A dynamic dimming mode selection control device and method

Abstract

The present application provides a dynamically adjustable optical mode selection control device and method, belonging to the field of optical communication technology. The method includes a silicon waveguide layer and a PWB layer. Each coupling region of the silicon waveguide layer and the PWB layer is provided with a stacked absorption layer. The stacked absorption layer is deposited inside the PWB, and there is a distance between the stacked absorption layer and the silicon waveguide layer. A MEMS cantilever beam is connected to the top of the stacked absorption layer. The MEMS cantilever beam is disposed outside the PWB layer and is used to adjust the distance between the stacked absorption layer and the silicon waveguide layer. The distance between the stacked absorption layer and the silicon waveguide layer is maintained within a set critical range. The MEMS cantilever beam is used to adjust the distance between the stacked absorption layer and the silicon waveguide layer, thereby achieving selective absorption of a specific mode and selecting the required coupling efficiency for mode transmission, thereby better performing mode selective absorption of silicon-based photonic devices and helping to optimize absorption efficiency.

Description

Technical Field

[0001] The present application relates to the field of optical communication technology, and in particular to a device and method for dynamically adjustable light mode selection control. Background Art

[0002] In the fields of optical communications and photonic computing, optical mode selection and control devices are core components for realizing multimode fiber communications, vortex light field control, and photonic neural networks.

[0003] In traditional silicon-based photonic devices, mode absorption control faces significant technical bottlenecks. Existing solutions typically use an absorption layer design that is fixed in position or directly integrated into the waveguide surface, resulting in an inability to effectively distinguish between the substrate mode and the target mode. For example, the spacing between the conventional absorption layer and the waveguide is limited to within 50 nanometers. Although this can improve light absorption efficiency, its broadband absorption characteristics will indiscriminately lose all modes. When the waveguide width is reduced to 200 nanometers, the leakage rate of high-order modes can reach over 90%, but the substrate TE mode will still be absorbed synchronously, resulting in at least 15% non-selective loss.

[0004] The root cause of this non-selective absorption lies in the lack of spatial resolution of the mode field distribution in traditional designs. When the waveguide enters the tapered region, the mode coupling process exhibits complex three-dimensional field reconstruction characteristics: the effective refractive index of the substrate mode differs significantly from that of the leaky mode, but the position of the fixed absorption layer cannot match the spatial attenuation length of a specific mode. In addition, the refractive index drift caused by temperature changes will change the mode coupling path, further exacerbating the non-selective interference of the absorption layer. Research shows that the mode extinction ratio of traditional devices is generally less than 10dB, which is difficult to meet the extinction ratio requirement of more than 25dB for high-speed optical communications.

[0005] In summary, in order to better perform mode-selective absorption and optimize the absorption efficiency of silicon-based photonic devices, a new mode-selective control technology is urgently needed to achieve coupled absorption of specific modes and select coupling efficiency. Summary of the Invention

[0006] To address the aforementioned technical issues, the present invention provides a dynamically adjustable optical mode selection control device. Using a MEMS cantilever beam, the spacing between the stacked absorption layer and the silicon waveguide layer is adjusted to achieve selective absorption of specific modes and select the desired coupling efficiency for mode transmission. This improves mode-selective absorption in silicon-based photonic devices and helps optimize absorption efficiency. The present invention also provides a dynamically adjustable optical mode selection control method, enabling automated control of the device and improving control accuracy.

[0007] In one aspect, the present application provides a dynamically adjustable optical mode selection control device, comprising a silicon waveguide layer and a PWB layer. Each coupling region of the silicon waveguide layer and the PWB layer is provided with a stacked absorption layer, the stacked absorption layer being deposited within the PWB, with a spacing between the stacked absorption layer and the silicon waveguide layer. A MEMS cantilever beam is connected to the top of the stacked absorption layer, the MEMS cantilever beam being disposed outside the PWB layer and used to adjust the spacing between the stacked absorption layer and the silicon waveguide layer. The spacing between the stacked absorption layer and the silicon waveguide layer is maintained within a set critical range. The critical range is defined as follows: when the spacing between the stacked absorption layer and the silicon waveguide layer is less than a minimum value of the critical range, all modes within the silicon waveguide layer are completely absorbed by the stacked absorption layer; when the spacing between the stacked absorption layer and the silicon waveguide layer is greater than a maximum value of the critical range, the target mode cannot be effectively absorbed; and when the spacing between the stacked absorption layer and the silicon waveguide layer is within the critical range, the spacing can be adjusted to achieve absorption of the target mode and adjust the absorption efficiency.

[0008] In one embodiment, the silicon waveguide layer is a silicon-based SWG waveguide.

[0009] In one embodiment, the stacked absorption layers are sequentially a TiN layer, a HfO2 layer, an ITO layer, and a Cu layer from bottom to top.

[0010] In one embodiment, a low refractive index cladding layer is provided outside the PWB layer.

[0011] In one embodiment, a piezoelectric film is provided on the top of each MEMS cantilever beam.

[0012] On the other hand, the present application provides a method for dynamically adjusting light mode selection control device, comprising:

[0013] Obtaining sample data of parameters affecting optical mode coupling and corresponding coupling modes and mode coupling efficiencies to generate a data set;

[0014] Normalize the data in the dataset;

[0015] Constructing a light mode selection control model based on a fully connected feedforward neural network;

[0016] The normalized dataset is used to train the light mode selection control model;

[0017] Inputting actual parameters and target modes and target coupling efficiencies of the modes into the trained optical mode selection control model to generate a target spacing between the stacked absorption layer and the silicon waveguide layer;

[0018] The height position of the stacked absorption layer is adjusted using a MEMS cantilever beam according to the target spacing.

[0019] In one embodiment, sample data of parameters affecting optical mode coupling and corresponding coupling modes and coupling efficiencies of the modes are obtained to generate a data set, including: determining the parameters affecting optical mode coupling, and in the process of optical mode selection control test, controlling one of the parameters to change each time the test is conducted, while the other parameters remain unchanged, and when the actual mode coupling efficiency is measured to meet the set conditions, recording the corresponding parameter data and the mode and mode coupling efficiency, arranging them in a preset order as a vector data, until all parameters are traversed and completed to form a data set consisting of multiple vector data.

[0020] In one embodiment, parameters affecting optical mode coupling include geometric parameters, environmental parameters, and the refractive index of each material. The geometric parameters include the width and length of the silicon waveguide layer, the horizontal position of the stacked absorption layer, the spacing between the stacked absorption layer and the silicon waveguide layer, and the length and thickness of each layer of the stacked absorption layer. The environmental parameters include the ambient temperature, humidity, and pressure.

[0021] In one embodiment, training a light mode selection control model using the normalized data set includes:

[0022] Divide the normalized dataset into a training set and a validation set;

[0023] The spacing between the stacked absorption layer and the silicon waveguide layer is used as the output, and all other parameters and coupling modes and their coupling efficiencies are used as input. The optical mode selection control model is trained using the training set, and the number of hidden layers and nodes in the optimization model is updated using the control variable method. Model training is completed when the loss function used to evaluate the error between the model's predicted value and the true value for the test set converges.

[0024] In one embodiment, after adjusting the height position of the stacked absorption layer using the MEMS cantilever beam according to the target spacing, the method further includes: detecting the coupled mode and the coupling efficiency of the mode after adjustment, and adjusting the spacing between the stacked absorption layer and the silicon waveguide layer using the piezoelectric film if there is an error.

[0025] The present application provides a dynamic dimming mode selection control device and method, which has the following beneficial effects:

[0026] 1. The MEMS cantilever beam is used to dynamically adjust the spacing between the stacked absorption layer and the silicon waveguide layer according to the actual mode selection and coupling efficiency requirements. By adjusting the spacing, selective absorption of specific modes and the required coupling efficiency can be achieved for mode transmission, thereby better performing mode selective absorption of silicon-based photonic devices and helping to optimize absorption efficiency.

[0027] 2. When using the MEMS cantilever beam to adjust the distance between the stacked absorption layer and the silicon waveguide layer, the piezoelectric effect is used to cause the piezoelectric film to undergo slight deformation, thereby driving the MEMS cantilever beam to move slightly. The piezoelectric film is used as a secondary adjustment to coordinate control with the MEMS cantilever beam, which helps to offset the mechanical hysteresis caused by the MEMS cantilever beam regulation.

[0028] 3. Select silicon-based SWG waveguides and adopt a cascaded stepped structure to form a tapered silicon-based SWG waveguide structure, which helps prevent light field leakage, reduce structural transmission loss, and reduce structural size.

[0029] 4. A low-refractive-index cladding is provided outside the PWB layer, which can not only protect the PWB layer but also form a refractive-index difference with the buried oxide layer, thereby generating structural asymmetry in the PWB layer and inducing a mode hybridization effect, thereby converting the mode in the PWB layer into a mode that matches the few-mode fiber, thereby improving the subsequent coupling efficiency with the few-mode fiber.

[0030] 5. A fully connected feedback neural network is introduced to establish a target light mode selection control model. By taking the spacing as output and other parameters, modes, and mode coupling efficiencies as input, the spacing can be automatically adjusted, which helps to improve the accuracy of spacing control.

[0031] 6. Taking temperature as one of the input parameters so that the actual temperature is taken into account when adjusting the spacing helps to offset the refractive index drift caused by temperature changes, thereby improving the accuracy of mode selection. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 A schematic structural diagram of a dynamic dimming mode selection control device provided by an example of the present invention;

[0033] Figure 2 A schematic structural diagram illustrating the positional relationship between the stacked absorption layer and the MEMS cantilever beam provided in an embodiment of the present invention;

[0034] Figure 3 A schematic structural diagram of a silicon-based SWG waveguide with a step-wise reduction in width provided by an embodiment of the present invention;

[0035] Figure 4 A schematic diagram of a stacked absorption layer and its connection to a power supply provided by an embodiment of the present invention;

[0036] Figure 5 Schematic diagram of the connection of the MEMS cantilever beam and piezoelectric film to the power supply provided in the embodiment of the present invention;

[0037] Figure 6 This is a diagram of the neural network architecture of the present invention.

[0038] Explanation of the numbers in the figure: 1. Silicon substrate; 2. Buried oxide layer; 3. PWB layer; 4. Stacked absorption layer; 401. TiN layer; 402. HfO2 layer; 403. ITO layer; 404. Cu layer; 5. Silicon waveguide layer; 6. MEMS cantilever beam; 7. Piezoelectric film. DETAILED DESCRIPTION

[0039] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0040] In this application, the full name of PWB is Photonic Wire Bonding; the full name of MEMS is Micro-Electro-Mechanical System.

[0041] Reference Figures 1 to 5 The present application provides a mode selection control device, including a silicon waveguide layer 5 and a PWB layer 3. Each coupling region between the silicon waveguide layer 5 and the PWB layer 3 is provided with a stacked absorption layer 4. A MEMS cantilever beam 6 is connected to the top of the stacked absorption layer 4. The MEMS cantilever beam 6 is arranged outside the PWB layer 3 and is used to adjust the distance between the stacked absorption layer 4 and the silicon waveguide layer 5; wherein the distance between the stacked absorption layer 4 and the silicon waveguide layer 5 is maintained within a set critical range.

[0042] Specifically, a buried oxide layer 2 is disposed on the upper surface of a silicon substrate 1. A silicon waveguide layer 5, a PWB layer 3, and a stacked absorption layer 4 are all disposed on the upper surface of the buried oxide layer 2. Before constructing the mode selection control device, geometric parameters of the stacked absorption layer 4 are generated based on the width of the silicon waveguide layer 5, the refractive indices of the materials of the stacked absorption layer 4, and theoretical environmental factors. The mode selection control device is then constructed based on these geometric parameters, ensuring that the stacked absorption layer 4 has a predetermined thickness and length, and a predetermined distance between the stacked absorption layer 4 and the silicon waveguide layer 5, such that the stacked absorption layer 4 is at a predetermined height relative to the silicon waveguide layer 5. Furthermore, the height of the silicon waveguide layer 5 is within a predetermined critical range, thereby achieving absorption of a specific mode.

[0043] In one embodiment, the thickness is set to a range of 50 nm to 200 nm; the length is set to a range of 10 nm to 50 nm; and the critical range is set to a range of 100 nm to 500 nm.

[0044] When absorbing a specific mode, the MEMS cantilever beam 6 needs to be used to dynamically adjust the distance between the stacked absorption layer 4 and the silicon waveguide layer 5 according to actual conditions, thereby helping to offset the refractive index drift caused by temperature changes on the basis of absorbing the specific mode.

[0045] In one embodiment, the vertical displacement range of the MEMS cantilever beam 6 is 0 to 200 nm.

[0046] In one embodiment, the present application can achieve optimal coupling efficiency for a specific mode by adjusting the spacing between the stacked absorption layer 4 and the silicon waveguide layer 5 , and can also select a specific coupling efficiency for mode transmission.

[0047] In one embodiment, the critical range defined in this application is as follows: when the spacing between the stacked absorption layer 4 and the silicon waveguide layer 5 is less than the minimum value of the critical range, all modes in the silicon waveguide layer 5 are completely absorbed by the stacked absorption layer 4; when the spacing between the stacked absorption layer 4 and the silicon waveguide layer 5 is greater than the maximum value of the critical range, the target mode cannot be effectively absorbed; when the spacing between the stacked absorption layer 4 and the silicon waveguide layer 5 is within the critical range, the absorption of the target mode and the absorption efficiency can be adjusted by adjusting the spacing.

[0048] By adjusting the MEMS cantilever beam 6 , the absorption rate of the non-target mode in the silicon waveguide is ≤ 0.05, and the absorption rate of the mode in the PWB layer 3 is < 0.05.

[0049] In one embodiment, the silicon waveguide layer 5 is prepared by the following steps:

[0050] In step 1, a 5μm-thick single-crystal silicon wafer is used as the substrate material. Surface particles and organic residues are removed through ultrasonic cleaning and chemical treatment. A 2μm-thick buried oxide layer (BOX) is formed on the silicon substrate using a plasma-enhanced chemical vapor deposition (PECVD) process, providing bottom insulation isolation for optical signal transmission.

[0051] Step 2: Deposit a 220nm thick single-crystal silicon thin film on the BOX layer using PECVD to serve as the waveguide core layer. A spin coating process (3000-5000 rpm) is used to evenly coat the silicon surface with a photoresist layer, ensuring a film thickness uniformity error of ≤±5%.

[0052] Step 3: A 1:1 scale optical path mask is fabricated based on the pre-set optical path design. Contact exposure is performed using a UV lithography system. Light of a specific wavelength passes through the mask pattern, triggering a photochemical reaction in the photoresist and accurately replicating the waveguide pattern onto the silicon surface.

[0053] Step 4: The exposed substrate is sprayed with an alkaline developer to remove uncured photoresist. A closed-loop temperature control system maintains the development process within the range of 21-23°C ± 0.5°C. After development, an optical microscope is used to inspect the pattern integrity and line width accuracy.

[0054] In step 5, reactive ion beam etching (RIE) technology is applied to selectively remove silicon material not protected by the photoresist through the synergistic effect of physical ion bombardment and chemically active free radicals, ultimately forming a submicron high-precision waveguide structure with a sidewall verticality of ≥85°.

[0055] In one embodiment, the PWB layer 3 is prepared by the following steps:

[0056] Step 1: First, clean the chip surface thoroughly with acetone and alcohol, and then deposit photoresist on the buried oxide layer 2. The photoresist is a TPP acrylate polymer material doped with a certain proportion of high-refractive-index inorganic nanoparticles of TiO2.

[0057] Step 2: Use femtosecond laser as the excitation light source, set the wavelength of the femtosecond laser to 532 nm, the repetition frequency to 50 MHz, and the pulse width to 55 fs.

[0058] In step 3, the femtosecond laser beam passes through a beam expander and is focused by an objective lens into the acrylate polymer hybrid material. The laser is focused by a lens to print a PWB layer 3 on top of the silicon waveguide layer 5.

[0059] In one embodiment, the MEMS layer is prepared by the following steps:

[0060] Step 1: Spin-coat photoresist (1.5 μm thick) onto silicon substrate 1. Use mask exposure to define the cantilever profile (200-500 μm long, 50-150 μm wide). Deep silicon etching (5 μm / min) is performed using a Bosch process to form a hollow cantilever structure with sidewall verticality >88°.

[0061] Step 2: depositing a scandium-doped aluminum nitride (Sc-AlN) piezoelectric film 7 on the surface of the cantilever beam by radio frequency magnetron sputtering.

[0062] In step 3, the sacrificial oxide layer (thickness 2 μm) was etched using buffered hydrofluoric acid (BHF) to release the movable structure of the cantilever beam, and a bias voltage of 0-50 V was applied to verify the displacement linearity (0-200 nm, nonlinear error <3%).

[0063] In one embodiment, the silicon waveguide layer 5 is a silicon-based SWG waveguide.

[0064] The SWG waveguide has a large evanescent field and coupling intensity, which can greatly shorten the structural length required for mode evolution between the upper and lower waveguide layers, thereby reducing the structural transmission loss.

[0065] In one embodiment, referring to Figure 3 The silicon waveguide layer 5 has a geometric structure with a decreasing width, and the decreasing step size matches the effective refractive index difference of the target mode.

[0066] The silicon-based SWG waveguide is a tapered waveguide with a stepped width. It is used to generate mode hybridization and couple the higher-order modes in the multimode chip to PWB layer 3. The ideal width at the end of the silicon-based SWG waveguide is an infinitely thin needle tip, but in practice, due to limitations in machining accuracy, the tip width is ultimately controlled within the range of 60-120nm.

[0067] In this embodiment, a silicon-based SWG waveguide is selected, and a tapered silicon-based SWG waveguide structure is formed by adopting a cascaded stepped structure, which helps to prevent light field leakage, reduce structural transmission loss, and reduce structural size.

[0068] In one embodiment, referring to Figure 2 and Figure 4 The stacked absorption layer 4 is composed of a TiN layer 401, a HfO2 layer 402, an ITO layer 403 and a Cu layer 404 from bottom to top. The vertical distance between each layer is adjustable, and each layer is sequentially deposited and embedded in the PWB layer 3.

[0069] Cu layer 404 and TiN layer 401 serve as electrodes, HfO2 layer 402 serves as a high-k gate dielectric, and ITO layer 403 serves as a semiconductor layer. The ITO layer 403 can change its electron concentration and thus its permittivity when an electric field is applied. The HfO2 layer 402 enhances the ability to locally control the electric field in the ITO layer 403, improving the thermal stability of the structure.

[0070] It can be understood that when the permittivity is negative, the ITO layer 403 exhibits metallic properties; and when the permittivity is positive, the ITO layer 403 exhibits dielectric properties.

[0071] Each absorption layer is independently connected to an external power source via the Cu layer 404 and the TiN layer 401. By controlling the carrier concentration, the ITO layer 403 is tuned to have either metallic or dielectric properties, achieving a switching effect. When the ITO layer 403 is dielectric, the device is in the ON state, the refractive index of the stacked absorption layer 4 decreases, and absorption is reduced, allowing the mode to pass. When the ITO layer 403 exhibits metallic properties, the device is in the OFF state, the refractive index of the stacked absorption layer 4 increases, and absorption is enhanced, blocking mode coupling.

[0072] In this embodiment, the design standard of the device is to achieve optimal coupling efficiency transmission or specific coupling efficiency transmission when the device is in the ON state, and to achieve less than 5% coupling efficiency transmission when the device is in the OFF state.

[0073] In one embodiment, the refractive index of the PWB layer 3 is in the range of 1.5 to 1.59, preferably 1.57. The refractive index of the silicon waveguide layer 5 is 3.47.

[0074] The refractive index difference between the PWB layer 3 and the silicon waveguide layer 5 is relatively large, and the mode hybridization effect is induced by the structural asymmetry, thereby achieving coupling of high-order modes to the PWB layer 3.

[0075] In one embodiment, a low-refractive-index cladding layer is provided outside the PWB layer 3 , and the refractive index of the low-refractive-index cladding layer is 1.3.

[0076] The low-refractive-index cladding can not only protect the PWB layer 3 but also form a refractive index difference with the buried oxide layer 2, thereby generating structural asymmetry in the PWB layer 3 and inducing a mode hybridization effect, thereby converting the mode in the PWB layer 3 into a mode that matches the few-mode fiber, thereby improving the subsequent coupling efficiency with the few-mode fiber.

[0077] In one embodiment, referring to Figure 5 A piezoelectric film 7 is provided on the top of each MEMS cantilever beam 6 .

[0078] When the MEMS cantilever beam 6 is used to adjust the distance between the stacked absorption layer 4 and the silicon waveguide layer 5, the piezoelectric effect is used to cause the piezoelectric film 7 to undergo a slight deformation, thereby driving the MEMS cantilever beam 6 to undergo a slight movement, thereby offsetting the mechanical hysteresis caused by the regulation of the MEMS cantilever beam 6 and the refractive index drift caused by temperature changes, thereby achieving coordinated control of the piezoelectric film 7 and the MEMS cantilever beam 6, which helps to offset the temperature drift and mechanical hysteresis of the device.

[0079] In one embodiment, the MEMS cantilever beam 6 can achieve vertical displacement within a set range through electrostatic driving.

[0080] In one embodiment, the material of the piezoelectric film 7 is AlN.

[0081] The piezoelectric film 7 made of AlN material can generate a deformation of ±1 nm by using the piezoelectric effect, thereby making slight adjustments to the MEMS cantilever beam 6 and helping to offset temperature drift and mechanical hysteresis.

[0082] On the other hand, refer to Figure 6 , the present application provides a dynamic dimming mode selection control method, comprising:

[0083] S100 , obtaining sample data of parameters affecting optical mode coupling and corresponding coupling modes and mode coupling efficiencies, and generating a data set.

[0084] Specifically, step S100 includes: determining the parameters that affect the coupling of optical modes. During the optical mode selection control test, one of the parameters is changed in each test, and the other parameters remain unchanged. When the actual mode coupling efficiency is measured to meet the set conditions, the corresponding parameter data and the mode-mode coupling efficiency are recorded, and arranged in a preset order as a vector data until all parameters are traversed to form a data set consisting of multiple vector data.

[0085] In one embodiment, the application sets the following conditions: a threshold value of 0.05 is set for the non-target mode coupling efficiency. If it exceeds this threshold, the data set is discarded. Only structural parameter combinations that meet the threshold condition and have a target mode coupling efficiency within the range of 0-1 are recorded to generate a parameter data set. For example, if the non-target mode efficiency is 0.12 and the target mode efficiency is 0.5, it is considered invalid data; if the non-target mode efficiency is 0.03 and the target mode efficiency is 0.45, it is recorded. Data collection and storage are completed according to this rule.

[0086] In one embodiment, parameters affecting optical mode coupling include geometric parameters, environmental parameters, and the refractive index of each material. The geometric parameters include the width and length of the silicon waveguide layer 5, the horizontal position of the stacked absorption layer 4, the spacing between the stacked absorption layer 4 and the silicon waveguide layer 5, and the length and thickness of each layer of the stacked absorption layer 4. The environmental parameters include the ambient temperature, humidity, and pressure.

[0087] S200, normalizing the data in the data set.

[0088] S300, builds a light mode selection control model based on a fully connected feedforward neural network.

[0089] S400: Training a light mode selection control model using the normalized data set.

[0090] Specifically, step S400 includes: dividing the normalized data set into a training set and a validation set; using the spacing between the stacked absorption layer 4 and the silicon waveguide layer 5 as output, and all other parameters and coupling modes and mode coupling efficiencies as input, training the optical mode selection control model using the training set, and updating the number of hidden layers and nodes in the optimization model using a control variable method. Model training is completed when the loss function used to evaluate the error between the model's predicted value and the true value for the test set converges.

[0091] S500 , inputting actual parameters, target mode, and target coupling efficiency of the mode into the trained optical mode selection control model to generate a target spacing between the stacked absorption layer 4 and the silicon waveguide layer 5 .

[0092] S600 , adjusting the height position of the stacked absorption layer 4 using the MEMS cantilever beam 6 according to the target spacing.

[0093] A fully connected feedback neural network is introduced to establish a target optical mode selection control model. By using spacing as the output and other parameters, modes, and mode coupling efficiencies as inputs, automatic spacing adjustment is achieved, helping to improve spacing control accuracy. Temperature is also included as an input parameter, allowing the actual temperature to be taken into account when adjusting spacing, helping to offset refractive index drift caused by temperature changes and further improving mode selection accuracy.

[0094] In one embodiment, after adjusting the height position of the stacked absorption layer 4 using the MEMS cantilever beam 6 according to the target spacing, the method further includes: detecting the coupled mode and the coupling efficiency of the mode after adjustment, and adjusting the spacing between the stacked absorption layer 4 and the silicon waveguide layer 5 using the piezoelectric film 7 if there is an error.

[0095] When the MEMS cantilever beam 6 is used to adjust the distance between the stacked absorption layer 4 and the silicon waveguide layer 5, the piezoelectric effect is used to cause the piezoelectric film 7 to undergo a slight deformation, thereby driving the MEMS cantilever beam 6 to undergo a slight movement. The piezoelectric film 7 is used as a secondary adjustment to coordinate control with the MEMS cantilever beam 6, which helps to offset the mechanical hysteresis caused by the regulation of the MEMS cantilever beam 6.

[0096] When the device is completed and the environmental parameters are stable in the preset working range, the host computer system obtains the actual temperature, humidity and air pressure data in real time through the environmental perception module integrated with the multi-source sensor, and inputs the above data and the geometric parameters of the device as well as the mode to be coupled and the coupling efficiency into the target light mode selection control model. According to the output predicted spacing, the height of the stacked absorption layer 4 is mainly adjusted by the MEMS cantilever beam 6, thereby adjusting the spacing between the stacked absorption layer 4 and the silicon waveguide layer 5, and then performing mode selective absorption and achieving the set efficiency absorption. At the same time, the piezoelectric film 7 integrated with the MEMS cantilever beam 6 is used to construct a stress feedback network, and dynamic compensation is performed through the piezoelectric effect to form a master-slave collaborative control architecture. This architecture realizes the dynamic mapping of environmental parameters and driving voltage through fuzzy logic algorithm, and suppresses the refractive index drift caused by temperature fluctuations, and finally realizes closed-loop adaptive displacement precision control within the tolerance range of the standard MEMS manufacturing process.

[0097] In this application, after obtaining the data set through parameter traversal, the parameters in the data set can provide guidance for the structural parameters of the manufacturing device, so that relevant personnel can select the preferred parameters in the data set to manufacture the device, thereby realizing static adjustment of the geometric parameters of the device, further helping to improve the accuracy of mode selection.

[0098] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0099] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present application. It should be noted that a person of ordinary skill in the art may make various modifications and improvements without departing from the spirit of the present application, and these modifications and improvements fall within the scope of protection of the present application. Therefore, the scope of protection of the present application shall be determined by the appended claims.

Claims

1. A dynamically adjustable light mode selection control device, comprising a silicon waveguide layer and a PWB layer, wherein each coupling region of the silicon waveguide layer and the PWB layer is provided with a stacked absorption layer, and the stacked absorption layer is deposited inside the PWB, characterized in that: A distance exists between the stacked absorption layer and the silicon waveguide layer. A MEMS cantilever beam is connected to the top of the stacked absorption layer. The MEMS cantilever beam is arranged outside the PWB layer and is used to adjust the distance between the stacked absorption layer and the silicon waveguide layer. The distance between the stacked absorption layer and the silicon waveguide layer is maintained within a set critical range. The critical range is defined as follows: when the distance between the stacked absorption layer and the silicon waveguide layer is less than the minimum value of the critical range, all modes in the silicon waveguide layer are completely absorbed by the stacked absorption layer; when the distance between the stacked absorption layer and the silicon waveguide layer is greater than the maximum value of the critical range, the target mode cannot be effectively absorbed; and when the distance between the stacked absorption layer and the silicon waveguide layer is within the critical range, the absorption of the target mode and the absorption efficiency can be adjusted by adjusting the distance value.

2. The device according to claim 1, characterized in that The silicon waveguide layer is a silicon-based SWG waveguide.

3. The device according to claim 1, characterized in that The stacked absorption layers are TiN layer, HfO2 layer, ITO layer and Cu layer from bottom to top.

4. The device according to claim 1, characterized in that A low refractive index cladding layer is provided outside the PWB layer.

5. The device according to claim 1, characterized in that Each MEMS cantilever has a piezoelectric film on top.

6. A method for dynamically adjusting light mode selection and control, applied to the device according to any one of claims 1 to 5, characterized in that: include: Obtaining sample data of parameters affecting optical mode coupling and corresponding coupling modes and mode coupling efficiencies to generate a data set; Normalize the data in the dataset; Constructing a light mode selection control model based on a fully connected feedforward neural network; The normalized dataset is used to train the light mode selection control model; inputting actual parameters and target modes and target coupling efficiencies of the modes into the trained optical mode selection control model to generate a target spacing between the stacked absorption layer and the silicon waveguide layer; The height position of the stacked absorption layer is adjusted using a MEMS cantilever beam according to the target spacing.

7. The method according to claim 6, characterized in that The method obtains sample data of parameters affecting optical mode coupling and corresponding coupling modes and mode coupling efficiencies to generate a data set, including: determining the parameters affecting optical mode coupling; during the optical mode selection control test, controlling one of the parameters to change each time the test is conducted, while the other parameters remain unchanged; when the actual mode coupling efficiency is measured to meet the set conditions, recording the corresponding parameter data and mode and mode coupling efficiency, arranging them in a preset order as a vector data, and forming a data set consisting of multiple vector data after all parameters are traversed.

8. The method according to claim 6, characterized in that Parameters that affect optical mode coupling include geometric parameters, environmental parameters, and the refractive index of each material. Geometric parameters include the width and length of the silicon waveguide layer, the horizontal position of the stacked absorption layer, the spacing between the stacked absorption layer and the silicon waveguide layer, and the length and thickness of each layer of the stacked absorption layer. Environmental parameters include the ambient temperature, humidity, and pressure.

9. The method according to claim 8, characterized in that The normalized dataset is used to train the light mode selection control model, including: Divide the normalized dataset into a training set and a validation set; The spacing between the stacked absorption layer and the silicon waveguide layer is used as the output, and all other parameters and coupling modes and their coupling efficiencies are used as input. The optical mode selection control model is trained using the training set, and the number of hidden layers and nodes in the optimization model is updated using the control variable method. Model training is completed when the loss function used to evaluate the error between the model's predicted value and the true value for the test set converges.

10. The method according to claim 6, characterized in that After adjusting the height position of the stacked absorption layer using the MEMS cantilever beam according to the target spacing, the method also includes: detecting the coupled mode and the coupling efficiency of the mode after adjustment, and adjusting the spacing between the stacked absorption layer and the silicon waveguide layer using the piezoelectric film when there is an error.

Images