IP Library › Granted Patent US 12,326,644
Granted Patent B2
US 12,326,644 · App. 17/963,027 · Granted Jun 10, 2025

Reconfigurable all-optical nonlinear activation functions on silicon-integrated platform

Inventors: Yiwei Peng (Houston, TX); Yuan Yuan (Milpitas, CA); Stanley Cheung (Milpitas, CA)
Assignee: Hewlett Packard Enterprise Development LP
G02F1/2257G02F1/212G02F1/365G06N3/048G06N3/067G02F2203/15
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 12,326,644
App. No.
17/963,027
Granted
Jun 10, 2025
Kind
B2
Abstract

Systems, devices, and methods are provided for all-optical reconfigurable activation devices for realizing various activations functions using low input optical power. The device and systems disclosed herein include a directional coupler comprising a first phase-shift mechanism and an interferometer coupled to the directional coupler. The interferometer comprises at least one microring resonator and a second phase-shift mechanism coupled to thereto. The interferometer and the directional coupler comprise waveguides formed of a first material, while the microring resonator comprises a waveguide formed of a second material and a third phase-shift mechanism. The second material is provided as a low-loss material having a high Kerr effect and large bandgaps, to generate various nonlinear activation functions. The first, second, and third phase-shift mechanisms are configured to control biases within the disclosed systems and devices to achieve a desired activation function.

Claims (40)

1. An optical device comprising:

a directional coupler comprising a first phase-shift mechanism;

an interferometer coupled to the directional coupler, the interferometer comprising:

a first branch comprising a first waveguide formed of a first material and a second waveguide formed of a second material dissimilar to the first material,

a second branch comprising a third waveguide formed of the first material, and

a second phase-shift mechanism coupled to the first waveguide; and

a microring resonator coupled to the second waveguide, the microring resonator formed of the second material and comprising a third phase-shift mechanism,

wherein the first, second, and third phase-shift mechanisms are configured to control biases of the optical device to achieve a desired activation function.

2. The optical device of claim 1 , wherein the second material is a low loss material having a linear loss coefficient that is less than 200 dB/m.

3. The optical device of claim 1 , wherein the second material has a bandgap that is greater than 1.9 eV.

4. The optical device of claim 1 , wherein the second material has a nonlinear refractive index that is greater than 3×10 −20 m 2 /W.

5. The optical device of claim 1 , wherein the second material is aluminum gallium arsenide (AlGaAs).

6. The optical device of claim 1 , wherein the second material is tantalum pentoxide (Ta 2 O 5 ).

7. The optical device of claim 1 , wherein a ratio of a nonlinear refractive index to a linear loss coefficient of the second material is larger than 3×10 −20 .

8. The optical device of claim 1 , wherein the first waveguide comprises a junction end coupled to a junction end of the second waveguide, the optical device further comprising:

a first inverse taper pair comprising a first taper at the junction end of the first waveguide and a second taper at the junction end of the second waveguide, wherein the first taper is in an inverse direction relative to the second taper.

9. The optical device of claim 1 , wherein at least the third phase-shift mechanism comprises at least one of: a heterogeneous metal oxide semiconductor (MOS) phase shifter and a heater.

10. The optical device of claim 1 , wherein at least the third phase-shift mechanism comprises a heterogeneous metal oxide semiconductor (MOS) capacitor.

11. The optical device of claim 1 , wherein the directional coupler is a Mach-Zehnder coupler and the interferometer is a Mach-Zehnder interferometer.

12. A nonlinear activation device, comprising:

a Mach-Zehnder coupler (MZC) comprising a first phase-shift mechanism;

a Mach-Zehnder interferometer (MZI) coupled to the MZC, the MZI comprising a first branch that includes a first waveguide formed of a first material and a second waveguide formed of a second material that is dissimilar to the first material, and a second phase-shift mechanism coupled to the first waveguide; and

a microring resonator (MRR) coupled to the second waveguide, the microring resonator formed of the second material,

wherein the first and second phase-shift mechanisms are configured to control biases of the nonlinear activation device to switchably achieve a desired activation function,

wherein the second material comprises at least one of: (i) a bandgap over 1.9 eV; (ii) a nonlinear refractive index that is greater than 3×10 −20 m 2 /W, and (iii) a linear loss coefficient that is less than 200 dB/m.

13. The nonlinear activation device of claim 12 , wherein a ratio of a nonlinear refractive index to a linear loss coefficient of the second material is larger than 5×10 −20 .

14. The nonlinear activation device of claim 12 , wherein the second material is aluminum gallium arsenide (AlGaAs).

15. The nonlinear activation device of claim 12 , wherein the second material is tantalum pentoxide (Ta 2 O 5 ).

16. The nonlinear activation device of claim 12 , wherein the first waveguide comprises a first junction end coupled to a second junction end of the second waveguide, the optical device further comprising:

a first inverse taper pair comprising a first taper at the first junction end of the first waveguide and a second taper at the second junction end of the second waveguide, wherein the first taper is in an inverse direction relative to the second taper.

17. The nonlinear activation device of claim 16 , wherein the first waveguide comprises a third junction end coupled to a fourth junction end of the second waveguide, the optical device further comprising:

a second inverse taper pair comprising a third taper at the third junction end of the first waveguide and a fourth taper at the fourth junction end of the second waveguide, wherein the third taper is inverse relative to the fourth taper and inverse relative to the first taper.

18. A method for operating a nonlinear activation device, the method comprising:

adjusting a first bias of a Mach-Zehnder interferometer (MZI) included in the nonlinear activation device by controlling a first phase-shift mechanism of the MZI to tune a phase difference between branches of the MZI, wherein the first phase-shift mechanism is coupled to a first waveguide of a first branch of the MZI;

adjusting a second bias of a microring resonator (MRR), coupled to a second waveguide of the first branch of the MZI, such that the nonlinear activation device functions at approximately a resonance frequency of the MRR by controlling a second phase-shift mechanism of the MRR, the MRR is formed of a second material comprising at least one of: (i) a bandgap over 1.9 eV; (ii) a nonlinear refractive index that is greater than 3×10 −20 m 2 /W, and (iii) a linear loss coefficient that is less than 200 dB/m; and

adjusting a third bias of a Mach-Zehnder coupler (MZC), coupled to the MZI, to tune amplitudes of the branches of the MZI relative to one another by controlling a third phase-shift mechanism of the MZC,

wherein the first, second, and third bias are controlled to achieve a desired activation function, and

wherein the first waveguide is formed of a first material and the second waveguide is formed of the second material that is dissimilar to the first material.

19. The method of claim 18 , wherein the first, second, and third phase-shift mechanisms each comprise one of a heterogeneous metal oxide semiconductor (MOS) phase shifter or a heater.

20. The method of claim 18 , wherein the second material of the MRR is one of aluminum gallium arsenide (AlGaAs) and tantalum pentoxide (Ta 2 O 5 ).

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Oct 10, 2022
From: PENG, YIWEI; YUAN, YUAN; CHEUNG, STANLEY
To: HEWLETT PACKARD ENTERPRISE DEVELOPMENT LP
Reel/Frame 061368/0219 →
Continuity (1)
Related Publication 20240118582A1 · Apr 11, 2024
References Cited (43)
US 11187963B2 · Huang · 2021 [cited by examiner]
US 20040208421A1 · Kitagawa · 2004 [cited by examiner]
US 20120039560A1 · Mazur · 2012 [cited by examiner]
US 20200401012A1 · Xu · 2020 [cited by examiner]
US 20210278743A1 · Huang · 2021 [cited by examiner]
US 20220229316A1 · Cohen · 2022 [cited by examiner]
US 20240120706A1 · Papp · 2024 [cited by examiner]
US 20240231180A9 · Yuan · 2024 [cited by examiner]
US 20240289600A1 · Xiao · 2024 [cited by examiner]
US 20240311627A1 · Tossoun · 2024 [cited by examiner]
US 20250068032A1 · Karabchevsky · 2025 [cited by examiner]
John E. Heebner and Robert W. Boyd, “Enhanced all-optical switching by use of a nonlinear fiber ring resonator,” Opt. Lett. 24, 847-849 (1999), 3 pages. (Year: 1999). [cited by examiner]
D. Liang et al., “A Tunable Hybrid III-V-on-Si MOS Microring Resonator with Negligible Tuning Power Consumption,” in Optical Fiber Communication Conference, OSA Technical Digest (online) (Optica Publishing Group, 2016),… [cited by examiner]
Y. Zuo et al., “All-optical neural network with nonlinear activation functions,” Optica 6, 1132-1137 (2019), 6 pages. (Year: 2019). [cited by examiner]
C. Huang et al., “On-Chip Programmable Nonlinear Optical Signal Processor and Its Applications,” IEEE Journal of Selected Topics in Quantum Electronics 27(2), 1-11 (Mar.-Apr. 2021) doi:10.1109/JSTQE.2020.2998073, 11 pag… [cited by examiner]
“Artificial Neural Network Market Research Report by Component, Organization Size, Application, Deployment Mode, Industry Vertical, Region—Global Forecast to 2027—Cumulative Impact of COVID-19”, available online at <htt… [cited by applicant]
Batta Mahesh, “Machine Learning Algorithms—A Review”, International Journal of Science and Research (IJSR), 2018, 7 pages. [cited by applicant]
Belt et al., “Ultra-low-loss Ta 2 O 5-core/SiO 2-clad planar waveguides on Si substrates”, vol. 4, No. 5, May 2017, pp. 532-536. [cited by applicant]
Cheung et al., “Comparison of Al2O3 and HfO2 MOSCAP III-V/Si Power Splitters and (De-) Interleavers for DWDM Optical Links”, Optica Publishing Group, 2022, 3 pages. [cited by applicant]
Chiu et al., “Nonlinear and low-loss tantalum pentoxide based micro-ring resonator by ion-assisted electron-beam deposition”, vol. 3, No. 12, 2020, 12 pages. [cited by applicant]
Christian et al., “Low-loss plasmon-assisted electro-optic modulator”, ETH Library, 2018, 12 pages. [cited by applicant]
Du et al., “On the Power of Over-parametrization in Neural Networks with Quadratic Activation”, In International conference on machine learning, 2018, pp. 1329-1338. [cited by applicant]
Fard et al., “Experimental realization of arbitrary activation functions for optical neural networks”, vol. 28, No. 8, Apr. 13, 2020, 11 pages. [cited by applicant]
Fathpour et al., “Heterogeneous Nonlinear Integrated Photonics”, IEEE Journal of Quantum Electronics, vol. 54, No. 6, Dec. 2018, pp. 6300716-6300716. [cited by applicant]
Feldmann et al., “All-optical spiking neurosynaptic networks with self-learning capabilities”, Nature 569, No. 7755, Feb. 2021, 26 pages. [cited by applicant]
Jean et al., “Universal micro-trench resonators for monolithic integration with silicon waveguides”, vol. 11, No. 9, Sep. 1, 2021, pp. 2753-2767. [cited by applicant]
Jha et al., “Programmable, high-speed all-optical nonlinear activation functions for neuromorphic photonics”, IEEE, 2021, 3 pages. [cited by applicant]
Jha et al., “Reconfigurable all-optical nonlinear activation functions for neuromorphic photonics”, vol. 45, No. 17, Sep. 2020, 4 pages. [cited by applicant]
Jung et al., “Tantala Kerr nonlinear integrated photonics”, vol. 8, No. 6, Jun. 2021, pp. 811-817. [cited by applicant]
Kiani, F., et al., “A fully hardware-based memristive multilayer neural network”, Nov. 24, 2021, 10 pages. [cited by applicant]
Kurczveil et al., “Hybrid Silicon Quantum Dot Comb Laser with Record Wide Comb Width”, Frontiers in Optics / Laser Science, 2020, 2 pages. [cited by applicant]
Miscuglio et al., “All-optical nonlinear activation function for photonic neural networks [Invited]”, vol. 8, No. 12, Dec. 1, 2018, 13 pages. [cited by applicant]
Miscuglio et al., “Roadmap on Material-Function Mapping for PhotonicElectronic Hybrid Neural Networks”, Oct. 10, 2019; 22 pages. [cited by applicant]
Nahmias et al., “A Leaky Integrate-and-Fire Laser Neuron for Ultrafast Cognitive Computing”, IEEE Journal of Selected Topics in Quantum Electronics, vol. 19, No. 5, Sep./Oct. 2013, pp. 1800212-1800212. [cited by applicant]
Pérez et al., “Multipurpose silicon photonics signal processor core”, Nature Communications, 2017, 10 pages. [cited by applicant]
Pu et al., “Ultra-Efficient and Broadband Nonlinear AlGaAs-on-Insulator Chip for Low-Power Optical Signal Processing”, DTU, Jul. 28, 2022, 34 pages. [cited by applicant]
Shastri et al., “Principles of Neuromorphic Photonics”, 2017, 28 pages. [cited by applicant]
Tait et al., “A silicon photonic modulator neuron”, Physical Review Applied 11, No. 6, 2019, 16 pages. [cited by applicant]
Totovic et al., “Femtojoule per MAC Neuromorphic Photonics: An Energy and Technology Roadmap”, 2019, 17 pages. [cited by applicant]
Williamson et al., “Reprogrammable Electro-Optic Nonlinear Activation Functions for Optical Neural Networks”, IEEE, 2019, 12 pages. [cited by applicant]
Woods et al., “Supercontinuum generation in tantalum pentoxide waveguides for pump wavelengths in the 900 nm to 1500 nm spectral region”, vol. 28, No. 21, Optics Express , 2020, 12 pages. [cited by applicant]
Wu et al., “Low-threshold all-optical nonlinear activation function based on a Ge/Si hybrid structure in a microring resonator”, Optical Materials Express 12, No. 3 , 2022, 11 pages. [cited by applicant]
Xie et al., “Silicon-integrated nonlinear III-V photonics”, vol. 10, No. 2, Feb. 2022, pp. 535-541. [cited by applicant]
Cited By (1)
US 12,607,887