Cross waveguide structure based on reverse design and design method thereof

Abstract

The invention discloses a reverse design-based crossed waveguide structure, which comprises an integral lower cladding and a crossed waveguide main body structure, and is characterized in that the crossed waveguide main body structure is arranged above the integral lower cladding; the cross waveguide main body structure comprises a coupling area, an input waveguide and three output waveguides are distributed around the coupling area, the input waveguide and the three output waveguides are arranged perpendicular to the end face of the coupling area and connected with the center position of the coupling area, and the cross waveguide structure is high in integration level, high in transmittance and low in transmission loss and crosstalk. The invention further discloses a design method of the crossed waveguide, a direct binary search algorithm is adopted for the crossed waveguide structure, and the design method not only improves the integration level of the crossed waveguide structure and the transmissivity of the output waveguide, but also achieves lower transmission loss and crosstalk when different modes are input.

Description

Technical Field

[0001] The present invention belongs to the technical field of optical communication, and in particular relates to a cross-waveguide structure based on reverse design, and also relates to a design method of the cross-waveguide structure. Background Art

[0002] Traditional wireless optical communication technology has certain limitations in future development due to its strong channel interference. However, coherent optical communication technology has high bandwidth, low latency and better anti-interference performance, providing a sustainable solution for future optical communication needs.

[0003] In coherent optical communication systems, cross waveguides are important optical devices in the interconnection of optical devices. Compared with forward-designed cross waveguides, reverse-designed cross waveguides will greatly improve the integration of devices. The refractive index difference between the core and cladding of the thin-film lithium niobate platform is large, and the transmittance increases, propagation loss and crosstalk decrease when light is transmitted in a multimode waveguide. Therefore, on the thin-film lithium niobate platform, reverse design is used to optimize the coupling area of ​​the cross waveguide, and a cross waveguide with low loss, low crosstalk, high transmittance and TE0 or TM0 mode transmission can be designed.

[0004] Based on this, the present application proposes a cross waveguide based on reverse design. Summary of the invention

[0005] The object of the present invention is to provide a cross-waveguide structure based on reverse design, which has high integration, high transmittance, low transmission loss and crosstalk compared with the cross-waveguide in the prior art.

[0006] Another object of the present invention is to provide a design method for the above-mentioned cross-waveguide structure, which improves the transmittance of the device and reduces the transmission loss and crosstalk of the device by optimizing the coupling area.

[0007] The technical solution adopted by the present invention is that the cross-waveguide structure based on reverse design includes an overall lower cladding layer and a cross-waveguide main body structure, and the cross-waveguide main body structure is arranged above the overall lower cladding layer;

[0008] The cross waveguide main structure includes a coupling region, and one input waveguide and three output waveguides are distributed around the coupling region. The input waveguide and the three output waveguides are arranged perpendicular to the end face of the coupling region and are connected to the center of the coupling region.

[0009] The upper end surface of the coupling region is also provided with an etched hole.

[0010] The present invention is also characterized in that:

[0011] The overall lower cladding thickness is 3μm~7μm; the initial length of the coupling region is 1μm~3μm, the initial width is 1μm~3μm, and the thickness is 0.25μm~0.35μm; the input waveguide width is 0.3μm~0.7μm, the length is 1.3μm~1.7μm, and the thickness is 0.25μm~0.35μm; the output waveguide width is 0.3μm~0.7μm, the length is 1.3μm~1.7μm, and the thickness is 0.25μm~0.35μm.

[0012] The overall lower cladding is made of silicon dioxide material, the size of the overall lower cladding is not less than the size of the cross-waveguide main structure, and the coupling area is made of thin-film lithium niobate material.

[0013] The input mode of the input waveguide is a TE0 or TM0 light source.

[0014] The second technical solution adopted by the present invention is: a design method of a cross-waveguide structure based on reverse design, and the specific steps are as follows:

[0015] Step 1: Design the initial cross-waveguide structure on the FDPD software;

[0016] Step 2, calculating the quality factor FOM of the initial cross-waveguide structure;

[0017] Step 3: Use a direct binary search algorithm to optimize the initial cross-waveguide structure to obtain a cross-waveguide structure with optimal performance.

[0018] The present invention is also characterized in that:

[0019] The initial cross-waveguide structure in the step is as follows: it includes an overall lower cladding and a cross-waveguide main structure, the cross-waveguide main structure is arranged above the overall lower cladding, the cross-waveguide main structure includes a coupling region, one input waveguide and three output waveguides are distributed around the coupling region, the input waveguide and the three output waveguides are arranged perpendicular to the end face of the coupling region and are connected to the center of the coupling region;

[0020] The overall lower cladding thickness is 3μm to 7μm; the initial length of the coupling region is 1μm to 3μm, the initial width is 1μm to 3μm, and the thickness is 0.25μm to 0.35μm; the input waveguide width is 0.3μm to 0.7μm, the length is 1.3μm to 1.7μm, and the thickness is 0.25μm to 0.35μm; the output waveguide width is 0.3μm to 0.7μm, the length is 1.3μm to 1.7μm, and the thickness is 0.25μm to 0.35μm;

[0021] The overall lower cladding is made of silicon dioxide; the coupling region is made of thin-film lithium niobate material; the input mode of the input waveguide is TE0 or TM0 light source;

[0022] The upper end surface of the coupling region is also provided with an etched hole.

[0023] The calculation process of the quality factor FOM of the initial cross-waveguide structure is as follows:

[0024] Step 2.1, set the input light source of the input waveguide of the initial cross waveguide structure to TE0, and calculate the transmittance T of the output waveguide 4 that is colinear with the input waveguide OE ;

[0025] Step 2.2: Set the light source of the initial cross-waveguide structure to TM0 and calculate the transmittance T of the output waveguide that is collinear with the input waveguide. OM ;

[0026] Step 2.3: T OE With T OM The product of gets the quality factor FOM.

[0027] The initial cross-waveguide structure optimization steps are as follows:

[0028] Step 3.1, divide the coupling area of ​​the initial cross-waveguide structure into M pixels with a length and width of Xμm and a height of Yμm, and record the initial states of all pixels in the designed coupling area by encoding "0" and "1", where "0" means no hole is etched and "1" means a hole is etched;

[0029] Step 3.2, setting all pixels in the coupling area of ​​the initial cross-waveguide structure to "0";

[0030] Step 3.3, change the state of the first pixel point in the coupling region to "1", and calculate the quality factor FOM1 of the cross-waveguide structure in the current state according to the calculation process of the quality factor FOM of the initial cross-waveguide structure in step 2;

[0031] Step 3.4, compare the quality factor FOM1 with the quality factor FOM obtained in step 2.3. If the quality factor FOM1 is greater than the quality factor FOM, the state of the first pixel in the coupling area is kept as 1, that is, the first pixel is etched. If the quality factor FOM1 is not greater than the quality factor FOM, the state of the first pixel in the coupling area is set to 0, that is, the first pixel is not etched with a hole;

[0032] Step 3.5, then set the state of the second pixel point in the coupling area to "1", repeat step 2 to calculate the current state quality factor FOM2, compare the quality factor FOM2 with the quality factor FOM1, and determine the state of the second pixel point in the coupling area;

[0033] Step 3.6, traverse M pixel points in the coupling area according to step 3.5 until the "quality factor" no longer changes, then stop the iteration. If after traversing M pixel points in the coupling area, the "quality factor" has not yet stopped changing, it is necessary to return to the first pixel point for the next round of iteration until the "quality factor" no longer changes, then stop the iteration. The cross-waveguide structure in this state is the optimal cross-waveguide structure.

[0034] In step 3.1, the number of pixels in the coupling area is 400, and the length and width of a single pixel are both 0.1 μm and the height is 0.22 μm.

[0035] The beneficial effects of the present invention are as follows: in the reverse-designed cross-waveguide structure of the present invention, the coupling region of the cross-waveguide body is optimized by adopting a direct binary search algorithm, so that the shape of the coupling region is changed into a two-dimensional code structure, and matching with the TE0 or TM0 light field is achieved through this structure, and lower loss and crosstalk are achieved through three output ports; the coupling region adopts a thin-film lithium niobate material, so that the waveguide structure has the advantages of high bandwidth, low insertion loss, low power consumption, and easy integration.

[0036] The present invention is based on a design method for a cross-waveguide structure by reverse design. For the first time, a direct binary search algorithm is introduced into the cross-waveguide structure optimization process. The cross-waveguide is optimized by the direct binary search algorithm, which not only improves the integration of the cross-waveguide structure and the transmittance of the output waveguide, but also achieves lower transmission loss and crosstalk when different modes are input. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 It is a schematic diagram of the three-dimensional initial structure of the cross-waveguide structure in the cross-waveguide structure and the design method thereof based on reverse design of the present invention;

[0038] Figure 2 It is a top view of the cross-waveguide structure and design method thereof after optimization based on reverse design of the present invention;

[0039] Figure 3 It is a side view after optimization of the cross-waveguide in the cross-waveguide structure and design method thereof based on reverse design of the present invention;

[0040] Figure 4 It is a flow chart of a direct binary search algorithm used in the cross-waveguide structure and design method thereof based on inverse design of the present invention;

[0041] Figure 5 It is a transmittance diagram of the overall structure of the cross-waveguide in the embodiment 1 of the cross-waveguide structure and design method thereof based on reverse design of the present invention, in which the light source is in the TE0 or TM0 mode;

[0042] Figure 6 It is a loss diagram of the overall structure of the cross-waveguide in the embodiment 1 of the cross-waveguide structure and design method thereof based on reverse design of the present invention, in which the light source is in the TE0 or TM0 mode;

[0043] Figure 7 It is a crosstalk diagram of the overall structure of the cross-waveguide in TE0 or TM0 mode in which the light source in Example 1 of the cross-waveguide structure and design method thereof based on reverse design of the present invention is a cross-waveguide.

[0044] Figure 8 It is a transmittance diagram of the overall structure of the cross-waveguide in TE0 or TM0 mode in Example 2 of the cross-waveguide structure and design method thereof based on reverse design of the present invention;

[0045] Fig. 9 It is a loss diagram of the overall structure of the cross-waveguide in TE0 or TM0 mode in Example 2 of the cross-waveguide structure and design method thereof based on reverse design of the present invention;

[0046] Fig.10 It is a crosstalk diagram of the overall structure of the cross-waveguide in TE0 or TM0 mode in which the light source in Example 2 of the cross-waveguide structure and design method thereof based on reverse design of the present invention is a cross-waveguide structure.

[0047] Fig.11 It is a transmittance diagram of the overall structure of the cross-waveguide in the embodiment 3 of the cross-waveguide structure and design method thereof based on reverse design of the present invention, in which the light source is in the TE0 or TM0 mode;

[0048] Fig.12 It is a loss diagram of the overall structure of the cross-waveguide in TE0 or TM0 mode in Example 3 of the cross-waveguide structure and design method thereof based on reverse design of the present invention;

[0049] Fig.13 It is a crosstalk diagram of the overall structure of the cross-waveguide in TE0 or TM0 mode in which the light source in Example 3 of the cross-waveguide structure and design method thereof based on reverse design of the present invention is a cross-waveguide structure.

[0050] In the figure, 1. Overall lower cladding, 2. Coupling region, 3. Input waveguide, 4. Output waveguide, 5. Etched holes; 6. Cross waveguide main structure. DETAILED DESCRIPTION

[0051] The present invention is described in detail below with reference to the accompanying drawings and specific embodiments.

[0052] The present invention is based on a reverse-designed cross-waveguide structure, the structure is as follows Figure 1As shown, it includes an overall lower cladding 1 and a cross-waveguide main structure 6, and the cross-waveguide main structure 6 is arranged above the overall lower cladding 1; the cross-waveguide main structure 6 includes a coupling region 2, and the upper end surface of the coupling region 2 can be a square; Figure 2 As shown, the coupling region 2 of the cross-waveguide main structure 6 of the present invention is optimized by adopting a direct binary search algorithm, so that the shape of the coupling region 2 is changed into a two-dimensional code structure; this structure is used to achieve matching with the TE0 or TM0 light field, and through three output ports, lower loss and crosstalk are achieved.

[0053] An input waveguide 3 and three output waveguides 4 are distributed around the coupling region 2. The input waveguide 3 and the three output waveguides 4 are arranged perpendicular to the end face of the coupling region 2 and are connected to the center of the coupling region 2. The input waveguide 3 and the three output waveguides 4 are distributed on the four end faces of the coupling region 2. The input mode of the input waveguide 3 is a TE0 or TM0 light source. An etched hole 5 is also arranged on the upper end face of the coupling region 2.

[0054] The overall lower cladding 1 has a thickness of 3 μm to 7 μm; the initial length of the coupling region 2 is 1 μm to 3 μm, the initial width is 1 μm to 3 μm, and the thickness is 0.25 μm to 0.35 μm; the input waveguide 3 has a width of 0.3 μm to 0.7 μm, a length of 1.3 μm to 1.7 μm, and a thickness of 0.25 μm to 0.35 μm; the output waveguide 4 has a width of 0.3 μm to 0.7 μm, a length of 1.3 μm to 1.7 μm, and a thickness of 0.25 μm to 0.35 μm. The overall lower cladding upper 1 is made of silicon dioxide material, the size of the overall lower cladding 1 is not less than the size of the cross waveguide main structure 6, and the coupling region 2 is made of thin film lithium niobate material.

[0055] The design method of the above-mentioned cross-waveguide structure is as follows:

[0056] Step 1: Design the initial cross-waveguide structure on the FDPD software;

[0057] The initial cross-waveguide structure in step 1 is as follows: it includes an overall lower cladding 1 and a cross-waveguide main structure 6, the cross-waveguide main structure 6 is arranged above the overall lower cladding 1, the cross-waveguide main structure 6 includes a coupling region 2, one input waveguide 3 and three output waveguides 4 are distributed around the coupling region 2, the input waveguide 3 and the three output waveguides 4 are arranged perpendicular to the end face of the coupling region 2 and are connected to the center of the coupling region 2;

[0058] The overall lower cladding 1 has a thickness of 3 μm to 7 μm; the initial length of the coupling region 2 is 1 μm to 3 μm, the initial width is 1 μm to 3 μm, and the thickness is 0.25 μm to 0.35 μm; the input waveguide 3 has a width of 0.3 μm to 0.7 μm, a length of 1.3 μm to 1.7 μm, and a thickness of 0.25 μm to 0.35 μm; the output waveguide 4 has a width of 0.3 μm to 0.7 μm, a length of 1.3 μm to 1.7 μm, and a thickness of 0.25 μm to 0.35 μm;

[0059] The overall lower cladding 1 is made of silicon dioxide material; the coupling region 2 is made of thin film lithium niobate material; the input mode of the input waveguide 3 is a TE0 or TM0 light source;

[0060] The upper end surface of the coupling region 2 is also provided with an etched hole 5 .

[0061] Step 2, calculating the quality factor FOM of the initial cross-waveguide structure;

[0062] The calculation process of the quality factor FOM of the initial cross-waveguide structure is as follows:

[0063] Step 2.1, set the input light source of the input waveguide 3 of the initial cross waveguide structure to TE0, and calculate the transmittance T of the output waveguide 4 colinear with the input waveguide 3 OE ;

[0064] Step 2.2: When the light source of the initial cross-waveguide structure is set to TM0, calculate the transmittance T of the output waveguide collinear with the input waveguide 3. OM ;

[0065] Step 2.3: T OE With T OM The product of gets the quality factor FOM.

[0066] Step 3: Use a direct binary search algorithm to optimize the initial cross-waveguide structure to obtain a cross-waveguide structure with optimal performance.

[0067] like Figure 4 As shown, the initial cross-waveguide structure optimization steps are as follows:

[0068] Step 3.1, divide the coupling region 2 of the initial cross-waveguide structure into M pixels with a length and width of X μm and a height of Y μm, and record the initial states of all pixels in the designed coupling region 2 by encoding "0" and "1", where "0" means no hole is etched and "1" means a hole is etched;

[0069] Step 3.2, setting all pixels in the coupling region 2 of the initial cross-waveguide structure to "0";

[0070] Step 3.3, change the state of the first pixel point in the coupling region 2 to "1", and calculate the quality factor FOM1 of the cross-waveguide structure in the current state according to the calculation process of the quality factor FOM of the initial cross-waveguide structure in step 2;

[0071] Step 3.4, compare the quality factor FOM1 with the quality factor FOM obtained in step 2.3. If the quality factor FOM1 is greater than the quality factor FOM, the state of the first pixel point of the coupling area 2 is kept as 1, that is, the first pixel point is etched. If the quality factor FOM1 is not greater than the quality factor FOM, the state of the first pixel point of the coupling area 2 is set to 0, that is, the first pixel point is not etched with a hole;

[0072] Step 3.5, then set the state of the second pixel point in coupling area 2 to "1", repeat step 2 to calculate the current state quality factor FOM2, compare the quality factor FOM2 with the quality factor FOM1, and determine the state of the second pixel point in coupling area 2;

[0073] Step 3.6, traverse the M pixel points in the coupling area 2 according to step 3.5 until the "quality factor" no longer changes, then stop the iteration. If after traversing the M pixel points in the coupling area 2, the "quality factor" has not yet stopped changing, it is necessary to return to the first pixel point for the next round of iteration until the "quality factor" no longer changes, then stop the iteration. The cross-waveguide structure in this state is the optimal cross-waveguide structure.

[0074] In step 3.1, the number of pixels in coupling region 2 is 400, and the length and width of a single pixel are both 0.1 μm, and the height is 0.22 μm. The number of pixels is set, and the length, width, and height of a single pixel are not unique. In actual operation, the number of pixels and the length, height, and width of the pixels can be selected according to actual conditions.

[0075] The cross waveguide based on reverse design of the present application is described below through three embodiments:

[0076] In Examples 1-3, the light source uses a mode light source in the 1550nm band, the overall lower cladding 1 material is silicon dioxide, the refractive index is 1.446, the coupling region 2, the refractive index is 2.2111, the top view and side view of the main body after the cross waveguide optimization are as follows Figure 2 and Figure 3 shown.

[0077] Embodiment 1:

[0078] The cross-waveguide structure based on reverse design of the present invention includes an overall lower cladding 1 and a cross-waveguide main structure 6, and the cross-waveguide main structure 6 is arranged above the overall lower cladding 1; the cross-waveguide main structure 6 includes a coupling region 2, and the upper end face of the coupling region 2 is a square; one input waveguide 3 and three output waveguides 4 are distributed around the coupling region 2, and the input waveguide 3 and the three output waveguides 4 are arranged perpendicular to the end face of the coupling region 2 and are connected to the center position of the coupling region 2, and the input waveguide 3 and the three output waveguides 4 are distributed on the four end faces of the circumference of the coupling region 2. The input mode of the input waveguide 3 is a TE0 or TM0 light source. The upper end face of the cross-waveguide main structure 6 is also provided with an etched hole 5.

[0079] The thickness of the overall lower cladding 1 is 5 μm, the initial length of the coupling region is 2 μm, the initial width is 2 μm, and the thickness is 0.3 μm; the width of the input waveguide 3 is 0.5 μm, the length is 1.5 μm, and the thickness is 0.3 μm; the width of the three-way output waveguide 4 is 0.5 μm, the length is 1.5 μm, and the thickness is 0.3 μm.

[0080] The light source of this embodiment is a cross-waveguide structure with a transmittance of TE0 or TM0 mode. Figure 5 As shown, Figure 5 It is shown in Figure 2 that when the light source mode is TE0, the transmittance of the cross-waveguide structure of this embodiment increases first and then decreases as the wavelength increases. At 1.55μm, the transmittance is 0.9399. When the light source mode is TM0, the transmittance of the cross-waveguide structure of this embodiment increases first and then decreases as the wavelength increases. At 1.55μm, the transmittance is 0.9634.

[0081] The light source of this embodiment is a cross-waveguide structure with a loss of TE0 or TM0 mode. Figure 6 As shown in the figure, when the light source mode is TE0, the loss of the cross-waveguide structure of this embodiment first decreases and then increases with the increase of wavelength, and the loss at 1550nm is 0.23dB. When the light source mode is TM0, the loss of the cross-waveguide structure of this embodiment first decreases and then increases with the increase of wavelength, and the loss at 1550nm is 0.162dB.

[0082] The light source of this embodiment is the crosstalk of the cross-waveguide structure in TE0 or TM0 mode. Figure 7 As shown in the figure, when the light source mode is TE0, the cross-waveguide structure crosstalk of this embodiment gradually decreases with the increase of wavelength, and the overall crosstalk at 1450-1650nm is less than -26.5dB. When the light source mode is TM0, the cross-waveguide structure crosstalk of this embodiment gradually decreases with the increase of wavelength, and the overall crosstalk at 1450-1650nm is less than -30.2dB.

[0083] Embodiment 2:

[0084] The cross-waveguide structure based on reverse design of the present invention includes an overall lower cladding 1 and a cross-waveguide main structure 6, and the cross-waveguide main structure 6 is arranged above the overall lower cladding 1; the cross-waveguide main structure 6 includes a coupling region 2, and the upper end face of the coupling region 2 is a square; one input waveguide 3 and three output waveguides 4 are distributed around the coupling region 2, and the input waveguide 3 and the three output waveguides 4 are arranged perpendicular to the end face of the coupling region 2 and are connected to the center position of the coupling region 2, and the input waveguide 3 and the three output waveguides 4 are distributed on the four end faces of the circumference of the coupling region 2. The input mode of the input waveguide 3 is a TE0 or TM0 light source. The upper end face of the cross-waveguide main structure 6 is also provided with an etched hole 5.

[0085] The thickness of the overall lower cladding 1 is 7 μm, the initial length of the coupling region is 3 μm, the initial width is 3 μm, and the thickness is 0.35 μm; the width of the input waveguide 3 is 0.7 μm, the length is 1.7 μm, and the thickness is 0.35 μm; the width of the three-way output waveguide 4 is 0.7 μm, the length is 1.7 μm, and the thickness is 0.35 μm.

[0086] The light source of this embodiment is a cross-waveguide structure with a transmittance of TE0 or TM0 mode. Figure 8 As shown, Figure 8 It is shown in Figure 2 that when the light source mode is TE0, the transmission of the cross-waveguide structure of this embodiment increases first and then decreases as the wavelength increases. At 1.55μm, the transmittance is 0.9267. When the light source mode is TM0, the transmission of the cross-waveguide structure of this embodiment increases first and then decreases as the wavelength increases. At 1.55μm, the transmittance is 0.9214.

[0087] The light source of this embodiment is a cross-waveguide structure with a loss of TE0 or TM0 mode. Fig. 9 As shown in the figure, when the light source mode is TE0, the loss of the cross-waveguide structure of this embodiment first decreases and then increases with the increase of wavelength, and the loss at 1550nm is 0.29dB. When the light source mode is TM0, the loss of the cross-waveguide structure of this embodiment first increases and then decreases with the increase of wavelength, and the loss at 1550nm is 0.25dB.

[0088] The light source of this embodiment is the crosstalk of the cross-waveguide structure in TE0 or TM0 mode. Fig.10As shown in the figure, when the light source mode is TE0, the cross-waveguide structure crosstalk of this embodiment first decreases and then increases with the increase of wavelength, and the overall crosstalk at 1450-1650nm is less than -23dB. When the light source mode is TM0, the cross-waveguide structure crosstalk of this embodiment first decreases and then increases with the increase of wavelength, and the overall crosstalk at 1450-1650nm is less than -28.5dB.

[0089] Embodiment 3:

[0090] The cross-waveguide structure based on reverse design of the present invention includes an overall lower cladding 1 and a cross-waveguide main structure 6, and the cross-waveguide main structure 6 is arranged above the overall lower cladding 1; the cross-waveguide main structure 6 includes a coupling region 2, and the upper end face of the coupling region 2 is a square; one input waveguide 3 and three output waveguides 4 are distributed around the coupling region 2, and the input waveguide 3 and the three output waveguides 4 are arranged perpendicular to the end face of the coupling region 2 and are connected to the center position of the coupling region 2, and the input waveguide 3 and the three output waveguides 4 are distributed on the four end faces of the circumference of the coupling region 2. The input mode of the input waveguide 3 is a TE0 or TM0 light source. The upper end face of the cross-waveguide main structure 6 is also provided with an etched hole 5.

[0091] The thickness of the overall lower cladding 1 is 3 μm, the initial length of the coupling region is 1 μm, the initial width is 1 μm, and the thickness is 0.25 μm; the width of the input waveguide 3 is 0.3 μm, the length is 1.3 μm, and the thickness is 0.25 μm; the width of the three-way output waveguide 4 is 0.3 μm, the length is 1.3 μm, and the thickness is 0.25 μm.

[0092] The light source of this embodiment is a cross-waveguide structure with a transmittance of TE0 or TM0 mode. Fig.11 As shown, Fig.11 It is shown in the figure that when the light source mode is TE0, the transmission of the cross-waveguide structure of this embodiment increases first, then decreases, then increases, and then decreases as the wavelength increases. At 1.55μm, the transmittance is 0.9216. When the light source mode is TM0, the transmission of the cross-waveguide structure of this embodiment increases first, then decreases, then increases, and then decreases as the wavelength increases. At 1.55μm, the transmittance is 0.9611.

[0093] The light source of this embodiment is a cross-waveguide structure with a loss of TE0 or TM0 mode. Fig.12 As shown in the figure, when the light source mode is TE0, the loss of the cross-waveguide structure of this embodiment decreases first, then increases, then decreases, and then increases with the increase of wavelength, and the loss at 1550nm is 0.3dB. When the light source mode is TM0, the loss of the cross-waveguide structure of this embodiment increases first, then decreases, then increases, and then decreases with the increase of wavelength, and the loss at 1550nm is 0.15dB.

[0094] The light source of this embodiment is the crosstalk of the cross-waveguide structure in TE0 or TM0 mode. Fig.13 As shown in the figure, when the light source mode is TE0, the cross-waveguide structure crosstalk of this embodiment first decreases and then increases with the increase of wavelength, and the overall crosstalk at 1450-1650nm is less than -27dB. When the light source mode is TM0, the cross-waveguide structure crosstalk of this embodiment first decreases and then increases with the increase of wavelength, and the overall crosstalk at 1450-1650nm is less than -29.5dB.

[0095] The cross-waveguide using principle based on reverse design of the present application is as follows: after the TE0 or TM0 light beam is transmitted in the cross-waveguide structure, the output transmittance of the three-way output waveguide 4 directly reflects the loss and crosstalk information of the cross-waveguide main structure 6. At the same time, an optical power meter is connected to the three-way output waveguide 4 of the cross-waveguide structure to obtain the output power of the three-way output waveguide 4, and the loss and crosstalk are obtained through the loss and crosstalk formula.

[0096] The performance of the cross-waveguide structure based on the reverse design is evaluated by calculating the loss and crosstalk of the cross-waveguide structure. The loss and crosstalk calculation process is as follows:

[0097] Step 1: Connect the input waveguide 3 and the output waveguide 4 of the cross waveguide structure to the output ports and input ports of two other devices, and check the stability of the connection between the ports.

[0098] Step 2: Detect input signal: Connect an optical power meter to the input waveguide 3 to detect the input power of the cross waveguide.

[0099] Step 3, detecting output signals: by connecting optical power meters to the ends of the three output waveguides 4 respectively, the output powers of the three output waveguides 4 are detected.

[0100] Step 4, calculate the loss: Substitute the input power of the input waveguide 3 and the output power of an output waveguide 4 colinear with the input waveguide 3 into the loss calculation formula to obtain the loss.

[0101] Step 5, calculating the crosstalk: the crosstalk is obtained by substituting the input power of the input waveguide 3 and the output power of the two output waveguides 4 arranged perpendicular to the input waveguide 3 into the crosstalk calculation formula.

Claims

1. A cross-waveguide structure based on reverse design, characterized in that: It comprises an overall lower cladding layer (1) and a cross-waveguide main body structure (6), wherein the cross-waveguide main body structure (6) is arranged above the overall lower cladding layer (1); The cross-waveguide main structure (6) comprises a coupling region (2), and one input waveguide (3) and three output waveguides (4) are distributed around the coupling region (2), and the input waveguide (3) and the three output waveguides (4) are all arranged perpendicular to the end face of the coupling region (2) and are all connected to the center of the coupling region (2); The upper end surface of the coupling region (2) is also provided with an etched hole (5).

2. The cross-waveguide structure based on reverse design according to claim 1, characterized in that: The overall lower cladding (1) has a thickness of 3 μm to 7 μm; the coupling region (2) has an initial length of 1 μm to 3 μm, an initial width of 1 μm to 3 μm, and a thickness of 0.20 μm to 0.24 μm; the input waveguide (3) has a width of 0.3 μm to 0.7 μm, a length of 1.3 μm to 1.7 μm, and a thickness of 0.20 μm to 0.24 μm; the output waveguide (4) has a width of 0.3 μm to 0.7 μm, a length of 1.3 μm to 1.7 μm, and a thickness of 0.20 μm to 0.24 μm.

3. The cross-waveguide structure based on reverse design according to claim 2, characterized in that: The overall lower cladding (1) is made of silicon dioxide material, the size of the overall lower cladding (1) is not less than the size of the cross-waveguide main structure (6), and the coupling region (2) is made of thin-film lithium niobate material.

4. The cross-waveguide structure based on reverse design according to claim 3, characterized in that: The input mode passing through the input waveguide (3) is a TE0 or TM0 light source.

5. A design method for a cross-waveguide structure based on inverse design, characterized in that: The specific steps are as follows: Step 1: Design the initial cross-waveguide structure on the FDPD software; Step 2, calculating the quality factor FOM of the initial cross-waveguide structure; Step 3: Use a direct binary search algorithm to optimize the initial cross-waveguide structure to obtain a cross-waveguide structure with optimal performance.

6. The design method of the cross-waveguide structure based on inverse design according to step 5 is characterized in that: The initial cross-waveguide structure described in step 1 is as follows: it comprises an overall lower cladding (1) and a cross-waveguide main structure (6), wherein the cross-waveguide main structure (6) is arranged above the overall lower cladding (1), and the cross-waveguide main structure (6) comprises a coupling region (2), and one input waveguide (3) and three output waveguides (4) are distributed around the coupling region (2), and the input waveguide (3) and the three output waveguides (4) are all arranged perpendicular to the end face of the coupling region (2) and are all connected to the center position of the coupling region (2).

7. The design method of the cross-waveguide structure based on inverse design according to step 6 is characterized in that: The overall lower cladding (1) described in step 1 has a thickness of 3 μm to 7 μm; the initial length of the coupling region (2) is 1 μm to 3 μm, the initial width is 1 μm to 3 μm, and the thickness is 0.25 μm to 0.35 μm; the input waveguide (3) has a width of 0.3 μm to 0.7 μm, a length of 1.3 μm to 1.7 μm, and a thickness of 0.25 μm to 0.35 μm; the output waveguide (4) has a width of 0.3 μm to 0.7 μm, a length of 1.3 μm to 1.7 μm, and a thickness of 0.25 μm to 0.35 μm; The overall lower cladding (1) is made of silicon dioxide material; the coupling region (2) is made of thin film lithium niobate material; the input mode of the input waveguide (3) is a TE0 or TM0 light source; An etched hole (5) is also provided on the upper end surface of the cross-waveguide main body structure (7).

8. The design method of the cross-waveguide structure based on inverse design according to step 6 is characterized in that: The calculation process of the quality factor FOM of the initial cross-waveguide structure is as follows: Step 2.1, set the input light source of the input waveguide (3) of the initial cross-waveguide structure to TE0, and calculate the transmittance T of the output waveguide (4) collinear with the input waveguide (3) OE ; Step 2.2: Set the light source of the initial cross-waveguide structure to TM0, and calculate the transmittance T of the output waveguide (4) collinear with the input waveguide (3). OM ; Step 2.3: T OE With T OM The product of gets the quality factor FOM.

9. The design method of the cross-waveguide structure based on inverse design according to step 6 is characterized in that: The initial cross-waveguide structure optimization steps are as follows: Step 3.1, the coupling region (2) of the initial cross-waveguide structure is divided into M pixels with a length and width of X μm and a height of Y μm, and the initial state of all pixels in the designed coupling region (2) is recorded by encoding "0" and "1", where "0" means no hole is etched and "1" means a hole is etched; Step 3.2, set the coupling area (2) of the initial cross-waveguide structure to "0" for all pixels; Step 3.3, change the state of the first pixel point in the coupling region (2) to "1", and calculate the quality factor FOM1 of the cross-waveguide structure in the current state according to the calculation process of the quality factor FOM of the initial cross-waveguide structure in step 2; Step 3.4, compare the quality factor FOM1 with the quality factor FOM obtained in step 2.

3. If the quality factor FOM1 is greater than the quality factor FOM, the state of the first pixel point in the coupling area (2) is retained as 1, that is, the first pixel point is etched. If the quality factor FOM1 is not greater than the quality factor FOM, the state of the first pixel point in the coupling area (2) is set to 0, that is, the first pixel point is not etched with a hole; Step 3.5, then set the state of the second pixel point in the coupling area (2) to "1", repeat step 2 to calculate the quality factor FOM2 of the current state, compare the quality factor FOM2 with the quality factor FOM1, and determine the state of the second pixel point in the coupling area (2); Step 3.6: According to step 3.5, traverse M pixel points in the coupling region (2) until the "quality factor" no longer changes, then stop the iteration. If after traversing M pixel points in the coupling region (2), the "quality factor" has not yet changed, it is necessary to return to the first pixel point for the next round of iteration until the "quality factor" no longer changes, then stop the iteration. The cross-waveguide structure in this state is the optimal cross-waveguide structure.

10. The design method of the cross-waveguide structure based on inverse design according to step 9 is characterized in that: The number of pixels in the coupling region (2) described in step 3.1 is 400, and the length and width of a single pixel are both 0.1 μm and the height is 0.22 μm.

Images