中科微点-全球专利检索

  1. Login
  2. Sign Up

Search Results

23 records (took 0.017 seconds)

    Polariton-Stabilized Solid-State Spin Clock
    1.
    发明申请
    Polariton-Stabilized Solid-State Spin Clock 有权

    公开(公告)号：US20220197225A1

    公开(公告)日：2022-06-23

    申请号：US17504238

    申请日：2021-10-18

    Applicant: Matthew Edwin TRUSHEIM ,  Kurt JACOBS ,  Jonathan HOFFMAN ,  Donald FAHEY ,  Dirk Robert ENGLUND

    Inventor： Matthew Edwin TRUSHEIM ,  Kurt JACOBS ,  Jonathan HOFFMAN ,  Donald FAHEY ,  Dirk Robert ENGLUND

    IPC: G04F5/14 ,  H03L7/26

    Abstract: An ensemble of spin defect centers or other atom-like quantum systems in a solid-state host can be used as a compact alternative for an atomic clock thanks to an architecture that overcomes magnetic and temperature-induced systematics. A polariton-stabilized solid-state spin clock hybridizes a microwave resonator with a magnetic-field-insensitive spin transition within the ground state of a spin defect center (e.g., a nitrogen vacancy center in diamond). Detailed numerical and analytical modeling of this polariton-stabilized solid-state spin clock indicates a potential fractional frequency instability below 10-13 over a 1-second measurement time, assuming present-day experimental parameters. This stability is a significant improvement over the state-of-the-art in miniaturized atomic vapor clocks.

    Slot Antennas for Graphene Mid-IR Imaging Arrays as well an Approach for CMOS Implementation Thereof
    3.
    发明申请
    Slot Antennas for Graphene Mid-IR Imaging Arrays as well an Approach for CMOS Implementation Thereof 审中-公开

    公开(公告)号：US20200295075A1

    公开(公告)日：2020-09-17

    申请号：US16684917

    申请日：2019-11-15

    Applicant: Jordan Goldstein ,  Dirk Robert ENGLUND

    Inventor： Jordan Goldstein ,  Dirk Robert ENGLUND

    IPC: H01L27/146 ,  H04N5/33 ,  H01L29/16 ,  H01Q21/06

    Abstract: A filter-based color imaging array that resolves N different colors detects only 1/Nth of the incoming light. In the thermal infrared wavelength range, filtering loss is exacerbated by the lower sensor detectivity at infrared wavelengths than at visible wavelengths. To avoid loss due to filtering, most spectral imagers use bulky optics, such as diffraction gratings or Fourier transform interferometers, to resolve different colors. Fortunately, it is possible to avoid filtering loss without bulky optics: detect light with interleaved arrays of sub-wavelength-spaced antennas tuned to different wavelengths. An optically sensitive element inside each antenna absorbs light at the antenna's resonant wavelength. Metallic slot antennas offer high efficiency, intrinsic unidirectionality, and lower cross-talk than dipole or bowtie antennas. Graphene serves at the optically active material inside each antenna because its 2D nature makes it easily adaptable to this imager architecture.

    Scalable, Ultra-Low-Latency Photonic Tensor Processor
    4.
    发明申请
    Scalable, Ultra-Low-Latency Photonic Tensor Processor 有权

    公开(公告)号：US20220337333A1

    公开(公告)日：2022-10-20

    申请号：US17673268

    申请日：2022-02-16

    Applicant: Liane Sarah Beland Bernstein ,  Alexander Sludds ,  Dirk Robert ENGLUND

    Inventor： Liane Sarah Beland Bernstein ,  Alexander Sludds ,  Dirk Robert ENGLUND

    IPC: H04J14/02 ,  G06N3/067 ,  H04B10/54 ,  H04B10/61

    Abstract: Deep neural networks (DNNs) have become very popular in many areas, especially classification and prediction. However, as the number of neurons in the DNN increases to solve more complex problems, the DNN becomes limited by the latency and power consumption of existing hardware. A scalable, ultra-low latency photonic tensor processor can compute DNN layer outputs in a single shot. The processor includes free-space optics that perform passive optical copying and distribution of an input vector and integrated optoelectronics that implement passive weighting and the nonlinearity. An example of this processor classified the MNIST handwritten digit dataset (with an accuracy of 94%, which is close to the 96% ground truth accuracy). The processor can be scaled to perform near-exascale computing before hitting its fundamental throughput limit, which is set by the maximum optical bandwidth before significant loss of classification accuracy (determined experimentally).

    Error Correction for Programmable Photonics
    5.
    发明申请
    Error Correction for Programmable Photonics 有权

    公开(公告)号：US20220269972A1

    公开(公告)日：2022-08-25

    申请号：US17556033

    申请日：2021-12-20

    Applicant: Saumil Bandyopadhyay ,  Ryan HAMERLY ,  Dirk Robert ENGLUND

    Inventor： Saumil Bandyopadhyay ,  Ryan HAMERLY ,  Dirk Robert ENGLUND

    IPC: G06N10/70 ,  H04B10/70 ,  G02F1/21 ,  G06N10/40

    Abstract: Programmable photonic circuits of reconfigurable interferometers can be used to implement arbitrary operations on optical modes, providing a flexible platform for accelerating tasks in quantum simulation, signal processing, and artificial intelligence. A major obstacle to scaling up these systems is static fabrication error, where small component errors within each device accrue to produce significant errors within the circuit computation. Mitigating errors usually involves numerical optimization dependent on real-time feedback from the circuit, which can greatly limit the scalability of the hardware. Here, we present a resource-efficient, deterministic approach to correcting circuit errors by locally correcting hardware errors within individual optical gates. We apply our approach to simulations of large-scale optical neural networks and infinite impulse response filters implemented in programmable photonics, finding that they remain resilient to component error well beyond modern day process tolerances. Our error correction process can be used to scale up programmable photonics within current fabrication processes.

    APPARATUSES AND METHODS FOR INCREASING MAGNETIC FLUX DENSITY USING SUPERCONDUCTORS
    6.
    发明申请
    APPARATUSES AND METHODS FOR INCREASING MAGNETIC FLUX DENSITY USING SUPERCONDUCTORS 有权

    公开(公告)号：US20210057135A1

    公开(公告)日：2021-02-25

    申请号：US16907741

    申请日：2020-06-22

    Applicant: Hyeongrak CHOI ,  Dirk Robert ENGLUND

    Inventor： Hyeongrak CHOI ,  Dirk Robert ENGLUND

    IPC: H01F7/02 ,  H01B12/02

    Abstract: Using the Meissner effect in superconductors, demonstrated here is the capability to create an arbitrarily high magnetic flux density (also sometimes referred to as “flux squeezing”). This technique has immediate applications for numerous technologies. For example, it allows the generation of very large magnetic fields (e.g., exceeding 1 Tesla) for nuclear magnetic resonance (NMR), magnetic resonance imaging (MRI), the generation of controlled magnetic fields for advanced superconducting quantum computing devices, and/or the like. The magnetic field concentration/increased flux density approaches can be applied to both static magnetic fields (i.e., direct current (DC) magnetic fields) and time-varying magnetic fields (i.e., alternating current (AC) magnetic fields) up to microwave frequencies.

    Low-Power Edge Computing with Optical Neural Networks via WDM Weight Broadcasting
    9.
    发明公开
    Low-Power Edge Computing with Optical Neural Networks via WDM Weight Broadcasting 审中-公开

    公开(公告)号：US20230274156A1

    公开(公告)日：2023-08-31

    申请号：US18247129

    申请日：2021-07-29

    Applicant: Ryan HAMERLY ,  Dirk Robert ENGLUND ,  Massachusetts Institute of Technology

    Inventor： Ryan HAMERLY ,  Dirk Robert ENGLUND

    IPC: G06N3/098 ,  G06N5/04

    CPC classification number: G06N3/098 ,  G06N5/04

    Abstract: NetCast is an optical neural network architecture that circumvents constraints on deep neural network (DNN) inference at the edge. Many DNNs have weight matrices that are too large to run on edge processors, leading to limitations on DNN inference at the edge or bandwidth bottlenecks between the edge and server that hosts the DNN. With NetCast, a weight server stores the DNN weight matrix in local memory, modulates the weights onto different spectral channels of an optical carrier, and distributes the weights to one or more clients via optical links. Each client stores the activations, or layer inputs, for the DNN and computes the matrix-vector product of those activations with the weights from the weight server in the optical domain. This multiplication can be performed coherently by interfering the spectrally multiplexed weights with spectrally multiplexed activations or incoherently by modulating the weight signal from the weight server with the activations.

    Large-Scale Artificial Neural-Network Accelerators Based on Coherent Detection and Optical Data Fan-Out
    10.
    发明申请
    Large-Scale Artificial Neural-Network Accelerators Based on Coherent Detection and Optical Data Fan-Out 有权

    公开(公告)号：US20210357737A1

    公开(公告)日：2021-11-18

    申请号：US16681284

    申请日：2019-11-12

    Applicant: Ryan Hamerly ,  Dirk Robert ENGLUND

    Inventor： Ryan Hamerly ,  Dirk Robert ENGLUND

    IPC: G06N3/067 ,  G06N3/04

    Abstract: Deep learning performance is limited by computing power, which is in turn limited by energy consumption. Optics can make neural networks faster and more efficient, but current schemes suffer from limited connectivity and the large footprint of low-loss nanophotonic devices. Our optical neural network architecture addresses these problems using homodyne detection and optical data fan-out. It is scalable to large networks without sacrificing speed or consuming too much energy. It can perform inference and training and work with both fully connected and convolutional neural-network layers. In our architecture, each neural network layer operates on inputs and weights encoded onto optical pulse amplitudes. A homodyne detector computes the vector product of the inputs and weights. The nonlinear activation function is performed electronically on the output of this linear homodyne detection step. Optical modulators combine the outputs from the nonlinear activation function and encode them onto optical pulses input into the next layer.

Patent Agency Ranking