61 results about "Optical nonlinearity" patented technology

An all-optical neural network that utilizes light beams and optical components to implement layers of the neural network is disclosed herein. The all-optical neural network includes an input layer, zero or more hidden layers, and an output layer. Each layer of the neural network is configured to simulate linear and nonlinear operations of a conventional artificial neural network neuron on an optical signal. In an embodiment, the optical linear operation is performed by a spatial light modulator and an optical lens. The optical lens performs a Fourier transformation on the set of light beams and sums light beams with similar propagation orientations. The optical nonlinear operation is implemented utilizing a nonlinear optical medium having an electromagnetically induced transparency characteristic whose transmission of a probe beam of light is controlled by the intermediate output of a coupling beam of light from the optical linear operation.

A device for optically imaging a scene has at least one optical detector and an optical arrangement arranged in front of the optical detector, by means of which the scene can be imaged in a focal plane of the optical arrangement. The detector surface of the optical detector is arranged outside the focal plane and the optical arrangement has an amplitude or phase mask which enables improved algorithmic reconstruction of a sharp image of the scene from a blurred image acquired by the detector. In an alternative, the device additionally has an optically nonlinear filter arranged in the region of the focal plane which, when subjected to incident optical radiation with an intensity above a threshold intensity, changes its optical properties such that it blocks or at least attenuates this radiation.This device creates an optical sensor system that has an improved protective effect against interference radiation, in particular incident laser radiation, while maintaining almost unchanged size.

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.

An optical tomography device includes a sample optical path system for injecting light, scanned from a light source, into a sample corresponding to biological tissue; a reference optical path system for collecting the light scanned from the light source; and a detection unit for detecting a tomography image of the sample by interfering with sample light, transmitted from the sample optical path system and reflected from the sample, and reference light transmitted from the reference optical path system. The detection unit comprises: a nonlinear crystal which receives the sample light and the reference light and generates nonlinear light having a different wavelength than the sample light and the reference light; and an image sensor which detects a depth direction signal from the nonlinear light.

The application discloses a kind of diffractive light neural network computing systems and methods for realizing pure optical nonlinearity, belong to photonic computing and artificial intelligence hardware technical field.Its system includes coherent light source, linear light computing unit, nonlinear light activation unit and optical detection unit arranged in order along optical path, wherein linear light computing unit is programmable phase modulation using spatial light modulator, nonlinear light activation unit is programmable binary amplitude modulation using digital micromirror device, and both are directly coupled to form pure optical computing path without photoelectric conversion on optical path.Its method trains neural network through customized loss function, drives nonlinear activation function output to binary convergence, and maps the parameters obtained by training into phase map and binary mask respectively and loads to hardware.The application realizes true all-optical nonlinear computation, with the advantages of high energy efficiency, extremely low delay, compact structure and strong parallel processing capability.

The invention discloses an optical nonlinear activator chip based on a graphene photoelectric device and a preparation method. Belongs to the technical field of photoelectric devices and optical signal processing. The chip comprises a detector optical path and a modulator optical path, a substrate is made of a standard SOI material, and the chip structurally comprises an optical coupling module, a waveguide, a graphene detector part and a graphene modulator part. The graphene detector and the modulator are integrated on the same chip, the detector is used for converting an input optical signal into a voltage signal, then the voltage signal is input into the modulator to modulate continuous light, and an optical signal after nonlinear conversion is output and used for a nonlinear activation unit of an optical neural network. Compared with a traditional nonlinear activator of a silicon-based integrated photon chip, the method gives full play to the advantages of excellent photoelectric property and easy integration of graphene, has the advantages of high integration level, high speed, low power consumption and the like, and has better compatibility with a silicon-based optoelectronic process.

The invention discloses a middle infrared band microstructure optical fiber polarization beam splitter, the polarization beam splitter comprises a centrosymmetric cladding and two fiber cores, the cladding and the fiber cores are composed of a substrate material GaS and air holes, and the cladding is located on the outer side of the fiber cores. The cladding is composed of first air circular holes and second air circular holes, and the plurality of first air circular holes are periodically arranged in a triangular lattice structure. The fiber core is composed of a first elliptical air hole, a second elliptical air hole, a third air round hole and a fourth air round hole, and the fiber core is located between the first elliptical air hole and the second elliptical air hole and is a solid core filled with a GaS substrate material. The length of the micro-structure optical fiber polarization beam splitter of the middle-infrared band is only 40 microns, and the extinction ratio at the wavelength of 4.42 microns can reach-70 dB; and single polarization output can be realized in a range of 4.29 [mu] m to 4.59 [mu] m. The device has the advantages of small size, high extinction ratio, large bandwidth and the like, adopts a single material, effectively reduces the manufacturing difficulty of the device, is easy to integrate and miniaturize, and has potential application value in the fields of mid-infrared optical fiber sensing, mid-infrared band communication, mid-infrared band environment detection, optical nonlinearity and the like.

The present invention provides an on-chip photonic circuit for generation of a polarization-entangled photon-pair state by utilizing the optical nonlinearities in a novel configuration comprising dual micro-resonators, with nearly degenerate transverse-electric- and transverse-magnetic-polarized cavity resonances for the two resonators coupled in series to a common bus waveguide, wherein, the pump laser and the generated signal and idle photons can be coupled in and out the photonic circuit.

An optical parametric multi-pass apparatus for generating laser radiation through optical parametric amplification or generation converts pump radiation into signal and idler radiation. It includes a multi-pass device with multiple mirror elements arranged to provide several beam paths from an input section to an output section of the device. The input section receives pump radiation, and the output section delivers signal radiation as the generated laser output. At least one optically nonlinear crystal is positioned within the device to enable the parametric process using the pump radiation. The mirror configuration ensures multiple passes of the beam paths through the crystal. The device is further designed to filter along the beam paths so that idler radiation is suppressed after at least one pass. A corresponding laser apparatus and method for generating laser radiation are also described.

An optical nonlinearity measurement method according to the present disclosure utilizes photon pair generation through a spontaneous four-wave mixing process, to observe photon pairs using an optical waveguide loaded with a two-dimensional material. Compared with the Z-scan method, the influence of free carriers on nonlinear refractive indexes is only indirect. With a parameter being the length of the attached two-dimensional material in the optical waveguide direction, a theoretical value of the coincidence rate of the photon pairs based on the coupled wave equation is fitted to a measured value of the coincidence rate of the photon pairs. For the coincidence rate of the photon pairs, the theoretical value based on the coupled wave equation is fitted to the measured value in a state reflecting the structure of the optical waveguide loaded with the two-dimensional material, and nonlinear coefficients γ1 and γ2 at that time are obtained.

The invention relates to an all-optical nonlinear neural network system based on a linear system, and belongs to the technical field of optical neural networks. The bottleneck problem that an optical neural network depends on a nonlinear material or photoelectric conversion to achieve nonlinear operation is solved. According to the technical scheme, an input module loads an input signal to the resonant wavelength detuning amount of an annular resonant cavity; the coupling module regulates and controls the interlayer weight; the neuron module linearly transmits signals through the multi-layer annular resonant cavity; and the output module loads the output signal to the detection optical power. Input and output signals are loaded on different physical quantities, so that a linear system realizes a nonlinear function. The method has the beneficial effects that the bottleneck of optical nonlinear operation is broken through, the system complexity and energy consumption are remarkably reduced, the compatibility and practicability are enhanced, and the parallel computing efficiency and the system expandability are improved.

An optical logic gate is provided which comprises signal providing means for providing a first optical signal and a second optical signal having the same phases. The optical logic gate further comprises phase modulating means and interference means. The optical logic gate also comprises an optically non-linear element configured to interact with a pump signal and with the first and second optical signals shifted with respect to their phases by the phase modulating means and caused to interfere by the interference means and, as a result of this non-linear interaction, to couple out an optical output signal as a logic output signal.

The present invention provides scalable manufacturing approaches to add compound semiconductor (CS) materials and devices to large-diameter substrate platforms. The method includes forming a CS buffer layer overlying the large-diameter substrate wafer, then a stop etch layer, and finally a device layer structure for high carrier mobility, optical gain or absorption, or optical nonlinearity. This device wafer is then bonded, topside down, to a photonics substrate platform. The underlying substrate is removed with grinding, polishing, or etching processes. Some of the CS layers, including the buffer and the stop etch layer, may be removed thereby leaving only the final device layer. This final layer may be patterned to form waveguides or device structures; may be exposed to ion implantation steps to form p-n junctions; and may be subjected to other common fabrication steps to form devices. These device structures may also be encapsulated and addressed through top-level metal contacts and vias.

The embodiment of the invention provides a frequency conversion device based on a lithium niobate ridge waveguide on an insulator and a preparation method, and belongs to the field of integrated optical, nonlinear optical and optoelectronic devices. The device comprises a silicon substrate layer, a silicon dioxide buffer layer, an x-cut lithium niobate film layer and a silicon dioxide cladding from bottom to top, a ridge waveguide with width periodically changing is arranged on the x-cut lithium niobate layer, and light in the waveguide is propagated in the y-axis direction. The waveguide with the width periodically changing is prepared on the lithium niobate film, quasi-phase matching is achieved, and therefore the second harmonic conversion efficiency is greatly improved. By changing the period of the waveguide, second-order nonlinear phase matching conditions under different wavelengths, such as phase matching conditions of second harmonic generation, sum frequency generation and difference frequency generation, can be met. Correspondingly, according to the preparation method, the waveguide structure with the periodically changed width can be formed through a single etching process, and a complex ferroelectric domain inversion step is not needed. The preparation method has the advantages of simple process, controllable morphology and flexible design.

The invention discloses a non-reciprocal light transmission device based on optical nonlinearity and PT symmetry principles, and relates to the technical field of nonlinear photonics. The low-refractive-index broadband low-pass filter comprises a low-refractive-index dielectric interlayer, a first rectangular metal cavity and a second rectangular metal cavity are formed in the two sides of the low-refractive-index dielectric interlayer respectively, a loss dielectric block is embedded in the first rectangular metal cavity, and a gain dielectric block is embedded in the second rectangular metal cavity. The gain coefficient of the gain medium block is greater than the loss coefficient of the loss medium block. According to the invention, the self-bias modulation of the system is realized only by changing the intensity of an optical signal without any external modulation mechanism, so that a relatively high forward and reverse transmission ratio is obtained; secondly, the unique design principle of PT symmetry is effectively utilized, an optical gain-loss material is introduced, a non-Hermite system is formed, lossless total transmission and directional reflection of light can be achieved under linear response, and effective non-reciprocal light transmission can be achieved under nonlinear response.

The invention provides an optical nonlinear activator based on a two-dimensional material-polymer composite film and a preparation method and application thereof, and relates to the technical field of optical nonlinearity. A polymer material and a two-dimensional material are mixed to obtain a two-dimensional material-polymer composite film, and the nonlinear activator is prepared; by utilizing the saturable absorption characteristic of the two-dimensional material-polymer composite film, the nonlinear operation of input and output is completed, a photoelectric conversion process is not needed, the polymer film can be arranged in an optical fiber connector, standard single-mode optical fibers are arranged at the two ends of the polymer film, the polymer film can be directly connected with an existing optical calculation chip, and the coupling structure is simple and easy to implement. The method has flexibility and compatibility, can be used as a nonlinear activation layer of an optical neural network, does not need a photoelectric conversion process, improves the response rate, and reduces the system energy consumption.

The application provides a double-microring coupling system for realizing optical nonlinear response, and relates to the technical field of microwave photonics. The double-microring coupling system comprises: a first microring waveguide; a second microring waveguide coupled with the first microring waveguide, wherein a light splitting ratio control module is arranged at the coupling position of the first microring waveguide and the second microring waveguide, and is used for determining the optical coupling coefficient between the first microring waveguide and the second microring waveguide; the first microring waveguide and the second microring waveguide are respectively provided with gain control modules and phase control modules; wherein the parameters of the gain control modules and the phase control modules are adjusted; when the difference between the net gain of the first microring waveguide and the net loss of the second microring waveguide is greater than or equal to the optical coupling coefficient, or when the difference between the net loss of the first microring waveguide and the net gain of the second microring waveguide is greater than or equal to the optical coupling coefficient, the optical nonlinear response is generated between the first microring waveguide and the second microring waveguide.

An optical logic gate is provided, comprising signal providing means (1.1, 1.2) for providing first and second optical signals having the same phase. The optical logic gate further comprises phase modulation means (2.1, 2.2) and interference means (3). The optical logic gate also comprises an optical nonlinear element (1.3) configured to interact with a pump signal, and to interact with the first and second optical signals whose phases have been shifted by the phase modulation means (2.1, 2.2) and interfered by the interference means (3), and to combine an optical output signal as a result of this nonlinear interaction as a logic output signal.

A method of creating a local oscillator light beam LO for a phase-resolved spectroscopy measurement comprises the steps of providing a first measuring light beam (1) and a second measuring light beam (2) being aligned to each other, creating the local oscillator light beam LO by an optical non-linear interaction of a first portion (1A) of the first measuring light beam (1) and a first portion (2A) of the second measuring light beam (2) in an optical nonlinear medium (20), and superimposing the local oscillator light beam LO, a second portion (1B) of the first measuring light beam (1) and a second portion (2B) of the second measuring light beam (2) with a predetermined mutual phase relationship, for providing a sample light beam (3) for the phase-resolved spectroscopy measurement. The local oscillator light beam LO and the second portions (1B, 2B) of the first and second measuring light beams (1, 2) are superimposed with a displaced Sagnac interferometer (10).

A laser system (1) comprises a laser radiation source (7) for providing a pulsed laser radiation (9), wherein the pulsed laser radiation comprises laser pulses (11) having a pulse energy in the range of 1 mJ to 100 J or in the range of 10 mJ to 1 J and a pulse duration in the range of 10 fs to 5 ps or in the range of 500 fs to 1.5 ps. The laser system (1) further comprises an optical system (3) for spectral broadening of the pulsed laser radiation (9). The optical system (3) comprises: - a first polarization setting optics (19) setting a circular polarization state (17B) of the pulsed laser radiation (9), and - a multi-pass cell (5) having at least two mirrors (25A, 25B) through which the pulsed laser radiation (9) passes forming a plurality of intermediate focus regions (29), wherein the multi-pass cell (5) is filled with a filling gas (5A) having an optical nonlinearity, wherein the filling gas (5A) enables spectral broadening of the pulsed laser radiation (9) in the intermediate focus regions (29). Further, a pressure of the filling gas (5A) is set in a pressure range in which there is an ionization behavior of the filling gas (5A) in the regime of multiphoton ionization, and a focus diameter (d) of the intermediate focus regions (29) is set such that the pulsed laser radiation (9) passes through the multi-pass cell (5) without ionization of the filling gas (5).

The present invention relates to an optical sensor apparatus (1) comprising: a laser light source (2) for providing laser light (L); an optical ring resonator (3) which is arranged to receive part of the laser light via a coupler (4), the ring resonator (3) having an optically nonlinear material and / or an optically absorbing material such that a resonance curve (31) of the ring resonator (3) as a function of the wavelength of the laser light (L) to be received is asymmetrical with respect to a resonance wavelength (32), the resonance curve (31) having a side (31a) which rises gradually to the resonance wavelength and a side (31b) which rises steeply to the resonance wavelength (32); a detection device (5) which is designed to detect a portion of the laser light (L) not coupled into the ring resonator (3); a control device (6) which is designed to control the laser power of the laser beam source (2) on the basis of the detected portion of the laser light (L) not coupled into the ring resonator (3), such that laser light (L) coupled into the ring resonator (3) has a laser wavelength (30) which is located on the steeply rising side (31b) of the resonance curve (31).

Optical assembly for generating the second and third harmonic frequencies of a laser beam using two optically nonlinear crystals (C1 and C2) placed in a temperature-controlled environment. The first optically nonlinear crystal (C1) converts the fundamental harmonic frequency to the second harmonic, while the second optically nonlinear crystal (C2) converts the fundamental and second harmonics to the third harmonic. The system includes beam intensity attenuation mechanisms using a first half-wave plate (HWP1) and first and second polarizers (P1, P2) that direct the beam according to the user's requirements. Motorized linear shifts (T1, T2) provide switching between harmonic frequency generation modes. Different cuts of the optically nonlinear crystals (C1 and C2) allow for optimization of phase matching and conversion efficiency. This assembly is designed for highly accurate and flexible generation of harmonic laser beams with wide industrial applications, especially in laser micromachining.

The invention relates to a method and an arrangement for the nonlinear spectral broadening of temporally incoherent or partially coherent optical pulses of a pulsed laser beam, which have a pulse duration not limited by bandwidth. In one embodiment of the method, the pulsed laser beam is split into two partial beams, which are spectrally shifted relative to each other by a frequency difference Δf. The spectrally shifted partial beams are then at least partially superimposed such that they interfere in the time domain during or after superposition and form at least one modulated pulsed laser beam, the pulses of which exhibit a temporal modulation of intensity with a period Δt = 1 / Δf.The pulses of the modulated pulsed laser beam are then spectrally broadened by self- or cross-phase modulation in an optically nonlinear medium, achieving a large spectral broadening through this modulation. This method and setup are particularly advantageous for IFE driver lasers.

The present invention discloses a phase matching method and device for optical nonlinear frequency conversion. The method comprises the following steps: multiple light fields are incident along the X-axis onto a periodically poled lithium niobate crystal whose optical axis and positive and negative domain polarization directions are both along the Z-axis; an electric field is applied to the front and rear sides of the periodically poled lithium niobate crystal in the Y-axis direction to rotate the polarization of the light field, so that the light field forms a sine or cosine oscillation form in the X-Z plane projection to provide a reciprocal lattice vector, which is used to compensate for momentum mismatch in the process of second-order nonlinear optical effects to achieve phase matching; the device does not need to rotate multiple layers of two-dimensional optical materials one by one, and can adjust the compensated reciprocal lattice vector to achieve rapid tuning.

The invention discloses a method for predicting optical soliton time domain evolution in an optical fiber based on an improved width learning system. The method comprises the following steps: constructing a width learning system, wherein the width learning system adopts a flattened network structure to realize structure horizontal expansion of feature nodes, enhanced nodes and the like; improving a width learning system, wherein the improvement comprises improving an incremental learning algorithm action range to synchronously adjust multiple layers of nodes so as to enable the system to learn more data features; the training system is suitable for optical soliton prediction. According to the method, the network structure can be dynamically updated in the training process, complex retraining is avoided, and the training efficiency and the prediction accuracy are effectively improved. In addition, the method utilizes a ridge regression algorithm to quickly solve pseudo-inverse from a hidden layer to an output layer, so that the training time is remarkably shortened, and quick and accurate prediction is realized. The method can be applied to rapid prediction of other optical nonlinear evolution processes, and a new thought is provided for application of the flattening network in physical field modeling and the optical field.

The invention discloses a passive multi-throw optical multistable all-optical switch based on an annular cavity system and a control method thereof, and the passive multi-throw optical multistable all-optical switch is characterized in that the passive multi-throw optical multistable all-optical switch comprises a one-way annular optical cavity and a rubidium bubble, the rubidium bubble is a cylindrical glass bubble which is located at the central position of the axis of the left upper cavity mirror and the central position of the axis of the right upper cavity mirror and filled with rubidium 85 atomic gas, and a monochromatic light field in the cavity can penetrate through the left side and the right side of the cylindrical glass bubble. Under the strong coupling condition of an atom-cavity system, the monochromatic cavity mode can excite multiple transition energy levels of rubidium atoms at the same time, so that strong optical nonlinear interaction is caused, and multiple different transmitted light intensity values can be achieved by modulating the light intensity of signal light. The multi-throw all-optical switch disclosed by the invention is based on an optical multistable state excited by signal light, does not need an additional auxiliary light field, a magnetic field and the like, and has the advantages of wide band, passivity, low power consumption, multiple channels, simple structure, easiness in integration and miniaturization and the like.

Lithium niobate (LiNbO3, LN) is a ferroelectric crystal of interest for integrated photonics owing to its large second-order optical nonlinearity and the ability to impart periodic poling via an external electric field. However, on-chip device performance based on thin-film lithium niobate (TFLN) is presently limited by propagation losses arising from surface roughness on the nano- and microscale. Atomic layer etching (ALE) can smooth these features and thereby increase photonic performance. In one embodiment disclosed herein, an isotropic ALE process for x-cut MgO-doped LN uses sequential exposures of H2 and SF6 / Ar plasmas. We observed an etch rate of 1.59±0.02 nm / cycle with a synergy of 96.9%. ALE can be achieved with SF6 / O2 or Cl2 / BCl3 plasma exposures in place of the SF6 / Ar plasma step with synergies of 99.5% and 91.5% respectively. The process decreased the sidewall surface roughness of TFLN waveguides etched by physical Ar+ milling by 30% without additional wet processing.

The invention provides a double-micro-ring coupling system for realizing optical nonlinear response, and relates to the technical field of microwave photons. The double micro-ring coupling system comprises a first micro-ring waveguide; the second micro-ring waveguide is coupled with the first micro-ring waveguide, a splitting ratio regulation and control module is arranged at the coupling position of the first micro-ring waveguide and the second micro-ring waveguide and used for determining the optical coupling coefficient between the first micro-ring waveguide and the second micro-ring waveguide, and the first micro-ring waveguide and the second micro-ring waveguide are each provided with a gain regulation and control module and a phase regulation and control module; parameters of the gain regulation and control module and the phase regulation and control module are regulated, and when the difference value between the net gain of the first micro-ring waveguide and the net loss of the second micro-ring waveguide is larger than or equal to an optical coupling coefficient, or the difference value between the net loss of the first micro-ring waveguide and the net gain of the second micro-ring waveguide is larger than or equal to the optical coupling coefficient, the phase regulation and control module is started. An optical nonlinear response is generated between the first micro-ring waveguide and the second micro-ring waveguide.

The application discloses a kind of optical deep neural network chips based on time domain, belong to optical computing technical field.A kind of optical deep neural network chips based on time domain, including sequentially connected: optical signal input area, input data modulator area, convolution weight modulator area, first optical nonlinear unit area, pooling modulator area, fully connected weight modulator area, second optical nonlinear unit area, high-speed photoelectric detector area and time domain integrator area.Optical deep neural network chip includes the three kinds of functions of convolution, pooling and fully connected in classic convolutional neural network, covers input layer, hidden layer and output layer in depth network;Multiple optical nonlinear unit area cascades are used to realize multilayer depth optical neural network, and large-scale high-throughput optical computing is realized by time dimension.Multiple layers of optical deep neural network computing are realized, and are suitable for artificial intelligence big model training, semantic segmentation, image recognition and medical instrument modeling imaging and other optical computing fields.

The invention provides an electro-optical and all-optical compatible on-chip reconfigurable optical nonlinear activation function device, and relates to the field of optical signal processing. The device comprises an adjustable light splitting module, a photoelectric detection module and a nonlinear output module, wherein the adjustable light splitting module adopts 2 * 2MZI, and an output splitting ratio alpha is set by a micro heater; the photoelectric detection module is a Ge-Si PD working in a reverse bias mode. The nonlinear output module is an Add-drop MRM, receives the light current from the Ge-Si PD and is driven in a carrier injection mode, and a micro heater above the nonlinear output module regulates and controls the wavelength offset delta lambda. An electro-optical working state and an all-optical working state are achieved by adjusting alpha, and the direct connection end and the downloading end of the MRM show different nonlinear activation characteristics in cooperation with delta lambda tuning. Electro-optic and all-optical nonlinearity are integrated in the same structure, the device has the advantages of being low in threshold value, flexible in regulation and control, high in integration, low in power consumption and the like, and key technical support is provided for a reconfigurable optical neural network.