PCM-based integrated optical switch design simulation method and system

Abstract

The invention relates to the technical field of optical switches, in particular to a PCM-based integrated optical switch design simulation method and system. Comprising the following steps: providing an initial architecture of the integrated optical switch, wherein the initial architecture comprises a pixelated array of a * b specification; judging whether the number of the output ends is an even number; if yes, whether a symmetrical synchronous output end exists or not is judged, and if yes, the synchronous output end is divided into at least two simulation partitions; performing simulation for multiple times by adopting multiple groups of simulation conditions according to the simulation partitions, and calculating corresponding dominant parameters according to simulation results; selecting a local optimal solution of the current round, judging whether the local optimal solution is larger than the global optimal solution or not, if yes, updating the local optimal solution into the global optimal solution, and if not, keeping the current global optimal solution; and updating the at least two groups of simulation conditions according to the globally optimal solution, and performing simulation again. The invention provides a pixelated grouping simulation method suitable for an optical switch with multiple output ends. The pixelated grouping simulation method can reduce the simulation cost of pixelated design.

Description

Technical Field

[0001] The present invention relates to the technical field of optical switches, and in particular to a PCM-based integrated optical switch design simulation method and system. Background Art

[0002] In recent years, with the rapid development of cloud computing and artificial intelligence, the requirements for signal transmission bandwidth and chip computing power have also been increasing. To meet the growing business needs in the field of big data computing, there is an urgent need to achieve higher density and higher performance of complex functional integration on the same chip, that is, photonic integration.

[0003] However, the integration of higher density and higher performance features also places higher demands on device design. This is particularly true for optical switches, where improved performance will lead to exponential increases in device size, which in turn presents new challenges for the practical application of photonic devices. For example, an overly large optical switch will result in a bulky optical device, failing to meet current miniaturization requirements. Furthermore, the excessive size of an optical switch can overly compress the space available for other core components, increasing design complexity.

[0004] To address the technical challenges of miniaturizing optical switches, Kun Yin, Yang Gao, et al. (Inverse Design and Numerical Investigations of an Ultra-Compact Integrated Optical Switch Based on Phase Change Material, nanomaterials 13.10(2023)) disclose an ultra-compact integrated single-channel optical switch based on PCM (phase change material). Specifically, the single-channel optical switch device consists of a Ge2Sb2Te5 (GST) nanodisk and an inversely designed pixelated subwavelength structure. The pixelated subwavelength structure offers a customizable refractive index, unattainable with traditional materials or structures, thereby optimizing the device's insertion loss (IL) and extinction ratio (ER) performance. This pixelated structure enhances the interaction between the light field and the GST, reducing both device size and the inserted GST footprint. This enables, to a certain extent, the miniaturization of single-channel optical switches.

[0005] However, the applicant noted that in order to meet the requirements of miniaturized design, the pixelated sub-wavelength structure has a complex pixelated distribution morphology, so in the process of implementing the solution, it will face high design costs. Summary of the Invention

[0006] The object of the present invention is to provide a PCM-based integrated optical switch design simulation method to partially solve or alleviate the above-mentioned deficiencies in the prior art and reduce the simulation design cost of multi-channel optical switches.

[0007] In order to solve the above-mentioned technical problems, the present invention specifically adopts the following technical solutions: A first aspect of the present invention is a PCM-based integrated optical switch design simulation method, comprising the steps of: S101, providing an initial architecture of the integrated optical switch, the initial architecture comprising: an a×b pixel array, wherein two ends of the pixel array are respectively connected to at least one input end and a plurality of output ends, wherein the pixel array comprises: a×b pixel regions, wherein a plurality of the pixel regions are filled with waveguides to form a waveguide region, a plurality of the pixel regions are configured as spacer regions to form partitions between the waveguide regions, and a PCM region is further provided within the pixel array; S102, determining whether the number of the output terminals is an even number; If yes, proceed to step S103; S103, determining whether there are symmetrical synchronous output terminals, wherein when two output terminals are enabled at the same time, the output terminals are considered to be mutually synchronized output terminals; If yes, proceed to step S104; S104, dividing the plurality of output terminals into at least two simulation partitions according to the synchronous output terminals, wherein the simulation partitions include: at least two output terminals and pixel array areas corresponding to the output terminals; S105, performing at least two simulations using at least two sets of simulation conditions according to at least one of the simulation partitions, and calculating corresponding advantage parameters according to the simulation results; S106, selecting a local optimal solution of the current round from the at least two advantage parameters according to the advantage parameter type, and determining whether the local optimal solution is greater than the global optimal solution; if so, updating the local optimal solution to the global optimal solution; otherwise, maintaining the current global optimal solution; S107: Update the at least two sets of simulation conditions according to the global optimal solution, and return to S105.

[0008] In some embodiments, the simulation conditions include: a pixel matrix of scale a1×b1 formed according to the pixelated array of the simulation partition, wherein the parameters in the pixel matrix are respectively used to represent the area type on the xth row and yth column in the pixelated array, and the area type is a waveguide area or a spacer area, wherein x∈a1, y∈b1, a1<a, b1<b.

[0009] In some embodiments, S107 uses an optimization objective function to update the simulation conditions, and the optimization objective function includes: ; in, To optimize the goal, is the first weight, is the second weight, N is the number of output terminals of the simulation partition, is the sum of the output optical powers of the simulation partitions, is the sum of the input optical powers of the simulation partitions, is the optical power output by the output end enabled in the simulation partition, is the optical power of the remaining output ends of the simulation partition.

[0010] In some embodiments, the Maxwell method is used for simulation in S105, wherein the Maxwell method discretizes the Maxwell equations in space and time, and gradually promotes the electromagnetic field change in the time domain to solve the distribution of the electric field and the magnetic field in the signal switching area.

[0011] In some embodiments, S105 includes the steps of: For one of the simulation partitions, two adjacent pixel array sub-areas are identified from the pixel array area, and the two pixel array sub-areas have the same pixel specifications; Obtaining a degree of design symmetry of pixel arrays of the two pixel array sub-areas; When the degree of design symmetry is greater than a preset symmetry threshold, a pixel array sub-area and its corresponding output end are selected as simulation sub-areas; At least two simulations are performed using at least two sets of simulation conditions according to at least one of the simulation sub-partitions.

[0012] In some embodiments, during the first round of simulation, simulation is performed using a first pixel specification, wherein the pixel area size of the first pixel specification is smaller than the pixel area size of the a×b specification; during the mth round of simulation, simulation is performed using a second pixel specification, wherein the pixel area size of the second pixel specification is smaller than the pixel area size of the first pixel specification.

[0013] The present invention also provides a PCM-based integrated optical switch design simulation system, comprising: an architecture acquisition module, configured to provide an initial architecture of the integrated optical switch, the initial architecture comprising: an a×b pixel array, wherein two ends of the pixel array are respectively connected to at least one input end and a plurality of output ends, wherein the pixel array comprises: a×b pixel regions, wherein a plurality of the pixel regions are filled with waveguides to form a waveguide region, a plurality of the pixel regions are configured as spacer regions to form partitions between the waveguide regions, and a PCM region is further provided within the pixel array; A first judging module, configured to judge whether the number of the output terminals is an even number; If yes, then enter the second judgment module; a second determining module, configured to determine whether there are symmetrical synchronous output terminals, wherein when two output terminals are enabled at the same time, the output terminals are considered to be mutually synchronized output terminals; If yes, then enter the grouping module; a grouping module, configured to divide the plurality of output terminals into at least two simulation partitions according to the synchronous output terminals, wherein the simulation partitions include: at least two output terminals and pixel array areas corresponding to the output terminals; a simulation module, configured to perform at least two simulations using at least two sets of simulation conditions according to at least one of the simulation partitions, and calculate corresponding advantage parameters according to the simulation results; a first optimization module, configured to select a local optimal solution of a current round from the at least two advantage parameters according to the advantage parameter type, and determine whether the local optimal solution is greater than the global optimal solution; if so, update the local optimal solution to the global optimal solution; otherwise, maintain the current global optimal solution; The second optimization module is used to update the at least two groups of simulation conditions according to the global optimal solution and return the results to the simulation module.

[0014] In some embodiments, the simulation conditions include: a pixel matrix of scale a1×b1 formed according to the pixelated array of the simulation partition, wherein the parameters in the pixel matrix are respectively used to represent the area type on the xth row and yth column in the pixelated array, and the area type is a waveguide area or a spacer area, wherein x∈a1, y∈b1, a1<a, b1<b.

[0015] In some embodiments, the second optimization module uses an optimization function to update the simulation condition, and the optimization function includes: ; in, To optimize the goal, is the first weight, is the second weight, N is the number of output terminals of the simulation partition, is the sum of the output optical powers of the simulation partitions, is the sum of the input optical powers of the simulation partitions, is the optical power output by the output end enabled in the simulation partition, is the optical power of the remaining output ends of the simulation partition.

[0016] In some embodiments, the Maxwell method is used for simulation, wherein the Maxwell method discretizes the Maxwell equations in space and time, and gradually promotes the electromagnetic field change in the time domain to solve the distribution of the electric field and the magnetic field in the signal switching area.

[0017] Beneficial technical effects: For the pixelated design of a multi-channel optical switch with a symmetrical or quasi-symmetrical design, the present invention provides a local simulation scheme based on the symmetrical or quasi-symmetrical design to achieve grouping. This local simulation scheme can reduce simulation time, improve simulation efficiency, and reduce simulation costs.

[0018] Furthermore, for multi-level simulation partitioning, the present invention provides a gradient approach from local to global, further enhancing the method's versatility, specifically its applicability to quasi-symmetrical optical switches. This gradient approach can further reduce simulation costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following is a brief introduction to the drawings required for the embodiments or the description of the prior art. In all drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the various elements or parts are not necessarily drawn according to the actual scale. Obviously, the drawings described below are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can also be obtained based on these drawings without inventive work.

[0020] Figure 1 is a schematic structural diagram of a pixelated signal switching area in an exemplary embodiment of the present invention; Figure 2 is a schematic cross-sectional structural diagram of a signal switching area in an exemplary embodiment of the present invention; Figure 3 is a schematic structural diagram of a pixelated signal switching area of ​​an optical switch in another exemplary embodiment of the present invention; Figure 4 for Figure 3 A schematic diagram of a first partitioning scheme of an optical switch is shown; Figure 5 for Figure 3A schematic diagram of a second partitioning scheme of the optical switch shown; Figure 6 FIG. 4 is a flow chart of a simulation method in an exemplary embodiment of the present invention.

[0021] Summary of reference numerals: Input end 01, signal switching area 02, signal switching layer (or pixel array area) 02A, substrate 02B, spacer area 021, waveguide area 022, PCM area 023, output end 03, first output end 031, second output end 032, third output end 033, fourth output end 034, first simulation partition D1, second simulation partition D2, first simulation sub-partition D11, second simulation sub-partition D12. DETAILED DESCRIPTION

[0022] To make the purpose, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0023] Herein, suffixes such as "module," "component," or "unit" used to represent elements are only used to facilitate description of the present invention and have no specific meaning. Therefore, "module," "component," or "unit" may be used interchangeably.

[0024] As used herein, terms such as "upper," "lower," "inner," "outer," "front," "back," "one end," and "the other end" indicate positions or locations based on those shown in the accompanying drawings. These terms are intended solely to facilitate and simplify the description of the present invention and are not intended to indicate or imply that the devices or components referred to must have, be constructed, or operate in a specific orientation. Therefore, they should not be construed as limitations on the present invention. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0025] As used herein, unless otherwise expressly specified or limited, the terms "installed," "provided with," and "connected" should be understood broadly. For example, "connected" may refer to a fixed connection, a detachable connection, or an integral connection; it may refer to a mechanical connection, a direct connection, an indirect connection via an intermediate medium, or internal communication between two components. Those skilled in the art will understand the specific meanings of the above terms in the present invention on a case-by-case basis.

[0026] As used herein, "and / or" includes any and all combinations of one or more of the associated listed items.

[0027] Herein, "plurality" means two or more than two, ie, it includes two, three, four, five, etc.

[0028] As used in this specification, the term "about" typically means + / - 5% of the stated value, more typically + / - 4% of the stated value, more typically + / - 3% of the stated value, more typically + / - 2% of the stated value, even more typically + / - 1% of the stated value, and even more typically + / - 0.5% of the stated value.

[0029] In this specification, certain embodiments may be disclosed in a format that is within a range. It should be understood that this description of "within a range" is merely for convenience and brevity and should not be interpreted as a rigid limitation on the disclosed range. Therefore, the description of a range should be considered to have specifically disclosed all possible subranges and individual numerical values ​​within this range. For example, the description of a range of 1-6 should be considered to have specifically disclosed subranges such as from 1 to 3, from 1 to 4, from 1 to 5, from 2 to 4, from 2 to 6, from 3 to 6, etc., as well as individual numbers within this range, such as 1, 2, 3, 4, 5, and 6. Regardless of the breadth of the range, the above rules apply.

[0030] "Optical switch": In this article, a typical optical switch is based on a 1×N or N×N multimode interference coupler (MMI) structure. The optical switch includes at least one input port 01, multiple output ports 03 (also known as output channels, or simply channels), and a signal switching region 02 between the input and output ports. The signal switching region 02 is the core functional module of the optical switch, responsible for dynamically switching and controlling the optical signal path. For example, the signal switching region can adjust internal device parameters to control the output of the input optical signal from at least one of the multiple output ports, thereby controlling the transmission of the optical signal. For example, a 1×N optical switch has one input port and N output ports, while an N×N optical switch has N input ports and N output ports.

[0031] Unlike conventional optical switches, the signal switching region of the optical switch (e.g., integrated optical switch) in the present invention is configured as a pixelated multimode waveguide region (abbreviated as: multimode waveguide region, or pixelated subwavelength structure) having a phase change material. For example, see Figure 1As shown, the signal switching region 02 can use a silicon-on-insulator (SOI) substrate 01B. The substrate surface is divided into a×b rectangles (specifically, squares). Rectangles at different locations are filled with waveguide material (e.g., Si / SiN) or spacer material (e.g., air / SiO2, etc.) according to design requirements. The spacer material primarily has a certain difference in refractive index from the waveguide, making it difficult for light to transmit continuously between the waveguide and spacer materials. Each rectangular region can be referred to as a "pixel." Specifically, the rectangular region filled with Si or SiN is referred to as the waveguide region 022 (or the first pixel region), and the rectangular region filled with air or SiO2 is referred to as the spacer region 021 (or the second pixel region). The intersecting waveguide and spacer regions are also referred to as a pixelated array.

[0032] Furthermore, a region filled with phase change material is provided in the pixelated array, which is also referred to as a PCM region 023 .

[0033] For example, as an exemplary embodiment, the side width of each pixel can be approximately 100 nm and the depth can be approximately 220 nm. At least one PCM region is provided in the center of the signal switching region 02. For example, when N output terminals are provided, N PCM regions are correspondingly provided. For example, the PCM region can be completely filled with GST. Typically, at a wavelength of 1550 nm, when GST is in the amorphous state (a-GST), the corresponding refractive index is 3.98 + 0.0244i. The real part of the refractive index is close to that of silicon, and the imaginary part is low, indicating low absorption. In this case, due to the small refractive index difference between the two materials, light can be transmitted through the Si-GST-Si interface, and the switch state is on. When the GST changes to a crystalline state (c-GST) and the refractive index increases to 6.49 + 1.054i, the switch state is reversed, i.e., the switch state is switched to off.

[0034] For example, Figure 2 The cross-sectional structure diagram of the signal switching area is shown. The signal switching area 02 may include a Si substrate 02B, on which a pixelated signal switching layer 02A is provided. The signal switching layer 02A is provided with spacer areas 021, waveguide areas 022, and PCM areas 023 arranged at intervals.

[0035] The arrangement of pixels, such as the first and second pixel regions, affects the distribution of the effective refractive index, and thus the evolution of the light field. For example, to meet the application requirements of an optical switch, the material distribution in the signal switching region (i.e., the arrangement of the first and second pixel regions) must be optimized to maximize the transmittance of the input TE0 mode when the GST is in the amorphous state, while minimizing the transmittance when the GST is in the crystalline state. This minimizes the IL of the optical switch and maximizes the ER.

[0036] Currently, due to the relatively complex layout of the first and second pixel regions, simulations are often performed using a direct binary search (DBS) algorithm. For example, the DBS algorithm determines the material properties of each pixel (such as the type of material used), and then simulates the light transmission within the signal switching region under a specific layout. Specifically, the simulation can determine the final optical power of the optical signal output at each output port, and then calculate performance parameters such as the extinction ratio (ER) based on the optical power. The layout that corresponds to the appropriate performance parameters is selected; for example, the layout with the lowest extinction ratio can be selected as the recommended design.

[0037] The light in the optical signal switching area may be transmitted in the same mode or in different modes.

[0038] In this article, a "pixel region" can also be referred to as a pixel point or simply a pixel. The size of a pixel region (or pixel point) can be used to describe the degree of simulation refinement. Correspondingly, a pixel size can refer to the number of pixels set within a specific region. For example, a larger pixel size means more pixels within a specific region, and a smaller pixel size means a more refined simulation.

[0039] Herein, in different embodiments, for ease of understanding, “simulation conditions”, “simulation schemes” or “design schemes” may be used interchangeably, which mainly include pixel array arrangement schemes within the signal switching region.

[0040] In this article, "symmetrical design" means that the application states (e.g., enabled and disabled) of the output terminals of two partitions are symmetrical, and the pixel array layout is also completely symmetrical. "Quasi-symmetrical design" means that the application states of the output terminals of two partitions are symmetrical, and the pixel array layout may not be completely symmetrical, but the arrangement of most pixels is symmetrical. For example, when the number of symmetrical pixels exceeds a symmetry threshold, the pixel array is considered to have a "quasi-symmetrical design," e.g., at least approximately 85% of the pixels are symmetrical.

[0041] In this article, in conjunction with the partitioning method provided by the present invention, users can subdivide a "simulation partition," such as the first simulation partition D1 or the second simulation partition D2, once or multiple times to obtain two or more sub-partitions (that is, the sub-partitions can also be further divided based on design requirements). Partitions obtained at different division stages are also considered to be simulation partitions at different levels. Simulation processes can be performed separately for simulation partitions at different levels.

[0042] Especially for high-performance optical switches, which are typically designed with multiple output ports and signal switching zones that can have multiple different modes, this significantly increases the difficulty of simulation, especially the amount of simulation computation and the simulation time. Therefore, for multi-channel optical switches, this pixelated design will face huge design costs.

[0043] In this regard, the present invention proposes a group simulation method corresponding to the pixelated design simulation process of multiple channels, which significantly reduces the necessary computing resources required for simulation, thereby reducing the design cost of the multi-channel optical switch.

[0044] Example 1: See also Figure 6 As shown, the present invention also provides a PCM-based integrated optical switch design simulation method, comprising the steps of: S101, providing an initial architecture of the integrated optical switch, the initial architecture comprising: an a×b pixel array (e.g., a signal switching region), wherein two ends of the pixel array are respectively connected to at least one input end and a plurality of output ends, wherein the pixel array comprises: a×b pixel regions, a plurality of the pixel regions are filled with waveguides to form a waveguide region, and a plurality of the pixel regions are configured as spacer regions to form partitions between the waveguide regions; further, a plurality of PCM regions corresponding to the output ends are further provided within the pixel array; S102, determining whether the number of the output terminals is an even number; If yes, proceed to step S103; S103, determining whether there are symmetrical synchronous output terminals, wherein when two output terminals are enabled at the same time, the output terminals are considered to be mutually synchronized output terminals; wherein being enabled means that the output terminal outputs optical signals as a main output terminal, that is, is in an on state.

[0045] If yes, proceed to step S104; S104, dividing the plurality of output terminals into at least two simulation partitions according to the synchronous output terminals, wherein the simulation partitions include: at least two output terminals and pixel array areas corresponding to the output terminals; For example, in some embodiments, Figure 1 For example, an optical switch includes four output terminals: a first output terminal 031, a second output terminal 032, a third output terminal 033, and a fourth output terminal 034. For example, when the first output terminal and the fourth output terminal can be enabled simultaneously, i.e., when they simultaneously output optical switch signals, they can be considered as synchronous output terminals, and the two synchronous output terminals are symmetrical to each other. Similarly, when the second output terminal 032 and the third output terminal 033 are also enabled simultaneously, they can also be considered as synchronous output terminals that are symmetrical to each other.

[0046] Correspondingly, the grouping method is: the first output terminal 031 and the second output terminal 032 are grouped into one output group, and the third output terminal 033 and the fourth output terminal 034 are grouped into one output group. That is, the first output terminal 031 and the second output terminal 032 are grouped into the first simulation partition D1, and the third output terminal 033 and the fourth output terminal 034 are grouped into the second simulation partition D2.

[0047] S105, performing at least two simulations using at least two sets of simulation conditions according to at least one of the simulation partitions, and calculating corresponding advantage parameters according to the simulation results; In this embodiment, the Figure of Merit (FOM) refers to an indicator related to the performance indicators of the optical switch, which can be any one or more of the following: device insertion loss (IL), excess loss, extinction ratio (ER), return loss, switching speed, etc.

[0048] For example, in some embodiments, the advantage parameter may be the value of one performance indicator, or, in other embodiments, may be the combined value of two or more performance indicators.

[0049] S106, selecting a local optimal solution of the current round from the at least two advantage parameters according to the type of the advantage parameter, and determining whether the local optimal solution is greater than the global optimal solution; if so, updating the local optimal solution to the global optimal solution; otherwise, maintaining the current global optimal solution; In this embodiment, the process of performing at least two simulations on a group of output groups using at least two groups of simulation conditions may be referred to as a round.

[0050] For example, in some embodiments, when the dominant parameter type is a loss-related indicator, such as insertion loss (IL), excess loss, or return loss, the minimum dominant parameter is selected as the local optimal solution. This local optimal solution can be the minimum dominant parameter for the current round; correspondingly, the global optimal solution is the minimum dominant parameter across all rounds. After the first round of simulation, the current local optimal solution can be defined as the current global optimal solution.

[0051] For example, in some embodiments, when the dominant parameter is an efficiency indicator, such as switching speed or extinction ratio, the largest dominant parameter is selected as the local optimal solution. Correspondingly, the global optimal solution is the largest dominant parameter across all rounds. After the first round of simulation, the current local optimal solution can be defined as the current global optimal solution.

[0052] S107: Update the at least two sets of simulation conditions according to the global optimal solution, and return to S105.

[0053] Furthermore, the method includes the steps of: performing at least two iterative simulations, and outputting the simulation conditions corresponding to the final global optimal solution as a recommended design solution.

[0054] For example, in some embodiments, when the iterative simulation process converges, the loop is stopped and a recommended design solution is output.

[0055] In some embodiments, the simulation conditions include: a pixel matrix of scale a1×b1 formed according to the pixelated array of the simulation partition, wherein the parameters in the pixel matrix are respectively used to represent the area type on the xth row and yth column in the pixelated array, and the area type is a waveguide area or a spacer area, wherein x∈a1, y∈b1, a1<a, b1<b.

[0056] In some embodiments, the simulation in S105 is performed using the Maxwell method, which discretizes the Maxwell equations in space and time, gradually advancing the electromagnetic field changes in the time domain to solve the distribution of the electric and magnetic fields in the signal switching region. Preferably, 3DFDTD simulation software can be directly used for optical simulation. For example, in some embodiments, the simulation process primarily includes the following steps: using the FDTD method to replace the set continuous differential equations with finite difference formulas; dividing the simulation area into a three-dimensional grid (commonly known as a Yee grid); alternately calculating the electric and magnetic fields in the three-dimensional grid area, advancing the electromagnetic wave at each time step; and adding auxiliary simulation conditions such as excitation sources, boundary conditions, and monitors to facilitate the calculation of various performance parameters.

[0057] In this embodiment, when the judgment results of S102 and S103 are both yes, the optical switch is designed to be symmetrical along its axis L. Preferably, in this embodiment, only local simulation can be performed, and the design scheme of the unsimulated area can be obtained by using the mirror symmetry relationship. Figure 1 For example, two independent simulation areas are formed, namely, a first simulation area D1 and a second simulation area D2.

[0058] Furthermore, due to the presence of mutually symmetrical synchronous output terminals, either the first or second simulation partition can be selected as the simulation target. For example, the first simulation partition D1 (which includes the third and fourth output terminals, as well as the pixelated areas corresponding to the two output terminals) can be considered a relatively independent entity and simulated. Ultimately, a recommended design solution for the first simulation partition (such as the filler material type for each pixel area) can be obtained. The design solution for the second simulation partition is axially symmetrical with the first simulation partition and can therefore be derived directly based on the recommended design solution for the first simulation partition.

[0059] During the simulation optimization process, the main optimization index, that is, the advantage parameter, will be determined first, which can be a single one.

[0060] For example, in some embodiments, when a single round of calculations includes multiple simulation results, the maximum or minimum value among the simulation results is selected as the optimization result. For example, in a single round of simulations, the GST crystal state may have multiple states, but the optimization result is always the insertion loss of the optical switch. During the calculation, the insertion losses of these multiple states are compared, and the one with the lowest insertion loss is selected for the next round of iterations.

[0061] Alternatively, multiple metrics can be optimized. The simulation process not only addresses the individual metrics but also involves trade-offs among them. For example, during the simulation, the optical switch can be designed to sacrifice some loss in exchange for a higher extinction ratio while meeting a certain additional loss.

[0062] The simulation process is described below through an exemplary embodiment. The FOM in this embodiment is insertion loss: In the first round of design simulations, the waveguide region's area ratio in the pixelated array can be divided into 10 different layout conditions, such as 10%, 20%, ..., 80%, and 90%. These 10 simulation conditions can be generated in matrix form, for example, a 01 matrix of size a1 × b1, where 0 indicates the rectangular region is a spacer region and 1 indicates the rectangular region is a waveguide region. The main simulation conditions: matrix data and the state of each PCM region (e.g., crystalline, amorphous) are input into 3DFDTD simulation software for optical simulation. The simulation results (e.g., the optical power of the optical signal output by each output port) are then used to calculate 10 FOMs. It should be noted that the simulation target in this embodiment is one or more simulation partitions, and the sizes of a1 and b1 are determined by the specifications of the specific simulation partition.

[0063] It is understood that the current rectangular data is usually the initial condition set by the user. In order to perform a complete optical simulation, other types of auxiliary simulation conditions are also involved, such as the refractive index of different materials, the mode of the optical signal in the signal switching region (such as single-mode, multi-mode), etc. These auxiliary simulation conditions can usually be pre-set with the help of simulation software (such as initial default values) or manually set by the user.

[0064] Furthermore, a local optimal solution (which is now equivalent to the current global optimal solution) is selected from the 10 FOMs obtained in the first round. Simulation conditions corresponding to the local optimal solution (which may include both primary and secondary simulation conditions) are then updated, with the goal of reducing the FOM. In some embodiments, simulation condition adjustments may be prioritized, such as prioritizing primary simulation conditions (particularly matrix data) and then subsequently adjusting secondary simulation conditions based on this. Ultimately, updated simulation conditions are obtained.

[0065] To accelerate the acquisition of simulation results, 10 sets of simulation conditions can be updated according to different adjustment levels. These 10 updated sets of simulation conditions are then input into the 3DFDTD simulation software for 10 optical simulations. At this point, 10 FOMs are obtained again; the local and global optimal solutions for the current round are obtained. If the local optimal solution is less than the global optimal solution, the global optimal solution is updated; and the simulation conditions are updated again based on the latest global optimal solution. After multiple cycles, the simulation conditions corresponding to the final global optimal solution are used as the recommended design solution.

[0066] In some embodiments, a round refers to a process of multiple simultaneous simulations performed on the same partition.

[0067] In some embodiments, the pixelation simulation method used in existing single-channel optical switches can be directly adopted.

[0068] Alternatively, in some embodiments, the user may also freely select other pixelation simulation methods or software in combination with design requirements, and the present invention does not limit this.

[0069] Furthermore, in some embodiments, the GST crystal state may have multiple states in a single simulation, but the dominant parameter ultimately calculated is the optical switch insertion loss. During the calculation, the insertion losses from these multiple states are compared, and the one with the lowest insertion loss is selected for the next iteration.

[0070] In some embodiments, S107 uses an optimization objective function to update the simulation conditions, and the optimization objective function includes: ; in, is the first weight, is the second weight, N is the number of output terminals in the simulation partition, is the sum of the output optical powers in the simulation partition, is the sum of the input optical powers in the simulation partition, The optical power output by the output port enabled in the simulation partition (or, is the size of the output of the main output terminal of multiple output terminals), is the optical power of the remaining output ends in the simulation partition (that is, the output ends that are not activated at the current moment).

[0071] In this embodiment, insertion loss and extinction ratio are preferably used as comprehensive optimization targets, and a better performance design is achieved through weighted trade-offs.

[0072] Alternatively, in other embodiments, when a single performance indicator is used as the optimization objective function, exemplary embodiments are as follows: For example, when the optimization target is Insertion Loss (IL), the corresponding optimization objective function is: ; is the sum of the output optical powers, is the sum of the input optical powers.

[0073] For example, when the optimization target is crosstalk (XT), the corresponding optimization objective function is: ; P_leak / P_signal represent the power leaked to adjacent output ports and the original signal power, respectively, and are used to quantify the impact of crosstalk. The greater the crosstalk, the worse the signal integrity.

[0074] Furthermore, when a simulation partition has multi-channel outputs, performance indicators focusing on overall efficiency and uniformity can be further added, such as power uniformity: Uniformity = max(Pout) / min(Pout), where Pout refers to the output power of the enabled output port in the simulation partition.

[0075] Alternatively, in some embodiments, the performance indicator may also be the total transmission efficiency: η = ∑(Pout) / Pin, where, is the sum of all output optical powers in the simulation partition, is the sum of all input optical powers in the simulation partition.

[0076] Furthermore, in some embodiments, S105 includes the steps of: For one of the simulation partitions, two adjacent pixel array sub-areas are identified from the pixel array area, and the two pixel array sub-areas have the same pixel specifications; wherein the pixel specifications refer to the length and width dimensions of the pixel array; Obtaining a degree of design symmetry of pixel arrays of the two pixel array sub-areas; When the degree of design symmetry is greater than a preset symmetry threshold, a pixel array sub-area and its corresponding output end are selected as simulation sub-areas; At least two simulations are performed using at least two sets of simulation conditions according to at least one of the simulation sub-partitions.

[0077] Preferably, in some embodiments, when a symmetrical design or a quasi-symmetrical design still exists within a simulation partition, simulation may be performed in the form of multi-layer grouping.

[0078] Can be Figure 4 For example, an 8×8 optical switch is shown. The exemplary simulation process of the optical switch may be as follows: First, the design parameters of the optical switch are obtained. These parameters may include the design architecture of the optical switch at the signal switching layer (pixelated array) O2A (which can be used to determine the symmetry of the optical switch design), as well as the output states of each channel. Based on the design parameters, it is determined whether the optical switch is symmetrical about axis L. Specifically, the output states (on or off) of each channel of the optical switch are symmetrical along axis L, and the pixel area distribution patterns of the pixel arrays on both sides are also symmetrical along axis L. In this case, the optical switch can be divided into a first simulation zone D1 and a second simulation zone D2.

[0079] In some embodiments, the first simulation partition or the second simulation partition can be directly selected as a simulation object to directly perform simulation design.

[0080] Alternatively, in other embodiments, a deeper level of simulation area division may be performed. Figure 5 For example, the pixel array of the first simulation partition is further identified as two pixel array sub-partitions, and the degree of symmetry that the two need to meet is determined based on the design parameters. When the degree of symmetry between the two is greater than the symmetry threshold (for example, greater than approximately 85%, 90%, 95%), it can be regarded as a quasi-symmetrical design, thereby further classifying the two pixel array sub-partitions into the first simulation sub-partition D11 and the second simulation sub-partition D12.

[0081] Preferably, the output states of the output terminals of the two sub-partitions at the same level, such as enabled or disabled, are symmetrical to each other.

[0082] Therefore, in this embodiment, any one sub-partition from the first simulation sub-partition D11 or the second simulation sub-partition D12 can be selected as a simulation object, that is, a local simulation of a smaller unit can be performed.

[0083] Furthermore, in this embodiment, for the case where the two simulation sub-regions are not completely symmetrical, a segmented simulation process with gradient pixel density variation can be used as follows: During the first round of simulation, simulation is performed using a first pixel specification, wherein the pixel area size of the first pixel specification is smaller than the pixel area size of the a×b specification; during the mth round of simulation, simulation is performed using a second pixel specification, wherein the pixel area size of the second pixel specification is smaller than the pixel area size of the first pixel specification.

[0084] Specifically, in an exemplary embodiment, the implementation process of the above steps is as follows: (1) During the simulation of the simulation sub-area with the first iteration number m1, the simulation is performed using the first pixel specification; The size of the pixel area (or pixel point) of the first pixel size is smaller than the size of the pixel area of ​​the a×b size. In other words, this batch can perform a coarse simulation of large pixel points to quickly achieve a rough layout pattern in the local area.

[0085] (2) Generate a second type of design scheme based on the current first type of design scheme, where the first type of design scheme refers to the design scheme formed by the simulation sub-area, and the second type of design scheme refers to the simulation area formed by two simulation sub-partitions, or the second type of design scheme refers to the design scheme of the corresponding simulation sub-area after being updated based on the first type of design scheme.

[0086] (3) Using the second type of design solution, the simulation process of the second iteration number m2 is performed on the corresponding simulation area, and the simulation is performed using the second pixel specification. Preferably, the second pixel specification is larger than the first pixel specification.

[0087] Preferably, the pixel size of the second type of design solution is smaller than the pixel area size of the first type of design solution. In other words, the pixel scale of the second type of design solution is larger than the pixel scale of the first type of design solution.

[0088] In some embodiments, when the iterative simulation process of the object to be simulated, such as the first simulation sub-partition D11, is completed, a class of design solutions including the first sub-design solution is output. Furthermore, the class of design solutions may further include: forming a second sub-design solution symmetrically based on the first sub-design solution (corresponding to the design solution of the second simulation sub-partition).

[0089] Furthermore, in some embodiments, the second type of design solution can be a new simulation solution obtained by increasing the pixel scale of the first sub-design solution. In other words, for a simulated object, such as a sub-region, multiple simulation stages with gradually increasing pixel scales can be used.

[0090] Alternatively, in some embodiments, when the first simulation sub-partition D11 is completed, a design scheme for the first simulation partition D1 (i.e., a second type of design scheme) can be directly formed based on the first and second sub-design schemes (i.e., a first type of design scheme); and the first simulation partition D1 can be simulated using the second type of design scheme. Preferably, the multi-layer hierarchical simulation method can be used only for N×N optical switches.

[0091] Furthermore, in some embodiments, a loop iteration limit may be set, that is, the number of iterations is increased by 1 for each round of simulation. Correspondingly, the method further includes the steps of: Get the current iteration number; Determine whether the iteration number m is less than the preset iteration index M; If yes, continue the simulation; if no, end the simulation.

[0092] It is understandable that in actual optical switch design, due to different design architecture requirements and differences in optical switch application scenarios, it is difficult to ensure that all optical switches can achieve ideal symmetrical design. To this end, the present invention proposes a segmented simulation with gradient pixel density changes for quasi-symmetrical design. Therefore, small-pixel scale simulation of smaller units is conducive to quickly determining the approximate distribution scheme of the local area, and then gradually increasing the simulation precision through pixel scale adjustment on the preliminary distribution scheme, thereby greatly reducing simulation time and improving simulation efficiency.

[0093] In particular, for the simulation design scenario of an optical switch with a quasi-symmetrical design, the segmented simulation of pixel gradient changes proposed in the present invention can provide simulation design schemes with different pixel specifications for different simulation objects (such as simulation partitions at different levels). For example, when simulating the first simulation sub-partition D11, a large pixel point can be used for rapid simulation to roughly determine the design schemes of the first simulation sub-partition D11 and the second simulation sub-partition D12. At this time, the simulation of large pixel points can ignore the differences between the first and second simulation sub-partitions to a certain extent (that is, it is conducive to improving the applicability of the design scheme of the first simulation sub-partition on the second simulation sub-partition). Furthermore, combined with the current preliminary first-class simulation design, the overall design scheme of the first simulation sub-partition D11 and the second simulation sub-partition D12 is generated, that is, the second-class design scheme, and the pixel scale of the second-class design scheme is set to be larger than the first-class design scheme, so as to further explore the detailed design of the two.

[0094] It is understandable that in actual application, the grouping method of the simulation areas can be freely selected by the user according to the number of output ports, optical switch type or other simulation requirements.

[0095] In some embodiments, S106 includes the steps of: selecting a local optimal solution from the simulation results of at least two of the simulation partitions according to the type of the advantage parameter; correspondingly, in S107, updating the simulation conditions using the simulation conditions corresponding to the local optimal solution as the current conditions.

[0096] It should be noted that the local-to-global gradient simulation method proposed in the present invention can, for example, first select a local sub-region from the overall pixel array and then gradually expand the scale from the local to the overall region to implement the simulation solution design. Furthermore, it is worth noting that the scale expansion in the present invention has two levels of design: first, the gradual expansion of the pixel scale, and second, the gradual expansion of the simulation area.

[0097] It is important to understand that Figure 1-Figure 5 The pixelated array shown is only an example. As actual optical switch design requirements evolve, the density and distribution of pixels will vary significantly. This complex variation further increases the design cost of multi-channel optical switches, making pixelated technology difficult to apply in actual optical switch design. Consequently, the simulation computational complexity is also very high. The partitioning design proposed in this invention can effectively improve simulation efficiency.

[0098] In some embodiments, the number of output ports can be an even number. Alternatively, in some embodiments, the number of output ports can be an odd number, but this must meet symmetrical or quasi-symmetrical design requirements. Taking a 1×9 optical switch as an example, five output groups can be selected from the edge to the center of the switch as a simulation partition.

[0099] Therefore, in other embodiments, the method provided by the present invention may not include step S102 to determine whether the number of the output terminals is an even number, for example, the method may directly proceed to step S103.

[0100] Alternatively, in other embodiments, the method provided by the present invention preferably includes two steps S102 and S103, but the execution order of the two steps can be adjusted.

[0101] For example, in some embodiments, a PCM-based integrated optical switch design simulation method includes the steps of: S101, providing an initial architecture of the integrated optical switch, the initial architecture comprising: an a×b pixel array, wherein two ends of the pixel array are respectively connected to at least one input end and a plurality of output ends, wherein the pixel array comprises: a×b pixel regions, wherein a plurality of the pixel regions are filled with waveguides to form a waveguide region, a plurality of the pixel regions are configured as spacer regions to form partitions between the waveguide regions, and a PCM region is further provided within the pixel array; S102, determining whether the number of the output terminals is an even number; S103, determining whether there are symmetrical synchronous output terminals, wherein when two output terminals are enabled at the same time, the output terminals are considered to be mutually synchronized output terminals; If both are yes, then the process proceeds to S104, dividing the plurality of output terminals into at least two simulation partitions according to the synchronous output terminals, wherein the simulation partitions include: at least two output terminals and pixel array areas corresponding to the output terminals; S105, performing at least two simulations using at least two sets of simulation conditions according to at least one of the simulation partitions, and calculating corresponding advantage parameters according to the simulation results; S106, selecting a local optimal solution of the current round from the at least two advantage parameters according to the advantage parameter type, and determining whether the local optimal solution is greater than the global optimal solution; if so, updating the local optimal solution to the global optimal solution; otherwise, maintaining the current global optimal solution; S107: Update the at least two sets of simulation conditions according to the global optimal solution, and return to S105.

[0102] The packet emulation solution provided by the present invention can particularly meet the optical switch design requirements in the following application scenarios: For example, in some embodiments, multiplexing is employed in optical communications within an optical switch, requiring the transmission of multiple signals to multiple frequency-modulated channels. This requires simultaneous output of light from multiple channels. Preferably, a symmetrical output scheme involves symmetrically distributing two or more channels along the axis L of the optical switch.

[0103] For another example, in some embodiments, a redundant design may be implemented in the communication link of the optical switch, requiring two outputs to be provided, and the optical signals in the primary and backup channels need to be synchronized to implement an error-proof system.

[0104] For example, simulated matrix operations in optical computing or photonic neural networks also need to be output to multiple channels at the same time to perform column weight output of matrix-vector multiplication.

[0105] Another example is laser multiplexing during on-chip testing and operation. For example, a single laser can split two light paths to provide light source signals to two different chips simultaneously, enabling laser signal sharing and reducing the number of lasers.

[0106] Example 2 The present invention also provides a simulation system corresponding to the above method, such as a PCM-based integrated optical switch design simulation system, comprising: an architecture acquisition module, configured to provide an initial architecture of the integrated optical switch, the initial architecture comprising: an a×b pixel array, wherein two ends of the pixel array are respectively connected to at least one input end and a plurality of output ends, wherein the pixel array comprises: a×b pixel regions, wherein a plurality of the pixel regions are filled with waveguides to form a waveguide region, a plurality of the pixel regions are configured as spacer regions to form partitions between the waveguide regions, and a PCM region is further provided within the pixel array; A first judging module, configured to judge whether the number of the output terminals is an even number; If yes, then enter the second judgment module; a second determining module, configured to determine whether there are symmetrical synchronous output terminals, wherein when two output terminals are enabled at the same time, the output terminals are considered to be mutually synchronized output terminals; If yes, then enter the grouping module; a grouping module, configured to divide the plurality of output terminals into at least two simulation partitions according to the synchronous output terminals, wherein the simulation partitions include: at least two output terminals and pixel array areas corresponding to the output terminals; a simulation module, configured to perform at least two simulations using at least two sets of simulation conditions according to at least one of the simulation partitions, and calculate corresponding advantage parameters according to the simulation results; a first optimization module, configured to select a local optimal solution of a current round from the at least two advantage parameters according to the advantage parameter type, and determine whether the local optimal solution is greater than the global optimal solution; if so, update the local optimal solution to the global optimal solution; otherwise, maintain the current global optimal solution; The second optimization module is used to update the at least two groups of simulation conditions according to the global optimal solution and return the results to the simulation module.

[0107] In some embodiments, the simulation conditions include: a pixel matrix of scale a1×b1 formed according to the pixelated array of the simulation partition, wherein the parameters in the pixel matrix are respectively used to represent the area type on the xth row and yth column in the pixelated array, and the area type is a waveguide area or a spacer area, wherein x∈a1, y∈b1, a1<a, b1<b.

[0108] In some embodiments, the second optimization module uses an optimization function to update the simulation condition, and the optimization function includes: ; in, To optimize the goal, is the first weight, is the second weight, N is the number of output terminals of the simulation partition, is the sum of the output optical powers of the simulation partitions, is the sum of the input optical powers of the simulation partitions, is the optical power output by the output end enabled in the simulation partition, is the optical power of the remaining output ends of the simulation partition.

[0109] It should be noted that, in this document, the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or apparatus comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or apparatus comprising the element.

[0110] Through the description of the above embodiments, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus the necessary general hardware platform, and of course can also be implemented by hardware, but in many cases the former is a better embodiment. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), including a number of instructions for enabling a computer terminal (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods described in each embodiment of the present invention.

[0111] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the present invention and the claims, all of which are protected by the present invention.

Claims

1. A PCM-based integrated optical switch design simulation method, characterized in that: Including steps: S101, providing an initial architecture of the integrated optical switch, the initial architecture comprising: an a×b pixel array, wherein two ends of the pixel array are respectively connected to at least one input end and a plurality of output ends, wherein the pixel array comprises: a×b pixel regions, wherein a plurality of the pixel regions are filled with waveguides to form a waveguide region, a plurality of the pixel regions are configured as spacer regions to form partitions between the waveguide regions, and a PCM region is further provided within the pixel array; S102, determining whether the number of the output terminals is an even number; If yes, proceed to step S103; S103, determining whether there are symmetrical synchronous output terminals, wherein when two output terminals are enabled at the same time, the output terminals are considered to be mutually synchronized output terminals; If yes, proceed to step S104; S104, dividing the plurality of output terminals into at least two simulation partitions according to the synchronous output terminals, wherein the simulation partitions include: at least two output terminals and pixel array areas corresponding to the output terminals; S105, performing at least two simulations using at least two sets of simulation conditions according to at least one of the simulation partitions, and calculating corresponding advantage parameters according to the simulation results; S106, selecting a local optimal solution of the current round from the at least two advantage parameters according to the advantage parameter type, and determining whether the local optimal solution is greater than the global optimal solution; if so, updating the local optimal solution to the global optimal solution; otherwise, maintaining the current global optimal solution; S107: Update the at least two sets of simulation conditions according to the global optimal solution, and return to S105.

2. The method according to claim 1, characterized in that The simulation conditions include: a pixel matrix of a1×b1 scale formed according to the pixelated array of the simulation partition, wherein the parameters in the pixel matrix are respectively used to represent the area type on the xth row and yth column in the pixelated array, and the area type is a waveguide area or a spacer area, wherein x∈a1, y∈b1, a1<a, b1<b.

3. The method according to claim 1, characterized in that S107 updates the simulation conditions using an optimization objective function, where the optimization objective function includes: ;in, To optimize the goal, is the first weight, is the second weight, N is the number of output terminals of the simulation partition, is the sum of the output optical powers of the simulation partitions, is the sum of the input optical powers of the simulation partitions, is the optical power output by the output end enabled in the simulation partition, is the optical power of the remaining output ends of the simulation partition.

4. The method according to claim 1, wherein In S105 , the Maxwell method is used for simulation, wherein the Maxwell method discretizes the Maxwell equations in space and time, and gradually promotes the change of the electromagnetic field in the time domain to solve the distribution of the electric field and the magnetic field in the signal switching area.

5. The method according to claim 1, wherein S105 includes the steps of: For one of the simulation partitions, two adjacent pixel array sub-areas are identified from the pixel array area, and the two pixel array sub-areas have the same pixel specifications; Obtaining a degree of design symmetry of pixel arrays of the two pixel array sub-areas; When the degree of design symmetry is greater than a preset symmetry threshold, a pixel array sub-area and its corresponding output end are selected as simulation sub-areas; At least two simulations are performed using at least two sets of simulation conditions according to at least one of the simulation sub-partitions.

6. The method according to claim 5, characterized in that During the first round of simulation, simulation is performed using a first pixel specification, wherein the pixel area size of the first pixel specification is smaller than the pixel area size of the a×b specification; during the mth round of simulation, simulation is performed using a second pixel specification, wherein the pixel area size of the second pixel specification is smaller than the pixel area size of the first pixel specification.

7. A PCM-based integrated optical switch design simulation system, characterized in that: include: an architecture acquisition module, configured to provide an initial architecture of the integrated optical switch, the initial architecture comprising: an a×b pixel array, wherein two ends of the pixel array are respectively connected to at least one input end and a plurality of output ends, wherein the pixel array comprises: a×b pixel regions, wherein a plurality of the pixel regions are filled with waveguides to form a waveguide region, a plurality of the pixel regions are configured as spacer regions to form partitions between the waveguide regions, and a PCM region is further provided within the pixel array; A first judging module, configured to judge whether the number of the output terminals is an even number; If yes, then enter the second judgment module of step; a second determining module, configured to determine whether there are symmetrical synchronous output terminals, wherein when two output terminals are enabled at the same time, the output terminals are considered to be mutually synchronized output terminals; If yes, then enter the grouping module; a grouping module, configured to divide the plurality of output terminals into at least two simulation partitions according to the synchronous output terminals, wherein the simulation partitions include: at least two output terminals and pixel array areas corresponding to the output terminals; a simulation module, configured to perform at least two simulations using at least two sets of simulation conditions according to at least one of the simulation partitions, and calculate corresponding advantage parameters according to the simulation results; a first optimization module, configured to select a local optimal solution of a current round from the at least two advantage parameters according to the advantage parameter type, and determine whether the local optimal solution is greater than the global optimal solution; if so, update the local optimal solution to the global optimal solution; otherwise, maintain the current global optimal solution; The second optimization module is used to update the at least two groups of simulation conditions according to the global optimal solution and return the results to the simulation module.

8. The system according to claim 7, characterized in that The simulation conditions include: a pixel matrix of a1×b1 scale formed according to the pixelated array of the simulation partition, wherein the parameters in the pixel matrix are respectively used to represent the area type on the xth row and yth column in the pixelated array, and the area type is a waveguide area or a spacer area, wherein x∈a1, y∈b1, a1<a, b1<b.

9. The system according to claim 7, wherein: The second optimization module uses an optimization objective function to update the simulation conditions, and the optimization objective function includes: ； in, To optimize the goal, is the first weight, is the second weight, N is the number of output terminals of the simulation partition, is the sum of the output optical powers of the simulation partitions, is the sum of the input optical powers of the simulation partitions, is the optical power output by the output end enabled in the simulation partition, is the optical power of the remaining output ends of the simulation partition.

10. The system according to claim 8, wherein: The Maxwell method is used for simulation, wherein the Maxwell method discretizes the Maxwell equations in space and time, and gradually promotes the change of the electromagnetic field in the time domain to solve the distribution of the electric field and the magnetic field in the signal switching area.

Images