CN-119472087-B - Polarization control method and system based on jacobian matrix

CN119472087BCN 119472087 BCN119472087 BCN 119472087BCN-119472087-B

Abstract

The invention discloses a polarization control method and a system based on a jacobian matrix, wherein the method comprises the steps of obtaining the current output polarization state; the method comprises the steps of obtaining an error between a current output polarization state and a preset target polarization state, obtaining a jacobian matrix according to the current output polarization state, obtaining a target updating value of a current moment control signal according to the error and the jacobian matrix, and obtaining a next moment control signal according to the target updating value of the current moment control signal. The invention can realize high-speed polarization control without resetting, and can be widely applied to the technical field of polarization control.

Inventors

WANG DAWEI
XU YUXI

Assignees

中山大学

Dates

Publication Date: 20260505
Application Date: 20241028

Claims (10)

1. The polarization control method based on the jacobian matrix is characterized by comprising the following steps of: acquiring a current output polarization state; Acquiring an error between the current output polarization state and a preset target polarization state; Obtaining a jacobian matrix according to the current output polarization state; according to the error and the jacobian matrix, obtaining a target updated value of the control signal at the current moment comprises the following steps: Judging whether the polarization control signals exceed the first physical boundary, if the polarization control signals do not exceed the first physical boundary, calculating a control signal update value at the current moment by using a transpose of a jacobian matrix based on a gradient descent method, and if a certain level of the polarization control signals exceeds the first physical boundary, starting a gradient projection algorithm in a self-adaptive manner to carry out first constraint on the polarization control signals; Or alternatively Presetting a second physical boundary, judging whether the polarization control signal exceeds the second physical boundary, if the polarization control signal does not exceed the second physical boundary, calculating a control signal update value at the current moment through pseudo-inverse of the jacobian matrix, and if a certain level of the polarization control signal exceeds the second physical boundary and is When the maximum level control signal is obtained, the adaptive start gradient projection algorithm performs second constraint on the polarization control signal, and only the projection subitem of the level with the maximum voltage is calculated; and obtaining a next time control signal according to the target update value of the current time control signal.
2. The polarization control method based on jacobian according to claim 1, wherein the obtaining the current output polarization state comprises the steps of: Acquiring an input polarization state, polarization control signals of each stage and a transmission matrix of each stage; And converting the input polarization state according to the polarization control signals of each stage and the transmission matrix of each stage to obtain the current output polarization state.
3. The jacobian-based polarization control method of claim 1, wherein the obtaining the current output polarization state uses a formula comprising: ; In the formula, Representing the current output polarization state; representing an input polarization state; Representative of At the lower part of the time The transmission matrix of the stage rotation waveplate, ; Represents the first A stage control signal.
4. The polarization control method based on jacobian according to claim 1, wherein the step of obtaining the jacobian according to the current output polarization state includes the steps of: obtaining a partial derivative of the current output polarization state on a vector of a polarization control signal to obtain the jacobian matrix; wherein, the expression of the jacobian matrix is: ; In the formula, Representing a jacobian matrix of the matrix, ; Represents a set of real numbers, Representing the number of polarization control signals; representing the current output polarization state; Representative of The vector of the level polarization control signals, 。
5. The polarization control method based on jacobian according to claim 1, wherein the obtaining the target updated value of the control signal at the current time according to the error and the jacobian comprises the following steps: presetting a first physical boundary; when the polarization control signal does not exceed the first physical boundary, acquiring a first updated value of the control signal at the current moment according to the transpose of the jacobian matrix and the error; when the polarization control signal exceeds the first physical boundary, a first constraint is placed on the polarization control signal.
6. The polarization control method based on jacobian according to claim 1, wherein the obtaining the target update value of the control signal at the current time according to the error and the jacobian further comprises the following steps: Presetting a second physical boundary; when the polarization control signal does not exceed the second physical boundary, obtaining a second updated value of the control signal at the current moment according to the pseudo-inverse of the jacobian matrix and the error; and when the polarization control signal exceeds the second physical boundary, performing second constraint on the polarization control signal.
7. A jacobian-based polarization control system, comprising: The first module is used for acquiring the current output polarization state; the second module is used for acquiring an error between the current output polarization state and a preset target polarization state; the third module is used for obtaining a jacobian matrix according to the current output polarization state; A fourth module, configured to obtain a target update value of a control signal at a current time according to the error and the jacobian matrix, where the fourth module is specifically configured to: Judging whether the polarization control signals exceed the first physical boundary, if the polarization control signals do not exceed the first physical boundary, calculating a control signal update value at the current moment by using a transpose of a jacobian matrix based on a gradient descent method, and if a certain level of the polarization control signals exceeds the first physical boundary, starting a gradient projection algorithm in a self-adaptive manner to carry out first constraint on the polarization control signals; Or alternatively Presetting a second physical boundary, judging whether the polarization control signal exceeds the second physical boundary, if the polarization control signal does not exceed the second physical boundary, calculating a control signal update value at the current moment through pseudo-inverse of the jacobian matrix, and if a certain level of the polarization control signal exceeds the second physical boundary and is When the maximum level control signal is obtained, the adaptive start gradient projection algorithm performs second constraint on the polarization control signal, and only the projection subitem of the level with the maximum voltage is calculated; And a fifth module, configured to obtain a control signal at the next time according to the target update value of the control signal at the current time.
8. An electronic device comprising a processor and a memory; the memory is used for storing programs; the processor executing the program implements the method of any one of claims 1 to 6.
9. A computer-readable storage medium, characterized in that the storage medium stores a program that is executed by a processor to implement the method of any one of claims 1 to 6.
10. A computer program product comprising a computer program which, when executed by a processor, implements the method of any of claims 1 to 6.

Description

Polarization control method and system based on jacobian matrix Technical Field The invention relates to the technical field of polarization control, in particular to a polarization control method and system based on a jacobian matrix. Background At present, polarization control is widely applied in the fields of optical communication, quantum key communication, optical sensing and the like. In an optical fiber link, external interference can bring different degrees of change to a transmitted optical signal, and knocking the dispersion compensation optical fiber in a core network can cause random change with the polarization change rate of 280krad/s, so that signal distortion is caused. In a quantum key communication system, the random change of polarization states of a local optical field and a signal optical field in an optical fiber can influence the balanced homodyne detection result of a receiving end, so that the error rate is increased, and the safety and stability of the system are reduced. In photo-sensing systems, random fluctuations in the polarization state can lead to degradation of the sensing signal. Therefore, the high-speed optical dynamic polarization control technology plays an important role in maintaining the stable operation of the system. However, the existing polarization control algorithm has certain defects in research, such as complex hardware implementation, large loop delay and slow iteration speed, or a reset mechanism with low robustness, the polarization control speed is always limited, and the algorithms are designed aiming at specific polarization controller architectures and are difficult to expand to any-level systems. Disclosure of Invention In view of the above, the embodiment of the present invention is mainly aimed at providing a polarization control method and system based on jacobian matrix, so as to solve at least one problem in the prior art. To achieve the above object, an aspect of an embodiment of the present invention provides a polarization control method based on a jacobian matrix, including the steps of: acquiring a current output polarization state; Acquiring an error between the current output polarization state and a preset target polarization state; Obtaining a jacobian matrix according to the current output polarization state; Acquiring a target update value of a control signal at the current moment according to the error and the jacobian matrix; and obtaining a next time control signal according to the target update value of the current time control signal. In some embodiments, the obtaining the current output polarization state includes the steps of: Acquiring an input polarization state, polarization control signals of each stage and a transmission matrix of each stage; And converting the input polarization state according to the polarization control signals of each stage and the transmission matrix of each stage to obtain the current output polarization state. In some embodiments, the obtaining the current output polarization state uses a formula comprising: Sout＝Mn(θn)Mn-1(θn-1)…M1(θ1)Sin Where S out represents the current output polarization state, S in represents the input polarization state, M i(θi) represents the transmission matrix of the i-th stage rotated waveplate at k times, i=1, 2. In some embodiments, the obtaining the jacobian matrix according to the current output polarization state includes the following steps: obtaining a partial derivative of the current output polarization state on a vector of a polarization control signal to obtain the jacobian matrix; wherein, the expression of the jacobian matrix is: wherein J represents a Jacobian matrix, Represents a real set, n represents the number of polarization control signals, S out represents the current output polarization state, θ represents a vector of n-level polarization control signals, θ= [ θ 1,θ2,...,θn]T. In some embodiments, the obtaining the target updated value of the control signal at the current moment according to the error and the jacobian matrix includes the following steps: presetting a first physical boundary; when the polarization control signal does not exceed the first physical boundary, acquiring a first updated value of the control signal at the current moment according to the transpose of the jacobian matrix and the error; when the polarization control signal exceeds the first physical boundary, a first constraint is placed on the polarization control signal. In some embodiments, the obtaining the target update value of the control signal at the current moment according to the error and the jacobian matrix further includes the following steps: Presetting a second physical boundary; when the polarization control signal does not exceed the second physical boundary, obtaining a second updated value of the control signal at the current moment according to the pseudo-inverse of the jacobian matrix and the error; and when the polarization control signal exceeds the s