CN111352284B - Photon neuron realization method and device based on passive nano beam resonant cavity - Google Patents

Photon neuron realization method and device based on passive nano beam resonant cavity Download PDF

Info

Publication number
CN111352284B
CN111352284B CN202010360888.5A CN202010360888A CN111352284B CN 111352284 B CN111352284 B CN 111352284B CN 202010360888 A CN202010360888 A CN 202010360888A CN 111352284 B CN111352284 B CN 111352284B
Authority
CN
China
Prior art keywords
signal
power
perturbation
nano
resonant cavity
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
Application number
CN202010360888.5A
Other languages
Chinese (zh)
Other versions
CN111352284A (en
Inventor
郭旭涵
向金龙
苏翼凯
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shanghai Jiaotong University
Original Assignee
Shanghai Jiaotong University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Shanghai Jiaotong University filed Critical Shanghai Jiaotong University
Priority to CN202010360888.5A priority Critical patent/CN111352284B/en
Publication of CN111352284A publication Critical patent/CN111352284A/en
Application granted granted Critical
Publication of CN111352284B publication Critical patent/CN111352284B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • GPHYSICS
    • G02OPTICS
    • G02FOPTICAL DEVICES OR ARRANGEMENTS FOR THE CONTROL OF LIGHT BY MODIFICATION OF THE OPTICAL PROPERTIES OF THE MEDIA OF THE ELEMENTS INVOLVED THEREIN; NON-LINEAR OPTICS; FREQUENCY-CHANGING OF LIGHT; OPTICAL LOGIC ELEMENTS; OPTICAL ANALOGUE/DIGITAL CONVERTERS
    • G02F1/00Devices or arrangements for the control of the intensity, colour, phase, polarisation or direction of light arriving from an independent light source, e.g. switching, gating or modulating; Non-linear optics
    • G02F1/35Non-linear optics

Landscapes

  • Physics & Mathematics (AREA)
  • Nonlinear Science (AREA)
  • General Physics & Mathematics (AREA)
  • Optics & Photonics (AREA)
  • Optical Modulation, Optical Deflection, Nonlinear Optics, Optical Demodulation, Optical Logic Elements (AREA)
  • Lasers (AREA)

Abstract

A method and a device for realizing photon neurons based on a passive nano beam resonant cavity are characterized in that the nano beam resonant cavity is arranged between a bus waveguide for transmitting a pumping light signal and a signal waveguide for transmitting a perturbation light signal and is simultaneously coupled with the bus waveguide and the signal waveguide, the perturbation light pulse signal is applied to stimulate the nano beam resonant cavity to show nonlinear behavior, and then the typical response of the biological neurons to external stimulation is simulated. The invention realizes the typical stimulated response of the pulse neuron by utilizing the nonlinear behavior of the nano beam resonant cavity, can work at the gigahertz speed under the condition of only consuming microwatt power, is compatible with the traditional complementary metal oxide semiconductor process, and has the advantages of simple structure, ultralow power consumption, ultracompactness and easy large-scale integration.

Description

Photon neuron realization method and device based on passive nano beam resonant cavity
Technical Field
The invention relates to a technology in the field of semiconductor optical chips, in particular to a photon neuron realization method and device based on a passive nano beam resonant cavity.
Background
The optical chip has the advantages of large bandwidth, high capacity, low time delay, low energy consumption and the like, and is an excellent platform for realizing brain-like calculation. The existing neural optical network in the free space utilizes the interference characteristic of light, but the deep neural network based on the 3D printing technology and laser cooling atoms cannot realize on-chip integration; the schemes based on active devices such as semiconductor lasers, modulators, photodetectors and the like inevitably need to utilize optical-electrical-optical conversion to realize the interaction of photonic neurons and nerve synapses, which not only involves the hybrid integration of III-V materials and silicon-based materials, but also requires a plurality of active devices to simulate the function of one photonic neuron, which greatly increases the cost and process complexity of the neural-optical network.
Disclosure of Invention
Aiming at the defects in the prior art, the invention provides a photon neuron realization method and a device based on a passive nano beam resonant cavity, which realize the typical stimulated response of a pulse neuron by utilizing the nonlinear behavior of the nano beam resonant cavity, can work at the gigahertz speed under the condition of only consuming a few microwatts of power, is compatible with the traditional Complementary Metal Oxide Semiconductor (CMOS) process, and has the advantages of simple structure, ultra-low power consumption, ultra-compactness and easy large-scale integration.
The invention is realized by the following technical scheme:
the invention relates to a photon neuron realization method based on a passive nano beam resonant cavity, namely, the nano beam resonant cavity is arranged between a bus waveguide for transmitting a pumping light signal and a signal waveguide for transmitting a perturbation light signal and is simultaneously coupled with the bus waveguide and the signal waveguide, and the perturbation light pulse signal is applied to stimulate the nano beam resonant cavity to show nonlinear behavior so as to simulate the typical response of a biological neuron to external stimulation.
The quality factor of the nano beam resonant cavity is more than 50000, and the resonant wavelength is the wavelength of a C-band communication window.
The pump light signal is a continuous light signal.
The wavelength of the pump light signal is the same as that of the perturbation light signal and is smaller than the resonance wavelength of the nano beam resonant cavity.
Typical responses described include: an excited trigger characteristic, a threshold characteristic, a leakage integration characteristic, a refractory period characteristic, a cascade characteristic, and an excited inhibit characteristic.
In the stimulated trigger characteristic, the threshold characteristic, the leakage integration characteristic, the refractory period characteristic and the cascade characteristic of the perturbation optical signal, the power is 0 in a stable state, and the power is increased to a certain fixed value only within a specific time width; in the stimulated suppression characteristic, the continuous optical signal with constant power is in a stable state, and when an exciting perturbation pulse needs to be applied, the power of the continuous optical signal is increased to a certain constant high-order power within a specific time width; when an inhibitory perturbation pulse is applied, its power is reduced to zero for a certain time width. Throughout the process, there are several sets of excitatory perturbation pulses and inhibitory perturbation pulses, and different inhibitory effects are observed by varying the time interval between the two perturbation pulses.
The passive nano beam resonant cavity, the bus waveguide and the signal waveguide are preferably but not limited to be prepared through a standard silicon-based device processing flow.
The invention relates to a photon neuron structure based on a passive nano beam resonant cavity, which comprises a signal waveguide, at least one passive nano beam resonant cavity and a bus waveguide which are sequentially arranged in parallel, wherein: the perturbation optical signal and the pump optical signal are respectively incident from the input ends of the signal waveguide and the bus waveguide, and the perturbation optical signal is coupled with the nano beam resonant cavity and then stimulates the nano beam resonant cavity to show a stimulated response behavior so as to simulate the basic function of a photon neuron.
The invention relates to an application based on the photonic neuron structure, which is used for transmitting information.
The application realizes the typical response of the passive nano beam resonant cavity by regulating and controlling the power and wavelength of the pump light signal and the perturbation light signal, and comprises the following steps: an excited trigger characteristic, a threshold characteristic, a leakage integration characteristic, a refractory period characteristic, a cascade characteristic, and an excited inhibit characteristic, wherein:
the stimulated trigger characteristic is that: the initial optical power of the perturbation signal is zero, the pumping optical signal provides proper pumping conditions for the nano beam resonant cavity, and after the output signal power keeps a constant state, the perturbation pulse signal is applied to ensure that the output signal power is obviously attenuated and then gradually returns to a stable state, namely, relative negative pulses are output under the excitation of the perturbation signal.
The threshold characteristic means: the initial optical power of the perturbation signal is zero, the pumping optical signal provides proper pumping conditions for the nano beam resonant cavity, when the output signal power is kept in a constant state, when the perturbation pulse signal with the power smaller than the threshold value is applied, the shape and the intensity of the relative negative pulse of the output signal are obviously different from those of the relative negative pulse obtained when the perturbation pulse signal with the power higher than the threshold value is applied, otherwise, the shape and the intensity of the relative negative pulse are kept constant.
The leakage integration characteristic is that: the initial optical power of the perturbation signal is zero, the pumping optical signal provides proper pumping conditions for the nano beam resonant cavity, after the output signal power keeps a constant state, the output signal power is obviously attenuated and then gradually returns to a stable state by continuously applying a plurality of perturbation pulse signals with short time intervals and the power intensity lower than a threshold value, namely, complete relative negative pulses are output under the excitation of the perturbation signals.
The leakage integration characteristic is that the output signal has small amplitude rise back in power in the time interval between adjacent perturbation signals.
The refractory period characteristics refer to: the initial optical power of the perturbation signal is zero, the pumping optical signal provides proper pumping conditions for the nano beam resonant cavity, and after the output signal power is kept in a constant state, a perturbation pulse signal with the power higher than a threshold value is applied, and then a second identical perturbation pulse signal is applied, so that the attenuation degree of the output signal power is closely related to the time interval of the two perturbation pulse signals.
The cascade characteristic means that: two same passive nano-beam resonant cavities are sequentially arranged and are in cascade coupling with the same bus waveguide, the initial optical power of a perturbation signal is zero, a pumping optical signal provides proper pumping conditions for the nano-beam resonant cavities, and after the power of an output signal is kept in a constant state, the output signal presents two complete relative negative pulse signals only by guiding the signal wave to the first passive nano-beam resonant cavity to apply a perturbation pulse signal with the power higher than a threshold value.
In the two complete relative negative pulse signals, the first relative negative pulse corresponds to the stimulated response of the first passive nano-beam resonant cavity, and the second relative negative pulse corresponds to the stimulated response of the second passive nano-beam resonant cavity under the stimulation of the output signal of the first passive nano-beam resonant cavity.
The stimulated inhibition property refers to: the initial optical power of the perturbation signal is far lower than that of the pump optical signal, the perturbation signal and the pump optical signal simultaneously provide proper pumping conditions for the nano-beam resonant cavity, and after the output signal power keeps a constant state, the power of the perturbation signal is quickly reduced to form an inhibitory excitation signal or the power of the perturbation signal is quickly increased to form an excitatory excitation signal; after the output signal power is kept in a constant state: applying only a single excitatory excitation signal will cause the output signal to exhibit one complete relative negative pulse; ② applying a suppressing excitation signal shortly before or after applying an exciting excitation signal will cause the intensity of the relative negative pulse of the output signal to be suppressed.
The degree of suppression is closely related to the time interval of the two excitation signals.
Technical effects
The invention integrally solves the technical problems of complex preparation process, high production cost, difficulty in large-scale integration and the like caused by the fact that the conventional neural optical network chip adopts active devices such as a semiconductor laser, an optical modulator, an optical detector and the like. The invention utilizes the nano-beam resonant cavity to simulate typical stimulated response behavior of a photon neuron, including stimulated triggering characteristic, threshold characteristic, leakage integration characteristic, refractory period characteristic, cascade characteristic and stimulated suppression characteristic; the related structures are all standard passive devices and can be manufactured in a large scale by adopting a conventional silicon-based integration process; the power consumption source is mainly the pump light power which is only a few microwatts; can work efficiently at speeds up to gigahertz.
Drawings
FIG. 1 is a schematic diagram of the inventive concept;
in the figure: bus waveguide 1, signal waveguide 2, nano-beam resonant cavity 3, pump light PinSignal light PtrAnd output light Pout
FIG. 2 is a schematic structural diagram of a passive nanobeam resonator;
in the figure: a cavity 4 and a reflector 5;
FIG. 3 is a schematic diagram of stimulated triggering characteristics of photonic neurons based on passive nanobeam resonators;
FIG. 4 is a schematic diagram of threshold characteristics of photonic neurons based on passive nanobeam resonators;
FIG. 5 is a schematic diagram of the leakage integration characteristics of photonic neurons based on passive nanobeam resonators;
FIG. 6 is a schematic diagram of the refractory thermal behavior of a photonic neuron based on a passive nanobeam resonator;
FIG. 7 is a schematic diagram of cascadable properties of photonic neurons based on passive nanobeam resonators;
fig. 8 is a schematic diagram of stimulated suppression characteristics of photonic neurons based on resonant cavities that are passive nanobeam cavities.
Detailed Description
As shown in fig. 1, the present embodiment relates to a photonic neuron device based on a passive nanobeam resonant cavity, which includes a bus waveguide, a passive nanobeam resonant cavity, and a signal waveguide, which are sequentially arranged, wherein: one end of the bus waveguide is continuously input with a pump light signal, and the other end of the bus waveguide is an output signal; pulse perturbation optical signals are input at one end of the signal waveguide, and under a specific pumping condition, the perturbation signals excite the passive nano-beam resonant cavity to emit a relative negative pulse signal, so that the simulation of neurons is realized.
The bus waveguide and the signal waveguide are both silicon waveguides, and preferably, the silicon waveguides are 500 nanometers in width and 220 nanometers in height.
As shown in fig. 2, the passive nanobeam resonant cavity 3 includes: the width of the whole resonant cavity waveguide is 500 nanometers, the height is 220 nanometers, the length is 15 micrometers, and the whole structure is symmetrical about the center.
The reflector is provided with 9 round holes in total, the diameters of all round holes are 0.5 multiplied by a, and the distance between the centers of two adjacent round holes is 0.9 multiplied by a.
The cavity 4 is provided with 6 round holes in total, wherein the part close to the reflector is a first round hole, and the center position is a sixth round hole.
Preferably, the diameters of the first circular hole to the sixth circular hole are sequentially decreased to be 0.5 × 0.98 × a, 0.5 × 0.92 × a, 0.5 × 0.88 × a, 0.5 × 0.84 × a, 0.5 × 0.80 × a and 0.5 × 0.76 × a respectively; and the distance between each round hole and the center of the previous round hole is 0.9 multiplied by 0.98 multiplied by a, 0.9 multiplied by 0.92 multiplied by a, 0.9 multiplied by 0.88 multiplied by a, 0.9 multiplied by 0.84 multiplied by a, 0.9 multiplied by 0.80 multiplied by a and 0.9 multiplied by 0.76 multiplied by a in sequence, wherein: a is a constant, and a is preferably 453.2 nm.
The coupling distance between the nano beam resonant cavity and the bus waveguide and between the nano beam resonant cavity and the signal waveguide is preferably 300 nanometers.
The resonance wavelength of the nano-beam resonant cavity is preferably 1.55245 micrometers, and the quality factor is preferably 114400.
The specific pumping conditions are as follows: the pumping signal and the perturbation signal have the same optical wavelength, and the resonant wavelength of the nano beam resonant cavity meets the following requirements: delta lambda is lambda-lambdaresWherein: lambda and lambdaresThe wavelength of the input pump light and the resonance wavelength of the nano-beam resonant cavity are respectively.
The time width of the perturbation pulse signal is kept constant for 1 nanosecond.
The normalized intensity of the relative negative pulse signal is
Figure BDA0002475029970000041
Wherein: pminAnd PconstantRespectively representing the minimum power of the output pulse and the output signal power at steady state.
As shown in FIGS. 3 to 8, Pin、PtrAnd PoutRespectively pump light power, perturbation signal light power and output signal light power.
As shown in fig. 3, δ λ is set to-13 pm, Pin=5μW,PtrAfter a perturbation light pulse signal is applied at 40 ns, the output signal of the nano-beam resonant cavity shows a relatively negative pulse signal under the combined action of various nonlinear effects. This indicates that photonic neurons based on passive nanobeam resonators can exhibit stimulated triggering properties.
As shown in fig. 4, δ λ is set to-13 pm, PinKeeping the other conditions unchanged, the optical power of the perturbation pulse signal is gradually increased from zero, and the normalized intensity of the output relative to the negative pulse is gradually increased and finally becomes constant. When the optical power of the perturbation signal is less than 1.36 microwatts, the shape of the output relative negative pulse can be obviously changed; but slightly perturbs the letterWhen the signal light power is more than 1.36 microwatts, the shape and the intensity of the relative negative pulse are basically kept constant; and the normalized intensity of the output relative negative pulse is most strongly varied with the optical power of the perturbation signal around 1.36 microwatts. This indicates that photonic neurons based on passive nanobeam resonators may exhibit threshold characteristics.
As shown in fig. 5, δ λ is set to-12 pm, Pin=4.4μW,PtrApplying a perturbation pulse signal at 20 ns, and the output signal only undergoes a small amplitude attenuation; and applying two perturbation pulse signals with the time interval of 1 nanosecond at 40 nanoseconds, and gradually recovering to a stable state after the output signal is almost attenuated to zero. In addition, there is a small amplitude recovery of the output signal to a stationary state in the time gap between the two perturbation signals. This suggests that photonic neurons based on passive nanobeam resonators may exhibit leaky integration properties.
As shown in fig. 6, δ λ is set to-13 pm, Pin=5μW,PtrA perturbation pulse signal applied at 30 ns excites the nanobeam resonator to emit a relatively negative pulse signal, 2 μ W. If a second perturbation pulse signal is continuously applied in 42 ns, the nano-beam resonant cavity remains insensitive, and the output signal power only undergoes a very small attenuation; while the nanobeam resonator is reactivated if a second perturbation pulse signal is applied at 46 ns. This indicates that photonic neurons based on passive nanobeam resonators may exhibit refractory period characteristics.
As shown in fig. 7, δ λ is set to-18.5 pm, Pin=11μW,PtrTwo nano-beam resonant cavities are cascaded, and an output signal of one nano-beam resonant cavity is directly used as an input signal of the other nano-beam resonant cavity. A perturbation pulse signal was applied to the first nano-beam cavity at 150 nanoseconds and as a result the output signal from the second nano-beam cavity exhibited two opposing negative pulses. Wherein: the first relative negative pulse corresponds to the output signal of the first nanobeam resonator, and the second relative negative pulse is the stimulated output signal of the second nanobeam resonator. This indicates that the photon neuron based on the passive nano beam resonant cavity has scalabilityAnd (4) link characteristics.
As shown in fig. 8, δ λ is set to-12 pm, PinThe power to keep the perturbation signal at steady state is 0.35 microwatts, 4.4 muw. The high optical power of the excitatory perturbation pulse signal is 1 microwatt, while the low optical power of the inhibitory perturbation pulse signal is 0 microwatt. When the inhibitory perturbation pulse signal is gradually close to the excitatory perturbation pulse signal, the output relative negative pulse is gradually and completely inhibited; when the inhibitory perturbation pulse signal is gradually far away from the excitation pulse signal, the intensity of the output relative negative pulse is gradually restored to the strongest point. This indicates that photonic neurons based on passive nanobeam resonators can exhibit stimulated inhibition properties.
The above embodiments excite the nano-beam resonant cavity to exhibit a nonlinear behavior that can simulate the stimulated response of the photonic neuron by applying an additional perturbation pulse signal, i.e., the nano-beam resonant cavity can simulate the stimulated triggering characteristic, the threshold characteristic, the leakage integration characteristic, the refractory period characteristic, the cascadable characteristic and the stimulated suppression characteristic of the photonic neuron.
Compared with the prior art, the invention completely adopts passive devices, has simple structure and compact size, can be manufactured by adopting a standard silicon-based integrated process, is compatible with the traditional CMOS process and is easy to construct an on-chip large-scale neural optical network; the optical-electrical-optical conversion link is adopted in the neural optical network based on the active device, so that not only is the heat loss of the active device present, but also additional loss is inevitably introduced in the conversion process, and the power consumption is often in the order of milliwatt to microwatt. The photon neuron based on the passive nano beam resonant cavity adopts an all-optical link, the energy consumption mainly comes from pumping signals and is only a few microwatts; in addition, the working speed of the photon neuron based on the nano beam resonant cavity can reach the gigahertz order compared with the photon neuron based on the active device.
The foregoing embodiments may be modified in many different ways by those skilled in the art without departing from the spirit and scope of the invention, which is defined by the appended claims and all changes that come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein.

Claims (15)

1. A photon neuron realization method based on a passive nano beam resonant cavity is characterized in that the nano beam resonant cavity is arranged between a bus waveguide for transmitting a pump light signal and a signal waveguide for transmitting a perturbation pulse light signal and is simultaneously coupled with the bus waveguide and the signal waveguide, the perturbation pulse light signal is applied to stimulate the nano beam resonant cavity to show nonlinear behavior, and then the typical response of a biological neuron to external stimulation is simulated;
the typical responses include: an excited trigger characteristic, a threshold characteristic, a leakage integration characteristic, a refractory period characteristic, a cascade characteristic, and an excited inhibit characteristic.
2. The method for realizing the photonic neuron based on the passive nano-beam resonant cavity according to claim 1, wherein the pump light signal is a continuous light signal; the wavelength of the pump light signal is the same as that of the perturbation pulse light signal, and is smaller than the resonance wavelength of the nano beam resonant cavity.
3. A photonic neuron structure based on a passive nanobeam resonant cavity, comprising: signal waveguide, at least one passive nanometer roof beam resonant cavity and bus waveguide that set up side by side in proper order, wherein: the nano-beam resonant cavity is arranged between a bus waveguide for transmitting a pump light signal and a signal waveguide for transmitting a perturbation pulse light signal and is simultaneously coupled with the bus waveguide and the signal waveguide, the perturbation pulse light signal and the pump light signal are respectively incident from the input ends of the signal waveguide and the bus waveguide, and the perturbation pulse light signal is coupled with the nano-beam resonant cavity and then stimulates the nano-beam resonant cavity to show a stimulated response behavior so as to simulate the function of a photon neuron.
4. The photonic neuron structure of claim 3, wherein the passive nanobeam resonant cavity comprises: the aperture of the reflector is unchanged, the aperture of the cavity is gradually changed, and the whole nano-beam resonant cavity is symmetrical about the center.
5. Use of a photonic neuronal structure according to claim 1 or 2 or according to claim 3 or 4 for the transfer of information.
6. The application of claim 5, wherein the typical response of the passive nano-beam resonant cavity is realized by adjusting and controlling the power and wavelength of the pump light signal and the perturbation pulse light signal, and comprises the following steps: an excited trigger characteristic, a threshold characteristic, a leakage integration characteristic, a refractory period characteristic, a cascade characteristic, and an excited inhibit characteristic.
7. The use of claim 6, wherein said activated trigger characteristic is: the initial optical power of the perturbation pulse signal is zero, and the pumping optical signal provides proper pumping conditions for the nano beam resonant cavity, namely setting delta lambda to-13 pm, Pin=5μW,PtrAfter applying a perturbation pulse signal at 40 ns, after the output signal power is kept constant, by applying the perturbation pulse signal, the output signal power will be attenuated significantly and then gradually restored to a stable state, namely, a relatively negative pulse is output under the excitation of the perturbation pulse signal, wherein: delta lambda is the difference between the wavelength of the input pump light and the resonant wavelength of the nano-beam resonator, Pin、PtrRespectively pump light power and perturbation signal light power.
8. The use of claim 6, wherein the threshold characteristic is: the initial optical power of the perturbation pulse signal is zero, and the pumping optical signal provides proper pumping conditions for the nano beam resonant cavity, namely setting delta lambda to-13 pm, Pin5 muW, when the power of the output signal is kept constant, when the applied power is less than PtrWill be significantly different from applying a pulse higher than P, the shape and intensity of the relative negative pulse of the output signal will be significantly differenttrThe relative negative pulse obtained when perturbing the pulse signal, otherwise the shape and intensity of the relative negative pulse remains constant, wherein: with δ λ being input pump lightDifference between wavelength and resonant wavelength of the nano-beam resonator, Pin、PtrRespectively pump light power and perturbation signal light power.
9. The use of claim 6, wherein said leak integration characteristic is: the initial optical power of the perturbation pulse signal is zero, and the pumping optical signal provides proper pumping conditions for the nano beam resonant cavity, namely setting delta lambda-12 pm, Pin=4.4μW,PtrAfter applying a perturbation pulse signal at 20 ns, after the output signal power is kept constant, by applying perturbation pulse signals with a power intensity lower than a threshold value for a plurality of short time intervals continuously, the output signal power is obviously attenuated and then gradually returns to a stable state, namely, complete relative negative pulses are output under the excitation of the perturbation pulse signals, wherein: delta lambda is the difference between the wavelength of the input pump light and the resonant wavelength of the nano-beam resonator, Pin、PtrRespectively pump light power and perturbation signal light power.
10. The use of claim 9, wherein said leakage integration characteristic is that the output signal has a small amplitude rise back in power during the time interval between adjacent perturbation pulse signals.
11. The use of claim 6, wherein the refractory period characteristic is: the initial optical power of the perturbation pulse signal is zero, and the pumping optical signal provides proper pumping conditions for the nano beam resonant cavity, namely setting delta lambda to-13 pm, Pin=5μW,PtrAfter applying a perturbation pulse signal at 30 ns, and after the output signal power is kept in a constant state, applying a perturbation pulse signal with power higher than a threshold value and then applying a second identical perturbation pulse signal to make the output signal power attenuation degree closely related to the time interval of the two perturbation pulse signals, wherein: delta lambda is the difference between the wavelength of the input pump light and the resonant wavelength of the nano-beam resonator, Pin、PtrAre respectively asPump light power, perturbation signal light power.
12. The use of claim 6, wherein said cascade characteristic is: two same passive nano-beam resonant cavities are sequentially arranged and are in cascade coupling with the same bus waveguide, the initial optical power of a perturbation pulse signal is zero, and a pumping optical signal provides proper pumping conditions for the nano-beam resonant cavities, namely delta lambda is-18.5 pm, P is setin=11μW,PtrWhen the output signal power is kept constant, only a perturbation pulse signal with the power higher than a threshold value is applied to the first passive nano-beam resonant cavity through signal wave guide, and then the output signal shows two complete relative negative pulse signals, wherein: delta lambda is the difference between the wavelength of the input pump light and the resonant wavelength of the nano-beam resonator, Pin、PtrRespectively pump light power and perturbation signal light power.
13. The use of claim 12, wherein the two complete opposite negative pulse signals, the first opposite negative pulse corresponds to the stimulated response of the first passive nano-beam resonator, and the second opposite negative pulse corresponds to the stimulated response of the second passive nano-beam resonator upon excitation of the first passive nano-beam resonator output signal.
14. The use according to claim 6, wherein said stimulated inhibition property is: the initial optical power of the perturbation pulse signal is far lower than that of the pump optical signal, and the perturbation pulse signal and the pump optical signal simultaneously provide proper pumping conditions for the nano beam resonant cavity, namely, the delta lambda is set to be-12 pm, and P is setinWhen the output signal power is 4.4 muW, the power of the perturbation signal in a steady state is maintained to be 0.35 microwatt, the high-order optical power of the exciting perturbation pulse signal is 1 microwatt, and the low-order optical power of the inhibiting perturbation pulse signal is 0 microwattAfter keeping a constant state, quickly reducing the power of the perturbation pulse signal to form an inhibitory excitation signal or quickly increasing the power of the perturbation pulse signal to form an excitatory excitation signal; after the output signal power is kept in a constant state: applying only a single excitatory excitation signal will cause the output signal to exhibit one complete relative negative pulse; (iii) applying an inhibitory excitation signal shortly before or after applying the excitatory excitation signal will suppress the intensity of the relative negative pulses of the output signal, wherein: delta lambda is the difference between the wavelength of the input pump light and the resonant wavelength of the nano-beam resonator, Pin、PtrRespectively pump light power and perturbation signal light power.
15. Use according to claim 14, wherein the degree of suppression is closely related to the time interval between two excitation signals.
CN202010360888.5A 2020-04-30 2020-04-30 Photon neuron realization method and device based on passive nano beam resonant cavity Active CN111352284B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010360888.5A CN111352284B (en) 2020-04-30 2020-04-30 Photon neuron realization method and device based on passive nano beam resonant cavity

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010360888.5A CN111352284B (en) 2020-04-30 2020-04-30 Photon neuron realization method and device based on passive nano beam resonant cavity

Publications (2)

Publication Number Publication Date
CN111352284A CN111352284A (en) 2020-06-30
CN111352284B true CN111352284B (en) 2021-11-19

Family

ID=71193389

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010360888.5A Active CN111352284B (en) 2020-04-30 2020-04-30 Photon neuron realization method and device based on passive nano beam resonant cavity

Country Status (1)

Country Link
CN (1) CN111352284B (en)

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6445470B1 (en) * 1990-04-06 2002-09-03 University Of Southern California Apparatus for simultaneous spatial modulation of incoherent/coherent angulary multiplexed optical beams
CN104677872A (en) * 2015-02-27 2015-06-03 中国科学院自动化研究所 Multi-photon excited multidirectional lighting microscopy imaging system
CN105242395A (en) * 2015-08-31 2016-01-13 西北工业大学 Electromagnetic-driving micro-mechanical tunable Fabry-Perot filters and manufacturing method thereof
CN208551800U (en) * 2017-07-07 2019-03-01 中国人民解放军第三军医大学 Mouse head suspention and physiological function recovery system after microelectrode array implantation
CN110302471A (en) * 2019-06-13 2019-10-08 深圳市恒康泰医疗科技有限公司 Bio wave wave spectrum physical therapy technology based on plant wave
WO2019217836A1 (en) * 2018-05-10 2019-11-14 The Board Of Trustees Of The Leland Stanford Junior University Systems and methods for activation functions for photonic neural networks

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6445470B1 (en) * 1990-04-06 2002-09-03 University Of Southern California Apparatus for simultaneous spatial modulation of incoherent/coherent angulary multiplexed optical beams
CN104677872A (en) * 2015-02-27 2015-06-03 中国科学院自动化研究所 Multi-photon excited multidirectional lighting microscopy imaging system
CN105242395A (en) * 2015-08-31 2016-01-13 西北工业大学 Electromagnetic-driving micro-mechanical tunable Fabry-Perot filters and manufacturing method thereof
CN208551800U (en) * 2017-07-07 2019-03-01 中国人民解放军第三军医大学 Mouse head suspention and physiological function recovery system after microelectrode array implantation
WO2019217836A1 (en) * 2018-05-10 2019-11-14 The Board Of Trustees Of The Leland Stanford Junior University Systems and methods for activation functions for photonic neural networks
CN110302471A (en) * 2019-06-13 2019-10-08 深圳市恒康泰医疗科技有限公司 Bio wave wave spectrum physical therapy technology based on plant wave

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
All-optical nonlinear activation function for photonic neural networks;MARIO MISCUGLIO等;《OPTICAL MATERIALS EXPRESS》;20181231;全文 *
Bulk PZT Cantilever Based MEMS Acoustic Transducer for Cochlear Implant Applications;Aziz Koyuncuoğlu等;《Proceedings 》;20170811;全文 *
Investigation of vertical cavity surface emitting laser dynamics for neuromorphic photonic systems;A.Hurtado;《APPLIED PHYSICS LETTERS》;20120307;全文 *
光子脉冲神经元权重器件的研制;宋晓佳等;《中国激光》;20170430;全文 *

Also Published As

Publication number Publication date
CN111352284A (en) 2020-06-30

Similar Documents

Publication Publication Date Title
Mesaritakis et al. Artificial neuron based on integrated semiconductor quantum dot mode-locked lasers
US7245801B2 (en) Apparatus with a series of resonator structures situated near an optical waveguide for manipulating optical pulses
JP2531443B2 (en) All-optical device
CN103022867A (en) High-power high-efficiency supercontinuum source
CN102170082B (en) All-fiber pulsed fiber laser with composite chamber
CN109494552A (en) A kind of full optical fiber laser system and method generating high-energy rectangular pulse
CN103124044A (en) Frequency interval adjustable multi-wavelength anti-Stokes four-wave mixing (FWM) fiber laser
Maywar et al. Effect of chirped gratings on reflective optical bistability in DFB semiconductor laser amplifiers
Sysoliatin et al. All-optical fiber soliton processing using dispersion oscillating fiber
Wei et al. A wavelength tunable optical neuron based on a fiber laser
CN111352284B (en) Photon neuron realization method and device based on passive nano beam resonant cavity
CN203014156U (en) High-power high-efficiency super-continuous spectrum light source
Puts et al. Measurements and modeling of a monolithically integrated self-spiking two-section laser in InP
Xiang et al. Experimental demonstration of a photonic spiking neuron based on a DFB laser subject to side-mode optical pulse injection
Gao et al. Cascadable excitability and inhibition in DFB laser-based photonic spiking neurons
Kuriakose et al. Elements of optical solitons: An overview
CN110231058B (en) Chaotic Brillouin dynamic grating generation device and method with substrate noise suppression function
Han et al. Automatic mode-locked fiber laser based on adaptive genetic algorithm
Diamantopoulos et al. Ultrafast spiking membrane III-V laser neuron on Si
CN110445002B (en) Device and method for generating super-continuum spectrum by low-pumping few-mode photonic crystal fiber
Heinsalu et al. Design and Experiment of Silicon Racetrack-Loop Multi-Mode Waveguide Structure with Low-Loss and Adjustable Couplings for Compact Reservoir Computing Device
Wang et al. Silicon-based Fano resonance devices based on photonic crystal nanobeams
Nawi et al. Analytical treatment of parametric effects in a ring resonator
CN1555111A (en) Wavelength tunable 4 femtosecond super short optical pulse generator
Diamantopoulos et al. All-Optical Spiking Membrane III-V Laser on Si

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