Optimization method of power divider with sub-wavelength structure and power divider

By using sub-wavelength structure and genetic algorithm optimization design in the optical power distributor, the problems of process processing difficulties and insufficient design freedom in the existing technology are solved, and the optical power distributor with high transmittance and low insertion loss are realized. The effectiveness of the design.

CN119001957APending Publication Date: 2024-11-22SOUTHWEST UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411424925.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-10-12
Publication Date
2024-11-22

AI Technical Summary

Technical Problem

The existing optical power distributors have limitations in taking into account both process processing and design freedom, which makes it difficult to achieve the device and has large insertion losses.

Method used

The power splitter optimization method adopts a subwavelength structure, and divides multiple rectangular cells distributed in a rectangular array based on the substrate size, uses a genetic algorithm to perform reverse design, determine the state of each rectangular cell, and passes the arc shape of the cell structure. The edge and compensation structure are optimized to form an optical power divider with high design freedom.

Benefits of technology

It has achieved stable process manufacturing, high design freedom, small insertion loss and high transmittance. Especially in the c-band, the transmittance reaches 84.32%-99.05%, and the insertion loss is between 0.04-0.74dB, which proves that Effectiveness of the design.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119001957A_ABST
    Figure CN119001957A_ABST
Patent Text Reader

Abstract

The invention belongs to the technical field of communication, and particularly relates to an optimization method of a power divider of a sub-wavelength structure and the power divider, and the optimization method comprises the following steps: determining a design region based on the size of a substrate, and uniformly dividing the design region into a plurality of rectangular cells which are distributed in a rectangular array on a two-dimensional plane; reverse design is conducted on the states of the rectangular cells through an optimization algorithm, the state of each rectangular cell is determined, and the states of the rectangular cells include one or two states, namely whether the rectangular cells contain unit structures or not; a rectangular coordinate system is established by taking the center of the rectangular cell where the unit junction is located as an original point c, and when the states of other rectangular cells adjacent to the vertex of the rectangular cell where the original point c is located and the two side lengths sharing the vertex are all not, the part of the edge of the unit structure where the original point c is located in the corresponding quadrant is arc-shaped, otherwise, the part is a right angle; the design method has the advantages of high degree of freedom of design, stable process manufacturing and the like, and the optical power divider obtained by using the design method is smaller in insertion loss.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention belongs to the technical field of communications, and in particular relates to an optimization method for a power divider with a sub-wavelength structure and a power divider. Background Art

[0002] An optical power divider (abbreviated as power divider) is a power element that divides the energy of one input optical signal into n outputs. It distributes optical power by distributing optical signals to different channels to meet the needs of different channels for optical signals.

[0003] In recent years, there have been many reports on devices based on reverse design, but many devices have remained in theory due to difficult processing. The devices designed by people can be divided into analog and digital types. The digital type device is essentially an expression of a matrix composed of the values ​​"0" and "1" in the design area. Because of the different mappings between 0, 1 and the expression structure, it can be further divided into two-dimensional code type and photonic crystal type. In the former, 1 means there is a unit structure here (usually a square block), and 0 means etching (no unit structure); in the latter, 1 means there is a circular hole here (etching), and 0 means there is no circular hole here.

[0004] However, the above schemes have limitations in balancing process processing and high design freedom. The result of the two-dimensional code type scheme is that the device formed by the effective arrangement of the unit structure has right-angle edges, but right-angle edges are currently difficult to achieve in terms of technology; the circular hole type is easy to achieve in terms of technology, but because it only has circular holes, the design freedom is small, resulting in a large insertion loss of the device designed therefrom. Summary of the invention

[0005] The technical problem to be solved by the present invention is to provide an optimization method for a sub-wavelength structure power divider and a power divider, which are easier to process and realize, have a high degree of design freedom, low insertion loss and high transmittance.

[0006] The present invention provides a method for optimizing a sub-wavelength structure power divider, comprising the following steps: Determine a design area based on the size of the substrate, and evenly divide the design area into a plurality of rectangular cells distributed in a rectangular array on a two-dimensional plane; An optimization algorithm is used to reversely design the state of the rectangular cell to determine the state of each rectangular cell. There are two states of the rectangular cell, namely, whether it contains a unit structure; A rectangular coordinate system is established with the center of the rectangular unit cell where the unit node is located as the origin c. When the states of the other rectangular units adjacent to the vertex and the two sides of the common vertex of the rectangular unit cell where the origin c is located are both negative, the edge of the unit structure where the origin c is located in the corresponding quadrant is an arc, otherwise it is a right angle, so as to determine the shape of each unit structure.

[0007] Optionally, the optimization method of the sub-wavelength structure power divider further includes, after obtaining the structure of each unit structure, when the top angle of the rectangular unit cell without unit structure corresponds exactly to the right angle of two unit structures, using a compensation structure formed by a quarter arc tangent to the right angle to compensate the rectangular unit cell without unit structure.

[0008] Optionally, the optimization algorithm is a genetic algorithm, which is as follows: first, an initial population of m is randomly generated, the initial population is a collection of unit structures, genes are randomly exchanged between populations, the composition of the unit structure is changed, and a new population of 2m is generated, and this new population is brought into FDTD to calculate the fitness of the entire population, and individuals in the population are selected based on the fitness, and the top m individuals with the highest fitness are selected as the initial population for the next generation of iterations until the iteration ends.

[0009] Optionally, the optimization method of the sub-wavelength structure power divider further includes iterative optimization using a genetic algorithm combined with a calculated fitness function.

[0010] Optionally, the rectangular unit cell is a square, and the radius of the arc edge of the unit structure is half the length of one side of the right angle.

[0011] The present invention provides a power divider, comprising a substrate obtained by adopting the optimization method of the sub-wavelength structure power divider, and an input waveguide and an output waveguide coupled to two ends of the substrate.

[0012] Optionally, the substrate is a waveguide layer, an insulating substrate and a base layer stacked in sequence, and the designed area is located on the waveguide layer.

[0013] Optionally, the waveguide layer is made of single crystal silicon, polycrystalline silicon, indium phosphide or silicon nitride.

[0014] Optionally, the length and width of the substrate are 1-6 μm×1-6 μm.

[0015] Optionally, the input waveguide and the output waveguide are tapered waveguides.

[0016] The beneficial effect of the present invention is that the optimization method of the sub-wavelength structure power divider provided by the present invention is based on the unit structure array (a structure composed of a two-dimensional code array), and more unit structure structures are designed in the design area. Through 16 types of unit structure structures, these 16 unit structures are used to arrange and combine to form a unit structure distribution structure in the design area, which has the advantages of high design freedom and stable process manufacturing. The insertion loss of the optical power divider obtained by using the design method is smaller, and the transmittance range in the C-band 1530nm-1565nm is 84.32%-97.68%, the insertion loss range is: 0.1dB-0.74dB, the insertion loss at a wavelength of 1550nm is 0.2dB, and the transmittance is 95.436%.

[0017] The power divider was further optimized by using four-state compensation structures. The transmittance range of the C-band 1530nm-1565nm was 95%-99.05%, and the insertion loss was 0.04-0.15dB. At a wavelength of 1550nm, the transmittance in the upper, middle and lower output waveguides was 32.9%, 33.25% and 32.9%, respectively, that is, the total transmittance was 99.05%, and the insertion loss was 0.04dB. The ultra-low loss result confirmed that this scheme has a strong ability to regulate light, that is, it has a high degree of design freedom, and the power divider has no right-angle edges, which is easier to implement in terms of process. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a schematic diagram of the position of the rectangular unit cell of the present invention in a rectangular coordinate system; Figure 2 This is a schematic diagram of the unit structure of the first embodiment of the present invention; Figure 3 This is a schematic diagram of the design area unit structure distribution of the first embodiment of the present invention; Figure 4 This is a schematic diagram of an optical power divider according to a first embodiment of the present invention; Figure 5 A schematic diagram of power distribution of an optical power distributor according to a first embodiment of the present invention; Figure 6 is a curve diagram showing the variation of the total transmittance of the optical power divider as the number of iterations in the first embodiment of the present invention; Figure 7 This is a diagram showing the insertion loss and transmittance variation of the optical power divider in the c-band of the first embodiment of the present invention; Figure 8 It is the flow chart of the genetic algorithm of the present invention; Fig. 9 It is a schematic diagram of the unit structure and compensation structure of the second embodiment of the present invention; Fig.10This is a schematic diagram of the design area unit structure distribution of the second embodiment of the present invention; Fig.11 for Fig.10 A magnified view of area A in; Fig.12 This is a schematic diagram of an optical power distributor according to a second embodiment of the present invention; Fig.13 A schematic diagram of power distribution of an optical power distributor according to a second embodiment of the present invention; Fig.14 is a curve diagram showing changes in transmittance of an optical power divider according to the second embodiment of the present invention as the number of iterations increases; Fig.15 This is a diagram showing the insertion loss and transmittance variation of the optical power divider in the c-band according to the second embodiment of the present invention.

[0019] In the figure: 101, unit structure; 102, input waveguide; 103, output waveguide; 104, insulating substrate; 105, compensation structure. DETAILED DESCRIPTION

[0020] Embodiment 1 The present invention provides a method for optimizing a sub-wavelength structure power divider, comprising the following steps: Determine the design area based on the substrate size, and evenly divide the design area into a plurality of rectangular cells distributed in a rectangular array in a rectangular grid on a two-dimensional plane; A genetic algorithm is used to reversely design the state of the rectangular cells to determine the state of each rectangular cell, where the state of each rectangular cell has two types: yes or no, that is, whether the cell structure 101 is contained; A rectangular coordinate system is established with the center of the rectangular cell where the unit structure 101 is located as the origin c. When the states of the other rectangular cells adjacent to the vertex of the rectangular cell where the origin c is located and the two sides of the common vertex are both negative, the edge of the unit structure 101 where the origin c is located is an arc in the corresponding quadrant, otherwise it is a right angle, so as to determine the shape of each unit structure 101. The structure of a single unit structure 101 is as follows: Figure 2 There are 16 types shown, and these 16 unit structures 101 are arranged and combined.

[0021] It should be noted that the state of the rectangular cell is either yes or no, which is equivalent to binary 1 / 0. Figure 1As shown, a rectangular coordinate system is established at the origin c in the figure, and binary 1 / 0 is used to represent the presence or absence of the rectangular cell structure 101. Taking the first quadrant as an example, the current cell structure 101 is at the origin c in the figure. If the three rectangular cells corresponding to "1", "4" and "8" in the figure are all logical 0, then the edge of the current cell structure 101 in the first quadrant is an arc. If this condition is not met, it is a right angle. Similarly, the second quadrant ("1", "2", "5"), the third quadrant ("2", "3", "6") and the fourth quadrant ("3", "4", "7") all follow this method. In summary, using the above method, we can get Figure 2 Unit structure of all structures in 101.

[0022] In addition, the rectangular cells have two states, "yes" or "no", "yes" means not etched, and "no" means etched. In the process of preparing the optical power divider, the unit structure 101 is the portion retained after etching the waveguide layer, and the boundary line between adjacent unit structures 101 in the figure does not actually exist.

[0023] Taking the above method to design a 1x3 optical power divider as an example, a 220 nm thick standard SOI is used as a substrate. The substrate is a waveguide layer, an insulating substrate 104 and a base layer stacked in sequence. The design area is located on the waveguide layer. The design area is 3μm×3μm. The widths of the input waveguide 102 and the output waveguide 103 are both 400nm (narrow) and 800nm ​​(wide). There are 30×30 rectangular cells in the design area. The rectangular cells are square. The radius of the arc edge of the unit structure 101 is set to 50nm, which is half of the side length of the rectangular cell. Using this optimization method, we get Figure 3 The unit structure 101 of the design area is distributed as shown, and an input waveguide 102 and three output waveguides 103 are coupled to the design area to obtain an optical power divider, as shown in FIG. Figure 4 The optical power divider shown in the figure includes a substrate obtained by adopting the optimization method of the sub-wavelength structure power divider and an input waveguide and an output waveguide coupled to both ends of the substrate, such as Figure 5 The power distribution diagram of the optical power divider is shown in FIG. Figure 6 The curve of transmittance changing with the number of iterations is shown in Figure 7 The curve of transmittance and insertion loss in the c-band is shown, where the transmittance ranges from 84.32% to 97.68% and the insertion loss ranges from 0.1dB to 0.74dB in the c-band 1530nm-1565nm. The insertion loss at a wavelength of 1550nm is 0.2dB and the transmittance is 95.436%. It should be noted that 1550nm is the most commonly used wavelength for optical fiber communication.

[0024] See also Figure 8The optimization method of the genetic algorithm is: first, randomly generate an initial population of 50 (a set of matrices), randomly exchange genes between populations (the composition of transformation matrices), generate a new population of 100, bring this new population into FDTD (Finite Difference Time Domain) to calculate the fitness of the entire population, select individuals in the population based on the fitness, select the top 50 individuals, and use these 50 individuals as the initial population (parent generation) for iteration in the next generation until the iteration ends.

[0025] To realize a 1x3 optical power divider, the goal is to have the three output waveguides 102 evenly divide the power of the input waveguide 102. If the transmittances in the upper, middle, and lower waveguides 102 are marked as t1, t2, and t3, respectively, the fitness function is set to: FOM = abs(0.3333 – t1) + abs(0.3333 – t2) + abs(0.3333 – t3) The design object is an evenly distributed power divider, and the unit structure 101 is distributed symmetrically in the upper and lower parts. In this way, the number of parameters to be optimized is reduced by half, and the transmittance values ​​of the upper and lower waveguides 102 can be guaranteed to be consistent.

[0026] Compared with the prior art, the optimization method of the sub-wavelength structure power divider provided by the present invention is based on a unit structure 101 array (a structure composed of a two-dimensional code array), and more unit structures 101 structures are designed in a design area. Through 16 types of unit structures 101 structures, these 16 unit structures 101 are arranged and combined to form a unit structure 101 distribution structure in the design area. The method has the advantages of high design freedom and stable process manufacturing. The insertion loss of the optical power divider obtained using the design method is smaller.

[0027] Embodiment 2 Taking a 1x3 optical power divider as an example, a 220 nm thick standard SOI is used as a substrate. The substrate is a waveguide layer, an insulating substrate 104 and a base layer stacked in sequence. The design area is located on the waveguide layer. The design area is 3μm×3μm. The widths of the input waveguide 102 and the output waveguide 103 are both 400nm (narrow) and 800nm ​​(wide). There are 30×30 rectangular cells in the design area. The rectangular cells are squares. The radius of the arc edge of the unit structure 101 is set to 50nm, which is half of the side length of the rectangular cell. The optimization method includes using four states of compensation structures 105. The compensation structure 105 is the portion retained after etching the waveguide layer. The boundary between the compensation structure 105 and the unit structure 101 in the figure does not actually exist.

[0028] On the basis of the optimization method of the first embodiment, the optimization method of the power divider with a sub-wavelength structure further includes, after obtaining the structure of each unit structure 101, when the top angle of the rectangular unit cell of the unit-free structure 101 just corresponds to the right angle of the two unit structures 101, a structure formed by a quarter arc tangent to the right angle is used to compensate the rectangular unit cell of the unit-free structure 101, such as Fig. 9 The compensation structure of the four states shown in the figure is used to further Figure 3 The optical power divider shown is further optimized to obtain Fig.10 and Fig.11 The design area unit structure 101 of the sub-wavelength structure shown is a distributed structure. The waveguide 102 is coupled to the design area to obtain Fig.12 The optical power divider shown in the figure, wherein the gray part (unit structure 101) is made of silicon, and the white part is etched for testing: Fig.13 The power distribution results shown are Fig.14 The transmission of the three channels of the power divider shown and the curves of their variation with the number of iterations; see Fig.15 The transmittance of the power divider in the C-band 1530nm-1565nm ranges from 95% to 99.05%, and the insertion loss is 0.04-0.15dB. When the wavelength is 1550nm, the transmittance in the upper, middle, and lower output waveguides 102 is 32.9%, 33.25%, and 32.9%, that is, the total transmittance is 99.05%, and the insertion loss is 0.04dB. The ultra-low loss result proves that the scheme has a strong ability to regulate light, that is, the design freedom is high; at the same time, from Fig.12 It can be seen that the power divider expresses a white circular hole, that is, this scheme inherits some characteristics of the photonic crystal-like scheme, which further illustrates that this scheme has a high degree of design freedom; Fig.13 The power distribution of the device is demonstrated, with the main features of stable output mode field and even power distribution; Fig.14 The change of the total transmittance of the device with the number of iterations of the genetic algorithm is shown. At the same time, the genetic algorithm fits well with this scheme and converges after 300 iterations.

[0029] A person skilled in the art should understand that the discussion of any of the above embodiments is merely illustrative and is not intended to imply that the scope of protection of the present application is limited to these examples. In line with the concept of the present application, the technical features in the above embodiments or different embodiments may be combined, the steps may be implemented in any order, and there are many other variations of different aspects of one or more embodiments of the present application as described above, which are not provided in detail for the sake of simplicity.

[0030] One or more embodiments of the present application are intended to cover all such substitutions, modifications and variations that fall within the broad scope of the present application. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of one or more embodiments of the present application should be included in the protection scope of the present application.

Claims

1. A method for optimizing a sub-wavelength power divider, characterized in that: The following steps are involved: Determine a design area based on the size of the substrate, and evenly divide the design area into a plurality of rectangular cells distributed in a rectangular array on a two-dimensional plane; An optimization algorithm is used to reversely design the state of the rectangular cell to determine the state of each rectangular cell, wherein the rectangular cell has two states, namely, whether it contains a unit structure (101); A rectangular coordinate system is established with the center of the rectangular cell where the unit structure (101) is located as the origin c. When the states of the vertex of the rectangular cell where the origin c is located and the two sides of the common vertex of the other rectangular cells adjacent to each other are both negative, the edge of the unit structure (101) where the origin c is located is an arc in the corresponding quadrant, otherwise it is a right angle, so as to determine the shape of each unit structure (101).

2. The optimization method of the sub-wavelength structure power divider according to claim 1, characterized in that: The method further includes, after obtaining the structure of each unit structure (101), when the top angle of the rectangular unit cell of the unit-free structure (101) corresponds exactly to the right angle of the two unit structures (101), using a compensation structure (105) formed by a quarter arc tangent to the right angle to compensate the rectangular unit cell of the unit-free structure (101).

3. The optimization method of the sub-wavelength structure power divider according to claim 1 or 2, characterized in that: The optimization algorithm is a genetic algorithm, which is as follows: first, an initial population of m is randomly generated, and the initial population is a set composed of a unit structure (101). Genes are randomly exchanged between populations, and the composition of the unit structure (101) is changed to generate a new population of 2m. This new population is brought into FDTD to calculate the fitness of the entire population. The individuals in the population are selected based on the fitness, and the top m individuals with the highest fitness are selected as the initial population for the next generation of iterations until the iteration ends.

4. The optimization method of the sub-wavelength structure power divider according to claim 1 or 2, characterized in that: It also includes the use of genetic algorithms combined with calculation of fitness functions for iterative optimization.

5. The optimization method of the sub-wavelength structure power divider according to claim 1 or 2, characterized in that: The rectangular unit cell is a square, and the radius of the arc edge of the unit structure (101) is half the length of one side of the right angle.

6. A power divider, characterized in that: The invention comprises adopting the optimization method of the sub-wavelength structure power divider according to claim 1 or 2 to obtain a substrate and an input waveguide and an output waveguide coupled to two ends of the substrate.

7. The power divider according to claim 6, characterized in that: The substrate is a waveguide layer, an insulating substrate and a base layer stacked in sequence, and the designed area is located on the waveguide layer.

8. The power divider according to claim 7, characterized in that: The material of the waveguide layer is single crystal silicon, polycrystalline silicon, indium phosphide or silicon nitride.

9. The power divider according to claim 6, characterized in that: The length and width of the substrate are 1-6 μm×1-6 μm.

10. The power divider according to claim 6, characterized in that: The input waveguide and the output waveguide are tapered waveguides.