A deployment method and optical fiber communication system for reconfigurable optical fiber mode demultiplexing

Abstract

This invention relates to a deployment method and optical fiber communication system for reconfigurable optical fiber mode demultiplexing. The method includes the following steps: S101, signal generation; S102, multiplexing crosstalk compensation; S103, mode multiplexing; S104, optical fiber transmission; S105, mode demultiplexing; S106, demultiplexing crosstalk compensation; S107, signal reception and processing. This invention employs static compensation at both the transmitting and receiving ends, achieving lower inter-module crosstalk even for any non-ideal mode switching device, maintaining lower MIMO computational complexity, and reducing the design and manufacturing difficulty of the mode switching device. Simultaneously, this method allows for real-time and rapid evaluation of the crosstalk compensation matrix, making the spatial division multiplexing communication system more tolerant to optical path coupling errors.

Description

Technical Field

[0001] This invention relates to the field of optical fiber spatial division multiplexing communication and signal processing technology, and in particular to a deployment method and optical fiber communication system for reconfigurable optical fiber mode demultiplexing. Background Technology

[0002] With the increasing global information flow driving demand for higher-capacity optical fiber communication technologies, mode division multiplexing (MDM) technology, which utilizes multiple orthogonal transmission modes in few-mode or multimode fibers to enhance fiber transmission capacity, has received widespread research and attention. Combining MDM with space division multiplexing, arranging multiple few-mode or multimode fiber cores within a single fiber to achieve multi-core fiber space division multiplexing is also a technological trend for further improving the transmission capacity of a single optical fiber.

[0003] Few-mode or multi-mode optical fibers exhibit mode grouping characteristics. Based on mode degeneracy, each mode is divided into several mode groups. Within the same mode group, strong, time-varying, and random crosstalk inevitably occurs due to unavoidable disturbances such as fiber manufacturing errors and variations in the fiber laying environment. However, modes in different modes exhibit weak coupling due to low degeneracy and significant differential mode group delay, making crosstalk negligible. Therefore, after receiving a signal, the signal from each mode can be processed separately using a smaller-scale multi-in multi-out (MIMO) approach. Compared to few-mode or multi-mode optical fiber communication that does not utilize mode grouping characteristics, MIMO computational complexity is significantly reduced. Furthermore, multi-core optical fibers have larger core spacing, and inter-core crosstalk is always negligible. The mode grouping characteristics of each core can be used to process the signal from each core separately, allowing for an increase in the number of channels without the need to introduce a larger-scale MIMO approach.

[0004] However, achieving lower MIMO computational complexity by utilizing mode grouping characteristics presupposes the realization of ideal few-mode or multimode fiber mode multiplexing and demultiplexing, converting separated single-mode optical fields into specific modules within the fiber without crosstalk. In recent years, scholars have proposed various mode conversion mechanisms: spatial light-based mode transformation can achieve multiplexing and demultiplexing with low crosstalk, but these devices or implementation methods currently struggle to simultaneously achieve reconfigurability and compactness, and suffer from poor optical field alignment robustness; fiber optic device-based methods, such as photonic lanterns, offer lower optical path alignment difficulty, but suffer from significant inter-module crosstalk; furthermore, the devices involved in these methods are relatively complex in design, hindering the expansion of the number of channels. Summary of the Invention

[0005] To address the technical problems existing in the prior art, this invention provides a deployment method and optical fiber communication system for reconfigurable optical fiber mode demultiplexing. By applying static compensation to the transmitting and receiving ends respectively, it can achieve lower inter-module crosstalk, maintain lower MIMO computational complexity, and reduce the design and manufacturing difficulty of mode switching devices even for any non-ideal mode switching device.

[0006] The method of this invention is implemented using the following technical solution: a deployment method for reconfigurable fiber mode demultiplexing, comprising the following steps:

[0007] S101, Signal generation: The signal source emits an N-channel electrical signal matrix s TX0 By performing complex amplitude modulation on the light field, the optical signal matrix s is obtained. TX ;

[0008] S102. Multiplexing crosstalk compensation, the optical signal matrix s obtained in step S101 is... TX The input is fed into the all-optical matrix operation unit, multiplied by the N×N mode multiplexing crosstalk compensation matrix M′, and outputs N optical signal vectors s′. TX =M's TX ;

[0009] S103, Mode multiplexing, converting the N optical signal vectors s′ obtained in step S102 into multiplexed signals. TX Mode multiplexing is achieved by coupling the N modes into the space-division multiplexed fiber channel through a spatial mode converter, denoted as s. channel =Ms′ TX =MM's TX , where s channel This represents the signal matrix of N optical field modes loaded into the mode-division multiplexing fiber before transmission, where M is a random N×N mode multiplexing crosstalk matrix.

[0010] S104. Fiber optic transmission: The signal matrix s of the N optical field modes that have not yet been transmitted and are loaded into the mode-division multiplexing fiber obtained in step S103 is transferred. channel Propagated in few-mode or multimode optical fibers, the output at the receiving end is a signal matrix s′ of N optical field modes that has been transmitted in the mode-division multiplexing fiber. channel The crosstalk matrix between different modules of a few-mode or multimode optical fiber is represented as the optical fiber channel transmission crosstalk matrix U′.

[0011] S105. Mode demultiplexing: The N optical field mode signal matrices s′ obtained in step S104 and transmitted into the mode-division multiplexing fiber are demultiplexed. channel The N optical signals separated by another spatial mode converter are demultiplexed, denoted as s′. RX =Ds′ channel ; where s′RX This represents the N independent optical signals output from the mode-division multiplexing fiber and obtained after mode demultiplexing, where D is a random N×N mode demultiplexing crosstalk matrix.

[0012] S106. Demultiplexing crosstalk compensation: The N independent optical signals s′ obtained in step S105, which were output from the mode-division multiplexing fiber and demultiplexed, are then demultiplexed. RX The input is fed into another all-optical matrix operation unit, multiplied by the N×N mode demultiplexing crosstalk compensation matrix D′, and outputs N optical signal vectors s. RX =D′s′ RX =D′Ds′ channel ;

[0013] S107. Signal reception and processing: Coherent detection converts the N-channel optical signal vectors s obtained in step S106 into... RX Converted to N electrical signals s RX0 After signal processing, the original signal vector s can be recovered. TX0 .

[0014] The present invention is implemented using the following technical solution: an optical fiber communication system with a deployment method for reconfigurable optical fiber mode demultiplexing, comprising:

[0015] The optical transmitter array module S201 is used to implement step S101. It consists of N optical transmitters and generates complex amplitude modulated optical signals according to user definition. The carrier light sources of these N optical transmitters are the same coherent light source or different light sources. It is used to evaluate the demultiplexing crosstalk compensation matrix or actual communication.

[0016] The transmitting end all-optical matrix operation unit S202 is used to implement step S102, and is divided into two parts: the first part reconfigurably defines an N×N complex matrix such that the complex amplitude of the output optical signal vector is equal to the product of the matrix and the complex amplitude of the input optical signal vector generated by the optical transmitter array module S201; the second part reconfigurably defines an N×N pure phase diagonal matrix, which superimposes an additional phase on each of the output optical signals of the first part; the all-optical matrix operation unit is composed of a Mach-Zehnder interferometer network, a micro-ring array optical integrated device, or a spatial optical wavefront conversion device, as well as other units or modules with linear matrix conversion functions;

[0017] The transmitting end mode conversion device S203, used to implement step S103, is a passive device that converts the N optical signals output by the transmitting end all-optical matrix operation unit S202 into optical signals that are multiplexed and transmitted in multiple orthogonal mode channels in a space-division multiplexing fiber. This device can be a photonic lantern, a mode-selective exciter fiber-type device, or a vortex phase plate, an optical wavefront modulator, a multi-plane optical conversion module spatial optical device, as well as other devices or modules with linear mode conversion and multiplexing functions.

[0018] Space division multiplexing fiber S204 is used to implement step S104. The type of fiber is multimode or few-mode fiber, or multi-core fiber composed of multiple multimode or few-mode fiber cores. It supports the transmission of N optical signals output by the transmitting end mode conversion device S203 in N mode channels. The mode channel is in the form of an intrinsic vector mode, linear polarization LP mode, orbital angular momentum OAM mode, or other orthogonal basis mode supported in any fiber core. These channel modes are divided into different modules according to degeneracy, thus having the transmission characteristics of step S104.

[0019] The receiver mode converter S205, used to implement step S103, is a passive device that separates the beam output from the optical fiber in the space-division multiplexing fiber S204 into N optical signals. This device can be a photonic lantern, a mode-selective exciter fiber-type device, or a vortex phase plate, an optical wavefront modulator, a multi-plane optical conversion module spatial optical device, as well as other devices or modules with mode linear conversion and multiplexing functions.

[0020] The receiver-side all-optical matrix operation unit S206 is used to implement step S106 and is divided into two parts: the first part reconfigurably defines an N×N pure phase diagonal matrix to superimpose an additional phase on each of the N optical signals output by the receiver-side mode converter S205; the second part reconfigurably defines an N×N complex matrix such that the complex amplitude of the output optical signal vector is equal to the product of the matrix and the complex amplitude of the output optical signal vector in the first part; the all-optical matrix operation unit can be a Mach-Zehnder interferometer network, a micro-ring array optical integrated device, or a spatial optical wavefront converter, as well as other units or modules with linear matrix conversion functions;

[0021] The coherent receiving array module S207 is used to realize the signal reception in step S107. It consists of N single-polarization coherent optical receivers or N / 2 dual-polarization coherent optical receivers and a reference light source with the same frequency as the carrier. It converts the complex amplitude of the optical signal output by the all-optical matrix operation unit S206 at the receiving end into an electrical signal.

[0022] The signal processing and control unit S208 is used to implement the signal processing in step S107 and the processes described in steps S111 to S115; in actual communication, it is used to perform small-scale MIMO operations and other signal processing on the optical signals of each module to obtain the information transmitted by the source; when evaluating the demultiplexing crosstalk compensation matrix, it extracts the value of each information symbol, and then executes the algorithm in steps S111 to S115 to provide feedback control of the optical transmitter array module S201, the transmitting end all-optical matrix operation unit S202, and the receiving end all-optical matrix operation unit S206, which are components of the system.

[0023] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0024] 1. This invention employs static compensation at both the transmitting and receiving ends, which can achieve lower inter-module crosstalk for any non-ideal mode switching device, maintain lower MIMO computational complexity, and reduce the design and manufacturing difficulty of the mode switching device.

[0025] 2. The method of the present invention can evaluate the crosstalk compensation matrix in real time and quickly, making the spatial multiplexing communication system more tolerant to optical path coupling errors.

[0026] 3. The all-optical matrix operation unit used to deploy the crosstalk compensation matrix in the system of the present invention can be integrated with the transceiver, making the demultiplexing module more flexible, compact and easy to deploy; while the static, all-optical signal processing method reduces the pressure on the signal buffer of the electronic signal processing equipment and breaks through the limitations of the operation speed and signal bandwidth of the electronic signal processing equipment, resulting in lower photoelectric power consumption, insertion loss, and the advantages of lower signal processing latency and higher signal bandwidth. Attached Figure Description

[0027] Figure 1 This is a flowchart of the method of the present invention;

[0028] Figure 2 This is a flowchart of the method for evaluating the specific values ​​of the crosstalk compensation matrix according to the present invention;

[0029] Figure 3 This is a flowchart of the method for online optimization of the phase diagonal matrix according to the present invention;

[0030] Figure 4 This is a system structure block diagram of the present invention. Detailed Implementation

[0031] The present invention will be further described in detail below with reference to the embodiments and accompanying drawings, but the embodiments of the present invention are not limited thereto.

[0032] Example

[0033] like Figure 1 As shown in the figure, this embodiment of a deployment method for reconfigurable fiber mode demultiplexing includes the following steps:

[0034] S101, Signal generation: The signal source emits an N-channel electrical signal matrix s TX0 By performing complex amplitude modulation on the light field, the optical signal matrix s is obtained. TX ;

[0035] S102. Multiplexing crosstalk compensation, the optical signal matrix s obtained in step S101 is... TX The input is fed into the all-optical matrix operation unit, multiplied by the N×N mode multiplexing crosstalk compensation matrix M′, and outputs N optical signal vectors s′. TX=M's TX ;

[0036] S103, Mode multiplexing, converting the N optical signal vectors s′ obtained in step S102 into multiplexed signals. Tx Mode multiplexing is achieved by coupling the N modes into the space-division multiplexed fiber channel through a spatial mode converter, denoted as s. channel =Ms′ TX =MM's TX , where s channel This represents the signal matrix of N optical field modes loaded into the mode-division multiplexing fiber before transmission; M is a random N×N mode multiplexing crosstalk matrix, which basically does not change over time after the system is built.

[0037] S104. Fiber optic transmission: The signal matrix s of the N optical field modes that have not yet been transmitted and are loaded into the mode-division multiplexing fiber obtained in step S103 is transferred. channel Propagated in few-mode or multimode optical fibers, the output at the receiving end is a signal matrix s′ of N optical field modes that has been transmitted in the mode-division multiplexing fiber. channel The crosstalk matrix between modes of different modules in a few-mode or multimode fiber is represented as the transmission crosstalk matrix U′ of the fiber channel. The crosstalk between modes of different modules in a few-mode or multimode fiber is negligible and has differential module delay, while there is strong random and time-varying crosstalk between modes within the same module. Therefore, the transmission crosstalk matrix U′ of the fiber channel has diagonal block characteristics.

[0038] S105. Mode demultiplexing: The N optical field mode signal matrices s′ obtained in step S104 and transmitted into the mode-division multiplexing fiber are demultiplexed. channel The N optical signals separated by another spatial mode converter are demultiplexed, denoted as s′. RX =Ds′ channel ; where s′ RX This represents the N independent optical signals output from the mode-division multiplexing fiber and obtained after mode demultiplexing; D is a random N×N mode demultiplexing crosstalk matrix, which basically does not change over time after the system is built.

[0039] S106. Demultiplexing crosstalk compensation: The N independent optical signals s′ obtained in step S105, which were output from the mode-division multiplexing fiber and demultiplexed, are then demultiplexed. RX The input is fed into another all-optical matrix operation unit, multiplied by the N×N mode demultiplexing crosstalk compensation matrix D′, and outputs N optical signal vectors s. RX =D′s′ RX =D′Ds′ channel ;

[0040] S107. Signal reception and processing: Coherent detection converts the N-channel optical signal vectors s obtained in step S106 into... RX Converted to N electrical signals s RX0 After signal processing, the original signal vector s can be recovered. TX0 .

[0041] Specifically, in this embodiment, the mode multiplexing crosstalk compensation matrix M′ in step S102 should satisfy the following relationship: the product MM′ of the mode multiplexing crosstalk matrix M and M′ in step S103 is an N×N block diagonal matrix, which has the same distribution of non-zero sub-blocks as the optical fiber channel transmission matrix U′ in step S104.

[0042] Specifically, in this embodiment, the mode demultiplexing crosstalk compensation matrix D′ in step S106 should satisfy the following relationship: the product D′D of D′ and the mode demultiplexing crosstalk matrix D in step S105 is an N×N block diagonal matrix, which has the same distribution of non-zero sub-blocks as the optical fiber channel transmission matrix U′ in step S104.

[0043] Specifically, in this embodiment, the mode multiplexing crosstalk compensation matrix in step S102 In step S106, the mode demultiplexing crosstalk compensation matrix in, and It is the phase diagonal matrix to be optimized.

[0044] like Figure 2 As shown in the figure, in this embodiment, the specific process of evaluating the mode multiplexing crosstalk matrix M and the mode demultiplexing crosstalk matrix D is as follows:

[0045] S111. Define an identity matrix signal. Let the number of mode channels in a space-division multiplexing fiber be N. Define N signals, each containing N information symbols, represented by an N×N matrix. The element in row i and column j of the matrix represents the value of the j-th information symbol in the i-th signal. The total duration of the N information symbols is less than the minimum differential module delay in the fiber channel. The matrix signal does not overlap after transmission in different modules of the fiber. It is defined as a matrix signal I represented by an identity matrix.

[0046] S112. Online transmission to obtain the transmission matrix: Using the matrix signal I defined in step S111 as the original signal, implement steps S101 to S107. Since there is no overlap after transmission in different modules, the N×N transmission matrix {T} of each module can be extracted separately. (1) T (2) ... T (m)}, where T (i) ,i=1,2,…,m represents the transmission matrix of the i-th module;

[0047] S113. Transmission matrix analysis: For each T obtained in step S112... (i) Perform singular value decomposition on i = 1, 2, ..., m:

[0048]

[0049] in,(·) H Σ represents the conjugate transpose of a matrix. (i) It is a diagonal matrix composed of non-negative real numbers, with the diagonal elements arranged in descending order from top left to bottom right. All are unitary matrices;

[0050] S114. Construct the crosstalk compensation matrix. Let the i-th module contain N... i In each mode, the following operation is performed on all modules: The result obtained in step S113 is... The first N i The columns are arranged in modular order to form an N×N matrix. Take the result obtained in step S113 The first N i The columns are arranged in modular order to form an N×N matrix. Then, the conjugate transpose is taken to obtain the matrix.

[0051] S115. Online optimization of the phase diagonal matrix, using the matrix obtained in step S114. and First, the matrix is ​​deployed to the all-optical matrix processing units described in steps S102 and S106, and then, after several online transmissions and analysis calculations, the phase diagonal matrix is ​​obtained. Optimized value selection.

[0052] like Figure 3 As shown in this embodiment, the specific calculation process of online optimization of the phase diagonal matrix in step S115 is as follows:

[0053] S1151. Search parameter settings: Set the stopping threshold L for the phase optimization loss function. th and maximum number of iterations k max Set the phase search step size Δ;

[0054] If Optimize by defining the sender loss function L. M for: Among them, C (i) It is matrix U in step S104 (i) The indexes of columns where all values ​​are zero form a set;

[0055] If Optimize and define the receiver loss function L. D for: Among them, R (i) It is matrix U in step S104 (i) The indices of rows where all values ​​are zero form a set;

[0056] S1152. Randomly specify the phase transformation order, and randomly specify one... The arrangement of the N phase terms contained therein serves as the optimization order;

[0057] S1153. Adjust the optimized phase terms by adding the phase search step size Δ set in step S1151 to the optimized phase terms, while keeping other phase terms unchanged, and deploy them to the diagonal matrix of the pure phase terms in the all-optical matrix operation unit.

[0058] S1154. Online transmission and loss assessment: Perform steps S111 and S112 sequentially to obtain the system transmission matrix T, and calculate and save the loss function at this time.

[0059] If the phase term being optimized has changed by more than 2π, proceed to the next step; otherwise, repeat steps S1153 and S1154.

[0060] S1155. Select the optimal phase: find the minimum loss function that has appeared so far and its corresponding optimal phase combination;

[0061] If the minimum loss function is less than the stopping threshold L at this point th If the optimization ends, return the current optimal phase combination; otherwise, deploy the optimal phase combination to the diagonal matrix of the pure phase item in the corresponding all-optical matrix operation unit, optimize the next phase item according to the phase item optimization order determined in step S1152, and repeat steps S1153 to S1155 until all phase items have been optimized.

[0062] If all phase terms have completed one round of optimization, repeat steps S1152 to S1155 until the number of iterations exceeds the maximum number of iterations k. max The optimization process ends and the current optimal phase combination is returned.

[0063] like Figure 4 As shown, this invention also proposes an optical fiber communication system with a deployment method for reconfigurable optical fiber mode demultiplexing, comprising:

[0064] The optical transmitter array module S201 is used to implement step S101. It consists of N optical transmitters and generates complex amplitude modulated optical signals according to user definition. The carrier light sources of these N optical transmitters can be the same coherent light source, which can be used to evaluate the demultiplexing crosstalk compensation matrix or actual communication; or they can be different light sources for actual communication.

[0065] The transmitting end all-optical matrix operation unit S202 is used to implement step S102, and is divided into two parts: the first part reconfigurably defines an N×N complex matrix such that the complex amplitude of the output optical signal vector is equal to the product of the matrix and the complex amplitude of the input optical signal vector generated by the optical transmitter array module S201; the second part reconfigurably defines an N×N pure phase diagonal matrix, which superimposes an additional phase on each of the output optical signals of the first part; the all-optical matrix operation unit can be an optical integrated device such as a Mach-Zehnder interferometer network, a micro-ring array, or a spatial optical wavefront conversion device, as well as other units or modules with similar linear matrix conversion functions;

[0066] The transmitting end mode conversion device S203, used to implement step S103, is a passive device that can convert the N optical signals output by the transmitting end all-optical matrix operation unit S202 into optical signals that are multiplexed and transmitted in multiple orthogonal mode channels in a space-division multiplexing fiber. This device can be a fiber-type device such as a photonic lantern or a mode-selective exciter, or a spatial optical device such as a vortex phase plate, an optical wavefront modulator, or a multi-plane optical conversion module, as well as other devices or modules that can realize mode linear conversion and multiplexing functions.

[0067] Space division multiplexing fiber S204 is used to implement step S104. The type of fiber can be multimode or few-mode fiber, or multi-core fiber composed of multiple multimode or few-mode fiber cores, etc., to support the transmission of N optical signals output by the transmitting end mode conversion device S203 in N mode channels; wherein, the form of the mode channel can be the intrinsic vector mode, linear polarization LP mode, orbital angular momentum OAM mode, or other orthogonal basis mode supported in any fiber core, and these channel modes can be divided into different modules according to degeneracy, thereby having the transmission characteristics described in step S104;

[0068] The receiver mode converter S205, used to implement step S103, is a passive device that can separate the beam output from the optical fiber in the space-division multiplexing fiber S204 into N optical signals. This device can be a fiber-type device such as a photonic lantern or a mode-selective exciter, or a spatial optical device such as a vortex phase plate, an optical wavefront modulator, or a multi-plane optical conversion module, as well as other devices or modules that can realize mode linear conversion and multiplexing functions.

[0069] The receiver-side all-optical matrix operation unit S206 is used to implement step S106 and is divided into two parts: the first part reconfigurably defines an N×N pure phase diagonal matrix to superimpose an additional phase on each of the N optical signals output by the receiver-side mode converter S205; the second part reconfigurably defines an N×N complex matrix such that the complex amplitude of the output optical signal vector is equal to the product of this matrix and the complex amplitude of the output optical signal vector in the first part; the all-optical matrix operation unit can be an optical integrated device such as a Mach-Zehnder interferometer network, a micro-ring array, or a spatial optical wavefront conversion device, as well as other units or modules with similar linear matrix conversion functions;

[0070] The coherent receiving array module S207 is used to realize the signal reception in step S107. It consists of N single-polarization coherent optical receivers or N / 2 dual-polarization coherent optical receivers and a reference light source with the same frequency as the carrier. It can convert the complex amplitude of the optical signal output by the all-optical matrix operation unit S206 at the receiving end into an electrical signal.

[0071] The signal processing and control unit S208 is used to implement the signal processing in step S107 and the processes described in steps S111 to S115. In actual communication, it is used to perform small-scale MIMO operations and other signal processing on the optical signals of each module to obtain the information transmitted by the source. When evaluating the demultiplexing crosstalk compensation matrix, only the value of each information symbol is extracted, without performing MIMO operations and other signal processing. Then, the algorithm of steps S111 to S115 is executed to provide feedback control of the optical transmitter array module S201, the transmitting end all-optical matrix operation unit S202, and the receiving end all-optical matrix operation unit S206, which are system components. The function of this unit can be implemented by hardware such as field-programmable gate arrays, application-specific integrated circuits, microcontrollers, and all-optical signal processing chips, or by software programs on a computer, as well as other devices or modules that can implement this signal processing and control function.

[0072] Specifically, the N-channel optical delay tolerances connected at both the N-channel output terminals of the transmitting end all-optical matrix operation unit S202 to the input terminal of the transmitting end mode conversion device S203, and the output terminal of the receiving end all-optical matrix operation unit S206 to the input terminal of the coherent receiving array module S207, should be much smaller than the duration of one information symbol.

[0073] Specifically, when the space division multiplexing fiber S204 is a multi-core fiber, it is only necessary to expand the scale of each device and module from the optical transmitter array module S201 to the transmitting end mode converter S203, the receiving end mode converter S205 to the signal processing and control unit S208 according to the number of fiber cores and the number of modes accommodated in each fiber core, so that the method described in any one of steps S101 to S107 and steps S111 to S115 can be implemented independently for each fiber core.

[0074] The above embodiments are preferred embodiments of the present invention, but the embodiments of the present invention are not limited to the above embodiments. Any changes, modifications, substitutions, combinations, or simplifications made without departing from the spirit and principle of the present invention shall be considered equivalent substitutions and shall be included within the protection scope of the present invention.

Claims

1. A deployment method for reconfigurable fiber mode demultiplexing, characterized in that, Includes the following steps: S101, Signal generation: The signal source emits N electrical signal vectors s TX0 By performing complex amplitude modulation on the light field, the optical signal matrix s is obtained. TX ; S102. Multiplexing crosstalk compensation, the optical signal matrix s obtained in step S101 is... TX The input is fed into the all-optical matrix operation unit, multiplied by the N×N mode multiplexing crosstalk compensation matrix M′, and outputs N optical signal vectors s′. TX =M's TX ; S103, Mode multiplexing, converting the N optical signal vectors s′ obtained in step S102 into multiplexed signals. TX Mode multiplexing is achieved by coupling the N modes into the space-division multiplexed fiber channel through a spatial mode converter, denoted as s. channel =Ms′ TX =MM's TX , where s channel This represents the signal matrix of N optical field modes loaded into the mode-division multiplexing fiber before transmission, where M is a random N×N mode multiplexing crosstalk matrix. S104. Fiber optic transmission: The signal matrix s of the N optical field modes that have not yet been transmitted and are loaded into the mode-division multiplexing fiber obtained in step S103 is transferred. channel When propagating in few-mode or multimode optical fibers, the output at the receiving end is a signal matrix s′ of N optical field modes that has been transmitted in the mode-division multiplexing fiber. channel The crosstalk matrix between different modules of few-mode or multimode optical fibers is represented as the optical fiber channel transmission crosstalk matrix U. ′ ; S105. Mode demultiplexing: The N optical field mode signal matrices s′ obtained in step S104 and transmitted into the mode-division multiplexing fiber are demultiplexed. channel N optical signals are separated by another spatial mode converter, completing mode demultiplexing, denoted as s′. RX =Ds′ channel ; where s′ RX This represents the N independent optical signals output from the mode-division multiplexing fiber and obtained after mode demultiplexing, where D is a random N×N mode demultiplexing crosstalk matrix. S106. Demultiplexing crosstalk compensation: The N independent optical signals s′ obtained in step S105, which were output from the mode-division multiplexing fiber and demultiplexed, are then demultiplexed. RX The input is fed into another all-optical matrix operation unit, multiplied by the N×N mode demultiplexing crosstalk compensation matrix D′, and outputs N optical signal vectors s. RX =D′s′ RX =D′Ds′ channel ; S107. Signal reception and processing: Coherent detection converts the N-channel optical signal vectors s obtained in step S106 into... RX Converted to N electrical signals s RX0 After signal processing, it can be recovered to the original electrical signal vector s. TX0 .

2. The deployment method for reconfigurable fiber mode demultiplexing according to claim 1, characterized in that, In step S102, the mode multiplexing crosstalk compensation matrix M′ should satisfy the following relationship: The product MM′ of the mode multiplexing crosstalk matrix M and M′ in step S103 is an N×N block diagonal matrix, which has the same distribution of non-zero sub-blocks as the optical fiber channel transmission crosstalk matrix U′ in step S104.

3. The deployment method for reconfigurable fiber mode demultiplexing according to claim 1, characterized in that, In step S106, the mode demultiplexing crosstalk compensation matrix D′ should satisfy the following relationship: the product D′D of D′ and the mode demultiplexing crosstalk matrix D in step S105 is an N×N block diagonal matrix, which has the same distribution of non-zero sub-blocks as the optical fiber channel transmission crosstalk matrix U′ in step S104.

4. The deployment method for reconfigurable fiber mode demultiplexing according to claim 1, characterized in that, In step S102, the mode multiplexing crosstalk compensation matrix In step S106, the mode demultiplexing crosstalk compensation matrix in, and It is the phase diagonal matrix to be optimized.

5. The deployment method for reconfigurable fiber mode demultiplexing according to claim 1, characterized in that, The specific process for evaluating the mode multiplexing crosstalk matrix M and the mode demultiplexing crosstalk matrix D is as follows: S111. Define an identity matrix signal. Let the number of mode channels in a space division multiplexing fiber be N. Define N signals, each containing N information symbols, represented by an N×N matrix. The p-row and q-column elements of the matrix represent the value of the q-th information symbol in the p-th signal. The total duration of the N information symbols is less than the minimum differential module delay in the fiber channel. The matrix signal is defined as a matrix signal I represented by an identity matrix. S112. Online transmission to obtain the transmission matrix: Using the matrix signal I defined in step S111 as the original signal, implement steps S101 to S107 to extract the N×N transmission matrix {T} of each module. (1) T (2) ...T (m) }, where T (i) ,i=1,2,...,m represents the transmission matrix of the i-th module; S113. Transmission matrix analysis: For each T obtained in step S112... (i) Perform singular value decomposition on i = 1, 2, ..., m: in,(·) H Σ represents the conjugate transpose of a matrix. (i) It is a diagonal matrix composed of non-negative real numbers, with the diagonal elements arranged in descending order from top left to bottom right. All are unitary matrices; S114. Construct the crosstalk compensation matrix. Let the i-th module contain N... i In each mode, the following operation is performed on all modules: The result obtained in step S113 is... The first N i The columns are arranged in modular order to form an N×N matrix. Take the result obtained in step S113 The first N i The columns are arranged in modular order to form an N×N matrix. Then, the conjugate transpose is taken to obtain the matrix. S115. Online optimization of the phase diagonal matrix, using the matrix obtained in step S114. and First, the matrix is ​​deployed to the all-optical matrix processing unit described in steps S102 and S106, and then, after several online transmissions and analysis calculations, the phase diagonal matrix is ​​obtained. Optimized value selection.

6. The deployment method for reconfigurable fiber mode demultiplexing according to claim 5, characterized in that, The specific calculation process for online optimization of the phase diagonal matrix in step S115 is as follows: S1151. Search parameter settings: Set the stopping threshold L for the phase optimization loss function. th and maximum number of iterations k max Set the phase search step size Δ; If Optimize by defining the sender loss function L. M for: Among them, C (i) It is a set composed of the indices of all zero columns in matrix U′ in step S104; If Optimize and define the receiver loss function L. D for: Among them, R (i) It is a set composed of the indices of the rows in matrix U′ that are all zero in step S104; S1152. Randomly specify the phase transformation order, and randomly specify one... The arrangement of the N phase terms contained therein serves as the optimization order; S1153. Adjust the optimized phase terms by adding the phase search step size Δ set in step S1151 to the optimized phase terms, while keeping other phase terms unchanged, and deploy them to the diagonal matrix of the pure phase terms in the all-optical matrix operation unit. S1154. Online transmission and loss assessment: Perform steps S111 and S112 sequentially to obtain the system transmission matrix T, and calculate and save the loss function at this time. If the change in the current optimized phase term exceeds 2π, proceed to the next step; otherwise, repeat steps S1153 and S1154. S1155. Select the optimal phase and choose the minimum loss function and its corresponding optimal phase combination. If the minimum loss function is less than the stopping threshold L at this point th If the optimization ends, return the current optimal phase combination; otherwise, deploy the optimal phase combination to the diagonal matrix of the pure phase item in the corresponding all-optical matrix operation unit, optimize the next phase item according to the phase item optimization order determined in step S1152, and repeat steps S1153 to S1155 until all phase items are optimized. If all phase terms have completed one round of optimization, repeat iteration steps S1152 to S1155 until the number of iterations exceeds the maximum number of iterations k. max The optimization process ends and the current optimal phase combination is returned.

7. The optical fiber communication system of the deployment method for reconfigurable optical fiber mode demultiplexing according to claim 6, characterized in that, include: The optical transmitter array module S201 is used to implement step S101. It consists of N optical transmitters and generates complex amplitude modulated optical signals according to user definition. The carrier light sources of these N optical transmitters are the same coherent light source or different light sources. It is used to evaluate the demultiplexing crosstalk compensation matrix or actual communication. The transmitting end all-optical matrix operation unit S202 is used to implement step S102, which is divided into two parts: the first part reconfigurably defines an N×N complex matrix such that the complex amplitude of the output optical signal vector is equal to the product of the matrix and the complex amplitude of the input optical signal vector generated by the optical transmitter array module S201. The second part reconfigurably defines an N×N pure phase diagonal matrix, which superimposes an additional phase onto the output optical signal of the first part; the all-optical matrix operation unit is composed of a Mach-Zehnder interferometer network, a micro-ring array optical integrated device, or a spatial optical wavefront conversion device, as well as other units or modules with linear matrix conversion functions; The transmitting end mode conversion device S203, used to implement step S103, is a passive device that converts the N optical signals output by the transmitting end all-optical matrix operation unit S202 into optical signals that are multiplexed and transmitted in multiple orthogonal mode channels in a space-division multiplexing fiber. This device can be a photonic lantern, a mode-selective exciter fiber-type device, or a vortex phase plate, an optical wavefront modulator, a multi-plane optical conversion module spatial optical device, as well as other devices or modules with linear mode conversion and multiplexing functions. Space division multiplexing fiber S204 is used to implement step S104. The type of fiber is multimode or few-mode fiber, or multi-core fiber composed of multiple multimode or few-mode fiber cores. It supports the transmission of N optical signals output by the transmitting end mode conversion device S203 in N mode channels. The mode channel is in the form of an intrinsic vector mode, linear polarization LP mode, orbital angular momentum OAM mode, or other orthogonal basis mode supported in any fiber core. These channel modes are divided into different modules according to degeneracy, thus having the transmission characteristics of step S104. The receiver mode converter S205, used to implement step S105, is a passive device that separates the beam output from the optical fiber in the space-division multiplexing fiber S204 into N optical signals. This device can be a photonic lantern, a mode-selective exciter fiber-type device, or a vortex phase plate, an optical wavefront modulator, a multi-plane optical conversion module spatial optical device, as well as other devices or modules with mode linear conversion and multiplexing functions. The receiver all-optical matrix operation unit S206 is used to implement step S106, which is divided into two parts: the first part reconfigurably defines an N×N pure phase diagonal matrix, which superimposes an additional phase on each of the N optical signals output by the receiver mode converter S205. The second part reconfigurably defines an N×N complex matrix such that the complex amplitude of the output optical signal vector is equal to the product of the matrix and the complex amplitude of the output optical signal vector of the first part; the all-optical matrix operation unit can be a Mach-Zehnder interferometer network, a micro-ring array optical integrated device, or a spatial optical wavefront conversion device, as well as other units or modules with linear matrix conversion functions; The coherent receiving array module S207 is used to realize the signal reception in step S107. It consists of N single-polarization coherent optical receivers or N / 2 dual-polarization coherent optical receivers and a reference light source with the same frequency as the carrier. It converts the complex amplitude of the optical signal output by the all-optical matrix operation unit S206 at the receiving end into an electrical signal. The signal processing and control unit S208 is used to implement the signal processing in step S107 and the processes in steps S111 to S115. In actual communication, it is used to perform small-scale MIMO operations and other signal processing on the optical signals of each module to obtain the information transmitted by the source. When evaluating the demultiplexing crosstalk compensation matrix, it extracts the value of each information symbol and then executes steps S111 to S115 to provide feedback control of the optical transmitter array module S201, the transmitting end all-optical matrix operation unit S202, and the receiving end all-optical matrix operation unit S206, which are components of the system.

8. The optical fiber communication system according to the deployment method of reconfigurable optical fiber mode demultiplexing as described in claim 7, characterized in that, The N-channel output of the transmitting end all-optical matrix operation unit S202 is connected to the input of the transmitting end mode conversion device S203, and the output of the receiving end all-optical matrix operation unit S206 is connected to the input of the coherent receiving array module S207. The delay tolerance of the N-channel optical signals connected at both locations is less than the duration of one information symbol.

9. The optical fiber communication system of the deployment method for reconfigurable optical fiber mode demultiplexing according to claim 7, characterized in that, When the space division multiplexing fiber S204 uses multi-core fiber, the scale of each device and module, from the optical transmitter array module S201 to the transmitting end mode converter S203, and from the receiving end mode converter S205 to the signal processing and control unit S208, is expanded according to the number of fiber cores and the number of modes accommodated in each fiber core.

Images