# US 12,113,581 B2 · Photonic processing systems and methods

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 12,113,581 B2
- Document ID: US12113581B2
- Status: Granted
- Kind: B2
- Application number: 18/102,395
- Filed: 2023-01-27
- Published: 2023-12-21
- Granted: 2024-10-08
- Earliest claimed: 2018-05-15
- Assignees: Lightmatter, Inc.
- Applicants: Lightmatter, Inc.
- Inventors: Darius Bunandar; Nicholas C. Harris; Carl Ramey
- Web page: https://patentgrep.com/patent/US12113581B2

## CPC

- H04B 10/548: Phase or frequency modulation
- G02F 1/21: by interference
- G06F 17/16: Matrix or vector computation {, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization (matrix transposition G06F7/78)}
- G06N 3/0464: Convolutional networks [CNN, ConvNet]
- G06N 3/0675: {using electro-optical, acousto-optical or opto-electronic means}
- G06N 3/084: Backpropagation, e.g. using gradient descent
- G06N 3/09: Supervised learning
- H04B 10/70: Photonic quantum communication
- H04B 10/801: {using optical interconnects, e.g. light coupled isolators, circuit board interconnections}
- H04J 14/02: Wavelength-division multiplex systems
- H04J 14/0279: {WDM point-to-point architectures}
- G02F 1/212: {Mach-Zehnder type}
- G02F 2203/48: Variable attenuator
- G02F 2203/50: Phase-only modulation
- G06N 3/047: Probabilistic or stochastic networks
- G06T 1/20: Processor architectures; Processor configuration, e.g. pipelining
- H04B 10/61: Coherent receivers
- H04B: TRANSMISSION
- G02F: OPTICAL DEVICES OR ARRANGEMENTS FOR THE CONTROL OF LIGHT BY MODIFICATION OF THE OPTICAL PROPERTIES OF THE MEDIA OF THE ELEMENTS INVOLVED THEREIN; NON-LINEAR OPTICS; FREQUENCY-CHANGING OF LIGHT; OPTICAL LOGIC ELEMENTS; OPTICAL ANALOGUE/DIGITAL CONVERTERS
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- H04J: MULTIPLEX COMMUNICATION (transmission in general H04B; peculiar to transmission of digital information H04L5/00; systems for the simultaneous or sequential transmission of more than one television signal H04N7/08; in exchanges H04Q11/00; stereophonic systems H04S)
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL

## Abstract

Aspects relate to a photonic processing system, a photonic processor, and a method of performing matrix-vector multiplication. An optical encoder may encode an input vector into a first plurality of optical signals. A photonic processor may receive the first plurality of optical signals; perform a plurality of operations on the first plurality of optical signals, the plurality of operations implementing a matrix multiplication of the input vector by a matrix; and output a second plurality of optical signals representing an output vector. An optical receiver may detect the second plurality of optical signals and output an electrical digital representation of the output vector.

## Claims

### Claim 1

A photonic processing system, comprising:
an optical encoder configured to produce a plurality of input optical signals by optically encoding light with a plurality of input numeric values;
a photonic processor configured to produce a plurality of output optical signals by performing a matrix-matrix multiplication or matrix-vector multiplication based on the plurality of input optical signals; and
an optical receiver, comprising a plurality of homodyne detectors, configured to produce a plurality of output numeric values based on the plurality of output optical signals, wherein a first homodyne detector of the plurality of homodyne detectors comprises:
a differential amplifier having first and second inputs;
first and second photodetectors coupled to the first input of the differential amplifier; and
third and fourth photodetectors coupled to the second input of the differential amplifier.

### Claim 2 (depends on 1)

The photonic processing system of claim 1, wherein the first homodyne detector of the plurality of homodyne detectors is configured to produce one of the plurality of output numeric values by beating one of the plurality of output optical signals with a local oscillator signal derived from the light.

### Claim 3 (depends on 1)

The photonic processing system of claim 1, wherein an anode of the first photodetector is coupled to a cathode of the second photodetector.

### Claim 4 (depends on 3)

The photonic processing system of claim 3, wherein the anode of the first photodetector and the cathode of the second photodetector are coupled to the first input of the differential amplifier.

### Claim 5 (depends on 3)

The photonic processing system of claim 3, wherein a cathode of the third photodetector is coupled to an anode of the fourth photodetector.

### Claim 6 (depends on 5)

The photonic processing system of claim 5, wherein the cathode of the third photodetector and the anode of the fourth photodetector are coupled to the second input of the differential amplifier.

### Claim 7 (depends on 1)

The photonic processing system of claim 1, wherein the first, second, third and fourth photodetectors are formed monolithically on a common substrate.

### Claim 8 (depends on 1)

The photonic processing system of claim 1, wherein the first, second, third and fourth photodetectors have equal responsivities.

### Claim 9 (depends on 1)

The photonic processing system of claim 1, further comprising an analog-to-digital converter coupled to first and second outputs of the differential amplifier.

### Claim 10 (depends on 1)

The photonic processing system of claim 1, further comprising a photonic circuit configured to provide:
a first optical signal to the first and third photodetectors, and
a second optical signal to the second and fourth photodetectors.

### Claim 11 (depends on 10)

The photonic processing system of claim 10, wherein the photonic circuit is configured to generate the first and second optical signals by combining one of the plurality of output optical signals with the light.

### Claim 12 (depends on 1)

The photonic processing system of claim 1, further comprising a controller configured to perform a first iteration comprising:
receiving an input bit string;
producing, using the input bit string, input analog signals representing the plurality of input numeric values, and providing the input analog signals to the optical encoder;
programming the photonic processor based on a weight matrix;
receiving output analog signals representing the plurality of output numeric values from the optical receiver; and
producing, using the output numeric values, an output bit string.

### Claim 13 (depends on 12)

The photonic processing system of claim 12, wherein the controller is further configured to perform a second iteration comprising:
producing, using the output bit string, further input analog signals representing a further plurality of input numeric values, and providing the further input analog signals to the optical encoder.

### Claim 14 (depends on 1)

The photonic processing system of claim 1, wherein the optical receiver further comprises a low-pass filter coupled to the first homodyne detector of the plurality of homodyne detectors.

### Claim 15

A method for controlling a photonic processing system, comprising:
producing, using an optical encoder, a plurality of input optical signals by optically encoding light with a plurality of input numeric values;
producing, using a photonic processor, a plurality of output optical signals by performing a matrix-matrix multiplication or matrix-vector multiplication based on the plurality of input optical signals; and
producing, using an optical receiver comprising a plurality of homodyne detectors, a plurality of output numeric values based on the plurality of output optical signals, wherein a first homodyne detector of the plurality of homodyne detectors comprises:
a differential amplifier having first and second inputs;
first and second photodetectors coupled to the first input of the differential amplifier; and
third and fourth photodetectors coupled to the second input of the differential amplifier, and wherein producing the plurality of output numeric values comprises:
providing a first photocurrent produced by the first and second photodetectors to the first input of the differential amplifier; and
providing a second photocurrent produced by the third and fourth photodetectors to the second input of the differential amplifier.

### Claim 16 (depends on 15)

The method of claim 15, wherein producing the plurality of output numeric values comprises producing, using the first homodyne detector of the plurality of homodyne detectors, one of the plurality of output numeric values by beating one of the plurality of output optical signals with a local oscillator signal derived from the light.

### Claim 17 (depends on 15)

The method of claim 15, further comprising performing a first iteration comprising:
receiving an input bit string;
producing, using the input bit string, input analog signals representing the plurality of input numeric values, and providing the input analog signals to the optical encoder;
programming the photonic processor based on a weight matrix;
receiving output analog signals representing the plurality of output numeric values from the optical receiver; and
producing, using the output numeric values, an output bit string.

### Claim 18 (depends on 17)

The method of claim 17, further comprising perform a seconding iteration comprising:
producing, using the output bit string, further input analog signals representing a further plurality of input numeric values, and providing the further input analog signals to the optical encoder.
