An optical neural network is constructed based on photonic integrated circuits to perform neuromorphic computing. In the optical neural network, matrix multiplication is implemented using one or more optical interference units, which can apply an arbitrary weighting matrix multiplication to an array of input optical signals. Nonlinear activation is realized by an optical nonlinearity unit, which can be based on nonlinear optical effects, such as saturable absorption. These calculations are implemented optically, thereby resulting in high calculation speeds and low power consumption in the optical neural network.

Methods, devices, and systems for processing information are disclosed. An example device may comprise a metastructure comprising a plurality of physical features configured to transform an analog signal according to a kernel of an integral equation. The device may comprise one or more waveguides coupled to the metastructure and configured to recursively supply a transformed analog output signal of the metastructure to an input of the metastructure to iteratively cause one or more transformed analog signals output from the metastructure to converge to an analog signal representing a solution to the integral equation.

A system performing optical and/or electro-optical tensor operations and featuring a photonic dot product engine with a first input and a second input and summation to perform multiply-accumulate operations. The first and/or second input is a matrix, and/or a vector, and/or scalar. The system is a Photonic Tensor Core.

An optical neural network is constructed based on photonic integrated circuits to perform neuromorphic computing. In the optical neural network, matrix multiplication is implemented using one or more optical interference units, which can apply an arbitrary weighting matrix multiplication to an array of input optical signals. Nonlinear activation is realized by an optical nonlinearity unit, which can be based on nonlinear optical effects, such as saturable absorption. These calculations are implemented optically, thereby resulting in high calculation speeds and low power consumption in the optical neural network.

An optical neural network is constructed based on photonic integrated circuits to perform neuromorphic computing. In the optical neural network, matrix multiplication is implemented using one or more optical interference units, which can apply an arbitrary weighting matrix multiplication to an array of input optical signals. Nonlinear activation is realized by an optical nonlinearity unit, which can be based on nonlinear optical effects, such as saturable absorption. These calculations are implemented optically, thereby resulting in high calculation speeds and low power consumption in the optical neural network.

A co-processor for performing a matrix multiplication of an input matrix with a data matrix in one step may be provided. The co-processor receives input signals for the input matrix as optical signals. A plurality of photonic memory elements is arranged at crossing points of an optical waveguide crossbar array. The plurality of memory elements is configured to store values of the data matrix. Input signals are connected to input lines of the optical waveguide crossbar array. Output lines of the optical waveguide crossbar array represent a dot-product between a respective column of the optical waveguide crossbar array and the received input signals, and values of elements of the input matrix to be multiplied with the data matrix correspond to light intensities received at input lines of the respective photonic memory elements. Additionally, different wavelengths are used for each column of the input matrix optical signals.

An optical processing system comprises a first integrated optical waveguide array; a first bundle of optical fibres; the optical fibres being coupled to the first integrated optical waveguide array by a first coupler; the optical fibres being further coupled to an optical Fourier stage; a second bundle of optical fibres being coupled to the optical Fourier stage; a second integrated optical waveguide array; and a second coupler for coupling the second bundle of optical fibres to the second integrated optical waveguide array.

A co-processor for performing a matrix multiplication of an input matrix with a data matrix in one step may be provided. The co-processor receives input signals for the input matrix as optical signals. A plurality of photonic memory elements is arranged at crossing points of an optical waveguide crossbar array. The plurality of memory elements is configured to store values of the data matrix. Input signals are connected to input lines of the optical waveguide crossbar array. Output lines of the optical waveguide crossbar array represent a dot-product between a respective column of the optical waveguide crossbar array and the received input signals, and values of elements of the input matrix to be multiplied with the data matrix correspond to light intensities received at input lines of the respective photonic memory elements. Additionally, different wavelengths are used for each column of the input matrix optical signals.

An optically computed optical coherence tomography (OC-OCT) technology is disclosed. The OC-OCT system performs depth resolved imaging by computing the Fourier transform of the interferometric spectra optically. The OC-OCT system modulates the interferometric spectra with Fourier basis function projected to a spatial light modulator and detects the modulated signal without spectral discrimination. The optical computation strategy enables volumetric OCT imaging without performing mechanical scanning and without the need for Fourier transform in a computer. OC-OCT performs Fourier transform signal processing optically, without the need of mechanical scanning, and before data acquisition unlike traditional OCT methods and systems. The scan-less OCT imaging is achieved through the use of spatial light modulator (SLM) that precisely manipulates light wave to generate output with desired amplitude and phase.

A technology is described for a Photonic Integrated Circuit (PIC) radio frequency (RF) in-phase quadrature phase (I/Q) correlator. The PIC RF Correlator can comprise two optical waveguides to receive first and second optical signals that are modulated by first and second RF signals, respectively. Two 1 to M optical splitters can split the first and second RF modulated optical signals. Optical delay lines can delay the M split first RF modulated optical signals. M optical balanced couplers can receive and combine the M first delayed RF modulated optical signals with the M split second RF modulated optical signals. Balanced photodetectors can output a differential integration on the first and second combined RF modulated optical signals for in-phase and quadrature phase signals. A processor can add the outputs of the M optical balanced photodetectors to form a frequency domain correlated signal of the first and second RF signals with real and imaginary parts.