262results about "Optical computing devices" patented technology

The invention relates to the technical field of edge computing, and discloses an intelligent edge computing co-processing system based on an integrated circuit, which comprises a heterogeneous computing cluster module, a hardware computing unit set, a multi-core control processor based on RISC-V, a programmable pulse tensor computing array and a reconfigurable engine oriented to streaming processing, according to the intelligent edge computing cooperative processing system based on the integrated circuit, zero-delay data exchange is realized through silicon intermediate layer integration of the heterogeneous computing cluster module and a snakelike data channel, and traditional bus arbitration delay is eliminated through real-time operation code analysis and optimal optical communication path mapping of the hardware task routing matrix (TRF) module; the priority path distribution of the task scheduling subsystem is synchronously coordinated with the time-sensitive task, so that the collaborative efficiency of the computing unit is improved in multiple dimensions, the non-blocking transmission of the high-priority task is ensured, and the effect of enhancing the overall processing capability and response speed of the intelligent edge computing is achieved.

The invention provides a photon parallel matrix convolution operation chip which can be applied to the technical field of integrated photonics and optical calculation. The chip is integrated by a 1 * N array waveguide grating, a 1 * 2 beam splitter array, an input matrix modulator array, two N * 1 array waveguide gratings, two 1 * N beam splitters, a weight matrix modulator array, an M * M array waveguide grating detector array and a detector array. Main functional devices are integrated in a monolithic mode, the system is compact, the complexity and power consumption of the system can be effectively reduced, and the calculation efficiency is improved.

The invention discloses an optical computing array based on micro-ring resonators without a free spectral range, which comprises an on-chip wavelength division multiplexing beam splitting array, an on-chip intensity modulation array, two groups of on-chip wavelength division multiplexing beam combining arrays and an on-chip balance detector, the on-chip wavelength division multiplexing beam splitting array and the on-chip wavelength division multiplexing beam combining array are respectively composed of a group of multiple FSR-free MRRs which share the same bus optical waveguide, each resonator corresponds to an independent wavelength channel, and the on-chip intensity modulator array is composed of a plurality of groups of on-chip MZI modulator arrays. According to the optical calculation array implementation method, independent control and coding calculation of different wavelength channels in a broadband wavelength range are realized by utilizing the single harmonic peak characteristic of the FSR-free MRR array and the wide-spectrum intensity modulation function of the MZI unit, the limitation of a free spectrum range on the optical calculation wavelength channel is eliminated, the higher wavelength parallelism is achieved, and the optical calculation array implementation method has the advantages of being simple in structure and convenient to implement. And therefore, the calculation scale is remarkably improved.

A calculation device of the present disclosure is characterized by implementing an item, which expresses a magnetic field, at each site by adding, to the conventional coherent Ising machine, a function for simulating the magnetic field item. Specifically, light having the amplitude of a predetermined sign is input in the conventional coherent Ising machine by dislocating an operation point of the push-pull-type optical modulator. When the operation point is dislocated to a + direction, DOPO (Degenerate Optical Parametric Oscillator) having the same wavelength as that of DOPO of an optical resonator and having a fixed initial phase difference from pump light into which the light is injected considerably tends to oscillate at 0 phase, and becomes reversed when the operation point is dislocated to a minus direction.

The invention discloses an optical calculation error adjustment method and system, a medium and a program product, and relates to the technical field of optical calculation, and the method comprises the steps: determining a unified correction deviation value corresponding to the phase noise of an interferometer array; a target function based on a correction deviation direction variable is generated with the minimum difference sum of squares between a measured light intensity value of the photoelectric detector array and an ideal light intensity value as a target; solving the objective function to obtain the optimal value of the correction deviation direction variable of each phase shifter; and sending the correction deviation direction variable optimal value and the unified correction deviation value of each phase shifter to control equipment, so that the control equipment adjusts each phase shifter at the same time. In this way, collective adjustment of all the phase shifters can be realized through the determined unified correction deviation value and in combination with the correction deviation direction variable optimal value of each phase shifter, extra overhead caused by independent calibration is avoided, complexity of light calculation noise mitigation is reduced, a noise mitigation process is accelerated, and output precision of light calculation equipment is improved.

The invention provides an optical signal synchronous calibration device and method of an optical computing system. The device comprises an optical probe array, an AI coprocessor, an optical path tuning engine and a low-speed electric feedback line, wherein the optical probe array is arranged on a chip; the optical probe array detects optical signals of wavelength channels corresponding to the plurality of optical calculation units and converts the optical signals into electric signals; the AI coprocessor drives a corresponding optical calculation unit to emit a detection pulse according to a pre-distributed time slot index variable, extracts a phase error from the electric signal and calculates a final calibration amount based on the phase error; the optical path tuning engine performs an optical path tuning operation based on the final calibration amount to calibrate an error of the corresponding optical calculation unit; and the low-speed electric feedback line transmits the electric signal to the AI coprocessor and transmits the detection pulse to the chip. According to the invention, through cooperation of spatial distributed detection and board-level centralized processing, real-time calibration in a dynamic environment is realized.

Systems and methods for increasing throughput of a photonic processor by using photonic degrees of freedom (DOF) are provided. The photonic processor includes a multiplexer configured to multiplex, using at least one photonic DOF, multiple encoded optical signals into a multiplexed optical signal. The photonic processor also includes a detector coupled to an output of an optical path including the multiplexer, the detector being configured to generate a first current based on the multiplexed optical signal or a demultiplexed portion of the multiplexed optical signal. The photonic processor further includes a modulator coupled to and output of the detector, the modulator being configured to generate a second current by modulating the first current.

An optical computing device includes an optical computing section including a first transmissive optical modulation element having cells each having an independently set phase modulation amount. The first transmissive optical modulation element emits a signal light, in a signal direction, generated when transmitted light beams phase-modulated by the cells interfere with each other and a noise light, in a noise direction, transmitted through the first transmissive optical modulation element without being phase-modulated by the cells. The optical computing device includes an optical sensor that detects the signal light outputted from the optical computing section and generates an electrical signal indicating a result of the detection. The first transmissive optical modulation element is configured such that an impact that the noise light has on the electrical signal is smaller than the impact in a state where the signal direction aligns with the noise direction.

An optical computing device includes: one or more light-diffraction elements each of which includes microcells, wherein each of the microcells has an individually set thickness or refractive index; and an optical signal input section that simultaneously inputs an optical signal and a delayed optical signal obtained by delaying the optical signal to the one or more light-diffraction elements.

The invention relates to a reconfigurable metasurface analog calculator based on a lithium niobate film. The reconfigurable metasurface analog calculator comprises a substrate and a lithium niobate unit, the plurality of lithium niobate units are periodically arranged on the substrate to form a metasurface structure, and the cross section of each lithium niobate unit is an isosceles trapezoid; incident light vertically enters the metasurface structure, and the transmission coefficient and phase distribution of the metasurface structure are changed by changing external bias voltage applied to the lithium niobate units, so that conversion and reconstruction of a mathematical operation function are realized; the mathematical operation comprises first-order differential operation, second-order differential operation and integral operation. According to the invention, different external bias voltages are applied to the periodically arranged lithium niobate units, so that the transmission coefficient and phase distribution of incident light by the metasurface structure can be accurately regulated and controlled, and the same metasurface device can be flexibly switched among different mathematical operation functions such as first-order differential, second-order differential and integral. And various calculation functions can be realized without changing a physical structure.

An optical processing system comprises an Optical Fourier transform stage; and one or more photodetectors for receiving a reference optical signal to provide currents and / or voltages relating to the intensities and / or phases of the reference optical signal; the system further comprising an electronics feedback loop which feeds back the currents and / or voltages and modulates the intensities and / or phases of a subsequent optical signal.

The exponential growth in deep learning models is challenging existing computing hardware. Optical neural networks (ONNs) accelerate machine learning tasks with potentially ultrahigh bandwidth and nearly no loss in data movement. Scaling up ONNs involves improving scalability, energy efficiency, compute density, and inline nonlinearity. However, realizing all these criteria remains an unsolved challenge. Here, we demonstrate a three-dimensional spatial time-multiplexed ONN architecture based on dense arrays of microscale vertical cavity surface emitting lasers (VCSELs). The VCSELs, coherently injection-locked to a leader laser, operate at gigahertz data rates with a 7T-phase-shift voltage on the 10-millivolt level. Optical nonlinearity is incorporated into the ONN with no added energy cost using coherent detection of optical interference between VCSELs.

Described herein are embodiments of a photonic computing system comprising one or more processors in communication with disaggregated memory through one or more optical channels. The disaggregated memory comprises multiple memory units placed on a photonic substrate that includes a photonic network that can be programmed to configure which of the memory units can be accessed by each of the processor(s). The disaggregated memory includes a memory controller for reading and writing data to / from the memory units. The memory controller may be configured to perform processing in concert with the processor(s).

The invention relates to a method for matrix multiplication and / or vector multiplication, in particular for matrix-vector multiplication, with features of claim 1 and a device (30) for matrix multiplication and / or vector multiplication, in particular for matrix-vector multiplication, with features of the dependent claim.

The invention discloses an optical calculation chip matrix weight control precision measurement and calibration device and method, and the device mainly comprises a photoelectric detector array module which is used for carrying out the photoelectric signal conversion of an optical signal outputted by an optical calculation chip, a parameter extraction and analysis module which is used for collecting an electric signal, and a calibration module. Calculating the matrix weight control precision and the control rate of the optical calculation chip; and the environment compensation and closed-loop control module is used for monitoring the environment temperature of the optical computing chip and performing compensation calibration control on the bias voltage of the optical computing chip based on the change of the temperature so as to realize accurate regulation and control on the matrix weight. According to the device and the method, the matrix weight control precision and the control rate can be accurately measured, external environment change parameters are accurately sensed, voltage parameters in weight control are changed through a compensation algorithm, and high performance and reliable operation of an optical calculation chip are effectively guaranteed.

The invention belongs to the technical field of semiconductor optoelectronic devices, and particularly relates to an optical computing array based on a reverse bias germanium modulator, which comprises a cross matrix formed by crossing a plurality of rows of mutually parallel input optical waveguides and a plurality of columns of mutually parallel output optical waveguides, a photoelectric detector is arranged at the output end of each row of output optical waveguides; a plurality of optical splitters; each germanium modulator corresponds to one crossed waveguide, and the first input end of each germanium modulator is connected with the output end of the optical splitter through a transmission optical waveguide; each light combiner is connected with the output end of the germanium modulator and the output optical waveguide, through the above structure, the computing array which has a thermal compensation mechanism, can reduce light reflection and is high in responsivity is comprehensively provided, and therefore under the condition that frequent switching is met, the computing array has the advantages that the computing efficiency is improved, and the computing efficiency is improved. And a low-power-consumption large-scale integrated calculation array scheme is realized.

The application discloses an optoelectronic computer based on time domain all-optical signal serial-parallel hybrid computing, which is composed of an electrical transceiving module, a light source module, a linear matrix vector computing module, a summation separation module and a nonlinear activation module.The application adopts time sequence encoding serial input, and multiple input signals are simultaneously and parallelly calculated, so that all computing steps of a full connection neural network, i.e., linear calculation, summation operation and nonlinear activation, can be completed in the optical domain, and the host end only needs to send input data and receive the final computing result.The application has the following advantages compared with other comparative applications: multiple matrix-vector multiplication operation can be completed in parallel; only one optical element is used to complete summation and nonlinear activation operation on multiple multiplication vectors, thereby reducing the hardware deployment pressure; and the bandwidth loss of a pure serial time domain computing method is relieved through serial-parallel hybrid encoding in the time domain, and the computer computing speed is improved.

A light diffraction element unit includes an optically-transparent substrate; a light diffraction element disposed on a main surface of the optically-transparent substrate; and a three-dimensional alignment mark disposed on the main surface in a vicinity of the light diffraction element and having a thickness greater than a thickness of the light diffraction element.

This invention relates to an optical Ising calculation method and system based on dual digital micromirror devices (DMDs). First, the combinatorial optimization problem is mapped to an Ising model. Multiple mask vectors are constructed, and each mask vector is decomposed into multiple binary bit planes. A spin configuration matrix is ​​constructed and loaded into DMD-A. For each mask vector, its binary bit planes are sequentially loaded into DMD-B. The light intensity modulated by the two DMDs is collected and processed. The optical inner product is obtained by combining the bit-weighted components, and then the energy contribution value of the mask vector is calculated. The total Hamiltonian is obtained by summing all contributions. The energy difference is calculated by randomly flipping the spins and repeating the above steps. The probability of accepting the current flip is calculated based on the energy difference. This process is iterated until the maximum number of iterations is reached to obtain the final solution. The system is used to implement the above method. Compared with the prior art, this invention can achieve a high-speed, stable, and coherently independent optical Ising solution method.

A photoelectric computing unit, a photoelectric computing array and a photoelectric computing method. The photoelectric computing unit comprise a semiconductor multifunctional region structure, which comprises at least one carrier control region, at least one coupling region, and at least one photon-generated carrier collection region and readout region.

A computer-implemented method includes a computer-implemented method can include receiving telematics data comprising information related to one or more driving behaviors of a user. The telematics data is received from one or more sensors located in a vehicle when the user is operating the vehicle. The method also can include determining one or more user characteristics of the user based at least on the telematics data. The method additionally can include selecting an advertisement from a plurality of advertisements based at least on the one or more user characteristics. Other embodiments are described.

The invention discloses a method for optically realizing a restricted Boltzmann machine. According to the invention, analog calculation of the restricted Boltzmann machine and optical Gibbs sampling can be realized, and the system has the advantages of wide application range, fast information transmission, simple structure, low cost, fast calculation and the like. According to the invention, a coherent wide-spectrum light source is used as signal input, a light field is subjected to light splitting by using a light splitting device such as a grating and then enters a modulator, then spinning, interaction and magnetic field parameters are coded on light wavefront at different positions through a time or space light modulator, and optical Fourier transform is carried out by using a time or space lens system, so that the optical field is obtained. The light intensity after Fourier transform is measured through the detector, and finally the difference between the light intensity measured twice is calculated, so that optical Gibbs sampling is realized, the calculation complexity is reduced, and the calculation efficiency is improved. The method provided by the invention has important application prospects in the fields of optical neural networks and the like, can realize applications of content generation, classification and identification and the like, and is convenient to integrate in optical chips and the like.

The application discloses an optical multiplier system for processing high-speed data flow and a method thereof, and belongs to the technical field of semiconductors, comprising a stibonium quantum well laser, an intensity modulator, a phase modulator, a balanced homodyne detector and an oscilloscope; the stibonium quantum well laser is used as a light source, light signals output by the stibonium quantum well laser are evenly divided into two light beams by a Y-type beam splitter, and the two light beams are defined as x-vector light beams and y-vector light beams respectively; the two light beams sequentially pass through the intensity modulator and the phase modulator to complete real number and complex number coding; after the two light beams are coupled by a beam splitter / coupler, the two light beams are input to the balanced homodyne detector for detection, and the photocurrent output by the balanced homodyne detector is calculated to obtain a differential current, the differential current is collected and displayed in real time by the oscilloscope, and the differential current is the vector point multiplication operation result corresponding to the two vector light beams. The application has the characteristics of low cost, low energy consumption, high speed, expandability and the like.

Embodiments of the present disclosure are directed to a photonic implementation of a processor for message generation for digital currency (e.g., bitcoin) transactions. The processor includes an input photonic circuit and a message generation photonic circuit coupled to the input photonic circuit via a first set of optical connections. The input photonic circuit receives input data of a first size and splits the received input data into a plurality of input messages of a second size. The message generation photonic circuit receives the plurality of input messages from the input photonic circuit via the first set of optical connections, and generates a plurality of output messages of the second size based at least in part on the plurality of input messages.

The application provides an optical signal synchronous calibration device and method of an optical computing system, the device comprising: an optical probe array arranged on a chip, an AI coprocessor, an optical path tuning engine and a low-speed electrical feedback line arranged on a board card; the optical probe array detects optical signals of wavelength channels corresponding to a plurality of optical computing units and converts the optical signals into electrical signals; the AI coprocessor drives the corresponding optical computing unit to emit a detection pulse according to a pre-allocated time slot index variable, extracts a phase error from the electrical signal and calculates a final calibration quantity based on the phase error; the optical path tuning engine performs an optical path tuning operation based on the final calibration quantity to calibrate the error of the corresponding optical computing unit; and the low-speed electrical feedback line transmits the electrical signal to the AI coprocessor and transmits the detection pulse to the chip. The application realizes real-time calibration in a dynamic environment through spatial distributed detection and board-level centralized processing cooperation.

The invention discloses a photon data stream processing method for an image classification feature extraction stage, which is applied to the feature extraction stage of image classification, and realizes maximum multiplexing of an input feature map by partitioning and mapping the input feature map according to DNN convolutional layer parameters; the method comprises the following steps of: carrying out electro-optical conversion on an input feature block, modulating the input feature block into a multi-wavelength optical signal, and then carrying out wavelength division multiplexing (WDM) one-to-many broadcast distribution, so that full broadcast of the optical signal is realized; optical domain product accumulation operation is executed through the PCU, so that the throughput calculation speed is increased; and finally, obtaining an output characteristic pattern under the combined action of a photoelectric detector and an ADC (Analog to Digital Converter).

The embodiment of the invention provides a photon calculation chip, and the photon calculation chip comprises a multilayer diffraction neural network and a photoelectric detector, the multilayer diffraction neural network is used for carrying out the phase regulation of an input optical signal, and obtaining a mixed light field in an OAM mode; and the photoelectric detector is used for converting the mixed light field in the OAM mode into light currents adaptive to different ports. A mixed light field in an OAM mode is output after an optical signal is processed, and then the mixed light field is transmitted to a photoelectric detector for processing, so that information propagation and processing can be carried out by using the OAM degree of freedom of light, the information propagation and processing are not limited to traditional intensity, phase or polarization regulation and control, but by means of an inherent infinite-dimensional orthogonal space of the OAM mode, the information propagation and processing efficiency is greatly improved. And parallel transmission and high-dimensional coding with higher capacity are realized.