阅读说明：本技术 一种光计算设备、光运算方法以及计算系统 (Optical computing device, optical computing method and computing system ) 是由 褚雅妍 董晓文 于 2019-12-20 设计创作，主要内容包括：一种光计算设备、光运算方法以及计算系统,该光计算设备包括分光器、N-1个第一波导、一个第二波导以及一个合束器。分光器可以将接收的连续光分为N路光信号,并将N路光信号分别传输到N-1个第一波导和一个第二波导中,每个第一波导接收一路相干光,第二波导接收一路相干光。N-1个第一波导调整接收到的N-1路相干光的振幅和相位。第二波导调整接收的一路光信号的相位。合束器可以将N-1个第一波导以及第二波导输出的相干光合并为一路光信号。第一波导能够调整相干光的振幅和相位,使得该光计算设备能够实现带符号乘法运算,且由于该光计算设备中借助了第一波导和第二波导,能够有效减少光计算设备所占用的体积。(An optical computing device includes an optical splitter, N-1 first waveguides, a second waveguide, and a beam combiner. The optical splitter can split the received continuous light into N paths of optical signals, and transmit the N paths of optical signals to N-1 first waveguides and one second waveguide respectively, where each first waveguide receives one path of coherent light, and the second waveguide receives one path of coherent light. The N-1 first waveguides adjust the amplitude and phase of the received N-1 coherent light beams. The second waveguide adjusts the phase of the received optical signal. The beam combiner can combine the coherent light output by the N-1 first waveguides and the coherent light output by the second waveguides into one path of optical signal. The first waveguide can adjust the amplitude and the phase of coherent light, so that the optical computing equipment can realize signed multiplication, and the optical computing equipment can effectively reduce the occupied volume of the optical computing equipment by means of the first waveguide and the second waveguide.)

1. A light computing device, comprising:

the optical splitter is used for splitting the received continuous light into N paths of optical signals;

the optical splitter is used for receiving the N-1 optical signals in the N optical signals, adjusting the amplitude of the N-1 optical signals according to the set amplitude parameters, and outputting N-1 intermediate optical signals, wherein each first waveguide receives one optical signal, at least two amplitude parameters are set in each first waveguide, the at least two amplitude parameters are used for indicating at least two data to be subjected to multiplication calculation, and the intermediate optical signal output by each first waveguide is used for indicating the product of the at least two data corresponding to the at least two amplitude parameters set in the first waveguide;

and the beam combiner is connected with the N-1 first waveguides and is used for combining the N-1 paths of intermediate optical signals output by the N-1 first waveguides into a first optical signal, and the first optical signal is used for indicating the sum of products of data indicated by the N-1 paths of intermediate optical signals.

2. The light computing device of claim 1, wherein: each first waveguide in the N-1 first waveguides is further used for adjusting the phase of the optical signal output by the first waveguide according to the set phase parameter, and the phase of the optical signal output by the first waveguide is used for indicating the positive and negative values of the product of at least two data indicated by the intermediate optical signal output by the first waveguide;

the light computing device further comprises:

the second waveguide is connected with the optical splitter and used for receiving an nth optical signal in the N optical signals and adjusting the received nth optical signal into a second optical signal, wherein the phase of the second optical signal reaching the beam combiner is a first phase;

the beam combiner is further configured to connect to the second waveguide, and to combine the N-1 intermediate optical signals output by the N-1 first waveguides and the second optical signal output by the second waveguide into a third optical signal.

3. The light computing device of claim 2, wherein:

the second waveguide is further configured to adjust the received nth optical signal to a fourth optical signal, where a phase of the fourth optical signal reaching the beam combiner is a second phase;

the beam combiner is further configured to combine the N-1 intermediate optical signals output by the N-1 first waveguides and the fourth optical signal into a fifth optical signal.

4. The light computing device of claim 3, further comprising:

a detector for detecting the light intensity of the third optical signal and for detecting the light intensity of the fifth optical signal;

and the processing circuit is used for obtaining the sum of products of the data indicated by the N-1 paths of intermediate optical signals according to the light intensity of the third optical signal output by the detector and the light intensity of the fifth optical signal.

5. The apparatus of claim 1 or 2, wherein each of the first waveguides comprises at least two amplitude modulation units,

any one of the at least two amplitude modulation units is configured to adjust the amplitude of the one path of optical signal received by the first waveguide according to one of the amplitude parameters.

6. The apparatus of claim 5,

the at least two amplitude modulation units comprise a first amplitude modulation unit and a second amplitude modulation unit,

the first amplitude modulation unit is used for adjusting the amplitude of the optical signal according to a first amplitude parameter;

and the second amplitude modulation unit is used for adjusting the amplitude of the optical signal adjusted by the first amplitude modulation unit according to a second amplitude parameter.

7. The apparatus of claim 5 or 6, wherein each of the first waveguides further comprises the first phase modulation unit, the first phase modulation unit and the at least two amplitude modulation units being connected in series;

the first phase modulation unit is configured to adjust a phase of the optical signal according to the phase parameter, so that a phase difference between the intermediate optical signal and reference light when the intermediate optical signal and the reference light reach the beam combiner is 0 or pi, where the reference light is any one of the N optical signals.

8. The apparatus of claim 3, wherein the second waveguide comprises a second phase modulation unit;

the second phase modulation unit is configured to adjust a phase of the nth optical signal, so that a difference between the first phase and a phase when the reference light reaches the beam combiner is 0, and a difference between the second phase and a phase when the reference light reaches the beam combiner is pi, where the reference light is any one of the N optical signals.

9. The apparatus according to claim 5 or 6, characterized in that the amplitude modulation unit is an electro-absorption modulator EAM or a semiconductor optical amplifier SOA or a dimmable attenuator VOA.

10. The apparatus according to any of claims 1 to 8, wherein the beam combiner is a multi-mode interference coupler MMI or a cascaded Y-branch;

the light splitter is an MMI or a cascade Y branch.

11. An optical computing method, performed by an optical computing device comprising an optical splitter, N-1 first waveguides, and a beam combiner, the method comprising:

the optical splitter splits received light into N paths of optical signals;

the N-1 first waveguides receive N-1 optical signals in the N optical signals, adjust the amplitudes of the N-1 optical signals according to set amplitude parameters, and output N-1 intermediate optical signals, wherein each first waveguide receives one optical signal, at least two amplitude parameters are set in each first waveguide, the at least two amplitude parameters are used for indicating at least two data to be subjected to multiplication, and the intermediate optical signal output by each first waveguide is used for indicating the product of at least two data corresponding to the at least two amplitude parameters set in the first waveguide;

the beam combiner combines the N-1 paths of intermediate optical signals output by the N-1 first waveguides into a first optical signal, and the first optical signal is used for indicating the sum of products of data indicated by the N-1 paths of intermediate optical signals.

12. The method of claim 11, wherein the light computing device further comprises a second waveguide, the method further comprising:

each first waveguide in the N-1 first waveguides adjusts the phase of the optical signal output by the first waveguide according to the set phase parameter;

the second waveguide receives an nth optical signal in the N optical signals, and adjusts the received nth optical signal into a second optical signal, where a phase of the second optical signal reaching the beam combiner is a first phase;

and the beam combiner combines the N-1 paths of intermediate optical signals output by the N-1 first waveguides and the second optical signals output by the second waveguides into third optical signals.

13. The method of claim 12, wherein the method further comprises:

the second waveguide adjusts the received Nth optical signal into a fourth optical signal, and the phase of the fourth optical signal reaching the beam combiner is a second phase;

and the beam combiner combines the N-1 first waveguides outputting the N-1 paths of intermediate optical signals and the fourth optical signal into a fifth optical signal.

14. The method of claim 13, wherein the light computing device further comprises a detector and processing circuitry, the method further comprising:

the detector detects the light intensity of the third optical signal and detects the light intensity of the fifth optical signal;

and the processing circuit obtains the sum of products of the data indicated by the N-1 paths of intermediate optical signals according to the light intensity of the third optical signal and the light intensity of the fifth optical signal output by the detector.

15. A computing system comprising a processor and an optical computing device according to any of claims 1 to 10, wherein the processor is configured to send data to be subjected to multiply-add operations to the optical computing device.

Technical Field

The present application relates to the field of information technology, and in particular, to an optical computing device, an optical computing method, and a computing system.

Background

The multiply-add operation is a basic operation, in which a plurality of multiplication results are added to obtain a final result, the multiply-add unit is a calculation unit for realizing the multiply-add operation, and the variety of multiply-add accumulators is many, wherein the optical multiply-add unit has the advantages of high efficiency and low power consumption due to the optical method adopted by the optical multiply-add unit.

At present, the common optical multiplier-adder implementation is mainly based on the optical multiplier-adder of the optical fiber system. The optical multiplier-adder includes an acousto-optic modulator array composed of a plurality of acousto-optic modulators connected by optical fibers and a plurality of detectors. In the process of realizing multiplication and addition operation of the optical multiplier-adder based on the optical fiber, signals can be modulated on multiple paths of optical signals through the acousto-optic modulator array to realize multiplication operation. And then, respectively receiving the optical signals modulated by the acousto-optic modulator array by using the plurality of detectors, converting the received optical signals into electric signals, and superposing the plurality of electric signals obtained by the plurality of detectors to realize addition operation so as to obtain a calculation result.

The optical multiplier-adder in the prior art uses an optical fiber system as a transmission carrier of optical signals, so that the optical multiplier-adder has large volume and large power consumption. Also, since the addition portion in such an optical multiplier-adder is electrically completed, the operation efficiency is not high.

Disclosure of Invention

The application provides an optical computing device, an optical computing method and a computing system, which are used for providing a multiplier-adder with small volume and high expansibility.

In a first aspect, the present application provides a light computing device comprising an optical splitter, N-1 first waveguides, and a beam combiner. The beam splitter is respectively connected with the input ends of the N-1 first waveguides, and the beam combiner is respectively connected with the output ends of the N-1 first waveguides. The optical splitter may split the received continuous light into N optical signals, and transmit the N optical signals to N-1 first waveguides and one second waveguide, respectively, where each first waveguide receives one optical signal and the second waveguide receives one optical signal. The N-1 first waveguides have an amplitude modulation function, and can adjust the amplitude of the received N-1 optical signals according to the set amplitude parameters and output N-1 intermediate optical signals. Wherein each first waveguide is provided with at least two amplitude parameters. The at least two amplitude parameters are used to indicate the same amount of data to be multiplied, and the intermediate optical signal output by the first waveguide may indicate a product of the data indicated by the at least two amplitude parameters. The beam combiner receives the N-1 paths of intermediate optical signals and combines the N-1 paths of intermediate optical signals into a first optical signal.

In the device provided by the application, the amplitude of the optical signal is adjusted through the first waveguide, so that the optical computing device can realize multiplication. And the N-1 paths of intermediate optical signals output by the N-1 first waveguides pass through the beam combiner, and the beam combiner can combine the N-1 paths of intermediate optical signals, so that the optical computing equipment can realize addition operation, and the optical computing equipment can be ensured to realize multiply-add operation. Because the optical computing equipment adopts the waveguide to realize the multiply-add operation, the volume and the power consumption occupied by the optical computing equipment can be effectively reduced, the expansibility of the optical computing equipment is improved, and the data volume which can be calculated by the optical computing equipment is larger.

In one possible design, the optical computing device may further include a second waveguide and a combiner. The beam splitter may be connected to an input of the second waveguide and the beam combiner to an output of the second waveguide.

The optical splitter divides the received continuous light into N paths of optical signals, and the N paths of optical signals are respectively transmitted to N-1 first waveguides and a second waveguide, wherein the N-1 first waveguides receive the N-1 optical signals, and the second waveguide receives one path of optical signals except the N-1 optical signals.

After the N-1 first waveguides adjust the received N-1 optical signals, the amplitudes of the N-1 optical signals can be adjusted according to the set amplitude parameters, and the phases of the optical signals output by the first waveguides can be adjusted according to the set phase parameters, wherein the phase parameters can indicate the positive and negative values of the sum of the products of the data indicated by the intermediate optical signals output by the first waveguides.

After the second waveguide receives the optical signal, the phase of the optical signal can be adjusted, and a second optical signal is output, wherein the phase of the second optical signal reaching the beam combiner is a first phase.

The beam combiner can combine the N-1 paths output by the N-1 first waveguides and the second optical signals output by the second waveguides into third optical signals.

In the device provided by the application, the amplitude of the optical signal is adjusted through the first waveguide, so that the optical computing device can realize multiplication. Further, since the first waveguide also has a function of adjusting the phase of the optical signal, the phase of the optical signal can be changed and a sign can be added to each multiplication result. And then the optical signals output by the first waveguide and the second waveguide can be combined by the beam combiner, so that the optical computing equipment can realize signed addition operation. The second waveguide can adjust the phase of the optical signal, so that the light intensity of the optical signal output by the final beam combiner can be changed, and the amplitude (namely the result of the multiplication and addition operation) of the optical signal after the intermediate optical signal output by the first waveguide is combined can be determined by the second waveguide, thereby ensuring that the optical computing equipment can realize signed multiplication and addition operation. Similarly, because the optical computing device adopts the waveguide to realize the multiply-add operation, the occupied volume and the power consumption of the optical computing device can be effectively reduced, the expansibility of the optical computing device is ensured, and the calculation amount which can be realized by the optical computing device is increased.

In a possible design, the second waveguide may further adjust the phase of the received nth optical signal again, adjust the received nth optical signal to a fourth optical signal, and make the phase of the fourth optical signal reaching the beam combiner a second phase;

and the beam combiner combines the N-1 first waveguides outputting the N-1 paths of intermediate optical signals and the fourth optical signals into a fifth optical signal.

In the device provided by the application, the second waveguide can adjust the phase of the optical signal for multiple times, and can output multiple different optical signals (such as the second optical signal and the fourth optical signal), so that the light intensity of the optical signal output by the beam combiner is changed, and the optical computing device can realize signed multiplication and addition operation.

In one possible design, the light computing device may further include a detector and a processing circuit, and the detector may detect the light intensity of the light signals (e.g., the first light signal, the third light signal, and the fifth light signal) output by the beam combiner. The processing circuit can obtain the light intensity of the optical signal output by the beam combiner from the detector, and determine the sum of products of data indicated by the N-1 paths of intermediate optical signals according to the light intensity.

For example, the processing circuit may determine the sum of products of the data indicated by the N-1 intermediate optical signals, i.e. the result of the multiplication and addition operation without sign, directly from the light intensity of the first optical signal. The processing circuit may also determine the sum of the products of the data indicated by the N-1 intermediate optical signals, i.e. the result of the signed multiply-add operation, directly from the light intensity of the third optical signal and the light intensity of the fifth optical signal.

In the equipment that this application provided, the light intensity of the light signal that processing circuit can be comparatively convenient through the beam combiner output confirms the result of calculation of multiply-add operation, and the resource that consumes is less, and the definite mode is more convenient, can effectively improve the efficiency that optical computing equipment realized multiply-add operation.

In one possible design, each first waveguide includes at least two amplitude modulation units for implementing a set of multiplication operations in a multiply-add operation.

Any amplitude modulation unit may adjust the amplitude of the received optical signal according to an amplitude parameter, which may indicate one data of one multiplication operation to be implemented in the multiplication and addition operations to be implemented.

In the device provided by the application, the first waveguide can realize multiplication operation by adjusting the amplitude, so that the multiplication operation is simpler and more efficient.

In one possible design, each first waveguide includes two amplitude modulation units, namely a first amplitude modulation unit and a second amplitude modulation unit, and an input end of the second amplitude modulation unit is connected with an output end of the first amplitude modulation unit.

The first amplitude modulation unit may adjust the amplitude of the optical signal according to a first amplitude parameter. Then, the second amplitude modulation unit may adjust the amplitude of the optical signal adjusted by the first amplitude modulation unit according to the second amplitude parameter.

In one possible design, the first waveguide further comprises a first phase modulation unit, one first phase modulation unit and at least two amplitude modulation units being connected in series. The order of connection of the first phase modulation unit and the at least two amplitude modulation units is not limited here. Each cell in the first waveguide conditions the optical signal output by the previous cell.

The first phase modulation unit can adjust the phase of the received optical signal according to the phase parameter, so that the difference between the phase of the optical signal (i.e. the intermediate optical signal) after the phase adjustment and the phase of the reference light when the reference light reaches the beam combiner is 0 or pi, wherein the reference light is one optical signal of the N optical signals.

In the device provided by the application, the first waveguide can realize multiplication operation by adjusting the amplitude, and adds a sign to the result of the multiplication operation by adjusting the phase, so that the multiplication operation is simpler and more efficient.

In one possible design, the first phase modulation unit may be calibrated in advance to determine a first phase parameter capable of adjusting the phase difference between the optical signal and the reference light when reaching the beam combiner to 0 or pi. Then, when the multiplication and addition operation is actually performed, the phase difference between the intermediate optical signal and the reference light when the intermediate optical signal and the reference light reach the beam combiner can be adjusted to be 0 or pi directly according to the first phase parameter.

In the device provided by the application, the first phase parameter is predetermined, so that the first phase modulation unit can more accurately control the phase difference between the intermediate optical signal and the reference light when the intermediate optical signal and the reference light reach the beam combiner.

In one possible design, the second waveguide includes a second phase modulation unit. The second phase modulation unit may adjust a phase of the received optical signal, so that the phase of the optical signal after the phase adjustment when reaching the beam combiner is a first phase or a second phase, and the reference light is one of the N optical signals.

The specific values of the first phase and the second phase are not limited, and the first phase and the second phase need to satisfy one of the following conditions:

the first condition is that the difference between the first phase and the phase when the reference light reaches the beam combiner is 0, and the difference between the second phase and the phase when the reference light reaches the beam combiner is pi.

And under the second condition, the difference between the first phase and the phase when the reference light reaches the beam combiner is pi, and the difference between the second phase and the phase when the reference light reaches the beam combiner is 0.

In the optical computing device provided by the application, the second waveguide only needs to adjust the phase of the optical signal, and change the phase difference between the output optical signal and the reference light when reaching the beam combiner so as to change the light intensity of the optical signal output by the beam combiner. That is, the optical signal and the reference light reach different phase differences when reaching the beam combiner, so that the beam combiner outputs different optical signals, and the positive and negative values of the sum of the products of the indicated data of the optical signals after the N-1 intermediate optical signals are combined can be conveniently and directly determined through the light intensity of the optical signals output by the beam combiner under the different phase differences.

In one possible design, the second phase modulation unit may be calibrated in advance, and a second phase parameter capable of adjusting the phase difference between the optical signal and the reference light reaching the beam combiner to 0 or pi is determined. Then, when the multiplication and addition operation is actually performed, the phase difference between the optical signal and the reference light reaching the beam combiner can be adjusted to 0 or pi directly according to the second phase parameter.

In the device provided by the application, the second phase parameter is predetermined, so that the second phase modulation unit can more accurately control the phase difference between the optical signal and the reference light when the optical signal and the reference light reach the beam combiner.

In one possible design, the light computing device further includes a laser coupled to the beam splitter and capable of generating continuous light, the laser splitting the continuous light output into the beam splitter. Wherein, the laser is a single longitudinal mode semiconductor laser.

In the device provided by the application, the laser can be used for generating continuous light with a single frequency, so that the subsequent treatment is facilitated.

In one possible design, the amplitude modulation unit is an EAM, SOA or VOA.

In one possible implementation, the beam combiner is an MMI or a cascaded Y-branch; the beam splitter is an MMI or a cascade Y branch.

In a second aspect, the present application provides an optical operation method, and beneficial effects may refer to related descriptions of the first aspect, which are not described herein again. The method is executed by an optical computing device, and an optical splitter divides received continuous light into N paths of optical signals; the N-1 first waveguides receive N-1 optical signals in the N optical signals, adjust the amplitude of the N-1 optical signals according to the set amplitude parameters, and output N-1 intermediate optical signals, wherein each first waveguide receives one optical signal, at least two amplitude parameters are set in each first waveguide, and the at least two amplitude parameters can indicate at least two data to be subjected to multiplication calculation, so that the intermediate optical signal output by each first waveguide can indicate the product of the at least two data corresponding to the at least two amplitude parameters set in the first waveguide;

the beam combiner receives the N-1 intermediate optical signals and combines the N-1 intermediate optical signals into a first optical signal, where the first optical signal may indicate a sum of products of data indicated by the N-1 intermediate optical signals.

In one possible design, each of the N-1 first waveguides may further adjust a phase of the optical signal output by the first waveguide according to the set phase parameter; the phase of the optical signal output by the first waveguide can indicate the positive and negative values of the product indicated by the intermediate optical signal.

The second waveguide receives the nth optical signal in the N optical signals, modulates the phase of the nth optical signal, and can adjust the received nth optical signal into a second optical signal, wherein the phase of the second optical signal reaching the beam combiner is a first phase;

then, the beam combiner may combine the N-1 intermediate optical signals and the second optical signal into a third optical signal.

In a possible design, the second waveguide may further adjust the phase of the nth optical signal again, adjust the received nth optical signal to a fourth optical signal, and make the phase of the fourth optical signal reaching the beam combiner a second phase;

and then, the beam combiner combines the N-1 paths of intermediate optical signals and the fourth optical signal into a fifth optical signal.

In one possible design, the detector may detect the light intensity of the first optical signal; the processing circuit may obtain a sum of products of the data indicated by the N-1 intermediate optical signals based on the light intensity of the first optical signal output by the detector.

The detector can also detect the light intensity of the third optical signal and detect the light intensity of the fifth optical signal; the processing circuit obtains the sum of products of data indicated by the N-1 paths of intermediate optical signals according to the light intensity of the third optical signal and the light intensity of the fifth optical signal output by the detector.

In a possible design, the difference between the first phase and the phase when the reference light reaches the beam combiner is 0, the difference between the second phase and the phase when the reference light reaches the beam combiner is pi, and the reference light is one of the N optical signals. Or the difference between the first phase and the phase when the reference light reaches the beam combiner is pi, the difference between the second phase and the phase when the reference light reaches the beam combiner is 0, and the reference light is one optical signal of the N optical signals.

In a third aspect, the present application provides a computing system that may include a processor and a light computing device as described in the first aspect or any one of the possible implementations of the first aspect.

The processor may send data to be subjected to multiply-add operations to the optical computing device. Specifically, the data to be subjected to multiply-add operation includes N-1 groups of data, and each group of data includes at least two data to be subjected to multiply-add operation in the multiply-add operation. After receiving the data to be subjected to multiply-add operation, the optical computing device may set an amplitude parameter of the first waveguide according to a set of data, and optionally, may set a phase parameter. Then, the optical computing device performs optical computation, outputs the result of the multiply-add operation, and feeds back the result of the multiply-add operation to processing.

Drawings

FIG. 1 is a schematic diagram of a light computing device according to the present application;

FIG. 2 is a schematic block diagram of another optical computing device provided herein;

FIG. 3 is a schematic diagram of a first waveguide structure provided herein;

FIG. 4 is a schematic diagram of a second waveguide according to the present application;

FIG. 5 is a schematic diagram of an optical computing method according to the present application;

FIG. 6 is a schematic diagram of an optical computing method according to the present application;

fig. 7 is a schematic diagram of a computing system provided in the present application.

Detailed Description

The application provides an optical computing device and an optical computing method, which are used for providing an optical multiplier-adder with small volume and high expansibility.

Fig. 1 shows a light computing device provided by the embodiment of the present application, which includes an optical splitter 110, N-1 first waveguides 120, and a beam combiner 130.

The optical splitter 110 is connected to the N-1 first waveguides 120, respectively, and the optical splitter 110 may split the received continuous light into N optical signals, and then transmit the N-1 optical signals of the N optical signals to the N-1 first waveguides 120, respectively; one of the first waveguides 120 receives one of the N-1 optical signals.

The N-1 first waveguides 120 may adjust the amplitude of the received N-1 optical signals according to the amplitude parameter, and output N-1 intermediate optical signals.

For any one of the N-1 first waveguides 120, the first waveguide 120 receives one of the N-1 optical signals, adjusts the amplitude of the one optical signal according to at least two set amplitude parameters, and outputs one intermediate optical signal. The at least two amplitude parameters may be used to indicate at least two data for multiplication by the tape, one amplitude parameter being at least one data, such that an output of the intermediate optical signal may indicate a product of the two data.

The beam combiner is respectively connected to the N-1 first waveguides, receives N-1 intermediate optical signals output by the N-1 first waveguides 120, and combines the N-1 intermediate optical signals into a first optical signal, where the N first optical signal is used to indicate a sum of products of data indicated by the N-1 intermediate optical signals.

In the embodiment of the present application, the first waveguide 120 has an amplitude modulation function, and is capable of adjusting the amplitude of the received optical signal. Optionally, the first waveguides 120 may also have a phase modulation function, and the N-1 first waveguides 120 may adjust the phase of the N-1 optical signals, so that the phases of the intermediate optical signals output by the first waveguides 120 are the same.

The multiplication and addition operation to be realized by the optical computing equipment is as follows:(wherein, N-1X_iW_iAre the same in sign, may be both positive signs, and may also be both negative signs), the optical computing device adjusts the amplitude of the N-1 optical signals by using N-1 first waveguides 120, and implements N-1 multiplication operations in the multiplication and addition operations, wherein each first waveguide 120 adjusts the amplitude of one optical signal in the N-1 optical signals according to an amplitude parameter, and the amplitude parameter indicates X_iW_iSuch that the amplitude of the intermediate optical signal is indicative of one of the N-1 multiplications, X_iW_i. Then, the intermediate optical signals output from the N-1 first waveguides 120 are combined in the beam combiner 140, and then the first optical signal having an amplitude ofThe sum of the N-1 products is indicated.

In the embodiment of the present application, since the first waveguide 120 can adjust the amplitude of the optical signal, the optical computing device can implement multiplication without sign, and then the intermediate optical signals output by the first waveguide 120 can be combined by the beam combiner 140, so that the optical computing device can implement addition, which ensures that the optical computing device can implement multiplication without sign (the sign of the result of each multiplication is the same). Because the first waveguide 120 is used in the optical computing device, the volume occupied by the optical computing device can be effectively reduced, the expansibility of the optical computing device is improved, and further, the data volume which can be calculated by the optical computing device is larger.

In order to realize signed multiply-add operations (each multiplication result may be a positive number or a negative number), the optical computing apparatus shown in fig. 1 is modified, as shown in fig. 2, to provide another optical computing apparatus according to an embodiment of the present application, which includes an optical splitter 110, N-1 first waveguides 120, a second waveguide 140, and a beam combiner 130.

The optical splitter 110 is connected to the N-1 first waveguides 120 and the second waveguide 140, respectively, and the optical splitter 110 may split the received continuous light into N optical signals, and then transmit the N optical signals to the N-1 first waveguides 120 and the second waveguide 140, respectively; one first waveguide 120 receives one optical signal of the N optical signals, N-1 first waveguides 120 receive the N-1 optical signals, and the second waveguide 140 receives one optical signal of the N optical signals except the N-1 optical signals, that is, the nth optical signal.

After receiving one optical signal of the N optical signals, any first waveguide 120 of the N-1 first waveguides 120 may adjust the amplitude of the optical signal according to the amplitude parameter (the manner in which the first waveguide 120 adjusts the amplitude of the optical signal may refer to the foregoing content, and is not described here again), and may also adjust the phase of the received optical signal according to the phase parameter. After that, the first waveguide 120 outputs the optical signal with the adjusted phase and amplitude, and in the optical computing apparatus shown in fig. 2, the optical signal output by the first waveguide 120 may also be referred to as an intermediate optical signal. Wherein each first waveguide 120 has a phase parameter disposed thereon, the phase parameter being indicative of at least two data values indicated by the intermediate optical signalPositive and negative values of the product. The multiplication and addition operation to be realized by the optical computing equipment is as follows:for example, a first waveguide 120 is arranged with a phase parameter indicative of X_iW_iPositive and negative values of (c).

After receiving the nth optical signal, the second waveguide 140 may adjust the phase of the received optical signal, and output a phase-adjusted optical signal, where the phase of the optical signal reaching the beam combiner may be the first phase or the second phase. That is, the second waveguide 140 may adjust only the phase of the received optical signal, unlike the first waveguide 120.

The beam combiner 130 is respectively connected to the N-1 first waveguides 120 and the second waveguide 140, receives the N-1 intermediate optical signals output by the N-1 first waveguides 120 and the optical signals output by the second waveguide 140, and can combine the N-1 intermediate optical signals and the optical signals output by the second waveguide 140 into one optical signal.

Optionally, the light computing device may further include a detector 150, and the detector 150 may detect the light intensity of the light signal output by the beam combiner. The detector 150 may be a photovoltaic type detector or a photoconductive detector.

The multiplication and addition operation to be realized by the optical computing equipment is as follows:is (herein, X is not limited)_iW_iThe sign of (b) may be a positive sign or a negative sign), the optical computing device implements N-1 multiplication operations in the multiplication and addition operations by using the N-1 first waveguides 120 to adjust the amplitude and phase of the N-1 optical signals, wherein each first waveguide 120 adjusts the amplitude of one optical signal in the N-1 optical signals to implement one multiplication operation X in the N-1 multiplication operations_iW_iThat is, the amplitude of the optical signal whose amplitude has been adjusted is the result of the multiplication. Each first waveguide adjusts the phase of one of the N-1 optical signals to the multiplication operation X_iW_iFront addingSigned (e.g., positive, or negative). Then, after the N-1 intermediate optical signals are combined in the beam combiner 130, an optical signal a is output, and the amplitude of the optical signal a can indicateSince the light intensity of the optical signal detected by the detector 150 is the square of the amplitude of the optical signal, and the sign of the amplitude of the optical signal cannot be distinguished, for this reason, the second waveguide 140 is added in the optical computing device, the second waveguide 140 can adjust the phase of the nth optical signal, the optical computing device can adjust the phase of one optical signal through the second waveguide 140, and further change the amplitude of one optical signal after the N-1 intermediate optical signals and the second waveguide 140 are combined in the beam combiner 130, and further change the light intensity of the one optical signal, and determine the sign of the optical signal a through the change of the light intensity (the specific way of determining the sign can be referred to in the following description).

In the embodiment of the present application, since the first waveguide 120 can adjust the amplitude of the optical signal, the optical computing apparatus can implement multiplication; since the first waveguide 120 also has a function of adjusting the phase of the optical signal, the phase of the optical signal can be changed, and a sign is added to the result of each multiplication; the optical signals output by the first waveguide 120 and the second waveguide 140 can then be combined by the combiner 130, so that the optical computing device can implement signed addition. The second waveguide 140 can adjust the phase of the optical signal, so that the light intensity of the optical signal output by the beam combiner can be changed, and the second waveguide 140 can determine the positive and negative of the amplitude of the optical signal after the optical signal output by the first waveguide 120 is combined, thereby ensuring that the optical computing equipment can realize signed multiply-add operation. Similarly, because the first waveguide 120 and the second waveguide 140 are used in the optical computing device, the volume occupied by the optical computing device can be effectively reduced, the expansibility of the optical computing device can be improved, and the data amount calculated by the optical computing device can be increased.

In the optical computing device shown in fig. 2, the second waveguide 140 may perform two phase adjustment operations on the received nth optical signals, and accordingly, the beam combiner may combine two different optical signals.

First phase adjustment operation:

the second waveguide 140 receives an nth optical signal of the N optical signals, and adjusts the received nth optical signal into a second optical signal, where a phase of the second optical signal reaching the beam combiner is a first phase.

The beam combiner 130 combines the N-1 intermediate optical signals output by the N-1 first waveguides 120 and the second optical signal output by the second waveguide into a third optical signal.

Second phase adjustment operation:

the second waveguide 140 receives the nth optical signal of the N optical signals, and adjusts the received nth optical signal into a fourth optical signal, where the phase of the third optical signal reaching the beam combiner is a second phase.

The beam combiner 130 combines the N-1 intermediate optical signals output by the N-1 first waveguides 120 and the fourth optical signal output by the second waveguide into a fifth optical signal.

In order to determine the result of the signed multiply-add operation, i.e. the amplitude of the optical signal after the combination of the N-1 intermediate optical signals.

A detector 150 may also be included in the optical computing device. After the detector 150 detects the light intensities of the third optical signal and the fifth optical signal, the detector 150 determines the light intensities of the third optical signal and the fifth optical signal through the detector, and obtains the sum of products of the data indicated by the N-1 intermediate optical signals according to the light intensities of the third optical signal and the fifth optical signal (for a specific determination, see the following description).

It should be noted that, as is clear from the optical computing apparatus shown in fig. 1 and 2, the optical computing apparatus shown in fig. 2 may also realize the functions that can be realized by the optical computing apparatus shown in fig. 1, and may control the second waveguide not to receive the nth optical signal (e.g., adjust the amplitude of the nth optical signal to 0).

The following takes the optical computing device shown in fig. 2 as an example, and the following describes the components of the optical computing device with reference to the drawings:

(1) a beam splitter 110

In the embodiment of the present application, the optical splitter 110 has an optical splitting function, and the embodiment of the present application does not limit the specific type of the optical splitter 110, for example, the optical splitter 110 may be a multimode interference (MMI) or a cascade Y-branch.

The light received by the optical splitter 110 is continuous light, which means that there is no break in the optical signal. The embodiment of the present application does not limit the source of the continuous light, and any device capable of generating the continuous light is suitable for the embodiment of the present application.

As a possible implementation, a laser 170 may also be included in the light computing device, and the laser 170 may generate the continuous light. Illustratively, the laser 170 may be a single longitudinal mode semiconductor laser that produces continuous light having a single frequency. Single longitudinal mode semiconductor lasers include, but are not limited to: a Distributed Feedback (DFB) laser, a Distributed Bragg Reflector (DBR) laser, or a micro-ring laser.

For the N optical signals generated after the optical splitter 110 splits the light, the N optical signals may be the same, for example, the phase and amplitude of the N optical signals are the same; the N optical signals may be different, for example, the N optical signals may have the same amplitude and different phases.

It should be noted that, when the optical splitter 110 splits light, the amplitudes of the N optical signals may differ due to the principle of splitting light by the optical splitter 110 itself or due to environmental factors, and in order to ensure that the amplitudes of the optical signals are consistent, the amplitudes of the N-1 optical signals may be compensated by using the first waveguide 120. Similarly, due to the principle of splitting by the optical splitter 110 itself or environmental factors, the phases of the N optical signals may also be different, and in order to ensure that the phases of the N optical signals are consistent, the phases of the N optical signals may be compensated by using the first waveguide 120 and the second waveguide 140.

(2) A first waveguide 120

In the embodiment of the present application, the first waveguide 120 has both an amplitude modulation function and a phase modulation function, and as shown in fig. 3, the first waveguide 120 includes one first phase modulation unit 121 and at least two amplitude modulation units. One first phase modulation unit 121 and at least two amplitude modulation units are connected in series. In the embodiment of the present application, the order of one first phase modulation unit and at least two amplitude modulation units is not limited, and in the first waveguide 120 shown in fig. 2, the first waveguide 120 includes two amplitude modulation units and one first phase modulation unit 121, and the example that two amplitude modulation units are located before the first phase modulation unit is described. For the sake of easy distinction, the two amplitude modulation units are respectively represented by the amplitude modulation unit 122A and the amplitude modulation unit 122B.

One of the N optical signals generated by the optical splitter 110 enters the first waveguide 120, passes through the amplitude modulation unit 122A, then passes through the amplitude modulation unit 122B, and finally passes through the first phase modulation unit 121.

When the optical signal passes through the amplitude modulation section 122A and the amplitude modulation section 122B, the amplitude modulation section 122A and the amplitude modulation section 122B can adjust the amplitude of the received optical signal according to the amplitude parameter. The optical signal passes through the amplitude modulation section 122A and the amplitude modulation section 122B in succession, and a set of multiplication operations in the multiplication and addition operations can be realized.

For example, the multiply-add operation to be implemented by the light computing device is:wherein each first waveguide 120 is used to implement a set of multiplication operations X in a multiply-add operation_iW_i. The amplitude parameter may indicate X_iAnd W_iOne amplitude modulation unit 122A in the first waveguide 120 may be according to the indication X_iAdjusts the amplitude of the received optical signal, and then another amplitude modulation unit 122B modulates the amplitude of the received optical signal in accordance with the instruction W_iAmplitude parameter adjusting amplitude modulating section 122A adjusts the amplitude of the optical signal after the amplitude is adjusted. One amplitude modulation unit 122A in the first waveguide 120 may also be based on the indication W_iAdjusts the amplitude of the received optical signal, and then another amplitude modulation unit 122B modulates the amplitude of the received optical signal in accordance with the instruction X_iAmplitude parameter adjusted amplitude modulation section 122A of (1) adjusts the amplitude of the lightThe amplitude of the signal. Here, X is not taken into account_iAnd W_iPositive and negative of the multiplication operation can be realized by the first waveguide 120 by adjusting the phase of the optical signal.

It should be noted that, taking the example that the first waveguide 120 includes two amplitude modulation units, the first waveguide 120 may also include three amplitude modulation units, or even more amplitude modulation units; the number of amplitude modulation units is related to the amount of data in one set of multiplication operations to be performed by the optical computing device, and when the amount of data is three, three amplitude modulation units may be included in the first waveguide 120. That is, the number of amplitude modulation units may be equal to the number of operation parameters.

In the embodiment of the present application, the specific form of the amplitude modulation unit is not limited, and any device capable of adjusting the amplitude of the optical signal may be used as the amplitude modulation unit. For example, the amplitude modulation unit may be an electro-absorption modulator (EAM), a Semiconductor Optical Amplifier (SOA), and a Variable Optical Attenuator (VOA).

When the amplitude modulation unit adopts an EAM or VOA, the absorption coefficient of the optical signal is adjusted by controlling the reverse bias voltage applied to the EAM or VOA, so that the attenuation of the optical signal is realized, and the amplitude of the optical signal is adjusted; wherein the reverse bias voltage is determined by the amplitude parameter.

When the amplitude modulation unit adopts an SOA, the amplification factor of the optical signal is adjusted by controlling the forward bias current applied to the SOA, so that the attenuation or amplification of the optical signal is realized, and the amplitude of the optical signal is adjusted. Wherein, the forward bias current is the amplitude parameter; when the forward bias current is lower than the threshold value, the attenuation of the optical signal can be realized, and the amplitude of the optical signal is reduced; when the forward bias current is higher than the threshold value, amplification of the optical signal can be realized, and the amplitude of the optical signal is increased.

After the optical signal passes through the amplitude modulation unit 122A and the amplitude modulation unit 122B successively, the amplitude of the optical signal is changed, the optical signal with the adjusted amplitude passes through the first phase modulation unit 121, the first phase modulation unit 121 adjusts the phase of the optical signal output by the amplitude modulation unit 122B, that is, the phase of the optical signal output by the first waveguide 120 can be adjusted, and the sign adjustment (for example, positive or negative adjustment) of the result of the multiplication can be realized.

In an actual operation, one optical signal of the N optical signals may be used as reference light, and the first phase modulation unit 121 may adjust a phase of the first waveguide 120 receiving the one optical signal to make a phase difference between the reference light and the one optical signal be 0 or pi, so as to implement a sign adjustment of a result of the multiplication operation.

The phase difference between the reference light and the optical signal is 0, the signs of the amplitudes of the reference light and the optical signal are the same, and the sign of the result of the multiplication performed by the first waveguide 120 can be regarded as a positive sign, assuming that the product of two data indicated by the reference light is a positive sign.

The phase difference between the reference light and the optical signal is pi, the signs of the amplitudes of the reference light and the optical signal are different, and the sign of the result of the multiplication performed by the first waveguide 120 can be regarded as a negative sign, assuming that the product of the two data indicated by the reference light is a positive sign.

Since the optical signal needs to be transmitted continuously after passing through the first phase modulation unit 121 to reach the beam combiner 130, in order to ensure that the beam combiner 130 can accurately implement addition operation when combining the optical signal from the N-1 first waveguides 120 and the optical signal from the second waveguide 140, that is, the sign of the result of each multiplication operation is accurate, the first phase modulation unit 121 may adjust the phase of the optical signal, so that the phase difference between the phase-adjusted optical signal and the reference light when reaching the beam combiner 130 is 0 or pi.

In order to ensure that the first phase modulation unit 121 can accurately adjust the phase difference between the received optical signal and the reference light when reaching the beam combiner 130 to 0 or pi, the first phase modulation unit 121 may be calibrated in advance to determine a first phase parameter capable of adjusting the phase difference between the received optical signal and the reference light when reaching the beam combiner 130 to 0 or pi.

The following description will be made of a manner of calibrating the first phase modulation units 121 in N-1 first waveguides 120 in advance, taking the first phase modulation units 121 in one first waveguide 120 as an example:

the optical splitter 110 splits the received continuous light into N optical signals, and then determines the reference light in the N optical signals.

The N-1 first waveguides 120 receive the N-1 optical signals, and one first waveguide 120 receives one of the N optical signals. The second waveguide 140 receives the remaining optical signals of the N optical signals except the N-1 optical signals.

Taking the optical signal received by the second waveguide 140 as the reference light and the first waveguide 120 to be calibrated as the first waveguide 120B as an example, the amplitude modulation unit of the first waveguide 120 except the first waveguide 120B is adjusted to attenuate the amplitude of the optical signal received by the first waveguide 120 except the first waveguide 120B to zero, so that the first waveguide 120 except the first waveguide 120B does not output the optical signal.

The second waveguide 140 may not adjust the phase of the optical signal. The amplitude modulation unit 122 in the first waveguide 120B may not adjust the amplitude of the optical signal, gradually adjust the operating parameter of the first phase modulation unit 121 in the first waveguide 120B, adjust the phase of the optical signal received by the first waveguide 120B, receive the optical signals output by the second waveguide 140 and the first waveguide 120B, and combine the optical signals output by the second waveguide 140 and the first waveguide 120B into one optical signal.

For example, the optical signal received by the second waveguide 140 isWherein A is₁For the amplitude of the optical signal received by the second waveguide 140,is the phase of the optical signal. A. the₁Can be determined by measurement in advance.

The optical signal received by the first waveguide 120B isWherein A is₂For the amplitude of the optical signal received by the first waveguide 120B,is the phase of the optical signal. A. the₂Can be determined by measurement in advance.

The optical signal combined by the beam combiner 130 is

If the phase difference between the optical signals output from the second waveguide 140 and the first waveguide 120B and reaching the beam combiner 130 is 0,optical signal detected by detector 150Has a light intensity of (A)₁+A₂)²。

If the phase difference between the optical signals output from the second waveguide 140 and the first waveguide 120B reaching the beam combiner 130 is pi,optical signal detected by detector 150Has a light intensity of (A)₁-A₂)²。

Each time the operating parameter of the first phase modulation unit 121 in the first waveguide 120B is adjusted, the light intensity of the optical signal output by the beam combiner 130 can be detected by the detector 150, and the optical signal output by the beam combiner 130 can be determinedHas a light intensity of (A)₁+A₂)²A first operating parameter of the first phase modulating unit 121 of the first waveguide 120B; determining the optical signal output by the beam combiner 130Has a light intensity of (A)₁-A₂)²The second operating parameter of the first phase modulating unit of the first waveguide 120B. The first operating parameter and the second operating parameter may be used as the first phase parameter of the first phase modulation unit 121 of the first waveguide 120B.

In the calibration, the continuous light received by the beam combiner 130 is the same as the continuous light received by the beam combiner 130 when the multiplication and addition operation is performed by the subsequent optical computing device. Accordingly, when performing calibration, the N optical signals generated by the beam combiner 130 using the continuous light are the same as the N optical signals generated by the optical splitter 110 using the continuous light when performing multiply-add operation with a subsequent optical computing device. Therefore, when the first phase modulation unit performs calibration, the determined first phase parameter is accurate, and the subsequent first phase modulation unit can more accurately adjust the phase difference between the received optical signal and the reference light when the received optical signal and the reference light reach the beam combiner 130 to be 0 or pi. In addition, in the above-described method of calibrating the first phase modulation units in the first waveguides 120, only the first phase modulation units in one first waveguide 120 are calibrated as an example, and as a possible embodiment, the first phase modulation units in a plurality of first waveguides 120 may be calibrated at the same time, and the first phase parameters of the first phase modulation units in each of the plurality of first waveguides 120 may be determined.

It should be noted that in the above-described process of calibrating the first phase modulation unit 121 in the first waveguide 120, the influence of the amplitude modulation unit 122 on the phase of the optical signal is ignored. In fact, the amplitude modulation unit 122 in the first waveguide 120 may also change the phase of the optical signal when operating, and for this reason, when calibrating the first phase modulation unit 121 in the first waveguide 120, the amplitude modulation unit 122 in the first waveguide 120 may also be operated to adjust the amplitude of the optical signal; the calibration method of the first phase modulation unit 121 is only an example, and the embodiment of the present application does not limit the specific calibration method of the first phase modulation unit 121, and all calibration methods capable of determining the first phase parameter are applicable to the embodiment of the present application.

In the embodiment of the present application, the specific form of the first phase modulation unit 121 is not limited, and any device capable of adjusting the phase of the optical signal may be used as the first phase modulation unit 121. For example, the first phase modulation unit 121 may be a passive waveguide capable of transmitting an optical signal, and the refractive index of the passive waveguide is changed by carrier injection, carrier depletion, quantum confinement stark effect, or the like, so as to change the phase of the optical signal. The method for changing the refractive index of the passive waveguide by utilizing carrier injection is to adjust the magnitude of forward current applied to the passive waveguide so as to change the number of carriers in the passive waveguide and change the refractive index of the passive waveguide. Changing the refractive index of the passive waveguide by carrier depletion means adjusting the magnitude of a reverse voltage applied to the passive waveguide to change the number of carriers in the passive waveguide and change the refractive index of the passive waveguide. Changing the refractive index of the passive waveguide using the quantum confined stark effect means adjusting the magnitude of the reverse voltage applied to the passive waveguide, so that the energy band of the material of the passive waveguide is bent, thereby changing the refractive index of the passive waveguide. It should be noted that, when only the multiplication and addition operation without a symbol needs to be implemented, the first phase modulation units 121 in the N-1 first waveguides only need to adjust the phases of the N-1 optical signals to the same value, for example, the first phase modulation units 121 in the N-1 first waveguides may adjust the phase difference between the N-1 intermediate optical signals and the reference light reaching the beam combiner to 0.

As a possible implementation manner, since at least two amplitude modulation units 122 in the first waveguide 120 are connected in series with the first phase modulation unit 121, in order to ensure that the at least two amplitude modulation units and the first phase modulation unit can operate normally, electrical isolation is performed between the at least two amplitude modulation units and two adjacent units in the first phase modulation unit, where the electrical isolation is implemented in the two adjacent units to ensure that the operating voltage or current of the two adjacent units cannot crosstalk, and this embodiment of the application does not limit a specific manner of the electrical isolation, and for example, proton implantation, oxygen ion implantation, an isolation groove, or the like may be used.

(3) A second waveguide 140

In the embodiment of the present application, the second waveguide 140 may adjust only the phase of the received optical signal, and as shown in fig. 4, the second waveguide 140 includes a second phase modulation unit 141. The second phase modulation unit 141 has the same function as the first phase modulation unit 121, and can adjust the phase difference between the received optical signal and the reference light when the received optical signal and the reference light reach the beam combiner 130 to 0 or pi. For example, the second phase modulation unit 141 adjusts the phase difference between the second optical signal and the reference light reaching the beam combiner 130 to 0, and adjusts the phase difference between the fourth optical signal and the reference light reaching the beam combiner 130 to pi. For another example, second phase modulation section 141 adjusts the phase difference between the second optical signal and the reference light reaching beam combiner 130 to pi, and adjusts the phase difference between the fourth optical signal and the reference light reaching beam combiner 130 to 0.

In order to ensure that the second phase modulation unit 141 can accurately adjust the phase difference between the received optical signal and the reference light when reaching the beam combiner 130 to 0 or pi, the second phase modulation unit 141 may be calibrated in advance to determine a second phase parameter capable of adjusting the phase difference between the received optical signal and the reference light when reaching the beam combiner 130 to 0 or pi. When calibrating the second phase modulation unit 141, the optical signal received by the calibrated first phase modulation unit 121 may be used as the reference light, and the way of calibrating the second phase modulation unit 141 is the same as the way of calibrating the first phase modulation unit 121, which may be referred to in the foregoing specifically, and is not described herein again.

Optionally, the second waveguide 141 includes an amplitude modulation unit 142, and when the optical computing device is configured to implement a multiplication and addition operation without a symbol, the amplitude modulation unit 142 may adjust the optical intensity of the received nth optical signal to zero, that is, adjust the amplitude of the nth optical signal to zero.

Since the detector 150 can only detect the light intensity of the optical signal output by the beam combiner 130, the light intensity is equal to the square of the amplitude of the optical signal, that is, the phase of the optical signal cannot be determined, that is, the sign of the multiply-add operation of the final output is determined. In view of this, the second waveguide 140 is provided in the embodiment of the present application, and the light intensity of the optical signal output by the beam combiner can be changed through the second waveguide 140, so that the subsequent processing circuit 160 can determine the sum of the products of the data indicated by the N-1 intermediate optical signals according to the light intensities of the different optical signals.

(4) Combiner 130

In the embodiment of the present application, the beam combiner 130 has a beam combining function, the operation performed by the beam combiner 130 is a reverse operation of the operation performed by the optical splitter 110, and the embodiment of the present application does not limit the specific type of the beam combiner 130, for example, the beam combiner 130 may be an MMI, or may be a cascade Y branch.

(5) And a processing circuit 160.

The following describes how the processing circuit 160 determines the result of the multiply-add operation that is finally output:

suppose that the amplitude Y of the output optical signal after the N-1 intermediate optical signals output by the N-1 first waveguides 120 are combined is:

suppose that the second waveguide 140 outputs an optical signal ofThe optical signal has an amplitude A_N。A_NThe determination may be measured in advance when calibrating the second phase modulation unit.

If the phase difference between the optical signal output by the second waveguide 140 and the reference light reaching the beam combiner 130 is 0, taking the optical signal output by the second waveguide 140 as the second optical signal and the optical signal output by the beam combiner as the third optical signal as an example, the third optical signal detected by the detector 150 isHas a light intensity of M₀＝(Y+A_N)²。

If the phase difference between the optical signal output by the second waveguide 140 and the reference light reaching the beam combiner 130 is pi, taking the optical signal output by the second waveguide 140 as the fourth optical signal and the optical signal output by the beam combiner as the fifth optical signal as an example, the detection is performedFifth optical signal detected by the detector 150Has a light intensity of M_π＝(Y-A_N)²。

Thus, it can be seen that:

as can be seen from the above process, the second phase modulation unit 140 respectively performs two phase adjustment operations on the optical signal, respectively adjusts the phase difference between the optical signal output by the second waveguide 140 and the reference light reaching the beam combiner 130 to 0 and pi, and the detector 150 respectively determines the light intensity M of the third optical signal output by the beam combiner 130₀And the light intensity M of the fifth optical signal output by the beam combiner 130_π(ii) a Then according to M₀And M_πThe sum of products of data indicated by the N-1 intermediate optical signals, that is, the operation result of the multiply-add operation, is determined.

The embodiment of the present application is not limited to a specific type of the processing circuit 160, and the processing circuit 160 may be a simple logic circuit, such as an adder, as a possible implementation manner, and the functions of the processing circuit 160 may also be implemented by other devices, for example, the functions of the processing circuit 160 may be implemented by a Central Processing Unit (CPU), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA), an Artificial Intelligence (AI) chip, a system on chip (SoC), or a Complex Programmable Logic Device (CPLD). Any circuit or device capable of determining the sum of products of data indicated by N-1 paths of intermediate optical signals according to the light intensity of the optical signals output by the beam combiner is suitable for the embodiment of the present application.

In order to make the description of the scheme clearer, the workflow of the optical computing device provided in the embodiment of the present invention will be generally described below with reference to the foregoing embodiments, by taking the optical computing device shown in fig. 1 and the optical computing method shown in fig. 5 as an example. In this embodiment, after receiving the continuous light, the optical splitter 110 splits the continuous light into N optical signals (optionally, the optical splitter may also split the continuous light into N-1 optical signals), and N-1 first waveguides 120 receive the N-1 optical signals in the N optical signals, where one first waveguide 120 receives one optical signal in the N-1 optical signals. Each first waveguide 120 adjusts the amplitude of the received optical signal according to the set amplitude parameter, and each first waveguide outputs an intermediate optical signal. The beam combiner 130 receives the intermediate optical signals output by the N-1 first waveguides 120, and combines the N-1 intermediate optical signals into one optical signal. The detector 150 detects the light intensity of the optical signal and the processing circuitry 160 determines the sum of the products of the data indicated by the N-1 intermediate optical signals based on the light intensity of the optical signal.

In order to make the description of the scheme clearer, the workflow of the optical computing device provided in the embodiment of the present invention will be generally described below with reference to the foregoing embodiments, by taking the optical computing device shown in fig. 2 and the optical computing method shown in fig. 6 as an example. In this embodiment, after receiving the continuous light, the optical splitter 110 splits the continuous light into N optical signals, and one second waveguide 140 and N-1 first waveguides 120 receive the N optical signals, where one first waveguide 120 receives one optical signal of the N optical signals, and one second waveguide 140 receives an nth optical signal of the N optical signals. Each first waveguide 120 adjusts the amplitude of the received optical signal according to the set amplitude parameter, adjusts the phase of the optical signal according to the set phase parameter, and outputs an intermediate optical signal. The second waveguide 140 adjusts the phase of the received optical signal so that the phase of the optical signal with the adjusted phase reaching the beam combiner 130 is the first phase or the second phase. The beam combiner 130 receives the N-1 optical signals output by the first waveguide 120 and the second waveguide 140, and combines the N-1 intermediate optical signals and the optical signal output by the second waveguide 140 into one optical signal. The detector 150 detects the light intensity of the optical signal output by the beam combiner 130, and the processing circuit 160 determines the sum of the products of the data indicated by the N-1 intermediate optical signals according to the light intensity of the optical signal output by the beam combiner 130.

As a possible implementation, when the phase of the second optical signal output by the second waveguide 140 reaching the beam combiner 130 is a first phase, the beam combiner 130 receives the N-1 intermediate optical signals and the second optical signal output by the second waveguide 140, and combines the N-1 intermediate optical signals and the second optical signal output by the second waveguide 140 into a third optical signal; the detector 150 detects the light intensity of the third light signal.

When the phase at which the fourth optical signal output by the second waveguide 140 reaches the beam combiner 130 is the second phase, the beam combiner 130 receives the N-1 intermediate optical signals and the fourth optical signal output by the second waveguide 140, and combines the N-1 intermediate optical signals and the fourth optical signal output by the second waveguide 140 into a fifth optical signal; the detector 150 detects the light intensity of the fifth light signal.

The processing circuit 160 obtains the sum of the products of the data indicated by the N-1 intermediate optical signals according to the light intensity of the third optical signal and the light intensity of the fifth optical signal output by the detector 150.

The difference between the first phase and the phase when the reference light reaches the beam combiner is 0, the difference between the second phase and the phase when the reference light reaches the beam combiner is pi, and the reference light is one of the N optical signals. Or the difference between the first phase and the phase when the reference light reaches the beam combiner is pi, the difference between the second phase and the phase when the reference light reaches the beam combiner is 0, and the reference light is one optical signal of the N optical signals.

As shown in fig. 7, a computing system 10 provided for embodiments of the present application, the computing system 10 includes a light computing device 100 and a processor 200. The structure of the optical computing device 100 can be seen in the optical computing device shown in fig. 1 or fig. 2.

The processor 200 is connected to the optical computing device 100, and the processor 200 may transmit data to be subjected to multiply-add operation to the optical computing device 100 to instruct the optical computing device 100 to perform optical computation on the received data. The data to be subjected to multiply-add operation comprises N-1 groups of data, and each group of data comprises at least two data to be subjected to multiply-add operation in the multiply-add operation.

After receiving the data to be subjected to multiply-add operation, the optical computing device 100 may set an amplitude parameter of the N-1 first waveguide, and optionally, may set a phase parameter of the N-1 first waveguide.

For any first waveguide, light computing device 100 may set at least two amplitude parameters of the first waveguide according to a set of data, and set a phase parameter of the first waveguide according to the positive and negative values of the set of data. Thereafter, the light calculation apparatus 100 may perform light calculation, output the calculation result of the multiply-add operation, and feed back the calculation result to the processor 200. In practical applications, the calculation result may be determined by (the processing circuit in) the light calculation apparatus 100 according to the first light signal, or may be determined according to the light intensity of the third light signal and the light intensity of the fifth light signal.

It should be noted that the examples provided in this application are only illustrative. It will be apparent to those skilled in the art that, for convenience and brevity of description, the description of the various embodiments has been focused on, and for parts of one embodiment that are not described in detail, reference may be made to the description of other embodiments. The features disclosed in the embodiments of the invention, in the claims and in the drawings may be present independently or in combination. Features described in hardware in embodiments of the invention may be implemented by software and vice versa. And are not limited herein.