A design method for polarization-insensitive wavelength division multiplexing demultiplexer

Abstract

The present invention discloses a design method for a polarization-insensitive wavelength division multiplexing (WDM) demultiplexer, comprising the following steps: selecting the core thickness of the WDM demultiplexer to obtain the variation of the effective refractive index with waveguide width in polarization mode TE and polarization mode TM; solving the grating equation to obtain the waveguide width and the diffraction order of polarization mode TE and the diffraction order of polarization mode TM, so that polarization mode TE and polarization mode TM both satisfy the grating equation at the same center wavelength; and if the wavelength spacing ratio of polarization mode TE to polarization mode TM is found to be between 0.98 and 1, then using the waveguide width and the diffraction order of polarization mode TE and the diffraction order of polarization mode TM as design parameters of the polarization-insensitive WDM demultiplexer to complete the design. The present invention enables the WDM to operate simultaneously in TE and TM polarization modes, thereby resolving the polarization sensitivity issue of WDM devices.

Description

Technical Field

[0001] The present invention relates to the technical field of multiplexing, and in particular to a design method for a polarization-insensitive wavelength division multiplexing demultiplexer. Background Art

[0002] Wavelength division multiplexing (WDM) technology significantly increases the transmission capacity of optical fibers by transmitting multiple independent data streams within the same fiber, meeting the ever-increasing demand for communications. Advances in WDM technology not only provide a solid communications foundation for high-speed internet, video transmission, and telemedicine, but also drive the rapid development of the information society and lay the foundation for the future of optical communications.

[0003] In WDM systems, single-mode fiber is the preferred choice for long-distance communications. However, the unstable polarization state of light can cause polarization-dependent loss and wavelength errors, affecting signal transmission. Therefore, WDM devices must function properly in all polarization states to avoid energy loss and signal dropout.

[0004] Although the planar lightwaveguide (PLC) platform based on silica can realize polarization-independent WDM devices, its device size is large and it is difficult to achieve monolithic integration with detectors and modulators. In contrast, the InP platform, although it can achieve monolithic integration and is polarization-insensitive, still faces size and cost limitations.

[0005] Silicon-based photonic integration is considered an ideal platform for implementing wavelength division multiplexing (WDM) systems due to its low cost, high density, monolithic integration, and compatibility with CMOS processes. However, for silicon-based optical integrated devices, the high refractive index difference between the silicon layer and the cladding leads to significant birefringence, which in turn causes serious polarization-dependent issues. This limits typical planar integrated WDM devices (such as etched diffraction gratings, arrayed waveguide gratings, Mach-Zehnder cascade structures, and microring resonator filters) to operation in a single polarization state. Therefore, constructing a polarization-compatible receiver demultiplexing architecture has become a pressing challenge for realizing monolithic silicon-based WDM systems.

[0006] In silicon-based platforms, due to the large refractive index difference between TE (Transverse Electric Mode) and TM (Transverse Magnetic Mode) modes, the demultiplexing devices are usually polarization-dependent.

[0007] To address this problem, planar integrated demultiplexing devices usually have the following solutions:

[0008] The first is a polarization-independent approach. Increasing the thickness of the waveguide core reduces the birefringence effect of the waveguide, reducing the difference in the effective refractive index under the two polarizations. By using thick waveguide cores and square waveguides, the effective refractive indices under the two polarizations are made as close as possible, constructing a polarization-insensitive demultiplexing device and achieving wavelength demultiplexing of TE and TM polarizations. However, this approach often requires a thicker waveguide cross-section, which is inconsistent with the thickness dimensions of conventional waveguides, resulting in poor adaptability and integration.

[0009] The second polarization diversity solution involves separating the TE and TM modes of the input signal using a polarization separation rotator (PSR). This separates the TM0 mode into TE0 mode, splitting the two polarizations into two identically polarized beams that are then demultiplexed using a demultiplexer designed for the TE mode. The PSR device in this solution often requires a second etching step, increasing the manufacturing process complexity.

[0010] The other is to separate the TE mode and TM mode of the input signal through a polarization separation device (PSR) and input them into WDM devices operating in TE mode and TM mode respectively.

[0011] This solution has obvious disadvantages. First, it requires additional integrated polarization separation devices. Second, this solution requires the design of WDM devices for TE and TM modes separately. The design is complex and requires two WDM devices, resulting in a large layout area, low efficiency and low integration.

[0012] In addition, the use of polarization beam splitters can also separate the TE0 mode and TM0 mode in real time. TE and TM signals are input from different input angles and output from a single wavelength division multiplexer, enabling devices such as EDGs and AWGs to operate simultaneously in TE and TM modes. However, this design still requires a front-end cascaded polarization beam splitter, making it difficult to achieve polarization-independent wavelength division multiplexing with a single device.

[0013] The above solutions often require additional components, have complex processes, and are not very universal. Summary of the Invention

[0014] The present invention provides a design method for a polarization-insensitive wavelength division multiplexing (WDM) demultiplexer. By utilizing the different effective refractive indices of the TE and TM mode waveguides, the polarization sensitivity of WDM devices can be addressed by selecting appropriate TE and TM mode diffraction orders and adjusting the optical path difference.

[0015] A polarization-insensitive wavelength division multiplexing (WDM) demultiplexer design method is proposed. The polarization-insensitive demultiplexer is designed to adjust the effective refractive index of each polarization and the diffraction order of different polarizations to construct the optical path difference according to the different effective refractive indexes under different polarization states and wavelengths. This ensures that both polarizations satisfy the grating equation under different diffraction orders and the same central wavelength, thereby achieving consistent wavelength separation under different polarizations.

[0016] This design is universal and applicable to arrayed waveguide gratings, microring resonators, Mach-Zehnder interferometers, and other devices. The optical path difference in arrayed waveguide gratings, microring resonators, and Mach-Zehnder interferometers primarily originates in straight or curved waveguides, which are constructed by adjusting the thickness, width, and length of the waveguides.

[0017] A design method for a polarization-insensitive wavelength division multiplexing demultiplexer comprises the following steps:

[0018] S1. Select the core thickness of the wavelength division multiplexing demultiplexer and obtain the variation of the effective refractive index with the waveguide width in polarization mode TE and polarization mode TM;

[0019] S2. According to the variation law of the effective refractive index with the waveguide width in polarization mode TE and polarization mode TM obtained in step S1, the waveguide width and the diffraction order of polarization mode TE and the diffraction order of polarization mode TM are obtained by solving the grating equation, so that the polarization mode TE and the polarization mode TM both satisfy the grating equation at the same center wavelength;

[0020] S3. The wavelength intervals under polarization mode TE and polarization mode TM are tested based on the waveguide width obtained in step S2 and the diffraction order of polarization mode TE and the diffraction order of polarization mode TM. If the ratio of the wavelength intervals under polarization mode TE and polarization mode TM is 0.98-1, the waveguide width and the diffraction order of polarization mode TE and the diffraction order of polarization mode TM obtained in step S2 are used as design parameters of the polarization-insensitive wavelength division multiplexing demultiplexer, and the design of the polarization-insensitive wavelength division multiplexing demultiplexer is completed.

[0021] The present invention utilizes the different effective refractive indices of TE mode and TM mode waveguides. By selecting appropriate TE mode diffraction orders and TM mode diffraction orders and adjusting the optical path difference, both TE polarization and TM polarization satisfy the grating equation at the same wavelength, thereby enabling the wavelength division multiplexer to operate simultaneously under TE polarization and TM polarization, and solving the polarization sensitivity problem of wavelength division multiplexing devices.

[0022] In step S1, the wavelength division multiplexing demultiplexer is an arrayed waveguide grating, a microring resonator or a Mach-Zehnder interferometer.

[0023] In step S2, the grating equation is as follows:

[0024]

[0025]

[0026] in, is the diffraction order of polarization mode TE, is the diffraction order of polarization mode TM, is the center wavelength, w is the waveguide width, is the effective refractive index of the waveguide in TE mode, is the effective refractive index of the waveguide in the TM mode, is the waveguide length difference.

[0027] In step S3, the wavelength intervals under the polarization mode TE and the polarization mode TM are checked using the following formula:

[0028]

[0029] in, is the wavelength interval ratio, and Respectively represent the wavelength intervals output at the same output port under polarization mode TE and polarization mode TM, and is the effective refractive index of the slab waveguide in polarization mode TE and polarization mode TM, and is the group refractive index of the array waveguide in polarization mode TE and polarization mode TM.

[0030] Furthermore, the grating equation is

[0031]

[0032] Where m is the diffraction order of the wavelength division multiplexer, is the central wavelength, is the effective refractive index of the waveguide in TE mode, is the waveguide length difference.

[0033] In an arrayed waveguide grating, the design steps are as follows:

[0034] 1) Selecting an appropriate core thickness to obtain the variation of the effective refractive index with waveguide width in both TE and TM polarization modes;

[0035] 2) Calculate and select the appropriate waveguide width and diffraction orders for TE and TM polarizations using the following formulas;

[0036]

[0037]

[0038] in, is the diffraction order of polarization mode TE, is the diffraction order of polarization mode TM, is the center wavelength, w is the waveguide width, is the effective refractive index of the waveguide in TE mode, is the effective refractive index of the waveguide in the TM mode, is the waveguide length difference.

[0039] 3) After selecting the diffraction order and waveguide width in step 2), check the wavelength separation under TE polarization and TM polarization. The two satisfy the following relationship:

[0040]

[0041] in, and Respectively represent the wavelength intervals output at the same output port under TE polarization and TM polarization, and is the effective refractive index of the slab waveguide under TE polarization and TM polarization, and is the group refractive index of the array waveguide in polarization mode TE and polarization mode TM.

[0042] The wavelength interval ratio under TE polarization and TM polarization was verified to be 0.98~1, and a polarization-insensitive device with good performance was obtained.

[0043] Compared with the prior art, the present invention has the following advantages:

[0044] The structure is simple in design and does not require any special design.

[0045] No need to cascade polarization separation devices, and the structure is simple.

[0046] It has wide applicability and is widely applicable to various wavelength devices without the need for redundant structural design. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 is the spectrum diagram of the output port;

[0048] Figure 2 The TE mode and TM mode in the straight waveguide vary with the effective refractive index and cladding thickness;

[0049] Figure 3 Schematic diagram of the structure of the polarization-insensitive AWG in the present invention;

[0050] Figure 4 The output of polarization-insensitive AWG in design case 1 under TE polarization and TM polarization;

[0051] Figure 5 For the design case 2, the output of the polarization-insensitive AWG under TE polarization and TM polarization; DETAILED DESCRIPTION

[0052] The implementation method, principle design and technical effects of the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0053] Among them, taking the polarization-insensitive arrayed waveguide grating in the silicon nitride (SiN) material platform as an example,

[0054] In principle, an AWG consists of three main components: the input and output slab waveguides, and an arrayed waveguide. The input and output slab waveguides adhere to a Rowland circle structure. Polarization issues primarily arise in the arrayed waveguide.

[0055] Take the input position of light at the center of the Rowland circle as an example. Light enters the input slab waveguide area, disperses in the slab waveguide area, and couples into the arrayed waveguide. Since the light travels the same distance in the first free propagation area, the optical path difference of each light is 0 when entering the arrayed waveguide. The arrayed waveguide is generally a group of waveguides with equal length differences and equal widths. When light passes through these waveguides, it will produce phase differences at equal intervals. Light enters the output slab waveguide from the arrayed waveguide port and diffracts. Light of different wavelengths converges on the Rowland circle. Since the optical path differences of different wavelengths when converging on the Rowland circle are different, light of different wavelengths converges at different positions on the Rowland circle.

[0056] If a beam of light is input from the center of the input slab waveguide, the output wavelength of an output port on both sides of the center satisfies the following equation:

[0057]

[0058] where m is the diffraction order, is the effective refractive index of the array waveguide, is the effective refractive index of the slab waveguide, is the length difference of the array waveguide, is the output angle of light, da is the spacing between array waveguides, and the grating equation satisfied by the light output from the center of the output slab waveguide is:

[0059]

[0060] It can be seen that due to the birefringence effect, the effective refractive index of different polarizations is different, and the optical path difference is different, resulting in different central wavelengths.

[0061] However, by utilizing the polarization refractive index difference, we can achieve polarization insensitivity by selecting the appropriate TE mode diffraction order, TM mode diffraction order and appropriate waveguide width, so that the mi-th diffraction peak of the TM mode coincides with the m-th diffraction peak of the TE mode. The output spectrum diagram of a certain output port under polarization mode TE and polarization mode TM is shown as follows: Figure 1 As shown, the first polarization mode TE The diffraction peaks are related to the first diffraction peaks in polarization mode TM. The diffraction peaks coincide with each other, which is an effect of linear polarization insensitivity.

[0062] The variation of the effective refractive index of SiN straight waveguide with waveguide width in TE mode and TM mode was simulated.

[0063] like Figure 2 As shown, considering that the effective refractive index of the waveguide is a function of the waveguide width, for the central wavelength,

[0064] TE mode meets the following requirements:

[0065]

[0066] TM mode satisfies:

[0067]

[0068] in, is the diffraction order of the arrayed waveguide grating in TE mode, is the center wavelength, w is the waveguide width, is the effective refractive index of the waveguide in polarization mode TE, is the effective refractive index of the waveguide in polarization mode TM, is the length difference between adjacent array waveguides.

[0069] Depend on Figure 2 The effective refractive index of the same waveguide in polarization mode TM is smaller than that in polarization mode TE. The waveguide width w and the diffraction order that satisfy the two equations are solved. and , and then the length difference of adjacent waveguides is obtained.

[0070] The AWG designed this time is a saddle-shaped structure made of silicon nitride. The array waveguide part provides different optical path differences for adjacent optical paths. In this design, the optical path difference is provided by the straight waveguide part, which is a strip waveguide. After calculation, the central wavelength of the AWG designed this time is selected as 1310nm, and the diffraction order of case 1 under polarization mode TE is 54, the diffraction order under polarization mode TE is 51, and the array waveguide width is 2.022um. The calculated length difference of adjacent waveguides is 42.64um; the diffraction order of case 2 is 37, the diffraction order under polarization mode TE is 35, the array waveguide width is 1.6145um, and the calculated length difference of adjacent waveguides is 29.4221um; the obtained AWG structure diagram is as follows Figure 3 shown.

[0071] Since TE polarization and TM polarization share the same output port, the wavelength interval of each output channel under TE polarization is

[0072]

[0073] Lian Li De

[0074]

[0075] In case one, When the wavelength interval of the TE mode is 3.2nm, the wavelength interval of the TM mode is 3.172nm, which meets the design requirements. With a limited number of channels, the error can be approximately ignored.

[0076] In case 2, When the wavelength interval of the TE mode is 3.2nm, the wavelength interval of the TM mode is 3.1718nm, which meets the design requirements. With a limited number of channels, the error can be approximately ignored.

[0077] like Figure 4 The figure shows the simulated output of the AWG in Case 1 under TE and TM polarizations. It can be seen that the output wavelengths of TE and TM polarizations match well, with the polarization-dependent wavelength offset of each channel ranging from 0.001 nm to 0.08 nm.

[0078] like Figure 5 The simulated output of the AWG in Case 2 under TE and TM polarizations is shown. The output wavelengths for TE and TM polarizations match well, with the polarization-dependent wavelength shift for each channel ranging from 0.001nm to 0.08nm. To minimize polarization-dependent loss, waveguide loss under different polarizations and the design of the wedge waveguide are considered, ensuring minimal loss for light of different polarizations.

[0079] The simulation results are consistent with the above deduction.

[0080] At the same time, the waveguide structure used in the example is a strip waveguide. If a ridge waveguide structure is adopted, due to the asymmetry of the ridge waveguide in its vertical direction, at certain widths, the TM0 mode is easily hybridized with the high-order TE mode, resulting in mode changes. In this case, the waveguide width should be avoided as much as possible within the hybridization range.

Claims

1. A design method for a polarization-insensitive wavelength division multiplexing demultiplexer, characterized in that: The following steps are involved: S1. Select the core thickness of the wavelength division multiplexing demultiplexer and obtain the variation of the effective refractive index with the waveguide width in polarization mode TE and polarization mode TM; S2. According to the variation law of the effective refractive index with the waveguide width in polarization mode TE and polarization mode TM obtained in step S1, the waveguide width and the diffraction order of polarization mode TE and the diffraction order of polarization mode TM are obtained by solving the grating equation, so that the polarization mode TE and the polarization mode TM both satisfy the grating equation at the same center wavelength; S3. The wavelength intervals under polarization mode TE and polarization mode TM are tested based on the waveguide width, the diffraction order of polarization mode TE, and the diffraction order of polarization mode TM obtained in step S2. If the ratio of the wavelength intervals under polarization mode TE and polarization mode TM is between 0.98 and 1, the waveguide width, the diffraction order of polarization mode TE, and the diffraction order of polarization mode TM obtained in step S2 are used as design parameters of the polarization-insensitive wavelength division multiplexing demultiplexer, and the design of the polarization-insensitive wavelength division multiplexing demultiplexer is completed.

2. The design method of a polarization-insensitive wavelength division multiplexing demultiplexer according to claim 1, wherein: In step S1, the wavelength division multiplexing demultiplexer is an arrayed waveguide grating, a microring resonator or a Mach-Zehnder interferometer.

3. The design method of a polarization-insensitive wavelength division multiplexing demultiplexer according to claim 1, wherein: In step S2, the grating equation is as follows: m TF λ0=n TE (w)ΔL m TM λ0=n TM (w)ΔL Among them, m TE is the diffraction order of polarization mode TE, m TM is the diffraction order of polarization mode TM, λ0 is the central wavelength, w is the waveguide width, n TE is the effective refractive index of the waveguide in polarization mode TE, n TM is the effective refractive index of the waveguide in polarization mode TM, and ΔL is the waveguide length difference.

4. The design method of a polarization-insensitive wavelength division multiplexing demultiplexer according to claim 1, wherein: In step S3, the wavelength intervals under the polarization mode TE and the polarization mode TM are checked using the following formula:

5. Among them, Refers to the wavelength interval ratio, Δλ TE and Δλ TM Respectively represent the wavelength intervals output at the same output port under polarization mode TE and polarization mode TM, n sTE With n sTM is the effective refractive index of the slab waveguide in polarization mode TE and polarization mode TM, n gTE With n gTM is the group refractive index of the array waveguide in polarization mode TE and polarization mode TM.

Images