IP Library › Granted Patent US 12,340,301
Granted Patent B2
US 12,340,301 · App. 17/596,090 · Granted Jun 24, 2025

Photonic neural network

Inventor: Lucas Zipp (Palo Alto, CA)
Assignee: SRI International
G06N3/067G06N3/08
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,340,301
App. No.
17/596,090
Granted
Jun 24, 2025
Kind
B2
Abstract

A photonic neural network device may include a planar waveguide; a layer having a changeable refractive index adjacent to the planar waveguide; and a plurality of electrodes. Each electrode may be electrically coupled to the layer having the changeable refractive index at a corresponding location of the layer having the changeable refractive index. Each electrode may be configured to apply a corresponding, configurable voltage to the corresponding location to affect a refractive index of the corresponding location of the layer having the changeable refractive index to induce an amplitude modulation or a phase modulation of a light waveform propagating through the photonic neural network device to configure a corresponding neuron of the photonic neural network device in order to perform a computation.

Claims (32)

1. A photonic neural network device comprising:

a planar waveguide;

a layer having a changeable refractive index adjacent to the planar waveguide; and

a plurality of electrodes, wherein each electrode is electrically coupled to the layer having the changeable refractive index at a corresponding location of the layer having the changeable refractive index, wherein each electrode is configured to apply a corresponding, configurable voltage to the corresponding location to affect a refractive index of the corresponding location of the layer having the changeable refractive index to induce an amplitude modulation or a phase modulation of a light waveform propagating through the photonic neural network device to configure a corresponding neuron of the photonic neural network device in order to perform a computation.

2. The photonic neural network device of claim 1 , wherein the plurality of electrodes are arranged in a plurality of columns of electrodes, wherein each column of electrodes of the plurality of columns of electrodes is a neural network layer of a neural network.

3. The photonic neural network device of claim 2 , further comprising a column of diffractive or refractive elements between a first column of electrodes of the plurality of columns of electrodes and a second column of electrodes of the plurality of columns of electrodes.

4. The photonic neural network device of claim 3 , wherein the column of diffractive or refractive elements comprises a plurality of planar waveguide lenses.

5. The photonic neural network device of claim 3 , wherein at least one of a spacing between the first column of electrodes and the second column of electrodes, a spacing between the first column of electrodes and the column of diffractive or refractive elements, a spacing between adjacent neurons within the second column of electrodes, or a property of the column of diffractive or refractive elements is selected to achieve a selected spread of light between the first column of electrodes and the second column of electrodes.

6. The photonic neural network device of claim 3 , wherein the column of diffractive or refractive elements comprises at least one tunable element tunable to control a spread of light between the first column of electrodes and the second column of electrodes.

7. The photonic neural network device of claim 1 , wherein the planar waveguide is a single mode planar waveguide.

8. The photonic neural network device of claim 1 , wherein the layer having a changeable refractive index comprises a liquid crystal.

9. The photonic neural network device of claim 1 , wherein optical nonlinearities of the planar waveguide are exploited to perform nonlinear computations.

10. The photonic neural network device of claim 1 , wherein respective shapes of the plurality of electrodes are selected to control optical effects.

11. The photonic neural network device of claim 1 , further comprising an optical modulator optically coupled to the planar waveguide, wherein the optical modulator is configured to provide an input waveform to the planar waveguide.

12. The photonic neural network device of claim 1 , further comprising a plurality of photodetectors coupled to the planar waveguide, wherein the plurality of photodetectors are configured to detect an output waveform from the planar waveguide.

13. A method comprising:

modelling a photonic neural network device, wherein the photonic neural network device comprises:

a planar waveguide;

a layer having a changeable refractive index adjacent to the planar waveguide; and

a plurality of electrodes, wherein each electrode is electrically coupled to the layer having the changeable refractive index at a corresponding location of the layer having the changeable refractive index, wherein each electrode is configured to apply a corresponding, configurable voltage to the corresponding location to affect a refractive index of the corresponding location of the layer having the changeable refractive index to configure a corresponding neuron of the photonic neural network device in order to perform a computation;

training the modelled photonic neural network, wherein training comprises selecting the corresponding, configurable voltages at corresponding locations so that the corresponding neurons produce a desired output waveform in response to an input waveform; and

physically implementing the trained photonic neural network in a device based on the training.

14. The method of claim 13 , wherein the plurality of electrodes are arranged in a plurality of columns of electrodes, wherein each column of the plurality of columns of electrodes is a neural network layer of a neural network, and wherein training the modelled photonic neural network further comprises selecting a spacing between adjacent columns of the plurality of columns of electrodes.

15. The method of claim 14 , wherein training the modelled photonic neural network further comprises selecting a spacing between adjacent electrodes within a column of the plurality of columns of electrodes.

16. The method of claim 13 , wherein the photonic neural network device further comprises a column of diffractive or refractive elements between a first column of electrodes of the plurality of columns of electrodes and a second column of electrodes of a plurality of columns of electrodes, and wherein training the modelled photonic neural network further comprises selecting a spacing between the column of diffractive or refractive elements and the second column of electrodes.

17. The method of claim 13 , wherein the photonic neural network device further comprises a column of diffractive or refractive elements between a first column of electrodes of a plurality of columns of electrodes and a second column of electrodes of a plurality of columns of electrodes, and wherein training the modelled photonic neural network further comprises selecting a property of the diffractive or refractive elements to achieve a selected spread of light between the first column of electrodes and the second column of electrodes.

18. The method of claim 13 , wherein training the modelled photonic neural network further comprises selecting respective shapes of the plurality of electrodes to control optical effects.

19. The method of claim 13 , wherein physically implementing the trained photonic neural network in the device based on the training comprises configuring corresponding electrodes and corresponding, configurable voltages of selected electrodes from an array of electrodes on a layer having a changeable refractive index adjacent to a planar waveguide to define a physical instantiation of the trained photonic neural network.

20. A photonic neural network device comprising:

a plurality of neural network layers for a neural network, wherein each neural network layer comprises:

a column of neurons, each neuron comprising a corresponding electrode adjacent to a layer having a changeable refractive index, wherein the layer having the changeable refractive index is adjacent to a planar optical waveguide; and

a column of refractive or diffractive elements adjacent to the planar optical waveguide, wherein the photonic neural network is configured to induce an amplitude or a phase modulation of a light waveform propagating through the photonic neural network device to configure a corresponding neuron of the photonic neural network device in order to perform a computation.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Dec 2, 2021
From: ZIPP, LUCAS
To: SRI INTERNATIONAL
Reel/Frame 058275/0039 →
Continuity (2)
Provisional Application 62856603 · Jun 3, 2019
Related Publication 20220300796A1 · Sep 22, 2022
References Cited (48)
US 5080464A · Toyoda · 1992 [cited by applicant]
US 5150242A · Fellows · 1992 [cited by applicant]
US 5388169A · Kobayashi et al. · 1995 [cited by applicant]
US 6819469B1 · Koba · 2004 [cited by applicant]
US 11836609B2 · Lamb · 2023 [cited by examiner]
US 12222680B1 · Raikov · 2025 [cited by examiner]
US 12229662B2 · Du · 2025 [cited by examiner]
US 20050111775A1 · Fridman et al. · 2005 [cited by applicant]
US 20100176278A1 · Yokoyama · 2010 [cited by applicant]
US 20170351293A1 · Carolan et al. · 2017 [cited by applicant]
US 20180211158A1 · Shainline et al. · 2018 [cited by applicant]
US 20180284834A1 · Abel et al. · 2018 [cited by applicant]
US 20190319633A1 · Zou · 2019 [cited by examiner]
US 20210133547A1 · Lin · 2021 [cited by examiner]
US 20210192342A1 · Hughes · 2021 [cited by examiner]
US 20220300796A1 · Zipp · 2022 [cited by examiner]
US 20230169339A1 · Ferreira de Lima · 2023 [cited by examiner]
US 20230306253A1 · Dobos · 2023 [cited by examiner]
US 20230409894A1 · Song · 2023 [cited by examiner]
US 20240086698A1 · Guo · 2024 [cited by examiner]
EP 3690804A1 · 2020 [cited by applicant]
JP S6033531A · 1985 [cited by applicant]
JP H03198031A · 1991 [cited by applicant]
JP H03256032A · 1991 [cited by applicant]
JP H06308452A · 1994 [cited by applicant]
JP 2012519311A · 2012 [cited by applicant]
JP 2019061496A · 2019 [cited by applicant]
WO 2010141122A1 · 2010 [cited by applicant]
Response to Office Action, and translation thereof, dated Jan. 10, 2023, from counterpart Japanese Application No. 2021-571577 filed Apr. 21, 2023, 19 pp. [cited by applicant]
Office Action, and translation thereof, from counterpart Japanese Application No. 2021-571577 dated Aug. 1, 2023, 6 pp. [cited by applicant]
Belt et al. “Ultra-low-loss Ta2O5-core / SiO2-clad planar waveguides on Si substrates,” Optica, vol. 4, No. 5, May 2017, pp. 532-536. [cited by applicant]
Cybenko “Approximation by Superpositions of a Sigmoidal Function,” Math. Control Signals Syst., vol. 2, Dec. 1989, pp. 303-314. [cited by applicant]
Desmet et al., “Silicon-on-insulator optical waveguides with liquid crystal cladding for switching and tuning,” Proc. Eur. Conf. Opt. Commun., vol. 3, No. 1, Sep. 2003, pp. 430-431. [cited by applicant]
Di Falco et al. “Luneburg lens in silicon photonics,” Opt. Express, vol. 19, No. 6, Mar. 2011, p. 5156-5162. [cited by applicant]
Hermann, “Integrated Optics With Liquid Crystals,” Technology, 2000, 34 pp. (Applicant points out, in accordance with MPEP 609.04(a), that the year of publication, 2000, is sufficiently earlier than the effective U.S. f… [cited by applicant]
International Preliminary Report on Patentability from International Application No. PCT/US2020/035750, dated Dec. 16, 2021, 9 pp. [cited by applicant]
International Search Report and Written Opinion of International Application No. PCT/US2020/035750, dated Aug. 27, 2020, 11 pp. [cited by applicant]
Kobayashi et al. “2×2 Optical Waveguide Matrix Switch Using Nematic Liquid Crystal,” IEEE Trans. Microw. Theory Tech., vol. 30, No. 10, Oct. 1982, pp. 1591-1598. [cited by applicant]
Lin et al., “All-optical machine learning using diffractive deep neural networks,” Science vol. 361, No. 6406, Sep. 2018, 20 pp. [cited by applicant]
Monkade et al., “SiO Evaporated films topography and nematic liquid crystal orientation,” J. Phys. Ilde Phys. II, vol. 7, Nov. 1997, pp. 1577-1596. [cited by applicant]
Shen et al. “Deep learning with coherent nanophotonic circuits,” Nature Photon 11, Jul. 2016, pp. 441-446. [cited by applicant]
Terui et al., “Refractive-index-adjustable SiO2—Ta2O5 films for integrated optical circuits,” Appl. Phys. Lett., vol. 32, No. 10, May 1978, 3 pp. [cited by applicant]
Zernike, “Luneburg lens for optical waveguide use,” Opt. Commun., vol. 12, No. 4, Dec. 1974, pp. 379-381. [cited by applicant]
Response to Extended Search Report dated Dec. 16, 2022, from counterpart European Application No. 20817696.6, filed Jul. 10, 2023, 16 pp. [cited by applicant]
Response to Office Action dated Aug. 1, 2023, from counterpart Japanese Application No. 2021-571577 filed Nov. 30, 2023, 8 pp. [cited by applicant]
Extended Search Report from counterpart European Application No. 20817696.6 dated Dec. 16, 2022, 13 pp. [cited by applicant]
Notification of Reason for Refusal, and translation thereof, from counterpart Japanese Application No. 2021-571577 dated Jan. 10, 2023, 8 pp. [cited by applicant]
Vlannes et al., “Hybrid optical/electronic nonlinear optical devices”, Optical Engineering, vol. 31, No. 1, Jan. 1992, pp. 121-133. [cited by applicant]