阅读说明：本技术 一种光学芯片设计方法、系统、设备及计算机存储介质 (Optical chip design method, system, equipment and computer storage medium ) 是由 吴睿振 王凛 陈静静 黄萍 于 2021-09-23 设计创作，主要内容包括：本申请公开了一种光学芯片设计方法、系统、设备及计算机可读存储介质,获取目标非线性运算函数；基于目标非线性运算函数中的自变量及MZI的结构,确定MZI的拟合连接关系；基于拟合连接关系,将目标非线性运算函数转换为MZI待拟合的拟合非线性运算函数；基于最小二乘法确定拟合非线性运算函数对应的拟合线性运算函数；基于拟合线性运算函数及拟合连接关系,确定目标光学芯片的结构,目标光学芯片的结构包括MZI的数量、连接关系及配置参数,以基于目标光学芯片的结构设计目标光学芯片。本申请中保证了目标光学芯片基于MZI实现目标非线性运算函数的准确性,且整个光学芯片设计过程中无需将光信号转换为电信号,结构简便,运算效率高。(The application discloses a method, a system and equipment for designing an optical chip and a computer readable storage medium, which are used for obtaining a target nonlinear operation function; determining a fitting connection relation of the MZI based on the independent variable in the target nonlinear operation function and the structure of the MZI; converting the target nonlinear operation function into a fitting nonlinear operation function to be fitted of the MZI based on the fitting connection relation; determining a fitting linear operation function corresponding to the fitting nonlinear operation function based on a least square method; and determining the structure of the target optical chip based on the fitting linear operation function and the fitting connection relation, wherein the structure of the target optical chip comprises the number of MZIs, the connection relation and the configuration parameters, so as to design the target optical chip based on the structure of the target optical chip. The accuracy of the target optical chip for realizing the target nonlinear operation function based on the MZI is guaranteed, the optical signal does not need to be converted into the electric signal in the whole optical chip design process, the structure is simple and convenient, and the operation efficiency is high.)

1. An optical chip design method, comprising:

obtaining a target nonlinear operation function to be realized by a target optical chip, wherein the target optical chip is designed based on MZI;

determining a fitting connection relation of the MZI based on an independent variable in the target nonlinear operation function and the structure of the MZI;

based on the fitting connection relation, converting the target nonlinear operation function into a fitting nonlinear operation function to be fitted of the MZI;

determining a fitting linear operation function corresponding to the fitting nonlinear operation function based on a least square method;

and determining the structure of the target optical chip based on the fitting linear operation function and the fitting connection relation, wherein the structure of the target optical chip comprises the number, the connection relation and the configuration parameters of the MZIs, so that the target optical chip is designed based on the structure of the target optical chip.

2. The method of claim 1, wherein determining the fitting connection relationship of the MZI based on the independent variables in the target nonlinear operational function and the structure of the MZI comprises:

using an independent variable in the target nonlinear operation function as an input of the MZI, and determining the connection relation of two outputs of the MZI based on the structure of the MZI;

and determining the output connection relation with the characteristics consistent with the target nonlinear operation function as the fitting connection relation.

3. The method of claim 2, wherein said determining a fitted linear operational function of said fitted non-linear operational function based on a least squares method comprises:

obtaining the design precision of the target optical chip;

and determining the fitted linear operation function corresponding to the fitted nonlinear operation function based on the design precision and the least square method.

4. The method of claim 3, wherein the design precision comprises a dataset range for the independent variables;

the determining the fitted linear operation function corresponding to the fitted nonlinear operation function based on the design accuracy and the least square method includes:

determining an initial linear operation function of the fitting nonlinear operation function at each power based on the least square method;

determining the similarity between each initial linear operational function and the fitting nonlinear operational function in the range of the data set;

and taking the initial linear operation function with the maximum similarity value as the fitting linear operation function.

5. The method of claim 4, wherein determining the structure of the target optical chip based on the fitted linear operation function and the fitted connection relationship comprises:

and determining the structure of the target optical chip based on the fitting linear operation function, the fitting connection relation and a singular value decomposition method.

6. The method of claim 5, wherein determining the structure of the target optical chip based on the fitted linear operation function, the fitted connection relation, and a singular value decomposition method comprises:

determining a link arithmetic matrix of the target optical chip based on the fitting linear operation function and the fitting connection relation;

decomposing the link arithmetic matrix based on the singular value decomposition method to obtain a matrix decomposition result;

determining a structure of the target optical chip based on the matrix decomposition result.

7. The method of claim 6, wherein the target non-linear operational function comprises a sigmoid function;

the structure of the target optical chip includes: the device comprises a first MZI, a photodiode, a first amplifier, a phase-shift controller, a second MZI and a target topological structure; a first optical input end of the first MZI is connected with a second optical input end, a second optical output end of the first MZI is suspended, and a first optical output end of the first MZI is respectively connected with an input end of the photodiode, a first optical input end of the second MZI and a second optical input end of the target topology structure; the output end of the photodiode is connected with the first amplifier; the first amplifier is connected with the phase-shifting controller; the phase shifting controller is connected with the phase shifter of the second MZI; a second optical input end and a second optical output end of the second MZI are both suspended; a first optical output of the second MZI is connected to a first optical input of the target topology.

8. An optical chip design system, comprising:

the device comprises a first acquisition module, a second acquisition module and a third acquisition module, wherein the first acquisition module is used for acquiring a target nonlinear operation function to be realized of a target optical chip, and the target optical chip is designed based on MZI;

the first determining module is used for determining the fitting connection relation of the MZI based on the independent variable in the target nonlinear operation function and the structure of the MZI;

the first conversion module is used for converting the target nonlinear operation function into a fitting nonlinear operation function to be fitted of the MZI based on the fitting connection relation;

the second determination module is used for determining a fitting linear operation function corresponding to the fitting nonlinear operation function based on a least square method;

a third determining module, configured to determine a structure of the target optical chip based on the fitted linear operation function and the fitted connection relationship, where the structure of the target optical chip includes the number of MZIs, the connection relationship, and configuration parameters, so as to design the target optical chip based on the structure of the target optical chip.

9. An optical chip designing apparatus, comprising:

a memory for storing a computer program;

a processor for implementing the steps of the optical chip design method according to any one of claims 1 to 7 when executing said computer program.

10. A computer-readable storage medium, in which a computer program is stored which, when being executed by a processor, carries out the steps of the optical chip design method according to any one of claims 1 to 7.

Technical Field

The present application relates to the field of electronic information technology, and more particularly, to a method, system, device and computer storage medium for designing an optical chip.

Background

As communication technology develops, the demand for optical computing technology is rapidly increasing due to the following reasons: firstly, with the gradual failure of moore's law and the continuous improvement of the requirements of the big data era on the power consumption and the speed of a computing system, the characteristics of high speed and low power consumption of an optical computing technology are more and more emphasized by people; secondly, the parallelism operation characteristic of the optical computing technology and the development of algorithms and hardware architectures such as an optical neural network provide the most potential solution for the demands of the artificial intelligence technologies such as image recognition, voice recognition, virtual reality and the like on computing power. The light calculation can be divided into an analog light calculation and a digital light calculation. The most typical example of the analog light calculation is fourier operation, and fourier transform related calculation, such as convolution calculation, needs to be applied in the field of image processing and the like. The calculation of the fourier transform with a conventional computer is very computationally expensive, and the passage of light through the lens is itself a fourier transform process, which requires almost no time at all. The digital optical calculation is to form a classic logic gate by combining light and an optical device, construct a calculation system similar to the traditional digital electronic calculation principle, and realize calculation through complex logic gate combination operation.

Although the optical chip generated by optical computation can quickly process data, the conventional optical chip needs to convert an optical signal into an electrical signal in the process of realizing nonlinear operation, so that the conventional optical chip has a complex structure and low operation efficiency.

In summary, the problem to be solved by those skilled in the art is how to improve the operation efficiency of the optical chip for implementing the nonlinear operation.

Disclosure of Invention

The present application aims to provide a method for designing an optical chip, which can solve the technical problem of how to improve the operation efficiency of the optical chip for realizing nonlinear operation to a certain extent. The application also provides an optical chip design system, an apparatus and a computer readable storage medium.

In order to achieve the above purpose, the present application provides the following technical solutions:

an optical chip design method, comprising:

obtaining a target nonlinear operation function to be realized by a target optical chip, wherein the target optical chip is designed based on MZI;

determining a fitting connection relation of the MZI based on an independent variable in the target nonlinear operation function and the structure of the MZI;

based on the fitting connection relation, converting the target nonlinear operation function into a fitting nonlinear operation function to be fitted of the MZI;

determining a fitting linear operation function corresponding to the fitting nonlinear operation function based on a least square method;

and determining the structure of the target optical chip based on the fitting linear operation function and the fitting connection relation, wherein the structure of the target optical chip comprises the number, the connection relation and the configuration parameters of the MZIs, so that the target optical chip is designed based on the structure of the target optical chip.

Preferably, the determining the fitting connection relationship of the MZI based on the independent variable in the target nonlinear operation function and the structure of the MZI includes:

using an independent variable in the target nonlinear operation function as an input of the MZI, and determining the connection relation of two outputs of the MZI based on the structure of the MZI;

and determining the output connection relation with the characteristics consistent with the target nonlinear operation function as the fitting connection relation.

Preferably, the determining a fitted linear operation function of the fitted nonlinear operation function based on the least square method includes:

obtaining the design precision of the target optical chip;

and determining the fitted linear operation function corresponding to the fitted nonlinear operation function based on the design precision and the least square method.

Preferably, the design precision comprises a dataset range of the independent variables;

the determining the fitted linear operation function corresponding to the fitted nonlinear operation function based on the design accuracy and the least square method includes:

determining an initial linear operation function of the fitting nonlinear operation function at each power based on the least square method;

determining the similarity between each initial linear operational function and the fitting nonlinear operational function in the range of the data set;

and taking the initial linear operation function with the maximum similarity value as the fitting linear operation function.

Preferably, the determining the structure of the target optical chip based on the fitted linear operation function and the fitted connection relationship includes:

and determining the structure of the target optical chip based on the fitting linear operation function, the fitting connection relation and a singular value decomposition method.

Preferably, the determining the structure of the target optical chip based on the fitting linear operation function, the fitting connection relationship, and a singular value decomposition method includes:

determining a link arithmetic matrix of the target optical chip based on the fitting linear operation function and the fitting connection relation;

decomposing the link arithmetic matrix based on the singular value decomposition method to obtain a matrix decomposition result;

determining a structure of the target optical chip based on the matrix decomposition result.

Preferably, the target nonlinear operation function comprises a sigmoid function;

the structure of the target optical chip includes: the device comprises a first MZI, a photodiode, a first amplifier, a phase-shift controller, a second MZI and a target topological structure; a first optical input end of the first MZI is connected with a second optical input end, a second optical output end of the first MZI is suspended, and a first optical output end of the first MZI is respectively connected with an input end of the photodiode, a first optical input end of the second MZI and a second optical input end of the target topology structure; the output end of the photodiode is connected with the first amplifier; the first amplifier is connected with the phase-shifting controller; the phase shifting controller is connected with the phase shifter of the second MZI; a second optical input end and a second optical output end of the second MZI are both suspended; a first optical output of the second MZI is connected to a first optical input of the target topology.

An optical chip design system, comprising:

the device comprises a first acquisition module, a second acquisition module and a third acquisition module, wherein the first acquisition module is used for acquiring a target nonlinear operation function to be realized of a target optical chip, and the target optical chip is designed based on MZI;

the first determining module is used for determining the fitting connection relation of the MZI based on the independent variable in the target nonlinear operation function and the structure of the MZI;

the first conversion module is used for converting the target nonlinear operation function into a fitting nonlinear operation function to be fitted of the MZI based on the fitting connection relation;

the second determination module is used for determining a fitting linear operation function corresponding to the fitting nonlinear operation function based on a least square method;

a third determining module, configured to determine a structure of the target optical chip based on the fitted linear operation function and the fitted connection relationship, where the structure of the target optical chip includes the number of MZIs, the connection relationship, and configuration parameters, so as to design the target optical chip based on the structure of the target optical chip.

An optical chip designing apparatus comprising:

a memory for storing a computer program;

a processor for implementing the steps of the optical chip design method as described in any one of the above when executing the computer program.

A computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of any of the optical chip design methods described above.

According to the design method of the optical chip, a target nonlinear operation function to be realized by the target optical chip is obtained, and the target optical chip is designed based on MZI; determining a fitting connection relation of the MZI based on the independent variable in the target nonlinear operation function and the structure of the MZI; converting the target nonlinear operation function into a fitting nonlinear operation function to be fitted of the MZI based on the fitting connection relation; determining a fitting linear operation function corresponding to the fitting nonlinear operation function based on a least square method; and determining the structure of the target optical chip based on the fitting linear operation function and the fitting connection relation, wherein the structure of the target optical chip comprises the number of MZIs, the connection relation and the configuration parameters, so as to design the target optical chip based on the structure of the target optical chip. In the application, the fitting connection relation of the MZI can be determined based on the independent variable in the target nonlinear operation function and the structure of the MZI, so that the fitting connection relation of the MZI is adaptive to the target nonlinear operation function; the target nonlinear operation function can be converted into a fitting nonlinear operation function to be fitted by the MZI based on the fitting connection relation, so that the fitting nonlinear operation function to be realized by the MZI is adaptive to the structure of the MZI; and the fitting linear operation function corresponding to the fitting nonlinear operation function can be determined based on a least square method, and the structure of the target optical chip is determined based on the fitting linear operation function and the fitting connection relation, so that the operation characteristic of the target optical chip is adapted to the connection framework of the MZI, the accuracy of the target optical chip for realizing the target nonlinear operation function based on the MZI is ensured, an optical signal does not need to be converted into an electric signal in the whole optical chip design process, the structure is simple and convenient, and the operation efficiency is high. The optical chip design system, the optical chip design equipment and the computer readable storage medium solve the corresponding technical problems.

Drawings

In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below, it is obvious that the drawings in the following description are only embodiments of the present application, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.

Fig. 1 is a flowchart of a method for designing an optical chip according to an embodiment of the present disclosure;

FIG. 2 is a schematic diagram of an MZI;

FIG. 3 is a schematic diagram of a direct connection structure of MZI;

FIG. 4 is a graph of a comparison of various fit results with a sigmoid function;

FIG. 5 is a schematic diagram of a partial structure of an optical chip implementing a sigmoid function;

FIG. 6 is a schematic diagram of a structure of an optical chip implementing a sigmoid function;

fig. 7 is a schematic structural diagram of a light chip design system according to an embodiment of the present disclosure;

fig. 8 is a schematic structural diagram of an optical chip designing apparatus according to an embodiment of the present disclosure;

fig. 9 is another schematic structural diagram of an optical chip designing apparatus according to an embodiment of the present disclosure.

Detailed Description

The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.

Referring to fig. 1, fig. 1 is a flowchart illustrating a method for designing an optical chip according to an embodiment of the present disclosure.

The optical chip design method provided by the embodiment of the application can comprise the following steps:

step S101: and acquiring a target nonlinear operation function to be realized by the target optical chip, wherein the target optical chip is designed based on the Mach-Zehnder interferometer.

In practical application, because the target optical chip is used for implementing a corresponding target operational function, the target nonlinear operational function to be implemented by the target optical chip needs to be obtained first, the type of the target nonlinear operational function can be determined according to actual needs, for example, the target nonlinear operational function can be a sigmoid function, and the like, and the target optical chip in the present application can be designed based on MZI (Mach-Zehnder interferometer), where the Mach-Zehnder interferometer described herein is an interferometer, and can be used for observing relative phase shift changes generated by a medium and two collimated light beams after a light beam emitted from a single light source is split into two collimated light beams through different paths.

Step S102: and determining the fitting connection relation of the Mach-Zehnder interferometer based on the independent variable in the target nonlinear operation function and the structure of the Mach-Zehnder interferometer.

In practical applications, since the MZI has two inputs and two outputs, and the connection relationship between the two outputs is different, after the target nonlinear operation function is obtained, the fitting connection relationship of the MZI needs to be determined based on the independent variable in the target nonlinear operation function and the structure of the MZI.

It should be noted that the structure of MZI can be as shown in FIG. 2, wherein、Two optical inputs of the MZI are shown,a phase shifter of the MZI is shown,、two optical outputs representing the MZI; the calculated relationship between the optical output and the optical input of the MZI can be as follows:

；

derivation process thereofThe following were used: suppose the amplitude of the light input to the MZI isAt a frequency ofAt a time ofThe initial phases of the two light input ends are respectivelyAndthen the expression for the two optical inputs of MZI can be:

；

since in the photoelectric conversion, only the real part of light can be identified, and the imaginary part represents the energy loss during transmission, the above equation can be further converted into:

；

whereinThe real part is represented by,the number of the units of the imaginary number is expressed,to representIs used to represent the complex state of the state,to representComplex state expressions of (2);

because of the fact thatAndafter entering the MZI, the energy contained in the light is transferred to the optical signals corresponding to the two output ports through the coupler, so that the amplitude of the optical signals is the original oneThus in MZI、And、the relationship between can be formulated as:

；

based on the coupler relationship, it can knowAndthe corresponding relation is as follows:

；

based on the working mode of the attenuator of the MZI and the structure of the MZI, the attenuator can be obtained、The expression of (a) is:

；

the calculation relationship between the optical output and the optical input of the MZI can be found by combining the above expressions as follows:

。

in addition, when the phase shifter of the MZI needs to be controlled, the programmable phase shifter can be implemented by, for example, plating a metal film on a section of waveguide material, and applying an external voltage to control the metal film heater to cause the waveguide temperature change to change the refractive index, so as to implement phase shift; for example, the phase shift is introduced by changing the refractive index of the waveguide by using the plasma dispersion effect (changing the concentration of electrons and holes) and the electro-optic effect; the present application is not specifically limited herein.

Step S103: and based on the fitting connection relation, converting the target nonlinear operation function into a fitting nonlinear operation function to be fitted by the Mach-Zehnder interferometer.

In practical application, because the target nonlinear operation function is not adapted to the MZI structure, after the fitting connection relationship of the MZI is determined based on the independent variable in the target nonlinear operation function and the MZI structure, the target nonlinear operation function is converted into the fitting nonlinear operation function to be fitted of the MZI based on the fitting connection relationship.

Step S104: and determining a fitting linear operation function corresponding to the fitting nonlinear operation function based on a least square method.

In practical application, because the MZI can only implement linear multiply-add operation, in the process of implementing a target nonlinear operation function based on a target optical chip, a fitting linear operation function corresponding to the fitting nonlinear operation function needs to be determined based on a least square method.

Step S105: and determining the structure of the target optical chip based on the fitting linear operation function and the fitting connection relation, wherein the structure of the target optical chip comprises the number, the connection relation and the configuration parameters of the Mach-Zehnder interferometers, and the target optical chip is designed based on the structure of the target optical chip.

In practical application, after determining the fitting linear operation function corresponding to the fitting nonlinear operation function based on the least square method, although the fitting linear operation function and the fitting connection relationship can be obtained, the fitting connection relationship only indicates a structure for implementing the target nonlinear operation function by the MZI, and the configuration parameters in the MZI are not determined, so that the structure of the target optical chip can be determined based on the fitting linear operation function and the fitting connection relationship, and the structure of the target optical chip can include the number, the connection relationship and the configuration parameters of the mach-zehnder interferometers, so that the target optical chip capable of implementing the target nonlinear operation function is designed based on the structure of the target optical chip.

According to the design method of the optical chip, a target nonlinear operation function to be realized by the target optical chip is obtained, and the target optical chip is designed based on MZI; determining a fitting connection relation of the MZI based on the independent variable in the target nonlinear operation function and the structure of the MZI; converting the target nonlinear operation function into a fitting nonlinear operation function to be fitted of the MZI based on the fitting connection relation; determining a fitting linear operation function corresponding to the fitting nonlinear operation function based on a least square method; and determining the structure of the target optical chip based on the fitting linear operation function and the fitting connection relation, wherein the structure of the target optical chip comprises the number of MZIs, the connection relation and the configuration parameters, so as to design the target optical chip based on the structure of the target optical chip. In the application, the fitting connection relation of the MZI can be determined based on the independent variable in the target nonlinear operation function and the structure of the MZI, so that the fitting connection relation of the MZI is adaptive to the target nonlinear operation function; the target nonlinear operation function can be converted into a fitting nonlinear operation function to be fitted by the MZI based on the fitting connection relation, so that the fitting nonlinear operation function to be realized by the MZI is adaptive to the structure of the MZI; and the fitting linear operation function corresponding to the fitting nonlinear operation function can be determined based on a least square method, and the structure of the target optical chip is determined based on the fitting linear operation function and the fitting connection relation, so that the operation characteristic of the target optical chip is adapted to the connection framework of the MZI, the accuracy of the target optical chip for realizing the target nonlinear operation function based on the MZI is ensured, an optical signal does not need to be converted into an electric signal in the whole optical chip design process, the structure is simple and convenient, and the operation efficiency is high.

In the method for designing an optical chip, in a process of determining a fitting connection relationship of a MZI based on an argument in a target nonlinear operation function and a structure of the MZI, because the argument of the target nonlinear operation function needs to be used as an input of the MZI, and the argument is input into the MZI, corresponding operation is performed according to the structure of the MZI, and only an output with an operation characteristic consistent with a characteristic of the target nonlinear operation function can be used as an analog output of the MZI to implement the target nonlinear operation function, the process may include the following steps: taking an independent variable in a target nonlinear operation function as an input of the MZI, and determining the connection relation of two outputs of the MZI based on the structure of the MZI; and determining the output connection relation with the consistent characteristics and the target nonlinear operation function as a fitting connection relation.

For the convenience of understanding, the target nonlinear operation function is still taken as a sigmoid function as an example, and the expression is as follows:(ii) a Because only one independent variable exists at this time, one input port of the MZI needs to be sealed, the structure of the MZI is shown in fig. 3, and the corresponding output of the sigmoid function after the independent variable is input into the MZI is:since only the output observed by the MZI is none, the corresponding output after the argument input MZI may be equivalent to(ii) a Finally, as can be seen from the curve characteristics of the sigmoid function, the function value cannot be less than 0, and only the function value in MZIThe connection relation can realize fitting, so that the fitting connection relation of the sigmoid function in the MZI is as follows:. At this time, in the process of converting the target nonlinear operation function into the fitting nonlinear operation function to be fitted through the MZI based on the fitting connection relationship, becauseIs equivalent toThe function may vary coefficients from the derived variable, so the fitted nonlinear operation function for the sigmoid function may be。

In the optical chip design method provided by the embodiment of the application, in the process of determining the fitting linear operation function of the fitting nonlinear operation function based on the least square method, the target optical chip cannot perfectly realize the target nonlinear operation function, and at this time, the precision of realizing the target nonlinear operation function by the target optical chip needs to be controlled, so that correspondingly, the fitting linear operation function can be determined according to the precision, that is, the design precision of the target optical chip can be obtained; and determining a fitting linear operation function corresponding to the fitting nonlinear operation function based on the design precision and the least square method.

In practical application, because the target nonlinear operation function is nonlinear and the fitting linear operation function is linear, the value range of the independent variable affects the fitting precision of fitting the target nonlinear operation function into the fitting linear operation function, and accordingly, the design precision can be represented by the range of the independent variable, namely the design precision can include the data set range of the independent variable;

correspondingly, in the process of determining the fitting linear operation function corresponding to the fitting nonlinear operation function based on the design precision and the least square method, the initial linear operation function of the fitting nonlinear operation function at each power can be determined based on the least square method; determining the similarity between each initial linear operation function and the fitting nonlinear operation function in the range of the data set; and taking the initial linear operation function with the maximum similarity value as a fitting linear operation function.

For ease of understanding, and again taking the sigmoid function described above as an example, in the data set [ -10,10 [ ]]Inner pairPerforming least square linear fitting, and assuming that the preset power is 1-5 powers, the fitting result can be:

；

similarity comparison is carried out on each fitting result and the sigmoid function, and the comparison result is shown in fig. 4, so that the most suitable fitting power is 3, and the fitting linear operation function of the sigmoid function can be as follows:。

in the optical chip design method provided in the embodiment of the present application, in the process of determining the structure of the target optical chip based on the fitting linear operation function and the fitting connection relationship, because in the application process of the MZI, the matrix multiplication operation of any dimension can be implemented in the optical domain by using a Singular Value Decomposition (SVD) method, the structure of the target optical chip can be determined based on the fitting linear operation function, the fitting connection relationship, and the Singular Value Decomposition method in the present application.

In practical application, in the process of determining the structure of the target optical chip based on the fitting linear operation function, the fitting connection relation and the singular value decomposition method, the link arithmetic matrix of the target optical chip can be determined based on the fitting linear operation function and the fitting connection relation; decomposing the link arithmetic matrix based on a singular value decomposition method to obtain a matrix decomposition result; the structure of the target optical chip is determined based on the matrix decomposition result.

Still taking the sigmoid function as an example, assume that the fitting linear operation function of the sigmoid function is:，for a composite structure, this can be achieved by designing other device structures, assuming that the structure is as shown in figure 5,the output of (a) is first converted into electrical energy by the photodiode, since the electrical energy is converted based on light energy, the parameter isSubjecting the obtained voltage toMagnification, then pass throughThe over-electricity is converted into heat of a phase-shift controller to control phase modulation, so that a phase-shift controller can be obtainedTo a phase angleThe adjustment relationship of (1) is as follows:(ii) a Further, the output of FIG. 5 is the calculation relationshipBased on trigonometric function operation relationshipIt follows that the amplification of the first amplifier may beSo the final mapping relationship of the target optical chip can be(ii) a Because of the final mapping of the target optical chipCan easily pass through the configurationRealized in the MZI structure of fig. 3, the-1 in the final mapping relationship of the target optical chip can be realized by introducing fixed optical information and performing standard multiply-add operation after SVD decomposition through the MZI; and in the final mapping of the target optical chipHaving been realized by the structure shown in fig. 5, the linked arithmetic matrix of the target optical chip can be:

；

after SVD is carried out on the obtained products, the following products can be obtained:

；

suppose thatIf the target topology is implemented by the existing MZI-based design, the structure of the target optical chip may be as shown in fig. 6, and may include: a first MZI11, a photodiode 12, a first amplifier 13, a phase-shift controller 14, a second MZI15, a target topology 16; a first optical input end of the first MZI is connected with a second optical input end, a second optical output end of the first MZI is suspended, and a first optical output end of the first MZI is respectively connected with an input end of a photodiode, a first optical input end of the second MZI and a second optical input end of a target topological structure; the output end of the photodiode is connected with the first amplifier; the first amplifier is connected with the phase-shifting controller; the phase shifting controller is connected with a phase shifter of the second MZI; a second optical input end and a second optical output end of the second MZI are both suspended; a first optical output end of the second MZI is connected with a first optical input end of the target topology; among them, the types of target topology may include but are not limited to: a Gridnet topology, a Reck topology, and the like, for example, the target topology may include a third MZI161, a fourth MZI162, a fifth MZI163, and a laser transmitter, where at this time, a first optical input end of the target topology is a first optical input end of the third MZI, a second optical input end of the target topology is a second optical input end of the third MZI, and an optical output end of the target topology is a first optical output end of the fifth MZI; a first optical output end of the third MZI is connected with a first optical input end of the fifth MZI, and a second optical output end of the third MZI is connected with a first optical input end of the fourth MZI; the laser transmitter is connected with a second optical input end of the fourth MZI; a first optical output of the fourth MZI and a second optical output of the fifth MZIThe input end is connected, and a second optical output end of the fourth MZI is suspended; the second optical output of the fifth MZI is suspended.

Referring to fig. 7, fig. 7 is a schematic structural diagram of a light chip design system according to an embodiment of the present disclosure.

An optical chip design system provided in an embodiment of the present application may include:

the first obtaining module 101 is configured to obtain a target nonlinear operation function to be implemented by a target optical chip, where the target optical chip is designed based on MZI;

a first determining module 102, configured to determine a fitting connection relationship of the MZI based on an independent variable in the target nonlinear operation function and a structure of the MZI;

the first conversion module 103 is configured to convert the target nonlinear operation function into a fitting nonlinear operation function to be fitted of the MZI based on the fitting connection relationship;

a second determining module 104, configured to determine a fitting linear operation function corresponding to the fitting nonlinear operation function based on a least square method;

and a third determining module 105, configured to determine a structure of the target optical chip based on the fitted linear operation function and the fitted connection relationship, where the structure of the target optical chip includes the number of MZIs, the connection relationship, and the configuration parameters, so as to design the target optical chip based on the structure of the target optical chip.

In an optical chip design system provided in an embodiment of the present application, the first determining module may include:

the first determining unit is used for taking an independent variable in a target nonlinear operation function as an input of the MZI and determining the connection relation of two outputs of the MZI based on the structure of the MZI;

and the second determining unit is used for determining the output connection relation with the consistent characteristic and the target nonlinear operation function as a fitting connection relation.

In an optical chip design system provided in an embodiment of the present application, the second determining module may include:

a first acquisition unit for acquiring a design accuracy of a target optical chip;

and the third determining unit is used for determining a fitting linear operation function corresponding to the fitting nonlinear operation function based on the design precision and the least square method.

According to the optical chip design system provided by the embodiment of the application, the design precision comprises the data set range of independent variables;

the third determination unit may specifically be configured to: determining an initial linear operation function of the fitting nonlinear operation function at each power based on a least square method; determining the similarity between each initial linear operation function and the fitting nonlinear operation function in the range of the data set; and taking the initial linear operation function with the maximum similarity value as a fitting linear operation function.

In an optical chip design system provided in an embodiment of the present application, the third determining module may include:

and the third determining unit is used for determining the structure of the target optical chip based on the fitting linear operation function, the fitting connection relation and the singular value decomposition method.

In an optical chip design system provided in an embodiment of the present application, the third determining unit may be specifically configured to: determining a link arithmetic matrix of the target optical chip based on the fitting linear operation function and the fitting connection relation; decomposing the link arithmetic matrix based on a singular value decomposition method to obtain a matrix decomposition result; the structure of the target optical chip is determined based on the matrix decomposition result.

In the optical chip design system provided by the embodiment of the application, the target nonlinear operation function comprises a sigmoid function; the structure of the target optical chip includes: the device comprises a first MZI, a photodiode, a first amplifier, a phase-shift controller, a second MZI and a target topological structure; a first optical input end of the first MZI is connected with a second optical input end, a second optical output end of the first MZI is suspended, and a first optical output end of the first MZI is respectively connected with an input end of a photodiode, a first optical input end of the second MZI and a second optical input end of a target topological structure; the output end of the photodiode is connected with the first amplifier; the first amplifier is connected with the phase-shifting controller; the phase shifting controller is connected with a phase shifter of the second MZI; a second optical input end and a second optical output end of the second MZI are both suspended; a first optical output end of the second MZI is connected with a first optical input end of the target topology;

wherein the types of target topologies include: gridnet topology, Reck topology.

The application also provides an optical chip design device and a computer readable storage medium, which have the corresponding effects of the optical chip design method provided by the embodiment of the application. Referring to fig. 8, fig. 8 is a schematic structural diagram of an optical chip design apparatus according to an embodiment of the present disclosure.

The optical chip design device provided by the embodiment of the application comprises a memory 201 and a processor 202, wherein a computer program is stored in the memory 201, and the processor 202 executes the computer program to realize the following steps:

obtaining a target nonlinear operation function to be realized by a target optical chip, wherein the target optical chip is designed based on MZI;

determining a fitting connection relation of the MZI based on the independent variable in the target nonlinear operation function and the structure of the MZI;

converting the target nonlinear operation function into a fitting nonlinear operation function to be fitted of the MZI based on the fitting connection relation;

determining a fitting linear operation function corresponding to the fitting nonlinear operation function based on a least square method;

and determining the structure of the target optical chip based on the fitting linear operation function and the fitting connection relation, wherein the structure of the target optical chip comprises the number of MZIs, the connection relation and the configuration parameters, so as to design the target optical chip based on the structure of the target optical chip.

The optical chip design device provided by the embodiment of the application comprises a memory 201 and a processor 202, wherein a computer program is stored in the memory 201, and the processor 202 executes the computer program to realize the following steps: taking an independent variable in a target nonlinear operation function as an input of the MZI, and determining the connection relation of two outputs of the MZI based on the structure of the MZI; and determining the output connection relation with the consistent characteristics and the target nonlinear operation function as a fitting connection relation.

The optical chip design device provided by the embodiment of the application comprises a memory 201 and a processor 202, wherein a computer program is stored in the memory 201, and the processor 202 executes the computer program to realize the following steps: obtaining the design precision of a target optical chip; and determining a fitting linear operation function corresponding to the fitting nonlinear operation function based on the design precision and the least square method.

The optical chip design device provided by the embodiment of the application comprises a memory 201 and a processor 202, wherein a computer program is stored in the memory 201, and the processor 202 executes the computer program to realize the following steps: the design precision includes a dataset range of independent variables; determining an initial linear operation function of the fitting nonlinear operation function at each power based on a least square method; determining the similarity between each initial linear operation function and the fitting nonlinear operation function in the range of the data set; and taking the initial linear operation function with the maximum similarity value as a fitting linear operation function.

The optical chip design device provided by the embodiment of the application comprises a memory 201 and a processor 202, wherein a computer program is stored in the memory 201, and the processor 202 executes the computer program to realize the following steps: and determining the structure of the target optical chip based on a fitting linear operation function, a fitting connection relation and a singular value decomposition method.

The optical chip design device provided by the embodiment of the application comprises a memory 201 and a processor 202, wherein a computer program is stored in the memory 201, and the processor 202 executes the computer program to realize the following steps: determining a link arithmetic matrix of the target optical chip based on the fitting linear operation function and the fitting connection relation; decomposing the link arithmetic matrix based on a singular value decomposition method to obtain a matrix decomposition result; the structure of the target optical chip is determined based on the matrix decomposition result.

The optical chip design device provided by the embodiment of the application comprises a memory 201 and a processor 202, wherein a computer program is stored in the memory 201, and the processor 202 executes the computer program to realize the following steps: the target nonlinear operation function comprises a sigmoid function; the structure of the target optical chip includes: the device comprises a first MZI, a photodiode, a first amplifier, a phase-shift controller, a second MZI and a target topological structure; a first optical input end of the first MZI is connected with a second optical input end, a second optical output end of the first MZI is suspended, and a first optical output end of the first MZI is respectively connected with an input end of a photodiode, a first optical input end of the second MZI and a second optical input end of a target topological structure; the output end of the photodiode is connected with the first amplifier; the first amplifier is connected with the phase-shifting controller; the phase shifting controller is connected with a phase shifter of the second MZI; a second optical input end and a second optical output end of the second MZI are both suspended; a first optical output end of the second MZI is connected with a first optical input end of the target topology; wherein the types of target topologies include: gridnet topology, Reck topology.

Referring to fig. 9, another optical chip design apparatus provided in the embodiment of the present application may further include: an input port 203 connected to the processor 202, for transmitting externally input commands to the processor 202; a display unit 204 connected to the processor 202, for displaying the processing result of the processor 202 to the outside; and the communication module 205 is connected with the processor 202 and is used for realizing the communication between the optical chip design device and the outside. The display unit 204 may be a display panel, a laser scanning display, or the like; the communication method adopted by the communication module 205 includes, but is not limited to, mobile high definition link technology (HML), Universal Serial Bus (USB), High Definition Multimedia Interface (HDMI), and wireless connection: wireless fidelity technology (WiFi), bluetooth communication technology, bluetooth low energy communication technology, ieee802.11s based communication technology.

A computer-readable storage medium is provided in an embodiment of the present application, in which a computer program is stored, and when the computer program is executed by a processor, the computer program implements the following steps:

obtaining a target nonlinear operation function to be realized by a target optical chip, wherein the target optical chip is designed based on MZI;

determining a fitting connection relation of the MZI based on the independent variable in the target nonlinear operation function and the structure of the MZI;

converting the target nonlinear operation function into a fitting nonlinear operation function to be fitted of the MZI based on the fitting connection relation;

determining a fitting linear operation function corresponding to the fitting nonlinear operation function based on a least square method;

and determining the structure of the target optical chip based on the fitting linear operation function and the fitting connection relation, wherein the structure of the target optical chip comprises the number of MZIs, the connection relation and the configuration parameters, so as to design the target optical chip based on the structure of the target optical chip.

A computer-readable storage medium is provided in an embodiment of the present application, in which a computer program is stored, and when the computer program is executed by a processor, the computer program implements the following steps: taking an independent variable in a target nonlinear operation function as an input of the MZI, and determining the connection relation of two outputs of the MZI based on the structure of the MZI; and determining the output connection relation with the consistent characteristics and the target nonlinear operation function as a fitting connection relation.

A computer-readable storage medium is provided in an embodiment of the present application, in which a computer program is stored, and when the computer program is executed by a processor, the computer program implements the following steps: obtaining the design precision of a target optical chip; and determining a fitting linear operation function corresponding to the fitting nonlinear operation function based on the design precision and the least square method.

A computer-readable storage medium is provided in an embodiment of the present application, in which a computer program is stored, and when the computer program is executed by a processor, the computer program implements the following steps: the design precision includes a dataset range of independent variables; determining an initial linear operation function of the fitting nonlinear operation function at each power based on a least square method; determining the similarity between each initial linear operation function and the fitting nonlinear operation function in the range of the data set; and taking the initial linear operation function with the maximum similarity value as a fitting linear operation function.

A computer-readable storage medium is provided in an embodiment of the present application, in which a computer program is stored, and when the computer program is executed by a processor, the computer program implements the following steps: and determining the structure of the target optical chip based on a fitting linear operation function, a fitting connection relation and a singular value decomposition method.

A computer-readable storage medium is provided in an embodiment of the present application, in which a computer program is stored, and when the computer program is executed by a processor, the computer program implements the following steps: determining a link arithmetic matrix of the target optical chip based on the fitting linear operation function and the fitting connection relation; decomposing the link arithmetic matrix based on a singular value decomposition method to obtain a matrix decomposition result; the structure of the target optical chip is determined based on the matrix decomposition result.

A computer-readable storage medium is provided in an embodiment of the present application, in which a computer program is stored, and when the computer program is executed by a processor, the computer program implements the following steps: the target nonlinear operation function comprises a sigmoid function; the structure of the target optical chip includes: the device comprises a first MZI, a photodiode, a first amplifier, a phase-shift controller, a second MZI and a target topological structure; a first optical input end of the first MZI is connected with a second optical input end, a second optical output end of the first MZI is suspended, and a first optical output end of the first MZI is respectively connected with an input end of a photodiode, a first optical input end of the second MZI and a second optical input end of a target topological structure; the output end of the photodiode is connected with the first amplifier; the first amplifier is connected with the phase-shifting controller; the phase shifting controller is connected with a phase shifter of the second MZI; a second optical input end and a second optical output end of the second MZI are both suspended; a first optical output end of the second MZI is connected with a first optical input end of the target topology; wherein the types of target topologies include: gridnet topology, Reck topology.

The computer-readable storage media to which this application relates include Random Access Memory (RAM), memory, Read Only Memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage media known in the art.

For a description of a relevant part in an optical chip design system, an optical chip design device, and a computer-readable storage medium provided in the embodiments of the present application, reference is made to detailed descriptions of a corresponding part in an optical chip design method provided in the embodiments of the present application, and details are not repeated here. In addition, parts of the above technical solutions provided in the embodiments of the present application, which are consistent with the implementation principles of corresponding technical solutions in the prior art, are not described in detail so as to avoid redundant description.

It is further noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.

The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.