Spike signal generator.Phase
Answering the selection of device will match with the resistance of memristor array and memristor 2, in the similar order of magnitude, be specifically shown in Fig. 4.
Cynapse array is memristor array, using Ni/Nb-SrTiO3/ Ti (nickel/niobium doping strontium titanates/titanium) device, property
It can be such as Fig. 2.The simulation result of resistance variation characteristic and resistance retentivity when here comprising device pulse stimulation.It is positive when applying
When scanning voltage, devices transition to low resistance state；When applying negative sense scanning voltage, devices transition to high-impedance state；It is continuous when applying
Direct impulse signal when, the resistance of device is gradually reduced；When continuous negative-going pulse signal can be applied, the resistance of device by
It is cumulative to add；Meanwhile the resistance state of device can slowly change with the time, be restored to one compared with high-impedance state by the spontaneous part of low-resistance.?
In magnetron sputtering apparatus, with Nb-SrTiO3Monocrystalline is matrix, plates Ti electrode in its bottom using magnetron sputtering, top plates
Ni electrode prepares the half volatibility Ni/Nb-SrTiO with class synaptic function3/ Ti memory resistor.
The memristor of the expression membrane potential of neurons is memristor 2, uses Pt/WO3/ Pt (platinum/tungsten oxide/platinum) is complete
Full volatibility bipolarity electric resistance changing device, performance such as Fig. 3.When applying forward scan voltage to 2V, device resistance can be dropped to
1M Ω, when further increasing forward scan voltage, device resistance can be reduced to smaller resistance value；When apply negative sense scanning voltage to-
When 1.5V, device resistance is converted to 100M Ω, and the holding of device low resistance state does not live, and spontaneous can be restored to high-impedance state.The memristor
Production are as follows: in magnetron sputtering apparatus, to be covered with the monocrystalline silicon piece of certain thickness oxide layer as matrix, with Ti to stick
Layer, Pt are hearth electrode and top electrode, WO3For functional layer, the Pt/WO with complete volatibility is prepared3/ Pt memory resistor.
By emulating to neuron circuit, integration and discharging function of the neuron to electric signal are realized.Fig. 4 is provided
A kind of embodiment of neuron circuit, the cynapse array being made of memristor array, adder, memristor 2 and electric discharge are single
Member etc. is constituted.Memristor array is indicated that memristor 2 is indicated by MEMRISTOR-2 by MEMRISTOR-1.Other are common electricity
Sub- device.
Neuron circuit in Fig. 4 works in this way: memristor array is as nerve synapse, Spike signal (these letters
Number from upper level neuron generate action potential) via 3 cynapses, passed through after entering reverse phase summation operation device by dendron
Phase inverter carries out real-time integration, and changes the resistance of memristor 2.Here, reverse phase sum operational amplifier is by the times magnification of signal
Number depends on resistance R2With the resistance value ratio of cynapse array, MOS transistor IRF450 is because of application -5V signal always, source electrode and drain electrode
Between be on state.When 2 resistance of memristor reduces to a certain extent, integrated signal is in resistance R6Partial pressure it is higher when, promote
Spike signal generator issues the action potential as upper level neuron circuit below, point three road signals, all the way as whole
It closes the action potential that electric discharge issues and is transmitted to next stage neuron circuit, control transistor switch IRF450 all the way, make its source electrode
It is disconnected with drain electrode, another way feeds back the cynapse at dendron, and interacts with input signal, adjusts the transmission efficiency W of cynapsein
(input signal of cynapse and the action potential of neuron generation are respectively acting on memristor array both ends, according to 1 gust of memristor
Each memristor electric resistance changing mechanism in column, changes the resistance value of device, and the application time difference of the two signals will affect the resistance of device
Value variation reflects the relevant plasticity of activity schedule, i.e. STDP well).Memristor 2 is complete volatibility second-order memristor,
I.e. the device changes low resistance state after electric pulse effect, and low resistance state holding does not live, and spontaneous with the time can be restored to originally
High-impedance state, such as local hierarchical current potential in neuron.Here realize that the device of comparator effect is contained in the generation of Spike signal
Device, such as Fig. 4 dotted box portion.As resistance R6Partial pressure it is higher when, Spike signal generator can emit action potential, otherwise not
Row.
Be presented in Fig. 5 three kinds of neuron circuits to upper level neuron circuit action potential through cynapse array by adder
Be added the example integrated in real time, wherein PLUS indicates that anode, MINUS indicate negative terminal in figure: setting, which is worked as, receives the prominent of signal
Touch it is more, such as 3, then neuronal cell membrane voltage can be more than threshold value, occur electric discharge (Fig. 5 a, wherein circuit indicates 3 cynapses
The artificial circuit of signal input, in the electric discharge figure of lower section three, electric signal size after first expression is integrated, second expression memristor
2 change in resistance situation of device, third indicate divider resistance (RC) voltage-to-ground situation of change)；When the cynapse for receiving signal is less,
Such as 2, then neuronal cell membrane voltage does not exceed threshold value, and electric discharge does not occur, and (Fig. 5 b, wherein circuit indicates 2 cynapses
The artificial circuit of signal input, in the electric discharge figure of lower section three, electric signal size after first expression is integrated, second expression memristor
2 change in resistance situation of device, third indicate divider resistance (RC) voltage-to-ground situation of change)；When the cynapse for receiving signal is less,
Such as 2, but synaptic connection strengths or transmission efficiency are higher, then neuronal cell membrane voltage can be more than threshold value, and electric discharge (figure occurs
5c, wherein circuit indicates the artificial circuit of 2 cynapse signals input, in the electric discharge figure of lower section three, after first expression is integrated
Electric signal size, second 2 change in resistance situation of expression memristor, third indicate divider resistance (RC) voltage-to-ground variation feelings
Condition).To the influence of the bonding strength of cynapse after the neuron generation action potential that Fig. 5 d is then provided, i.e., activity schedule is relevant can
Plasticity.Three electric pulses of Fig. 5 d left figure are respectively to indicate upper level neuron circuit action potential, neuron circuit electric discharge
Action potential, the voltage at the two superimposed actually applied memristor both ends in cynapse array because of having time difference, Fig. 5 d
Right figure be upper level neuron circuit action potential and the neuron circuit electric discharge action potential it is poor in different times when pair
The influence of the transmission efficiency of memristor in cynapse array.
Fig. 6 is the neuron circuit proposed by the present invention based on part volatibility bipolarity electric resistance changing device.Not with Fig. 1
Together, the memristor 2 of the circuit is turned using part volatibility bipolarity electric resistance changing device or complete non-volatile bipolarity resistance
When becoming device, since device resistance spontaneous cannot be restored to high-impedance state, so in integrated signal VoutIncrease between memristor 2
Reset signal completes the nerve of the function of neuron when neuron circuit electric discharge for device resistance to be reset to high-impedance state
First circuit.
The present embodiment combines practical memory resistor resistance switching performance, by being emulated to the neuron circuit of design,
The each essential characteristic for demonstrating neuron circuit design includes the number of cynapse, neuronal transmission efficiency to neuronal integration
Influence of electric discharge etc. is successfully realized.Illustrate that neuron circuit design is reasonably that being can be by actual physics device
What part was realized.
The above is presently preferred embodiments of the present invention, but the present invention should not be limited to the embodiment and attached drawing institute
Disclosure.So all do not depart from the lower equivalent or modification completed of spirit disclosed in this invention, guarantor of the present invention is both fallen within
The range of shield.
Claims (6)
1. a kind of neuron circuit based on memory resistor, which is characterized in that including cynapse array, switching tube (T), adder
(OP1), memristor, the comparator (OP2), Spike signal generator of membrane potential of neurons are expressed；Wherein:
The cynapse array is for receiving the action potential that upper level neuron circuit transmits comprising several parts arranged side by side
Volatibility memristor, each memristor one end are connected with each aixs cylinder of upper level neuron circuit, and other end tandem is warp all the way
Switch (T) is connected with adder (OP1) input terminal；
The adder (OP1) is used to integrate the action potential of each input terminal of cynapse array, for adjusting expression neuron membrane electricity
The resistance of the memristor of position, to realize the simulation to membrane potential of neurons；
The memristor of the expression membrane potential of neurons is complete volatibility memristor, the part for imictron cell membrane
Graded potential；One terminates the output end of the adder (OP1), and the other end is divided into two-way, meets divider resistance (R all the wayC) connect
Ground；Another way meets comparator (OP2), send after action potential partial pressure to comparator (OP2) input terminal after the integration that will acquire；
Another input of comparator (OP2) terminates reference voltage VR, for comparing divider resistance (RC) voltage-to-ground and reference electricity
Press VRSize；As divider resistance (RC) voltage-to-ground be greater than VRWhen, conduction level is exported, cut-off level is otherwise exported；
The Spike signal generator input terminal is connected with comparator (OP2) output end, and Spike signal generator exports three tunnels
Signal is connected with the control electrode of the switching tube (T) all the way, is used for the movement of control switch pipe (T)；Second road signal connects cynapse
The tandem end of array, for adjusting the transmission efficiency of cynapse array；Third road signal connects next stage neuron circuit, as next
Grade neuron circuit input signal；Usually when comparator (OP2) output cut-off level, Spike signal generator stops working, and makes
Switching tube (T) conducting；When comparator (OP2) exports conduction level, Spike signal generator exports a cut-off signals, makes
Switching tube (T) disconnects.
2. the neuron circuit according to claim 1 based on memory resistor, which is characterized in that when work, via cynapse
The upper level neuron signal of array input is added by adder (OP1), carries out electric signal real-time integration, telecommunications after integration
When number amplitude rises to threshold value, reduce the resistance value for expressing the memristor of membrane potential of neurons, so that divider resistance (RC) over the ground
Voltage increases；As divider resistance (RC) voltage-to-ground is more than reference voltage VRWhen, comparator (OP2) exports conduction level, notice
The electric signal electric discharge of Spike signal generator simulation biology according to the pre-stored data, realizes the sending of action potential, completes nerve
The integration discharging function of member；The reference voltage VRIt is the resistance value and divider resistance according to the memristor of expression membrane potential of neurons
(RC) resistance value size and neuron action potential amplitude mutually weigh depending on a constant pressure being arranged.
3. the neuron circuit according to claim 1 or 2 based on memory resistor, which is characterized in that the cynapse array
In each cynapse, using part volatibility memristor realize.
4. the neuron circuit according to claim 1 or 2 based on memory resistor, which is characterized in that the adder
(OP1) operational amplifier used.
5. the neuron circuit according to claim 1 or 2 based on memory resistor, which is characterized in that Spike signal occurs
Output end all the way in device is connected with the output end of the memristor of expression membrane potential of neurons and adder (OP1), in mind
When through first circuit discharging, the memristor resistance of resetting expression membrane potential of neurons makes it be restored to high-impedance state, at the beginning of realizing cell membrane
The expression of beginning current potential.
6. the neuron circuit according to claim 1 or 2 based on memory resistor, which is characterized in that the Spike signal
Adjusting of the generator to cynapse Array transfer efficiency is according to the relevant plasticity STDP principle of neuron activity timing to each
Cynapse resistance carries out respectively: after Spike signal generator issues electric signal, discovery upper level neuron is after a short time
Also an action potential has been sent out, the cynapse resistance being attached thereto becomes larger, and transmission efficiency becomes smaller；When Spike signal generator issues
After electric signal, discovery upper level neuron has also sent out an action potential before a bit of time, the cynapse resistance being attached thereto
Become smaller, transmission efficiency becomes larger.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611235356.9A CN106845634B (en) | 2016-12-28 | 2016-12-28 | A kind of neuron circuit based on memory resistor |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611235356.9A CN106845634B (en) | 2016-12-28 | 2016-12-28 | A kind of neuron circuit based on memory resistor |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106845634A CN106845634A (en) | 2017-06-13 |
| CN106845634B true CN106845634B (en) | 2018-12-14 |
Family
ID=59113084
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611235356.9A Active CN106845634B (en) | 2016-12-28 | 2016-12-28 | A kind of neuron circuit based on memory resistor |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106845634B (en) |
Families Citing this family (33)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109146073B (en) * | 2017-06-16 | 2022-05-24 | 华为技术有限公司 | Neural network training method and device |
| CN107742153B (en) * | 2017-10-20 | 2020-02-21 | 华中科技大学 | Memristor-based neuron circuit with steady-state plasticity |
| CN107909146B (en) * | 2017-11-13 | 2021-09-17 | 中国科学院微电子研究所 | Neuron circuit based on volatile threshold transition device |
| US11636316B2 (en) * | 2018-01-31 | 2023-04-25 | Cerfe Labs, Inc. | Correlated electron switch elements for brain-based computing |
| CN108664735B (en) * | 2018-05-11 | 2020-06-09 | 华中科技大学 | STDP pulse design method based on multivalued memristor and realization method of diversified STDP |
| CN110751273A (en) * | 2018-07-22 | 2020-02-04 | 徐志强 | Neuron and synapse simulation assembly |
| CN109165731B (en) * | 2018-08-09 | 2020-06-30 | 清华大学 | Electronic neural network and parameter setting method thereof |
| CN109255437B (en) * | 2018-08-17 | 2019-06-14 | 郑州轻工业学院 | A kind of memristor nerve network circuit of flexibly configurable |
| CN109447250B (en) * | 2018-09-14 | 2020-07-10 | 华中科技大学 | Artificial neuron based on battery effect in memristor |
| CN109449289B (en) * | 2018-11-01 | 2022-12-09 | 中国科学院宁波材料技术与工程研究所 | Light-excited nerve synapse bionic memristor and preparation method thereof |
| CN109977470B (en) * | 2019-02-20 | 2020-10-30 | 华中科技大学 | Circuit for sparse coding of memristive Hopfield neural network and operation method thereof |
| CN109978019B (en) * | 2019-03-07 | 2023-05-23 | 东北师范大学 | Image mode recognition analog and digital mixed memristor equipment and preparation thereof, and STDP learning rule and image mode recognition method are realized |
| CN110059816B (en) * | 2019-04-09 | 2022-08-16 | 南京邮电大学 | Memristor-based neural network unit circuit |
| CN110111234B (en) * | 2019-04-11 | 2023-12-15 | 上海集成电路研发中心有限公司 | Image processing system architecture based on neural network |
| WO2020219149A1 (en) * | 2019-04-25 | 2020-10-29 | Hrl Laboratories, Llc | Active memristor based spiking neuromorphic circuit for motion detection |
| CN110163364B (en) * | 2019-04-28 | 2022-08-30 | 南京邮电大学 | Neural network unit circuit based on memristor bridge synapse |
| CN110163365A (en) * | 2019-05-29 | 2019-08-23 | 北京科易达知识产权服务有限公司 | A kind of spiking neuron circuit applied to memristor synapse array |
| CN110309908B (en) * | 2019-06-11 | 2021-03-23 | 北京大学 | FeFET-CMOS mixed pulse neuron circuit based on ferroelectric transistor |
| CN110443356B (en) * | 2019-08-07 | 2022-03-25 | 南京邮电大学 | Current type neural network based on multi-resistance state memristor |
| CN110837253A (en) * | 2019-10-31 | 2020-02-25 | 华中科技大学 | Intelligent addressing system based on memristor synapse |
| CN110991610B (en) * | 2019-11-28 | 2022-08-05 | 华中科技大学 | Probability determination method for nondeterministic problem |
| CN111129297B (en) * | 2019-12-30 | 2022-04-19 | 北京大学 | Method and system for realizing diversity STDP of memristive synapse device |
| CN111275177B (en) * | 2020-01-16 | 2022-10-21 | 北京大学 | Full memristor neural network and preparation method and application thereof |
| CN111401540B (en) * | 2020-03-09 | 2023-04-07 | 北京航空航天大学 | Neuron model construction method and neuron device |
| JP6899024B1 (en) * | 2020-06-11 | 2021-07-07 | ウィンボンド エレクトロニクス コーポレーション | Resistance change type synapse array device |
| CN111958599A (en) * | 2020-08-17 | 2020-11-20 | 湖南大学 | Self-repairing control system based on astrocytes and intelligent robot arm |
| CN111967589B (en) * | 2020-08-21 | 2023-12-26 | 清华大学 | Neuron simulation circuit, driving method thereof and neural network device |
| CN112053726B (en) * | 2020-09-09 | 2022-04-12 | 哈尔滨工业大学 | Flash memory mistaken erasure data recovery method based on Er-state threshold voltage distribution |
| CN112598124B (en) * | 2020-12-28 | 2022-12-20 | 清华大学 | Neuron analog circuit and neural network device |
| CN112906880B (en) * | 2021-04-08 | 2022-04-26 | 华中科技大学 | Adaptive neuron circuit based on memristor |
| CN113191492B (en) * | 2021-04-14 | 2022-09-27 | 华中科技大学 | Synapse training device |
| CN115688897B (en) * | 2023-01-03 | 2023-03-31 | 浙江大学杭州国际科创中心 | Low-power-consumption compact Relu activation function neuron circuit |
| CN116663632B (en) * | 2023-08-02 | 2023-10-10 | 华中科技大学 | Intelligent sensing system integrating sensing, storage and calculation |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2012006471A1 (en) * | 2010-07-07 | 2012-01-12 | Qualcomm Incorporated | Methods and systems for memristor-based neuron circuits |
| WO2013044143A1 (en) * | 2011-09-21 | 2013-03-28 | Qualcomm Incorporated | Method and apparatus for structural delay plasticity in spiking neural networks |
| CN103941581A (en) * | 2014-04-17 | 2014-07-23 | 广西大学 | Single-neuron PID controller based on memory resistors |
| CN105701541A (en) * | 2016-01-13 | 2016-06-22 | 哈尔滨工业大学深圳研究生院 | Circuit structure based on memristor pulse nerve network |
-
2016
- 2016-12-28 CN CN201611235356.9A patent/CN106845634B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2012006471A1 (en) * | 2010-07-07 | 2012-01-12 | Qualcomm Incorporated | Methods and systems for memristor-based neuron circuits |
| WO2013044143A1 (en) * | 2011-09-21 | 2013-03-28 | Qualcomm Incorporated | Method and apparatus for structural delay plasticity in spiking neural networks |
| CN103941581A (en) * | 2014-04-17 | 2014-07-23 | 广西大学 | Single-neuron PID controller based on memory resistors |
| CN105701541A (en) * | 2016-01-13 | 2016-06-22 | 哈尔滨工业大学深圳研究生院 | Circuit structure based on memristor pulse nerve network |
Non-Patent Citations (3)
| Title |
|---|
| Neuromorphic Hardware System for Visual;Myonglae Chu;《IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS》;20150430;第62卷(第4期);第2410-2419页 * |
| 改进型细胞神经网络实现的忆阻器混沌电路;李志军;《物理学报》;20141231;第63卷(第1期);第1-9页 * |
| 新型忆阻细胞神经网络的建模及电路仿真;张小红;《系统仿真学报》;20160831;第28卷(第8期);第1715-1724页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106845634A (en) | 2017-06-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106845634B (en) | A kind of neuron circuit based on memory resistor | |
| TWI509537B (en) | Electronic learning synapse with spike-timing dependent plasticity using memory-switching elements | |
| Zhang et al. | An artificial neuron based on a threshold switching memristor | |
| US10650308B2 (en) | Electronic neuromorphic system, synaptic circuit with resistive switching memory and method of performing spike-timing dependent plasticity | |
| CN109447250B (en) | Artificial neuron based on battery effect in memristor | |
| Wu et al. | Homogeneous spiking neuromorphic system for real-world pattern recognition | |
| CN103778468A (en) | RRAM-based new type neural network circuit | |
| CN101997538B (en) | Pulse coupling based silicon-nanowire complementary metal oxide semiconductors (CMOS) neuronal circuit | |
| US10740672B2 (en) | Capacitative artificial neural networks | |
| CN103078054A (en) | Unit, device and method for simulating biological neuron and neuronal synapsis | |
| CN103580668A (en) | Associative memory circuit based on memory resistor | |
| KR20150034900A (en) | Synapse circuit for connecting neuron circuits, unit cell composing neuromorphic circuit, and neuromorphic circuit | |
| CN109787592B (en) | Random nerve pulse generator | |
| CN103078055A (en) | Unit, device and method for simulating biological neuronal synapsis | |
| CN112801284A (en) | Power-less neuron circuit using p-n-p-n diode | |
| Mahalanabis et al. | Demonstration of spike timing dependent plasticity in CBRAM devices with silicon neurons | |
| Garbin et al. | Probabilistic neuromorphic system using binary phase-change memory (PCM) synapses: Detailed power consumption analysis | |
| CN114792130A (en) | Artificial neuron with leakage-integration-emission function | |
| CN112906880B (en) | Adaptive neuron circuit based on memristor | |
| Yang et al. | Neuromorphic Artificial Vision Systems Based on Reconfigurable Ion‐Modulated Memtransistors | |
| Singha et al. | Analog memristive time dependent learning using discrete nanoscale RRAM devices | |
| CN109165731B (en) | Electronic neural network and parameter setting method thereof | |
| Parmar et al. | Short-term plasticity circuit device exploration in the MASTISK neuromorphic framework | |
| CN115906961A (en) | Self-adaptive artificial pulse neuron circuit based on volatile threshold resistance changing memristor | |
| CN109948792A (en) | A kind of simulation associative learning circuit and control method based on transistor and memristor |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |