Polarization locking device, method, optical communication equipment and optical chip

Abstract

The present invention provides a polarization locking device, method, optical communication equipment and optical chip. The polarization locking device includes a polarization beam splitter for orthogonally decomposing mixed light to obtain a first beam of light and a second beam of light; the first beam of light includes N first polarized lights of different wavelengths, and the second beam of light includes N second polarized lights of different wavelengths; the polarization adjustment structure group includes at least one polarization adjustment structure, and the polarization adjustment structure includes N micro-ring resonators and N phase shifters; each micro-ring resonator is used to select the first polarized light of the corresponding wavelength for gating, and output the gated light to the corresponding phase shifter for phase modulation; the phase shifter outputs the phase-modulated light to a multiplexer for mixing, and outputs the mixed light to a coupler; the coupler couples the N first polarized lights after phase modulation and the N second polarized lights of different wavelengths to obtain a modulated mixed light. The present invention can realize the locking of N input lights of different wavelengths to specific stable polarization states respectively.

Description

Technical Field

[0001] The present invention relates to the technical field of wavelength division multiplexing coherent optical communication networks, and particularly relates to a polarization locking device, method, optical communication device and optical chip. Background Art

[0002] During the propagation of light in an optical fiber, the polarization state is easily affected by external environmental factors. In fields such as quantum communication, coherent optical communication, and medical imaging, it is crucial to obtain a single and stable polarized light. The process of converting an arbitrarily changing polarization state into a stable polarization state is called polarization state locking, which is usually achieved by an adaptive polarization control device. However, the current adaptive polarization controller aims to achieve a stable output of a single polarization state, and it is difficult to achieve a stable output of an arbitrary polarization state of incident light with different wavelengths.

[0003] Due to the strong wavelength dependence of the fiber birefringence effect, the transmission matrices corresponding to lights with different wavelengths passing through the optical fiber are different, resulting in different polarization rotations; moreover, there is an accumulation effect for the polarization effect differences caused by wavelengths. When two beams of light with very small wavelength differences pass through a long optical fiber path, the polarization changes generated may be very different. When a polarization controller is applied to a multi-wavelength system, such as a wavelength division multiplexing optical communication system, it is difficult to simultaneously perform polarization tracking and locking for polarized lights with different wavelengths. If a set of polarization control systems is designed separately for the polarized lights of each wavelength, this method not only occupies a large volume but also causes waste of resources. Therefore, how to achieve a stable output of an arbitrary polarization state of incident lights with multiple different wavelengths is a technical problem that urgently needs to be solved at present. Summary of the Invention

[0004] The purpose of the present invention is to provide a polarization locking device, method, optical communication device and optical chip that can lock N input lights with different wavelengths to specific polarization states respectively.

[0005] The present invention provides a polarization locking device, including a polarization beam splitter and a polarization adjustment structure group;

[0006] The polarization beam splitter is used to orthogonally decompose the mixed light into a first beam of light and a second beam of light, and output the first beam of light to the polarization adjustment structure group; the mixed light includes N lights with different wavelengths, the first beam of light includes N first polarized lights with different wavelengths, the second beam of light includes N second polarized lights with different wavelengths, the polarization directions of the first polarized lights and the second polarized lights are perpendicular to each other, and perpendicular to the propagation direction of the mixed light; N is an integer greater than 1;

[0007] The polarization adjustment structure group includes at least one polarization adjustment structure, and the polarization adjustment structure includes a multiplexer, a coupler, N micro-ring resonators and N phase shifters; the N micro-ring resonators and the N phase shifters correspond one by one; each of the micro-ring resonators is configured to select and gate the first polarized light corresponding to the corresponding wavelength from the first beam of light, and output the gated light to the corresponding phase shifter for phase adjustment;

[0008] The phase shifter outputs the phase-adjusted light to the multiplexer;

[0009] The multiplexer mixes the N phase-adjusted first polarized lights and outputs them to the coupler;

[0010] The coupler couples the N phase-adjusted first polarized lights and N second polarized lights with different wavelengths to obtain the modulated mixed light.

[0011] Further, it further includes a mixer and a demultiplexer group;

[0012] One end of the mixer is connected to the polarization adjustment structure group, and is configured to convert a part of the mixed light output by the polarization adjustment structure group to obtain optical signals of six different output channels;

[0013] The other end of the mixer is connected to the demultiplexer group. The demultiplexer group includes six first demultiplexers, and the six first demultiplexers are respectively configured to decouple the optical signals of the six different output channels to obtain optical signals of six different output channels corresponding to each wavelength.

[0014] Further, it further includes a balanced photodetector and a drive control circuit;

[0015] The balanced photodetector is configured to obtain 3 differential optical signals based on the six optical signals output by the demultiplexer group to obtain 3 Stokes parameters; the six optical signals are divided into 3 pairs of optical signals, and each pair of optical signals performs a difference operation to obtain the corresponding difference signal;

[0016] Determine the current polarization state corresponding to each wavelength based on the 3 Stokes parameters;

[0017] The drive control circuit is configured to obtain a feedback value based on the current polarization state to update the drive voltage signal of the phase shifter corresponding to each wavelength; the feedback value is the square of the distance between the current polarization state and the preset target polarization state.

[0018] Further, the polarization adjustment structure group includes a first polarization adjustment structure, a second polarization adjustment structure, a third polarization adjustment structure, and a fourth polarization adjustment structure. The first polarization adjustment structure, the second polarization adjustment structure, the third polarization adjustment structure, and the fourth polarization adjustment structure are connected in series in sequence. The first polarization adjustment structure, the second polarization adjustment structure, the third polarization adjustment structure, and the fourth polarization adjustment structure are respectively the polarization adjustment structures. The phase shifter in the polarization adjustment structure is used to increase or decrease the phase of the input light to obtain a target phase;

[0019] The drive control circuit is used to obtain a first initial phase, a second initial phase, a third initial phase, a fourth initial phase, and a boundary factor, use the first initial phase, the second initial phase, the third initial phase, the fourth initial phase, and the boundary factor as inputs of a feedback function, and perform p iterations using a four-dimensional gradient descent algorithm to output a target feedback value, and update the drive voltage signals of the phase shifters in the first polarization adjustment structure, the second polarization adjustment structure, the third polarization adjustment structure, and the fourth polarization adjustment structure based on the target feedback value;

[0020] Among them, the first initial phase is the phase when the selected light enters the phase shifter in the first polarization adjustment structure, the second initial phase is the phase when the selected light enters the phase shifter in the second polarization adjustment structure, the third initial phase is the phase when the selected light enters the phase shifter in the third polarization adjustment structure, and the fourth initial phase is the phase when the selected light enters the phase shifter in the fourth polarization adjustment structure;

[0021] The boundary factor is determined based on the phase change boundary, the step size of the four-dimensional gradient descent algorithm, and the phase of any one of the polarization adjustment structures in the polarization adjustment structure group;

[0022] The target feedback value is the minimum value among the 2 4 function values obtained by the p-th iteration calculation; p is a positive integer; the feedback function is the square of the distance between the current polarization state and the target polarization state. The current polarization state is the Stokes parameters in the current period feedback by the balanced photodetector, and the target polarization state is the preset output polarization state; in each iteration, calculate the corresponding function value based on 2 4 reference polarization states. The reference polarization state is the polarization state after increasing or decreasing the step size of the light with the first initial phase, the second initial phase, the third initial phase, and the fourth initial phase.

[0023] Furthermore, the calculation formula of the boundary factor is

[0024] * ;

[0025] Among them, is the boundary factor. When the phase change boundary is set to , is the maximum phase shift, is the minimum phase shift, is the step size of the four-dimensional gradient descent algorithm, which is equal to the phase shift, is the phase shift of any polarization adjustment structure in the polarization adjustment structure group at the p-th iteration, where p represents the number of iterations and m represents any polarization adjustment structure in the polarization adjustment structure group.

[0026] Furthermore, the phase shifter is a lithium niobate electro-optic phase shifter or a silicon-based thermo-optic phase shifter.

[0027] Furthermore, it further includes a polarization beam combiner and a second demultiplexer;

[0028] One end of the polarization beam combiner is connected to the polarization adjustment structure group, and is used for orthogonally coupling the modulated mixed light output by the polarization adjustment structure group into polarized light;

[0029] The other end of the polarization beam combiner is connected to the second demultiplexer, and the second demultiplexer is used for decoupling the polarized light output by the polarization beam combiner to obtain stable polarized light of different wavelengths.

[0030] The present invention also provides a polarization locking method, which is applied to a polarization locking device. The method includes:

[0031] Orthogonally decomposing the mixed light to obtain a first light beam and a second light beam; the mixed light includes N lights of different wavelengths, the first light beam includes N first polarized lights of different wavelengths, the second light beam includes N second polarized lights of different wavelengths, the polarization directions of the first polarized lights and the polarization directions of the second polarized lights are perpendicular to each other, and perpendicular to the propagation direction of the mixed light; N is an integer greater than 1;

[0032] Selecting the first polarized light corresponding to the wavelength from the first light beam for gating, and performing phase modulation on the gated light to obtain N phase-modulated first polarized lights;

[0033] Mixing the N phase-modulated first polarized lights;

[0034] Coupling the mixed N phase-modulated first polarized lights and N second polarized lights of different wavelengths to obtain modulated mixed light.

[0035] Further, the polarization adjustment structure group includes a first polarization adjustment structure, a second polarization adjustment structure, a third polarization adjustment structure, and a fourth polarization adjustment structure. The first polarization adjustment structure, the second polarization adjustment structure, the third polarization adjustment structure, and the fourth polarization adjustment structure are connected in series in sequence. The first polarization adjustment structure, the second polarization adjustment structure, the third polarization adjustment structure, and the fourth polarization adjustment structure are respectively the polarization adjustment structures. The phase shifter in the polarization adjustment structure is used to increase or decrease the phase of the input light to obtain a target phase. The polarization locking device further includes a balanced photodetector and a drive control circuit. The method further includes:

[0036] Obtain a first initial phase, a second initial phase, a third initial phase, and a fourth initial phase. The first initial phase is the phase when the selected light enters the phase shifter in the first polarization adjustment structure. The second initial phase is the phase when the selected light enters the phase shifter in the second polarization adjustment structure. The third initial phase is the phase when the selected light enters the phase shifter in the third polarization adjustment structure. The fourth initial phase is the phase when the selected light enters the phase shifter in the fourth polarization adjustment structure.

[0037] Obtain a boundary factor, where the boundary factor is determined based on a phase change boundary, a step size of a four-dimensional gradient descent algorithm, and a phase of any one of the polarization adjustment structures in the polarization adjustment structure group.

[0038] Take the first initial phase, the second initial phase, the third initial phase, the fourth initial phase, and the boundary factor as inputs of a feedback function, and perform p iterations using a four-dimensional gradient descent algorithm to output a target feedback value. The target feedback value is the minimum value among the 2 function values calculated in the p-th iteration; p is a positive integer; the feedback function is the square of the distance between the current polarization state and the target polarization state. The current polarization state is the Stokes parameters in the current period feedback by the balanced photodetector, and the target polarization state is a preset output polarization state; in each iteration, calculate the corresponding function value based on 2 reference polarization states. The reference polarization state is the polarization state after increasing or decreasing the step size of the light with the first initial phase, the second initial phase, the third initial phase, and the fourth initial phase. 4 Output the target feedback value to the drive control circuit to update the drive voltage signals of the phase shifters in the first polarization adjustment structure, the second polarization adjustment structure, the third polarization adjustment structure, and the fourth polarization adjustment structure. 4 The present invention also provides an optical communication device including a polarization locking device.

[0039] Output the target feedback value to the drive control circuit to update the drive voltage signals of the phase shifters in the first polarization adjustment structure, the second polarization adjustment structure, the third polarization adjustment structure, and the fourth polarization adjustment structure.

[0040] The present invention also provides an optical communication device including a polarization locking device.

[0041] The present invention also provides an optical chip, including a polarization locking device.

[0042] Technical effect: The present invention can achieve simultaneous polarization locking of input lights with multiple different wavelengths. The input lights with multiple different wavelengths are input simultaneously, and after locking, polarized lights with stable polarization states of each wavelength are output.

[0043] The present invention can achieve the continuity of control while locking the polarization state at any position, avoid the reset operation while realizing continuous polarization tracking, and has strong robustness. High-speed polarization tracking and locking can be achieved only by improving the device performance and reducing the loop delay. Description of the Drawings

[0044] Figure 1 is a schematic structural diagram of a polarization locking device in an embodiment;

[0045] Figure 2 is a schematic structural diagram of a polarization beam splitter in an embodiment;

[0046] Figure 3 is a schematic structural diagram of a dual-bus microring resonator in an embodiment;

[0047] Figure 4 is a resonant power distribution diagram of a dual-bus microring resonator in an embodiment;

[0048] Figure 5 is a detuning power distribution diagram of a dual-bus microring resonator in an embodiment;

[0049] Figure 6 is a transmission response diagram of a dual-bus microring resonator in an embodiment;

[0050] Figure 7 is a schematic structural diagram of a 90° mixer in an embodiment;

[0051] Figure 8 is a schematic structural diagram of an arrayed waveguide grating in an embodiment;

[0052] Figure 9 is a schematic diagram of a polarization locking method in another embodiment;

[0053] Figure 10 is a logic block diagram of a four-dimensional gradient descent algorithm in another embodiment;

[0054] Figure 11 is a schematic diagram of the rotation of the polarization state around the equator of the Poincaré sphere;

[0055] Figure 12 is a schematic diagram of the output of the first polarization state after polarization locking;

[0056] Figure 13 is a phase change diagram during the polarization locking process;

[0057] Figure 14 It is a schematic diagram of the rotation of the polarization state around the 90° and 270° meridians of the Poincaré sphere;

[0058] Figure 15 It is a schematic diagram of the first polarization state output after polarization tracking;

[0059] Figure 16 It is a schematic diagram of the application scenario of the polarization locking device in a wavelength division multiplexing optical communication system. Specific embodiments

[0060] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below. Apparently, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention. Unless otherwise defined, the technical terms or scientific terms used herein shall have the ordinary meaning as understood by those of ordinary skill in the art to which the present invention belongs. The words such as "including" used herein mean that the elements or items appearing before the word cover the elements or items listed after the word and their equivalents, without excluding other elements or items.

[0061] In addition to solving the technical problems in the background art, the present invention also solves a secondary technical problem, that is, the technical problem that the continuity of polarization control makes it difficult for the device to reset. The existing algorithms introduce a high complexity to solve the problem of control continuity, find the optimal solution through matrix operations, or achieve control continuity by reversing the phase on the premise of introducing 3dB loss. However, these methods have a large amount of calculation and are difficult to implement high-speed negative feedback loop control, or need to rely on accurate phase boundary position calibration, are easily affected by environmental factors, and have poor robustness. Therefore, how to achieve continuous polarization locking using limited boundaries is also a technical problem that the present invention needs to solve.

[0062] As Figure 1 shown, the embodiments of the present invention provide a polarization locking device, including a polarization beam splitter 1 and a polarization adjustment structure group.

[0063] The polarization beam splitter 1 is used to orthogonally decompose the mixed light into a first beam of light and a second beam of light, and output the first beam of light to the polarization adjustment structure group; the mixed light includes N lights with different wavelengths, the first beam of light includes N first polarized lights with different wavelengths, the second beam of light includes N second polarized lights with different wavelengths, the polarization directions of the first polarized lights and the second polarized lights are perpendicular to each other, and perpendicular to the propagation direction of the mixed light; N is an integer greater than 1. Among them, 、 … represents light of different wavelengths; , … respectively represent the micro-ring radii of N micro-ring resonators, where n is a positive integer.

[0064] In some embodiments, as Figure 2 shown, the polarization beam splitter 1 mainly uses an asymmetric directional coupler, such that the effective refractive index of the TM mode (Transverse Magnetic mode) of one waveguide is equal to that of the TE mode (Transverse Electric Mode, referring to the wave in which the electric field component is perpendicular to the propagation direction during the propagation of the electromagnetic wave) of the other waveguide, and the two satisfy the phase matching condition, and the TM mode is converted into the TE mode in the other waveguide. The operating wavelength exceeds 80 nm, which can cover the C band and the L band, covering a relatively wide spectral range, and can support higher data transmission rates and larger communication capacities.

[0065] The polarization adjustment structure group includes at least one polarization adjustment structure 2, and the polarization adjustment structure 2 includes a multiplexer 21, a coupler 22, N micro-ring resonators 23 and N phase shifters 24; the N micro-ring resonators 23 and the N phase shifters 24 are in one-to-one correspondence; each micro-ring resonator 23 is used to select and gate the first polarized light of the corresponding wavelength from the first beam of light, and output the gated light to the corresponding phase shifter 24 for phase adjustment.

[0066] The one-to-one correspondence between the N micro-ring resonators 23 and the N phase shifters 24 realizes functions such as multi-channel gating, wavelength routing, and multiplexing. The one-to-one connection method can avoid a complex alignment process, simplify the operation steps, and improve the overall stability of the system.

[0067] In some embodiments, as Figure 3 shown, the micro-ring resonator 23 is a dual-bus micro-ring resonator structure, and its basic structure is composed of two straight waveguides and a ring waveguide. The input light is input from the input end 901 of the straight waveguide, and then is coupled into the ring waveguide through the straight waveguide. Light of a specific wavelength will resonate in the ring waveguide and finally be output from the download end 902 of one of the two straight waveguides, achieving the effect of wavelength selection; the light that cannot form resonance will be output from the output end 903 of one of the two straight waveguides, where represents the micro-ring radius.

[0068] Therefore, the first polarized light incident into the micro-ring resonator 23 needs to satisfy the formula: , where k is an integer multiple of the light wave's circulation in the micro-ring, is the wavelength of the input first polarized light, is the effective refractive index, is the micro-ring radius of the microring resonator, and n is a positive integer.

[0069] The resonance power distribution of the dual-bus microring resonator is as Figure 4 shown; the detuning power distribution of the dual-bus microring resonator is as Figure 5 shown; the transmission response of the dual-bus microring resonator is as Figure 6 shown, where 903 represents the output end and 902 represents the download end. Therefore, by controlling the micro-ring radius of the microring resonator, a wavelength resolution of 0.4 nm can be achieved. A wavelength resolution of 0.4 nm means that the microring resonator can very finely resolve different wavelengths.

[0070] In some embodiments, the microring resonator 23 is commonly used as devices such as filters, dispersion compensators, modulators, etc. in optical systems, and has the characteristics of low cost, simple structure, high integration, and low loss.

[0071] The phase shifter 24 outputs the phase-modulated light to the multiplexer 21.

[0072] The multiplexer 21 mixes N phase-modulated first polarized lights and outputs them to the coupler 22.

[0073] The coupler 22 couples N phase-modulated first polarized lights and N second polarized lights with different wavelengths to obtain the modulated mixed light.

[0074] The present invention can achieve polarization locking of N lights with different wavelengths input simultaneously through the polarization adjustment structure, and output polarized lights with stable polarization states for each wavelength after locking.

[0075] In some embodiments, the coupler 22 is a coupler with a splitting ratio of 50:50, that is, the input optical signal is evenly distributed to two output ports. Setting the splitting ratio to 50:50 can not only reduce the cross-interference between different channels, improve the purity of the signal, but also improve the signal-to-noise ratio of the system and reduce the loss.

[0076] In some embodiments, a mixer 3 and a demultiplexer group 4 are further included.

[0077] One end of the mixer 3 is connected to the polarization adjustment structure group, and is used to convert part of the mixed light output by the polarization adjustment structure group to obtain optical signals of six different output channels.

[0078] The other end of the mixer 3 is connected to the demultiplexer group 4. The demultiplexer group 4 includes six first demultiplexers, and the six first demultiplexers are respectively used to decouple the optical signals of six different output channels to obtain optical signals of six different output channels corresponding to each wavelength.

[0079] In some embodiments, the optical signals of six different output channels are respectively represented as , , , , , ; where

[0080] ;

[0081] ;

[0082] ;

[0083] ;

[0084] ;

[0085] ;

[0086] where represents the expected value in the first polarized light (i.e., the X direction), represents the conjugate expected value in the first polarized light (i.e., the X direction); represents the expected value in the second polarized light (i.e., the Y direction), represents the conjugate expected value in the second polarized light (i.e., the Y direction).

[0087] In some embodiments, the mixer 3 is a 90° mixer, and the specific structure of the 90° mixer is as shown in Figure 7 . represents the electric field of the polarization component in the X direction, represents the electric field of the polarization component in the Y direction. The expression for solving the optical intensity through the electric field is . and respectively pass through a 2:1 coupler, and then pass through the matrix of the first MMI (Multi-Mode Interference) multiplied by the transmission structure. The expression is: , and the optical intensity expressions of and are obtained; then, after phase shift and the second MMI for conjugate operation, the expression is: , and the optical intensity expressions of and are obtained; and do not pass through the MMI, where represents the imaginary unit, respectively represent the optical intensities of the optical signals of six different output channels.

[0088] In some embodiments, a balanced photodetector 5 and a drive control circuit 6 are further included;

[0089] The balanced photodetector 5 is configured to obtain 3 differential optical signals based on six optical signals output by the demultiplexer group 4, so as to obtain 3 Stokes parameters; the six optical signals are divided into 3 pairs of optical signals, and a differential operation is performed on each pair of optical signals to obtain a corresponding differential signal;

[0090] In some embodiments, the 3 Stokes parameters are respectively represented by , , , where , , , represents the intensity difference between horizontally and vertically linearly polarized light; represents the intensity difference between ±45-degree linearly polarized light; represents the intensity difference between right-handed and left-handed circularly polarized light. Regarding the Stokes parameters , , as the coordinate axes in a three-dimensional space, we can construct a geometric model, which is called a Poincaré sphere in polarization optics. Any polarization state can be represented as a point on the Poincaré sphere. For two points on the sphere, a transformation can be achieved by rotating a certain angle around the rotation axis passing through the center of the sphere. The output polarization state corresponding to each wavelength can be obtained as a feedback input to control the drive circuit.

[0091] The unit vector of the rotation axis is denoted as , and is represented in spherical coordinates as , where is the azimuth angle in spherical coordinates, is the elevation angle in spherical coordinates, represents the transpose of the object within the parentheses.

[0092] Assuming the rotation angle is , then the rotation matrix R can be represented as:

[0093] ;

[0094] where , , , , , represents the identity matrix.

[0095] The above R matrix contains 3 independent parameters , , , where the parameter , is used to describe the rotation axis, and the parameter is used to describe the rotation angle. Therefore, at least three degrees of freedom are required to describe the rotation between any polarization states. When the input and output polarization states are determined, and , , within the specified period [0, 2π], the rotation matrix R is unique. Therefore, no matter how the rotation matrix R is decomposed, ultimately three independent degrees of freedom are required for description, and when the decomposition method is determined, the parameters describing these three degrees of freedom are also unique. However, when there are four or more degrees of freedom, there are infinitely many ways to implement the polarization state transformation operation of the rotation matrix R. Therefore, the polarization controller with a four-level structure can use redundant design to avoid the reset problem caused by the control parameters falling into local optimal solutions during the polarization tracking process.

[0096] The Poincaré Sphere is a geometric model used to describe the polarization state, which was proposed by the French mathematician and physicist Henri Poincaré. The Poincaré Sphere is a unit sphere used to represent the polarization state of electromagnetic waves and their phase changes. There are the following three ways to represent the polarization state with the Poincaré Sphere: Linear polarization: located on the equator of the Poincaré Sphere, and the angle represents the polarization direction. Circular polarization: located at the north and south poles of the Poincaré Sphere, the north pole represents left-handed circular polarization, and the south pole represents right-handed circular polarization. Elliptical polarization: located in the middle region of the Poincaré Sphere, neither completely linearly polarized nor circularly polarized.

[0097] Based on three Stokes parameters, the current polarization state corresponding to each wavelength is determined; the Stokes parameters are a set of physical quantities used to describe the state of polarized light.

[0098] The drive control circuit 6 is used to obtain a feedback value based on the current polarization state to update the drive voltage signal of the phase shifter 24 corresponding to each wavelength.

[0099] In some embodiments, the polarization adjustment structure group includes a first polarization adjustment structure, a second polarization adjustment structure, a third polarization adjustment structure, and a fourth polarization adjustment structure. The first polarization adjustment structure, the second polarization adjustment structure, the third polarization adjustment structure, and the fourth polarization adjustment structure are connected in series in sequence. The first polarization adjustment structure, the second polarization adjustment structure, the third polarization adjustment structure, and the fourth polarization adjustment structure are respectively the polarization adjustment structure 2. The phase shifter 24 in the polarization adjustment structure 2 is used to increase or decrease the phase of the input light to obtain the target phase.

[0100] The driving control circuit 6 is used to obtain the first initial phase, the second initial phase, the third initial phase, the fourth initial phase, and the boundary factor, take the first initial phase, the second initial phase, the third initial phase, the fourth initial phase, and the boundary factor as the inputs of the feedback function, perform p iterations using the four-dimensional gradient descent algorithm, output the target feedback value, and update the driving voltage signals of the phase shifters 24 in the first polarization adjustment structure, the second polarization adjustment structure, the third polarization adjustment structure, and the fourth polarization adjustment structure based on the target feedback value.

[0101] Among them, the first initial phase is the phase when the selected light enters the phase shifter 24 in the first polarization adjustment structure, the second initial phase is the phase when the selected light enters the phase shifter 24 in the second polarization adjustment structure, the third initial phase is the phase when the selected light enters the phase shifter 24 in the third polarization adjustment structure, and the fourth initial phase is the phase when the selected light enters the phase shifter 24 in the fourth polarization adjustment structure.

[0102] The boundary factor is determined based on the phase change boundary, the step size of the four-dimensional gradient descent algorithm, and the phase of any one of the polarization adjustment structures in the polarization adjustment structure group.

[0103] The target feedback value is the minimum value among the 2 4 function values calculated in the p-th iteration; p is a positive integer; the feedback function is the square of the distance between the current polarization state and the target polarization state, the current polarization state is the Stokes parameters in the current period feedback by the balanced photodetector 5, and the target polarization state is the preset output polarization state; in each iteration, the corresponding function value is calculated based on 2 4 reference polarization states, and the reference polarization state is the polarization state after increasing or decreasing the step size of the light with the first initial phase, the second initial phase, the third initial phase, and the fourth initial phase.

[0104] Furthermore, the calculation formula of the boundary factor is

[0105] * ;

[0106] Among them, is the boundary factor. When the phase change boundary is set to , is the maximum phase shift, is the minimum phase shift, is the step size of the four-dimensional gradient descent algorithm, which is equal to the phase shift, is the phase shift of any one of the polarization adjustment structures in the polarization adjustment structure group at the p-th iteration, where p represents the number of iterations and m represents any one of the polarization adjustment structures in the polarization adjustment structure group.

[0107] The above four-dimensional gradient descent algorithm can achieve continuous polarization control without reset.

[0108] In some embodiments, the phase shifter 24 is a lithium niobate electro-optic phase shifter or a silicon-based thermo-optic phase shifter.

[0109] In some embodiments, a polarization beam combiner 7 and a second demultiplexer 8 are further included.

[0110] One end of the polarization beam combiner 7 is connected to the polarization adjustment structure group, and is used to orthogonally couple the modulated mixed light output by the polarization adjustment structure group into polarized light.

[0111] The other end of the polarization beam combiner 7 is connected to the second demultiplexer 8, and the second demultiplexer 8 is used to decouple the polarized light output by the polarization beam combiner 7 to obtain stable polarized light of different wavelengths.

[0112] The continuous co-directional changes of the polarization states in two orthogonal directions can be locked by finite phase modulation, which can prove the effectiveness of the four-dimensional gradient descent algorithm. At the same time, this algorithm has no precise requirement for the boundary position, so it is less affected by environmental factors and has strong robustness.

[0113] In some embodiments, the first demultiplexer and the second demultiplexer are implemented by an arrayed waveguide grating, as Figure 8 shown. It mainly consists of two Rowland circles and a series of waveguides with different lengths. Lights of different wavelengths enter the left Rowland circle from the same port and freely transmit therein. When entering the waveguide array, due to the difference in waveguide lengths, lights of different wavelengths will accumulate different phase differences. Finally, after passing through the right Rowland circle, they are transmitted to different channels. The two Rowland circle regions can be regarded as planar waveguides. The length difference between adjacent waveguides satisfies the grating equation. The insertion loss of the arrayed waveguide grating is generally about 2 - 3 dB, and the crosstalk is about -20 dB. The insertion loss of 2 - 3 dB means that the signal loss is small during transmission, which is beneficial to long-distance transmission. The crosstalk level of -20 dB means that the interference between different channels is very small, which helps to maintain the signal clarity and system performance. The arrayed waveguide grating can not only provide good channel uniformity, making the output power and bandwidth of each wavelength very close, but also has high wavelength selectivity, can accurately separate and combine different wavelengths, and can also increase or decrease the number of channels according to needs.

[0114] As Figure 9 shown, in another embodiment, a polarization locking method is provided, which is applied to a polarization locking device. The polarization locking method includes:

[0115] Step 100: Orthogonally decompose the mixed light to obtain a first light beam and a second light beam; the mixed light includes N lights with different wavelengths, the first light beam includes N first polarized lights with different wavelengths, the second light beam includes N second polarized lights with different wavelengths, the polarization directions of the first polarized lights and the polarization directions of the second polarized lights are perpendicular to each other and perpendicular to the propagation direction of the mixed light; N is an integer greater than 1.

[0116] Step 200: Select the first polarized light corresponding to the wavelength from the first light beam for gating, and perform phase modulation on the gated light to obtain N phase-modulated first polarized lights.

[0117] Step 300: Mix the N phase-modulated first polarized lights.

[0118] Step 400: Couple the mixed N phase-modulated first polarized lights and the N second polarized lights with different wavelengths to obtain the modulated mixed light.

[0119] In some embodiments, the polarization adjustment structure group includes a first polarization adjustment structure, a second polarization adjustment structure, a third polarization adjustment structure, and a fourth polarization adjustment structure. The first polarization adjustment structure, the second polarization adjustment structure, the third polarization adjustment structure, and the fourth polarization adjustment structure are connected in series in sequence. The first polarization adjustment structure, the second polarization adjustment structure, the third polarization adjustment structure, and the fourth polarization adjustment structure are respectively polarization adjustment structures. The phase shifter in the polarization adjustment structure is used to increase or decrease the phase of the input light to obtain the target phase; the polarization locking device further includes a balanced photodetector and a drive control circuit. This polarization locking method further includes:

[0120] Obtain a first initial phase, a second initial phase, a third initial phase, and a fourth initial phase. The first initial phase is the phase when the gated light enters the phase shifter in the first polarization adjustment structure, the second initial phase is the phase when the gated light enters the phase shifter in the second polarization adjustment structure, the third initial phase is the phase when the gated light enters the phase shifter in the third polarization adjustment structure, and the fourth initial phase is the phase when the gated light enters the phase shifter in the fourth polarization adjustment structure.

[0121] Obtain a boundary factor, which is determined based on the phase change boundary, the step size of the four-dimensional gradient descent algorithm, and the phase of any one of the polarization adjustment structures in the polarization adjustment structure group.

[0122] Take the first initial phase, the second initial phase, the third initial phase, the fourth initial phase, and the boundary factor as the input of the feedback function, and perform p iterations using the four-dimensional gradient descent algorithm to output the target feedback value; the target feedback value is the 2 obtained from the p-th iteration calculation 4The minimum value among the function values; p is a positive integer; the feedback function is the square of the distance between the current polarization state and the target polarization state, the current polarization state is the Stokes parameters within the current period obtained by the balanced photodetector feedback, and the target polarization state is the preset output polarization state; in each iteration, based on 2 4 reference polarization states, the corresponding function values are calculated, and the reference polarization states are the polarization states after adjusting the light of the first initial phase, the second initial phase, the third initial phase, and the fourth initial phase by a large or small step size.

[0123] Output the target feedback value to the drive control circuit to update the drive voltage signals of the phase shifters in the first polarization adjustment structure, the second polarization adjustment structure, the third polarization adjustment structure, and the fourth polarization adjustment structure.

[0124] In some embodiments, as Figure 10 shown is the logic block diagram of the four-dimensional gradient descent algorithm. First, obtain the first initial phase , the second initial phase , the third initial phase , and the fourth initial phase . The first initial phase is the phase when the selected light enters the phase shifter in the first polarization adjustment structure, the second initial phase is the phase when the selected light enters the phase shifter in the second polarization adjustment structure, the third initial phase is the phase when the selected light enters the phase shifter in the third polarization adjustment structure, and the fourth initial phase is the phase when the selected light enters the phase shifter in the fourth polarization adjustment structure.

[0125] Obtain the boundary factor b, which is determined based on the phase change boundary and the step size of the four-dimensional gradient descent algorithm; when the phase change boundary of the phase shifter is set to , the boundary factor b is defined as * ;

[0126] Among them, is the boundary factor. When the phase change boundary is set to , is the maximum phase shift, is the minimum phase shift, is the step size of the four-dimensional gradient descent algorithm, which is equal to the phase shift, is the phase shift of any polarization adjustment structure in the polarization adjustment structure group at the pth iteration, where p represents the number of iterations, and m represents any polarization adjustment structure in the polarization adjustment structure group. Preferably, in this embodiment, the value range of m is 1 to 4.

[0127] The first initial phase , the second initial phase , the third initial phase and the fourth initial phase and the boundary factor b are used as the inputs of the feedback function, and the four-dimensional gradient descent algorithm is used for p iterations to output the target feedback value; the target feedback value is the minimum value among the 2 4 function values calculated in the p-th iteration; p is a positive integer; the feedback function is the square of the distance between the current polarization state and the target polarization state, the current polarization state is the Stokes parameters in the current period obtained by the balanced photodetector feedback, and the target polarization state is the preset output polarization state; when the output polarization state is set to ( , , ) polarization state, the feedback function is defined as FB = ( ) 2 + ( ) 2 + ( ) 2 , where FB is the feedback function, , , are the Stokes parameters in the current period obtained by the balanced photodetector feedback, and the parameters , , represent different output polarization state components respectively. For the phase of the phase shifter in each polarization adjustment structure, there are two changing directions of increasing and decreasing in each algorithm period, so there are 2 4 = 16 changing directions in the four-level phase modulation structure.

[0128] In each iteration, the corresponding function value is calculated based on 2 4 reference polarization states, and the reference polarization states are the polarization states after adjusting the first initial phase, the second initial phase, the third initial phase, and the fourth initial phase by a large or small step.

[0129] When reaches near the boundary, the phase increment caused by the step size gradually fails and loses the effect on the feedback function FB, so that the algorithm actively selects the convergence path away from the boundary direction. Due to the existence of redundant degrees of freedom, there are infinitely many convergence paths, so there must be a path direction away from the phase boundary. After introducing the boundary factor b, "probe" the 16 directions, and mark the direction i corresponding to the minimum value of the feedback function FB. The direction i is the direction that simultaneously satisfies the fastest convergence speed and is relatively far from the phase boundary. According to the direction i, for the phases , , , Apply a step size greater than that used in the p-th iteration, and enter the (p + 1)-th iteration loop. For example Figure 10 the step size of the reference polarization state in is . 、 、 、 respectively represent the phase of the phase shifter in the first polarization adjustment structure, the phase of the phase shifter in the second polarization adjustment structure, the phase of the phase shifter in the third polarization adjustment structure, and the phase of the phase shifter in the fourth polarization adjustment structure during the p-th iteration.

[0130] Output the target feedback value to the drive control circuit to update the drive voltage signals of the phase shifters in the first polarization adjustment structure, the second polarization adjustment structure, the third polarization adjustment structure, and the fourth polarization adjustment structure.

[0131] Due to the redundant design of the four-stage structure compared to the two-stage structure, there must be more than one change direction for the feedback function FB to show a converging trend. In the p-th cycle, "probe" the 16 change directions in sequence and record the feedback function corresponding to each change direction , = 1, 2, …… 16. At the same time, introduce the boundary factor b to make the phase change near the boundary more inclined to change in the direction away from the boundary, while ensuring that the phase changes within the specified boundary. Among them, only one iteration is performed in each cycle, break means end; Min() means taking the minimum value.

[0132] This four-stage polarization adjustment structure provides four independent control degrees of freedom, which means that there are multiple paths to reach the same polarization state. When the system falls into a local optimal solution during the optimization process, the local optimal can be bypassed by changing other control parameters, so as to find the global optimal solution. It can use the redundant design to avoid the reset problem caused by the control parameters falling into the local optimal solution during the polarization tracking process, and at the same time add the boundary factor b to achieve continuous polarization locking within the limited boundary.

[0133] Generally speaking, this is an iterative optimization method that determines the most favorable direction through fine search in a small range, and then takes a larger step in that direction to quickly converge to the optimal solution.

[0134] Such as Figure 11As shown, assuming that the phase change range is [0, 2π], the initial circular polarization state rotates through a continuously varying phase in one direction, which is shown as a circle rotating around S1 on the Poincaré sphere. The purpose of the continuous phase rotation is to artificially simulate the boundary problem caused by the continuously co-directional polarization rotation. After polarization scrambling, the dynamic polarized light enters the algorithm loop, and the polarization tracking target is set to the first polarization state (X linearly polarized state).

[0135] As Figure 12 shown is the output state diagram of the first polarization state (X linearly polarized state) after polarization locking. As Figure 13 shown is the phase change during the polarization locking process. During the entire control process, no boundary "unlocking" problem occurs, indicating that the four-level gradient descent algorithm is reliable. Among them, 、 、 、 respectively represent the phases of the first polarization adjustment structure, the second polarization adjustment structure, the third polarization adjustment structure, and the fourth polarization adjustment structure. Similarly, as Figure 14 shown, the polarization scrambling method is changed to apply an azimuth angle change that rotates in the same direction to the initial 45° linearly polarized state, which is reflected as a continuously rotating ring around S3 on the Poincaré sphere. As Figure 15 shown is the output state diagram of the first polarization state (X linearly polarized state) after polarization tracking.

[0136] The co-directional continuous changes of the polarization states in two orthogonal directions can both be locked by finite phase modulation, which can prove the effectiveness of the four-dimensional gradient descent algorithm. At the same time, this algorithm has no requirements for the boundary position, so it is less affected by environmental factors and has strong robustness.

[0137] In another embodiment, an optical communication device is provided, including the polarization locking device mentioned in the above embodiment.

[0138] In another embodiment, an optical chip is provided, including the polarization locking device mentioned in the above embodiment.

[0139] As Figure 16 shown, another embodiment provides an application scenario of the above multi-wavelength polarization locking device in a wavelength division multiplexing optical communication system.

[0140] Polarization: It refers to the phenomenon of asymmetry of the vibration vector (including direction and amplitude) in the plane perpendicular to the propagation direction of a transverse wave (such as light).

[0141] Unpolarized light and polarized light: In the plane perpendicular to the propagation direction of the light, if the vibration vector of the light can point in all directions and has the same amplitude in each direction, such light is called unpolarized light; otherwise, it is called polarized light.

[0142] Although the embodiments of the present invention have been described in detail above, it will be apparent to those skilled in the art that various modifications and variations can be made to these embodiments. However, it should be understood that such modifications and variations are all within the scope and spirit of the present invention as described in the claims. Moreover, the present invention described herein can have other embodiments and can be implemented or realized in various ways.

Claims

1. A polarization locking device, characterized in that: It includes a polarization beam splitter, a polarization adjustment structure group and a drive control circuit; The polarization beam splitter is used to perform orthogonal decomposition on the mixed light to obtain a first beam of light and a second beam of light, and output the first beam of light to the polarization adjustment structure group; the mixed light includes N lights of different wavelengths, the first beam of light includes N first polarized lights of different wavelengths, the second beam of light includes N second polarized lights of different wavelengths, the polarization direction of the first polarized light and the polarization direction of the second polarized light are perpendicular to each other and perpendicular to the propagation direction of the mixed light; N is an integer greater than 1; The polarization adjustment structure group includes at least one polarization adjustment structure, and the polarization adjustment structure includes a multiplexer, a coupler, N microring resonators and N phase shifters; N microring resonators correspond to N phase shifters one by one; each of the microring resonators is used to select the first polarized light of the corresponding wavelength from the first beam of light for gating, and output the gated light to the corresponding phase shifter for phase modulation; The phase shifter outputs the phase-modulated light to the multiplexer; The multiplexer mixes the N phase-modulated first polarized lights and outputs the mixed light to the coupler; The coupler couples N phase-modulated first polarized lights and N second polarized lights of different wavelengths to obtain modulated mixed light; The driving control circuit is used to receive the current polarization state corresponding to each wavelength, and obtain a feedback value based on the current polarization state to update the driving voltage signal of the phase shifter corresponding to each wavelength; The feedback value is the square of the distance between the current polarization state and the preset target polarization state.

2. The device according to claim 1, characterized in that Also included are mixer and demultiplexer banks; One end of the mixer is connected to the polarization adjustment structure group, and is used to convert part of the mixed light output by the polarization adjustment structure group to obtain optical signals of six different output channels; The other end of the mixer is connected to a demultiplexer group, which includes six first demultiplexers. The six first demultiplexers are respectively used to decouple the optical signals of the six different output channels to obtain optical signals of six different output channels corresponding to each wavelength.

3. The device according to claim 2, characterized in that Also included is a balanced photodetector; The balanced photodetector is used to obtain three difference optical signals based on the six optical signals output by the demultiplexer group to obtain three Stokes parameters; the six optical signals are divided into three pairs of optical signals, and a difference operation is performed on each pair of optical signals to obtain a corresponding difference signal; The current polarization state corresponding to each wavelength is determined based on the three Stokes parameters.

4. The device according to claim 3, characterized in that The polarization adjustment structure group includes a first polarization adjustment structure, a second polarization adjustment structure, a third polarization adjustment structure and a fourth polarization adjustment structure, wherein the first polarization adjustment structure, the second polarization adjustment structure, the third polarization adjustment structure and the fourth polarization adjustment structure are sequentially connected in series, and the first polarization adjustment structure, the second polarization adjustment structure, the third polarization adjustment structure and the fourth polarization adjustment structure are respectively the polarization adjustment structures, and the phase shifter in the polarization adjustment structure is used to increase or decrease the phase of the input light to obtain the target phase; The drive control circuit is used to obtain a first initial phase, a second initial phase, a third initial phase, a fourth initial phase and a boundary factor, use the first initial phase, the second initial phase, the third initial phase, the fourth initial phase and the boundary factor as inputs of a feedback function, and use a four-dimensional gradient descent algorithm to perform p iterations, output a target feedback value, and update the drive voltage signal of the phase shifter in the first polarization adjustment structure, the second polarization adjustment structure, the third polarization adjustment structure and the fourth polarization adjustment structure based on the target feedback value; The first initial phase is the phase of the selected light when it enters the phase shifter in the first polarization adjustment structure, the second initial phase is the phase of the selected light when it enters the phase shifter in the second polarization adjustment structure, the third initial phase is the phase of the selected light when it enters the phase shifter in the third polarization adjustment structure, and the fourth initial phase is the phase of the selected light when it enters the phase shifter in the fourth polarization adjustment structure; The boundary factor is determined based on a phase change boundary, a step size of a four-dimensional gradient descent algorithm, and a phase of any polarization adjustment structure in the polarization adjustment structure group; The target feedback value is 2 calculated by the pth iteration. 4 The minimum value among the function values; p is a positive integer; the feedback function is the square of the distance between the current polarization state and the target polarization state, the current polarization state is the Stokes parameter in the current cycle obtained by feedback from the balanced photodetector, and the target polarization state is the preset output polarization state; each iteration, based on 2 4 The reference polarization state is the polarization state after increasing or decreasing the step size of the light of the first initial phase, the second initial phase, the third initial phase, and the fourth initial phase.

5. The device according to claim 4, characterized in that The calculation formula of the boundary factor is: * in, is the boundary factor. When the phase change boundary is set to hour, is the maximum phase shift, is the minimum phase shift, is the step size of the four-dimensional gradient descent algorithm, which is equal to the phase offset, is the phase shift of any polarization adjustment structure in the polarization adjustment structure group at the p-th iteration, where p represents the number of iterations, and m represents any polarization adjustment structure in the polarization adjustment structure group.

6. The device according to claim 1, characterized in that The phase shifter is a lithium niobate electro-optic phase shifter or a silicon-based thermo-optic phase shifter.

7. The device according to claim 2, characterized in that Also included is a polarization beam combiner and a second demultiplexer; One end of the polarization beam combiner is connected to the polarization adjustment structure group, and is used for orthogonally coupling the modulated mixed light output by the polarization adjustment structure group into polarized light; The other end of the polarization beam combiner is connected to the second demultiplexer, and the second demultiplexer is used to decouple the polarized light output by the polarization beam combiner to obtain stable polarized light of different wavelengths.

8. A polarization locking method, characterized in that: Applied to the polarization locking device according to any one of claims 1 to 7, the method comprising: The mixed light is orthogonally decomposed to obtain a first light beam and a second light beam; the mixed light comprises N lights of different wavelengths, the first light beam comprises N first polarized lights of different wavelengths, the second light beam comprises N second polarized lights of different wavelengths, the polarization directions of the first polarized lights and the second polarized lights are perpendicular to each other and perpendicular to the propagation direction of the mixed light; N is an integer greater than 1; Selecting a first polarized light of a corresponding wavelength from the first light beam for gating, and performing phase modulation on the gated light to obtain N phase-modulated first polarized lights; Mixing the N phase-modulated first polarized lights; The mixed N phase-modulated first polarized lights and N second polarized lights with different wavelengths are coupled to obtain modulated mixed light.

9. The method according to claim 8, characterized in that The polarization adjustment structure group includes a first polarization adjustment structure, a second polarization adjustment structure, a third polarization adjustment structure and a fourth polarization adjustment structure, wherein the first polarization adjustment structure, the second polarization adjustment structure, the third polarization adjustment structure and the fourth polarization adjustment structure are sequentially connected in series, and the first polarization adjustment structure, the second polarization adjustment structure, the third polarization adjustment structure and the fourth polarization adjustment structure are respectively the polarization adjustment structures, and the phase shifter in the polarization adjustment structure is used to increase or decrease the phase of the input light to obtain the target phase; The polarization locking device further includes a balanced photodetector and a drive control circuit, and the method further includes: Acquire a first initial phase, a second initial phase, a third initial phase, and a fourth initial phase, wherein the first initial phase is the phase of the selected light when it enters the phase shifter in the first polarization adjustment structure, the second initial phase is the phase of the selected light when it enters the phase shifter in the second polarization adjustment structure, the third initial phase is the phase of the selected light when it enters the phase shifter in the third polarization adjustment structure, and the fourth initial phase is the phase of the selected light when it enters the phase shifter in the fourth polarization adjustment structure; Acquire a boundary factor, where the boundary factor is determined based on a phase change boundary, a step size of a four-dimensional gradient descent algorithm, and a phase of any polarization adjustment structure in the polarization adjustment structure group; The first initial phase, the second initial phase, the third initial phase, the fourth initial phase and the boundary factor are used as inputs of the feedback function, and the four-dimensional gradient descent algorithm is used for p iterations to output a target feedback value; the target feedback value is 2 obtained by calculating the pth iteration 4 The minimum value among the function values; p is a positive integer; the feedback function is the square of the distance between the current polarization state and the target polarization state, the current polarization state is the Stokes parameter in the current cycle obtained by feedback from the balanced photodetector, and the target polarization state is the preset output polarization state; each iteration, based on 2 4 The reference polarization state is the polarization state after increasing or decreasing the step size of the light of the first initial phase, the second initial phase, the third initial phase, and the fourth initial phase; The target feedback value is output to the driving control circuit to update the driving voltage signals of the phase shifters in the first polarization adjustment structure, the second polarization adjustment structure, the third polarization adjustment structure, and the fourth polarization adjustment structure.

10. An optical communication device, characterized in that: Comprising the polarization locking device as claimed in any one of claims 1 to 7.

11. An optical chip, characterized in that: Comprising the polarization locking device as claimed in any one of claims 1 to 7.

Images