A method and system for optical fiber communication based on an adversarial mode chimpanzee algorithm

Abstract

The application discloses a kind of optical fiber communication method and system based on anti-pattern gorilla algorithm, comprising: obtaining original bit data to be sent;Original bit data is mapped to constellation diagram by quadrature amplitude modulation mapping algorithm and converted into optical fiber transmission signal;Optical fiber transmission signal is transmitted to receiving end by few-mode channel;The received optical fiber transmission signal is converted into receiving bit data by analog-digital conversion, and receiving bit data is decomposed into leading receiving information and text receiving information;Mode coupling matrix and mode transmission matrix are generated from leading receiving information using anti-pattern gorilla algorithm;Text receiving information is corrected using mode coupling matrix and mode transmission matrix to obtain correction signal, and correction signal is demapped and converted into original bit data again;Reduce the waste of computing resources caused by mode equalization technology and the equalization deviation caused, improve the quality of quadrature mode signal transmission.

Description

Technical Field

[0001] This invention belongs to the field of optical fiber communication technology, specifically relating to an optical fiber communication method and system based on the adversarial mode chimpanzee algorithm. Background Technology

[0002] In the field of optical communication, traditional single-mode fiber has excellent low-loss characteristics. However, with the rapid development of digitization leading to a continuous expansion of data traffic, single-mode fiber, due to its limited capacity, is increasingly unable to meet the growing transmission demands. Few-mode multiplexing technology has emerged to address this need, significantly improving the transmission capacity of optical fibers by simultaneously transmitting multiple modes within the same fiber.

[0003] However, in actual long-distance transmission, the inherent non-ideal characteristics of optical fibers, such as bending, temperature changes, and subtle differences in manufacturing processes, can lead to mode field degradation, thereby disrupting the orthogonality between modes. When the orthogonality between modes is disrupted, the signals carried by each mode interfere with each other, resulting in an increased data error rate and severe signal distortion, which greatly reduces the reliability of the communication system and the quality of signal transmission.

[0004] To address this challenge, mode equalization techniques are crucial. Traditional blind equalization techniques do not require training sequences, which saves spectrum resources, but the algorithm complexity is extremely high when dealing with mode interference in few-mode fibers. This not only consumes a large amount of computing resources and increases equipment costs, but also results in slow processing speeds, making it difficult to meet the real-time requirements of high-speed data transmission. Summary of the Invention

[0005] This invention provides an optical fiber communication method and system based on the adversarial mode chimpanzee algorithm, which reduces the waste of computing resources and the resulting equalization deviation caused by mode equalization technology, and improves the transmission quality of orthogonal mode signals.

[0006] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0007] The first aspect of this invention provides a fiber optic communication method based on the adversarial mode chimpanzee algorithm, comprising:

[0008] The process involves acquiring the raw bit data to be transmitted; mapping the raw bit data to a constellation diagram using an orthogonal amplitude modulation mapping algorithm, and then converting it into an optical fiber transmission signal; finally, transmitting the optical fiber transmission signal to the receiver via a few-mode channel.

[0009] The received fiber optic transmission signal is converted from analog to digital to obtain received bit data, and the received bit data is decomposed into preamble received information and text received information.

[0010] The adversarial mode chimpanzee algorithm is used to generate the mode coupling matrix and mode transmission matrix from the preamble reception information.

[0011] The received text information is corrected using the mode coupling matrix and the mode transmission matrix to obtain a correction signal, which is then demapped and converted back into the original bit data.

[0012] Furthermore, the original bit data is mapped to a constellation diagram using an orthogonal amplitude modulation mapping algorithm and then converted into an optical fiber transmission signal. Specifically, this includes:

[0013] The original bit data is orthogonally amplitude-modulated to obtain orthogonal frequency division multiplexing (OFDM) symbols; the OFDM symbols are then subjected to inverse fast Fourier transform (IFFT) to obtain the time-domain signal, expressed as:

[0014]

[0015] In the formula, For time-domain signals; j is the imaginary unit; N is the number of subcarriers; k is the discrete frequency index of the orthogonal frequency division multiplexing symbol block; and n is the discrete time index of the time-domain signal. This is an orthogonal frequency division multiplexing symbol; Pi;

[0016] The time-domain signal is subjected to amplitude limiting and parallel-to-serial conversion to obtain parallel transmission data. After adding a cyclic prefix to the parallel transmission data, it is converted from digital to analog to obtain the fiber optic transmission signal.

[0017] Furthermore, the adversarial mode chimpanzee algorithm is used to generate the mode coupling matrix and mode transfer matrix from the preamble reception information, specifically including:

[0018] Equalizer parameters are randomly generated using a generator. and channel estimates ; by equalizer parameters Construct the mode coupling matrix from the channel estimate. Construct a mode transmission matrix;

[0019] The chimpanzee population is obtained by combining the pattern coupling matrix and the pattern transfer matrix to represent individual chimpanzees, and the chimpanzee population contains M individual chimpanzees.

[0020] The preamble received information is denoised and used as a training sequence. Real labels are added to the training sequence based on the preamble preset information.

[0021] The training sequence is corrected using the mode coupling matrix and mode transfer matrix to obtain leading correction information. The fitness value MSE is then calculated using the leading correction information and the true label. The formula is as follows:

[0022]

[0023] In the formula, s represents the true label of the training sequence; Preceding correction information;

[0024] The preceding correction information is input into the discriminator to output the training evaluation value, and the training sequence is input into the discriminator to output the training true value; the adversarial loss is calculated based on the training evaluation value and the training true value.

[0025] Determine whether the adversarial loss has converged. If the adversarial loss has not converged, update the position of the chimpanzee individual based on the fitness value MSE. Repeat the update process of the chimpanzee population until the adversarial loss converges, and output the mode coupling matrix and mode transfer matrix corresponding to the chimpanzee individual with the best fitness value MSE.

[0026] Furthermore, the training sequence is corrected based on the mode coupling matrix and the mode transfer matrix to obtain leading correction information, specifically including:

[0027] The training sequence is input into the equalizer to construct a transfer compensation matrix that is the inverse of the mode transfer matrix, expressed by the following formula:

[0028]

[0029] In the formula, I is the identity matrix; For transmission compensation matrix; For the mode transfer matrix; This is the conjugate transpose of the mode transfer matrix H; For noise variance;

[0030] The amplitude and phase of the training sequence are compensated using a transmission compensation matrix to obtain preamble compensation information; the formula is as follows:

[0031]

[0032]

[0033] In the formula, r represents additive white Gaussian noise; y represents the preamble received information; s represents the preamble preset information; and z represents the preamble compensation information.

[0034] The leader correction information is obtained by eliminating the local coupling of the leader compensation information through the mode coupling matrix.

[0035] Furthermore, the leader correction information is obtained by eliminating local coupling of the leader compensation information through the mode coupling matrix, specifically including:

[0036] Based on the amplitude, the elements in the mode coupling matrix are divided into degenerate module sub-blocks, high- and low-order module sub-blocks, and inter-group coupling sub-blocks; the average coupling coefficient and coupling variance of the degenerate module sub-blocks, high- and low-order module sub-blocks, and inter-group coupling sub-blocks are calculated;

[0037] The sparsity order of degenerate module sub-blocks, high- and low-order module sub-blocks, and inter-group coupling sub-blocks is determined based on the average coupling coefficient and coupling variance, thereby obtaining the degenerate module sub-matrix, high- and low-order module sub-matrix, and inter-group sub-matrix.

[0038] The joint MIMO equilibrium matrix is ​​constructed from the degenerate module submatrix, high- and low-order module submatrix, and inter-group submatrix, and is expressed by the following formula:

[0039]

[0040] In the formula, For the joint MIMO equilibrium matrix; For degenerate module submatrices; These are high- and low-order module submatrices; and For inter-group submatrices;

[0041] The joint MIMO equalization matrix eliminates local coupling of the leader compensation information to obtain the leader correction information, expressed by the following formula:

[0042] , , ;

[0043] , , ;

[0044] In the formula, To eliminate the intermediate signal after strong coupling within the group of the degenerate mode receiver sub-signal; To suppress the intermediate signal after weak coupling within the high- and low-order mode receiving sub-signals; and For degenerate mode receiver sub-signals and high- and low-order mode receiver sub-signals in the preamble compensation information; and Inter-group interference; and This refers to the degenerate mode receiver sub-signal and the high- and low-order mode receiver sub-signal in the preamble correction information.

[0045] Furthermore, the location of individual chimpanzees is updated based on their fitness value (MSE), specifically including:

[0046] Chimpanzee individuals in the chimpanzee population were categorized into chasers, obstacle chasers, pursuers, and attackers based on their fitness values ​​(MSE). The positions of chasers were updated by adding perturbations.

[0047] The positions of the obstacle and the pursuer are updated based on the position of the chaser, expressed by the following formula:

[0048]

[0049]

[0050]

[0051]

[0052] In the formula, and Let be the position vectors of the obstacle in the (t+1)th and tth iterations; Let be the position vector of the driver in the t-th iteration; and Let A be the position vector of the obstacle in the (t+1)th and tth iterations; A and B are dynamic coefficients. To set a constant; represents the standard deviation of the fitness value MSE in the chimpanzee population; The bias weight of the obstacle relative to the driver; for Random numbers within a range; This represents the average fitness value (MSE) within the chimpanzee population.

[0053] The attacker's position is updated by performing a local search within the range from the driver to the obstacle, expressed as follows:

[0054]

[0055] In the formula, and Let be the attacker's position vector in the (t+1)th and tth iterations; This is a random disturbance term.

[0056] Furthermore, adversarial loss is calculated based on the training evaluation values ​​and the actual training values, specifically including:

[0057]

[0058] In the formula, To combat the losses; Let it be the expected function; The training ground truth values ​​output by the discriminator; The training evaluation value output by the discriminator.

[0059] Furthermore, the corrected signal is demapped and converted back into the original bit data, specifically including:

[0060] After removing the cyclic prefix from the correction signal, a serial-to-parallel conversion is performed to obtain the time-domain signal sequence.

[0061] The process of converting a time-domain signal sequence into an orthogonal frequency division multiplexing (OFDM) symbol block using Fast Fourier Transform is as follows:

[0062]

[0063] In the formula, is the time-domain signal sequence; j is the imaginary unit; N is the number of subcarriers; k is the discrete frequency index of the orthogonal frequency division multiplexing symbol block; and n is the discrete time index of the time-domain signal. For orthogonal frequency division multiplexing symbol blocks; Pi is the mathematical constant of a circle.

[0064] The original bit data is obtained by orthogonal amplitude modulation demapping of the orthogonal frequency division multiplexing symbol block.

[0065] A second aspect of the present invention provides an optical fiber communication system based on the adversarial mode chimpanzee algorithm, comprising:

[0066] The signal transmitting unit is used to acquire the raw bit data to be transmitted; after mapping the raw bit data to a constellation diagram using an orthogonal amplitude modulation mapping algorithm, it is converted into an optical fiber transmission signal; and the optical fiber transmission signal is transmitted to the receiving end through a few-mode channel.

[0067] The decomposition unit is used to perform analog-to-digital conversion on the received optical fiber transmission signal to obtain received bit data, and decompose the received bit data into preamble received information and text received information.

[0068] The correction unit is used to generate a mode coupling matrix and a mode transfer matrix from the preamble received information using the adversarial mode chimpanzee algorithm; and to obtain a correction signal by correcting the main text received information using the mode coupling matrix and the mode transfer matrix.

[0069] The receiving unit is used to demap the correction signal and convert it back into the original bit data.

[0070] A third aspect of the present invention provides an electronic terminal, characterized in that it includes a processor and a storage medium; the storage medium is used to store instructions; the processor is used to operate according to the instructions to execute the steps of the optical fiber communication method of the first aspect.

[0071] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0072] This invention performs analog-to-digital conversion on the received fiber optic transmission signal to obtain received bit data, which is then decomposed into preamble received information and text received information. An adversarial mode chimpanzee algorithm is used to generate a mode coupling matrix and a mode transmission matrix from the preamble received information. Based on the obtained mode transmission matrix, a first-stage non-blind joint equalization is performed on the received signal to compensate for mode power fading and mode nonlinearity impairments. Based on the mode coupling matrix, accurate MIMO (Multiple-Input Multiple-Output) equalization matrix order allocation and filter design can be achieved, improving the MIMO joint equalization matching degree. Simultaneously, precise compensation for impairments such as mode coupling is achieved, reducing the waste of computational resources and equalization deviations caused by blind large-scale MIMO.

[0073] This invention can recalculate the channel matrix in real time based on each data frame or periodically transmitted preamble signals, thereby dynamically tracking channel changes and achieving adaptive equalization. It maintains high performance and stronger robustness even in complex, time-varying real-world environments. By correcting the received text information using the mode coupling matrix and mode transmission matrix to obtain a correction signal, crosstalk is suppressed, the bit error rate is reduced to an acceptable level, and communication reliability is guaranteed. Attached Figure Description

[0074] Figure 1 This is a flowchart of optical fiber communication provided in Embodiment 1 of the present invention;

[0075] Figure 2 This is a flowchart of the non-blind joint MIMO equalization algorithm provided in Embodiment 1 of the present invention;

[0076] Figure 3 This is a flowchart of the adversarial mode chimpanzee algorithm provided in Embodiment 1 of the present invention. Detailed Implementation

[0077] The present invention will be further described below with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and should not be used to limit the scope of protection of the present invention.

[0078] Example 1

[0079] like Figure 1 As shown, this embodiment provides a hollow-core optical fiber communication method for reducing complexity, including:

[0080] The process involves acquiring the raw bit data to be transmitted; mapping the raw bit data to a constellation diagram using an orthogonal amplitude modulation mapping algorithm, and then converting it into an optical fiber transmission signal. Specifically, this includes:

[0081] The original bit data is mapped using orthogonal amplitude modulation (16QAM modulation in this embodiment) to obtain orthogonal frequency division multiplexing (OFDM) symbols; the OFDM symbols are then subjected to inverse fast Fourier transform (IFFT) to obtain the time-domain signal, expressed as follows:

[0082]

[0083] In the formula, For time-domain signals; j is the imaginary unit; N is the number of subcarriers; k is the discrete frequency index of the orthogonal frequency division multiplexing symbol block; and n is the discrete time index of the time-domain signal. This is an orthogonal frequency division multiplexing symbol; Pi;

[0084] The time-domain signal is subjected to amplitude limiting and parallel-to-serial conversion to obtain parallel transmission data, which is serial data transmitted at high speed on one line. This data is then distributed to multiple lines for parallel transmission at a lower rate. A cyclic prefix is ​​added to the parallel transmission data, and then digital-to-analog conversion is performed to obtain the fiber optic transmission signal. The length of the cyclic prefix is ​​usually greater than the maximum delay spread of the channel. The fiber optic transmission signal is then transmitted to the receiving end through a few-mode channel.

[0085] like Figure 2 As shown, the received fiber optic transmission signal is converted from analog to digital to obtain received bit data, and the received bit data is decomposed into preamble received information and text received information.

[0086] like Figure 3 As shown, the adversarial mode chimpanzee algorithm is used to generate the mode coupling matrix and mode transfer matrix from the preamble reception information, specifically including:

[0087] Equalizer parameters are randomly generated using a generator. and channel estimates ; by equalizer parameters Construct the mode coupling matrix from the channel estimate. Construct a mode transmission matrix;

[0088] The chimpanzee population is obtained by combining the pattern coupling matrix and the pattern transfer matrix to represent individual chimpanzees, and the chimpanzee population contains M individual chimpanzees.

[0089] The preamble received information is denoised and used as a training sequence. Real labels are added to the training sequence based on the preamble preset information.

[0090] The training sequence is corrected based on the mode coupling matrix and the mode transfer matrix to obtain leading correction information, specifically including:

[0091] The training sequence is input into the equalizer to construct a transfer compensation matrix that is the inverse of the mode transfer matrix, expressed by the following formula:

[0092]

[0093] In the formula, I is the identity matrix; For transmission compensation matrix; For the mode transfer matrix; This is the conjugate transpose of the mode transfer matrix H; For noise variance;

[0094] The amplitude and phase of the training sequence are compensated using a transmission compensation matrix to obtain preamble compensation information; the formula is as follows:

[0095]

[0096]

[0097] In the formula, r represents additive white Gaussian noise; y represents the preamble received information; s represents the preamble preset information; and z represents the preamble compensation information.

[0098] Based on the amplitude, the elements in the mode coupling matrix are divided into degenerate module sub-blocks, high- and low-order module sub-blocks, and inter-group coupling sub-blocks; the average coupling coefficient and coupling variance of the degenerate module sub-blocks, high- and low-order module sub-blocks, and inter-group coupling sub-blocks are calculated;

[0099] The sparsity order of degenerate module sub-blocks, high- and low-order module sub-blocks, and inter-group coupling sub-blocks is determined based on the average coupling coefficient and coupling variance, thereby obtaining the degenerate module sub-matrix, high- and low-order module sub-matrix, and inter-group sub-matrix.

[0100] The joint MIMO equilibrium matrix is ​​constructed from the degenerate module submatrix, high- and low-order module submatrix, and inter-group submatrix, and is expressed by the following formula:

[0101]

[0102] In the formula, For the joint MIMO equilibrium matrix; For degenerate module submatrices; These are high- and low-order module submatrices; and For inter-group submatrices;

[0103] The joint MIMO equalization matrix eliminates local coupling of the leader compensation information to obtain the leader correction information, expressed by the following formula:

[0104] , , ;

[0105] , , ;

[0106] In the formula, To eliminate the intermediate signal after strong coupling within the group of the degenerate mode receiver sub-signal; To suppress the intermediate signal after weak coupling within the high- and low-order mode receiving sub-signals; and For degenerate mode receiver sub-signals and high- and low-order mode receiver sub-signals in the preamble compensation information; and Inter-group interference; and This refers to the degenerate mode receiver sub-signal and the high- and low-order mode receiver sub-signal in the preamble correction information.

[0107] The fitness value MSE is calculated using prior correction information and the true label; the formula is as follows:

[0108]

[0109] In the formula, s represents the true label of the training sequence; Preceding correction information;

[0110] The preceding correction information is input to the discriminator to output the training evaluation value, and the training sequence is input to the discriminator to output the training ground value. The adversarial loss is calculated based on the training evaluation value and the training ground value, specifically including:

[0111]

[0112] In the formula, To combat the losses; Let it be the expected function; The training ground truth values ​​output by the discriminator; The training evaluation value output by the discriminator.

[0113] To determine if the adversarial loss has converged, if it has not, update the chimpanzee's position based on the fitness value MSE, specifically including:

[0114] Chimpanzee individuals in the chimpanzee population were categorized into chasers, obstacle chasers, pursuers, and attackers based on their fitness values ​​(MSE). The positions of chasers were updated by adding perturbations.

[0115] The positions of the obstacle and the pursuer are updated based on the position of the chaser, expressed by the following formula:

[0116]

[0117]

[0118]

[0119]

[0120] In the formula, and Let be the position vectors of the obstacle in the (t+1)th and tth iterations; Let be the position vector of the driver in the t-th iteration; and Let A be the position vector of the obstacle in the (t+1)th and tth iterations; A and B are dynamic coefficients. To set a constant; represents the standard deviation of the fitness value MSE in the chimpanzee population; The bias weight of the obstacle relative to the driver; for Random numbers within a range; This represents the average fitness value (MSE) within the chimpanzee population.

[0121] The obstacle acts as a limit or constraint, preventing signal distortion or excessive bit error rate during transmission; the chaser can quickly adjust and optimize the equalizer parameters based on the information provided by the chaser and the constraints set by the obstacle, and react quickly to channel changes to achieve accurate signal recovery.

[0122] The attacker's position is updated by performing a local search within the range from the driver to the obstacle, expressed as follows:

[0123]

[0124] In the formula, and Let be the attacker's position vector in the (t+1)th and tth iterations; This is a random disturbance term.

[0125] The attacker undertakes the key decision-making and optimization tasks, taking into account the channel information monitored by the driver and the constraints set by the obstacle, and optimizes and improves the equalization algorithm to enhance the signal recovery effect.

[0126] Repeat the update process of the chimpanzee population iteratively until the adversarial loss converges, and output the pattern coupling matrix and pattern transfer matrix corresponding to the chimpanzee individual with the best fitness value MSE.

[0127] The received text information is corrected using the mode coupling matrix and mode transmission matrix to obtain a correction signal. The correction signal is then demapped and converted back into the original bit data. Specifically, this includes:

[0128] After removing the cyclic prefix from the correction signal, a serial-to-parallel conversion is performed to obtain the time-domain signal sequence.

[0129] The process of converting a time-domain signal sequence into an orthogonal frequency division multiplexing (OFDM) symbol block using Fast Fourier Transform is as follows:

[0130]

[0131] In the formula, is the time-domain signal sequence; j is the imaginary unit; N is the number of subcarriers; k is the discrete frequency index of the orthogonal frequency division multiplexing symbol block; and n is the discrete time index of the time-domain signal. For orthogonal frequency division multiplexing symbol blocks; Pi is the mathematical constant of a circle.

[0132] The original bit data is obtained by orthogonal amplitude modulation demapping of the orthogonal frequency division multiplexing symbol block.

[0133] This invention enables the first-stage non-blind joint equalization of the received signal based on the obtained mode transmission matrix, compensating for mode power fading and mode nonlinearity impairments. Based on the mode coupling matrix, accurate MIMO (Multiple-Input Multiple-Output) equalization matrix order allocation and filter design can be achieved, improving the MIMO joint equalization matching degree. Simultaneously, it accurately compensates for impairments such as mode coupling, reducing the waste of computational resources and equalization deviations caused by blind large-scale MIMO.

[0134] This invention can recalculate the channel matrix in real time based on each data frame or periodically transmitted preamble signals, thereby dynamically tracking channel changes and achieving adaptive equalization. It maintains high performance and stronger robustness even in complex, time-varying real-world environments. By correcting the received text information using the mode coupling matrix and mode transmission matrix to obtain a correction signal, crosstalk is suppressed, the bit error rate is reduced to an acceptable level, and communication reliability is guaranteed.

[0135] Example 2

[0136] This embodiment provides an optical fiber communication system based on the adversarial mode chimpanzee algorithm. The optical fiber communication system is used to execute the optical fiber communication method described in Embodiment 1. The optical fiber communication system includes:

[0137] The signal transmitting unit is used to acquire the raw bit data to be transmitted; after mapping the raw bit data to a constellation diagram using an orthogonal amplitude modulation mapping algorithm, it is converted into an optical fiber transmission signal; and the optical fiber transmission signal is transmitted to the receiving end through a few-mode channel.

[0138] The decomposition unit is used to perform analog-to-digital conversion on the received optical fiber transmission signal to obtain received bit data, and decompose the received bit data into preamble received information and text received information.

[0139] The correction unit is used to generate a mode coupling matrix and a mode transfer matrix from the preamble received information using the adversarial mode chimpanzee algorithm; and to obtain a correction signal by correcting the main text received information using the mode coupling matrix and the mode transfer matrix.

[0140] The receiving unit is used to demap the correction signal and convert it back into the original bit data.

[0141] The correction unit utilizes the adversarial mode chimpanzee algorithm to generate a mode coupling matrix and a mode transmission matrix from the preamble reception information, specifically including:

[0142] Equalizer parameters are randomly generated using a generator. and channel estimates ; by equalizer parameters Construct the mode coupling matrix from the channel estimate. Construct a mode transmission matrix;

[0143] The chimpanzee population is obtained by combining the pattern coupling matrix and the pattern transfer matrix to represent individual chimpanzees, and the chimpanzee population contains M individual chimpanzees.

[0144] The preamble received information is denoised and used as a training sequence. Real labels are added to the training sequence based on the preamble preset information.

[0145] The training sequence is corrected using the mode coupling matrix and mode transfer matrix to obtain leading correction information. The fitness value MSE is then calculated using the leading correction information and the true label. The formula is as follows:

[0146]

[0147] In the formula, s represents the true label of the training sequence; Preceding correction information;

[0148] The preceding correction information is input into the discriminator to output the training evaluation value, and the training sequence is input into the discriminator to output the training true value; the adversarial loss is calculated based on the training evaluation value and the training true value.

[0149] Determine whether the adversarial loss has converged. If the adversarial loss has not converged, update the position of the chimpanzee individual based on the fitness value MSE. Repeat the update process of the chimpanzee population until the adversarial loss converges, and output the mode coupling matrix and mode transfer matrix corresponding to the chimpanzee individual with the best fitness value MSE.

[0150] Example 3

[0151] This embodiment provides an electronic terminal, characterized in that it includes a processor and a storage medium; the storage medium is used to store instructions; the processor is used to perform operations according to the instructions to execute the steps of the optical fiber communication method described in Embodiment 1.

[0152] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0153] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0154] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0155] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0156] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method of optical fiber communication based on an adversarial mode chimpanzee algorithm, characterized by, The application relates to a method for correcting mode coupling and mode transmission in a few-mode channel. The method comprises the following steps: acquiring original bit data to be transmitted; converting the original bit data into optical fiber transmission signals through a quadrature amplitude modulation mapping algorithm after mapping the original bit data to a constellation diagram; transmitting the optical fiber transmission signals to a receiving end through a few-mode channel; performing analog-digital conversion on the received optical fiber transmission signals to obtain received bit data, and decomposing the received bit data into leading received information and body received information; Generating equalizer parameters randomly by a generator and channel estimates ; constructing a mode coupling matrix from the equalizer parameters ; constructing a mode transmission matrix from the channel estimates ; generating a mode coupling matrix and a mode transmission matrix from the leading received information by using an adversarial mode chimpanzee algorithm, and the method comprises the following steps: combining the mode coupling matrix and the mode transmission matrix to represent a chimpanzee individual to obtain a chimpanzee population, wherein the chimpanzee population comprises M chimpanzee individuals; performing denoising processing on the leading received information as a training sequence, and adding a real label to the training sequence according to preset leading information; ； In the formula, I is a unit matrix; is a transmission compensation matrix; is a mode transmission matrix; is a conjugate transpose of the mode transmission matrix H; is a noise variance; inputting the training sequence into an equalizer to establish a transmission compensation matrix which is reciprocal to the mode transmission matrix, and the expression formula is: ； ； compensating the amplitude and the phase of the training sequence by using the transmission compensation matrix to obtain leading compensation information, and the expression formula is: in the formula, r is additive white Gaussian noise; y is the leading received information; s is the preset leading information; and z is the leading compensation information; dividing elements in the mode coupling matrix into degenerate mode group subblocks, high-low order mode group subblocks and inter-group coupling subblocks according to the amplitude; and calculating average coupling coefficients and coupling variances of the degenerate mode group subblocks, the high-low order mode group subblocks and the inter-group coupling subblocks; setting sparse orders of the degenerate mode group subblocks, the high-low order mode group subblocks and the inter-group coupling subblocks according to the average coupling coefficients and the coupling variances to obtain degenerate mode group submatrices, high-low order mode group submatrices and inter-group submatrices; ； In the formula, is the joint MIMO equalization matrix; is the degenerate module submatrix; is the high-low order module submatrix; and is the inter-group submatrix; constructing a joint MIMO equalization matrix from the degenerate mode group submatrices, the high-low order mode group submatrices and the inter-group submatrices, and the expression formula is: 、 、 ； 、 、 ； In the formula, is the intermediate signal after eliminating the strong coupling within the group for the degenerate mode received sub-signal; is the intermediate signal after suppressing the weak coupling within the group for the high-low order mode received sub-signal; and are the degenerate mode received sub-signal and the high-low order mode received sub-signal in the preamble compensation information; and is the inter-group interference; and are the degenerate mode received sub-signal and the high-low order mode received sub-signal in the preamble correction information; eliminating local coupling of the leading compensation information by using the joint MIMO equalization matrix to obtain leading correction information, and the expression formula is: ； In the formula, s is the real label of the training sequence; is the preamble correction information; calculating an adaptability value MSE by using the leading correction information and the real label, and the expression formula is: inputting the leading correction information into a discriminator to output a training evaluation value, and inputting the training sequence into the discriminator to output a training real value; and calculating an adversarial loss according to the training evaluation value and the training real value; judging whether the adversarial loss converges or not, updating positions of the chimpanzee individuals according to the adaptability value MSE when the adversarial loss does not converge; repeating the updating process of the chimpanzee population until the adversarial loss converges, and outputting mode coupling matrices and mode transmission matrices corresponding to chimpanzee population individuals with optimal adaptability values MSE; 2. The optical fiber communication method according to claim 1, characterized by, correcting the body received information by using the mode coupling matrices and the mode transmission matrices to obtain corrected signals, and re-converting the corrected signals into original bit data through demapping. The method for converting the original bit data into the optical fiber transmission signals through the quadrature amplitude modulation mapping algorithm comprises the following steps: ； In the formula, is a time domain signal; j is an imaginary unit; N is a number of subcarriers; k is a discrete frequency index of an orthogonal frequency division multiplexing symbol block; and n is a discrete time index of the time domain signal; is an orthogonal frequency division multiplexing symbol; is a constant of the circle; performing quadrature amplitude modulation mapping on the original bit data to obtain orthogonal frequency division multiplexing symbols; and performing inverse fast Fourier transform on the orthogonal frequency division multiplexing symbols to obtain time domain signals, and the expression formula is: performing amplitude limiting processing and parallel-serial conversion on the time domain signals to obtain parallel transmission data, adding a cyclic prefix to the parallel transmission data, and obtaining the optical fiber transmission signals through digital-analog conversion.

3. The optical fiber communication method according to claim 1, wherein, Updating the positions of the chimpanzee individuals according to the fitness value MSE, specifically comprising: According to the fitness value MSE, the chimpanzee individuals in the chimpanzee population are divided into drivers, obstacles, pursuers and attackers; the position of the driver is updated by adding a disturbance; According to the position of the driver, the positions of the obstacle and the pursuer are updated, and the expression formula is: ； ； ； ； wherein, and are the position vectors of the obstacle in the t+1 and t iterations; is the position vector of the driver in the t iteration; and are the position vectors of the obstacle in the t+1 and t iterations; A and B are dynamic coefficients; is a set constant; is the standard deviation of the fitness value MSE in the chimpanzee population; is the deviation weight of the obstacle relative to the driver; is a random number in the interval [0, 1]; is a random number in the interval [0, 1]; is the average of the fitness value MSE in the chimpanzee population; The position of the attacker is updated by local search in the range from the driver to the obstacle, and the expression formula is: ； In the formula, and is the position vector of the attacker in the t+1 and t iteration; is a random perturbation term.

4. The optical fiber communication method according to claim 1, characterized by, According to the training evaluation value and the training true value, the adversarial loss is calculated, specifically comprising: ； In the formula, is the adversarial loss; is the expected function; is the training true value of the discriminator output; is the training evaluation value of the discriminator output.

5. The optical fiber communication method according to claim 1, wherein, The correction signal is demapped and converted back to the original bit data, specifically comprising: After deleting the cyclic prefix in the correction signal, perform serial-parallel conversion to obtain a time domain signal sequence; The time domain signal sequence is converted into an orthogonal frequency division multiplexing symbol block by fast Fourier transform, and the specific process is: ； In the formula, is a time-domain signal sequence; j is an imaginary unit; N is the number of subcarriers; k is a discrete frequency index of an orthogonal frequency division multiplexing symbol block; and n is a discrete time index of a time-domain signal; is an orthogonal frequency division multiplexing symbol block; is a constant; and The original bit data is obtained by quadrature amplitude modulation demapping of the orthogonal frequency division multiplexing symbol block.

6. An optical fiber communication system based on the adversarial mode chimpanzee algorithm, characterized by, It includes: The signal sending unit is used for obtaining the original bit data to be sent; the original bit data is mapped to the constellation diagram by the quadrature amplitude modulation mapping algorithm, and then converted into an optical fiber transmission signal; The optical fiber transmission signal is transmitted to the receiving end through the few-mode channel; The decomposition unit is used for analog-digital conversion of the received optical fiber transmission signal to obtain received bit data, and the received bit data is decomposed into preamble reception information and text reception information; The correction unit is used for generating a mode coupling matrix and a mode transmission matrix from the preamble reception information by using the chimpanzee algorithm in the adversarial mode; the text reception information is corrected by the mode coupling matrix and the mode transmission matrix to obtain a correction signal, The response unit is used for demapping the correction signal and converting it back to the original bit data; The correction unit generates a mode coupling matrix and a mode transmission matrix from the preamble reception information by using the chimpanzee algorithm in the adversarial mode, specifically comprising: Generating equalizer parameters randomly by a generator and channel estimates ; constructing a mode coupling matrix from the equalizer parameters ; constructing a mode transmission matrix from the channel estimates ; The mode coupling matrix and the mode transmission matrix are combined to represent the chimpanzee individuals to obtain a chimpanzee population, and the chimpanzee population contains M chimpanzee individuals; The preamble reception information is denoised and treated as a training sequence, and a true label is added to the training sequence according to the preamble preset information; The training sequence is input into the equalizer to establish a transmission compensation matrix which is reciprocal to the mode transmission matrix, and the expression formula is: ； In the formula, I is a unit matrix; is a transmission compensation matrix; is a mode transmission matrix; is a conjugate transpose of the mode transmission matrix H; is a noise variance; The amplitude and phase of the training sequence are compensated by using the transmission compensation matrix to obtain preamble compensation information; the expression formula is: ； ； In the formula, r is an additive white Gaussian noise; y is the preamble reception information; s is the preamble preset information; z is the preamble compensation information; According to the amplitude, the elements in the mode coupling matrix are divided into degenerate mode group subblocks, high-low order mode group subblocks and inter-group coupling subblocks; the average coupling coefficient and the coupling variance of the degenerate mode group subblocks, the high-low order mode group subblocks and the inter-group coupling subblocks are calculated; According to the average coupling coefficient and the coupling variance, the sparsity order of the degenerate mode group subblocks, the high-low order mode group subblocks and the inter-group coupling subblocks is set to obtain degenerate mode group submatrices, high-low order mode group submatrices and inter-group submatrices; The joint MIMO equalization matrix is constructed from the degenerate mode group submatrices, the high-low order mode group submatrices and the inter-group submatrices, and the expression formula is: ； In the formula, is the joint MIMO equalization matrix; is the degenerate module submatrix; is the high-low order module submatrix; and is the inter-group submatrix; The joint MIMO equalization matrix is used to eliminate the local coupling of the preamble compensation information to obtain preamble correction information, and the expression formula is: 、 、 ； 、 、 ； In the formula, is the intermediate signal after eliminating the strong coupling within the group for the degenerate mode received sub-signal; is the intermediate signal after suppressing the weak coupling within the group for the high-low order mode received sub-signal; and are the degenerate mode received sub-signal and the high-low order mode received sub-signal in the preamble compensation information; and are the inter-group interference; and are the degenerate mode received sub-signal and the high-low order mode received sub-signal in the preamble correction information; The fitness value MSE is calculated by the preamble correction information and the real label, and the expression formula is: ； In the formula, s is the true label of the training sequence; is the preamble correction information; The training evaluation value is output by inputting the preamble correction information into the discriminator, and the training real value is output by inputting the training sequence into the discriminator; the adversarial loss is calculated according to the training evaluation value and the training real value; It is judged whether the adversarial loss converges or not, and the position of the chimpanzee individual is updated according to the fitness value MSE when the adversarial loss does not converge; the updating process of the chimpanzee population is repeated until the adversarial loss converges, and the mode coupling matrix and the mode transmission matrix corresponding to the chimpanzee population individual with the optimal fitness value MSE are output.

7. An electronic terminal, characterized in that The device comprises a processor and a storage medium; the storage medium is used to store instructions; the processor is used to operate according to the instructions to execute the steps of the optical fiber communication method in any one of claims 1 to 6.

Images