Forward Raman nonlinear suppression amplification method for OAM optical communication

Abstract

The invention discloses a forward Raman nonlinear suppression amplification method for OAM optical communication, and relates to the technical field of nonlinear optics, OAM composite signals are separated into multiple groups of intrinsic signals according to spiral phase characteristics and radial distribution characteristics, corresponding gain coefficients are adjusted through an order gain algorithm according to cooperative attenuation characteristics of orders and moduli of an OAM mode, and the OAM optical communication is realized. Pre-amplification is carried out on each group of intrinsic signals; collecting and calculating the dispersion amount of each group of amplified intrinsic signals, mapping orders and moduli according to a modulation type time sequence model to generate a control vector, taking the control vector as a modulation constraint, deducing a predicted dispersion amount, and superposing the predicted dispersion amount with an actual dispersion amount to carry out dispersion pre-compensation; the method comprises the following steps: determining the number of pumping wavelengths, collecting transmission loss, calculating target Raman gain, distributing injection point power and pause duration to realize distributed Raman amplification, inhibiting FWM clutter by using random phase modulation, and sending the FWM clutter to a receiving end for processing and demodulation to realize stable long-distance OAM optical communication.

Description

Technical Field

[0001] This invention relates to the field of nonlinear optics, and more specifically to a forward Raman nonlinear suppression and amplification method for OAM optical communication. Background Technology

[0002] OAM optical communication signals are generated by combining multiple intrinsic signals of different orders and moduli, following a Laguerre-Gaussian function. During fiber transmission, these signals experience transmission loss, and their strength continuously decreases with increasing transmission distance. Furthermore, chromatic dispersion exists, disrupting the signal's phase coherence. Without optical amplification to compensate for gain and correct dispersion, the signal-to-noise ratio at the receiver is significantly reduced, causing signal distortion and making stable long-distance OAM optical communication impossible. Therefore, implementing optical amplification technology is essential in OAM communication.

[0003] Existing optical amplification technology in OAM communication has shortcomings:

[0004] First, the gain was not adjusted by combining the synergistic attenuation characteristics of the OAM mode order and modulus. Instead, a uniform gain coefficient was used to amplify all intrinsic signals, which could not accurately match the attenuation requirements of different mode signals and was difficult to effectively compensate for signal loss.

[0005] Secondly, there is a lack of dynamic dispersion compensation mechanism for OAM signal characteristics, and the four-wave mixing (FWM) clutter is not effectively suppressed, resulting in dispersion distortion and clutter interference in the amplified signal, which affects communication quality. Summary of the Invention

[0006] This invention addresses the shortcomings of existing technologies by proposing a forward Raman nonlinear suppression and amplification method for OAM optical communication. It uses a step-path gain algorithm to accurately match the attenuation requirements of intrinsic signals of different orders and moduli in OAM mode to compensate for signal loss. It also implements dispersion pre-compensation based on a modulation-type timing model to solve the signal dispersion distortion problem. Finally, it combines distributed Raman amplification and random phase modulation to suppress FWM clutter interference, thereby achieving stable long-distance OAM optical communication.

[0007] The technical solution adopted to achieve the purpose of this invention is as follows:

[0008] A forward Raman nonlinear suppression and amplification method for OAM optical communication includes the following steps:

[0009] Acquiring OAM composite signal Separation is based on helical phase characteristics and radial distribution characteristics. The intrinsic signals are grouped, and the corresponding gain coefficients are adjusted according to the synergistic attenuation characteristics of the order and modulus of the OAM mode using the step-path gain algorithm. The intrinsic signals of each group are pre-amplified to obtain The group amplifies the intrinsic signal, among which, Radial coordinates, Angular coordinates, For axial coordinates, The total number of eigensets equals the total order. Multiply by the total modulus ;

[0010] Collect and calculate The dispersion of the amplified intrinsic signals is calculated. Based on a modulation-based timing model, the order and modulus of each intrinsic signal are used to generate a control vector through feature mapping, which serves as a modulation constraint. This process leads to the derivation of... The predicted dispersion of the amplified intrinsic signal is grouped, the corresponding dispersion is superimposed, and then applied to... The amplified intrinsic signal is pre-compensated for dispersion to obtain... Group compensation intrinsic signal;

[0011] Determine the total number of pump wavelengths And collect The transmission loss of the intrinsic signal is compensated by grouping and calculated separately. The target Raman gain of the group compensates for the intrinsic signal, combined with the preset total number of injection points. Assign each group of compensated intrinsic signals in The matching power and matching pause duration at each injection point were collected. The FWM clutter power at each injection point is used to disrupt the FWM nonlinear phase matching condition through random phase modulation. Distributed Raman amplification at each injection point yields... Raman amplification of intrinsic signals;

[0012] At the receiving end The Raman amplified intrinsic signal is then converted into photoelectric signal and subsequently passed through amplification, filtering, equalization, and demodulation techniques to obtain the final signal. The raw data is used to complete OAM optical communication.

[0013] Furthermore, OAM composite signal for The summation of the group of intrinsic signals follows the Laguerre-Gaussian function. Group number is intrinsic signals equals group number amplitude order Modulus radial field distribution function order Angular spiral phase With order Modulus Axial propagation phase The product of, where, For group number The propagation coefficient, and These are the exponential function and the imaginary number, respectively, with the natural constant as the base.

[0014] Furthermore, the OAM composite signal Separate into The process of generating intrinsic signals includes the following steps:

[0015] Using the electric domain spiral phase filtering algorithm, at order Select the target order within the range of values. And multiply by the imaginary number and angular coordinates The opposite of the number is used to obtain the digital spiral phase filter kernel. ;

[0016] OAM composite signal With angular spiral phase filter kernel Multiply and diagonally to coordinates Integrate to obtain the target order. intermediate signal ;

[0017] In modulus Select the target modulus within the range of values. Based on the target order Target Modulus radial field distribution function Constructing radial conjugate matched filter kernels in conjugate form ;

[0018] target order intermediate signal With digital radial matched filter kernel Multiply and relative to radial coordinates Integrate to obtain the target order. Target Modulus intermediate single-mode signal And multiply by the same order of angular spiral phase To obtain the target order Target Modulus intrinsic signals traversal order With modulus Separation Group intrinsic signals.

[0019] Furthermore, for group number... intrinsic signals Execute the path gain algorithm, based on the attenuation characteristic table, order... Modulus decay rate Equal to the reference decay rate with order 1 and modulus 1 With regard to order Modulus The product of the exponential correction terms, order Modulus The exponential correction term equals the order factor. Multiply by the order The absolute value of minus 1 plus the modulus factor Multiply by the modulus Decrease by 1 and add cross factor order The absolute value of minus 1 and the modulus The attenuation characteristic table, determined by multiplying by 1, is based on fitting the attenuation of intrinsic signals of different orders and moduli over the same transmission distance. The order is... Modulus decay rate With average decay rate Add 1 to the ratio and multiply by the base gain Results and Upper Limit Gain The smaller value in is used as the order. Modulus signal gain And for group number intrinsic signals Perform pre-amplification to obtain group number as Amplified intrinsic signal , , .

[0020] Furthermore, dispersion pre-compensation includes the following steps:

[0021] The data collection group number is Amplified intrinsic signal Pulse delay difference Divide by the total wavelength bandwidth and axial total length The group number is obtained as Amplified intrinsic signal Dispersion And defined as the dispersion amount in this case. ;

[0022] order absolute value, modulus and order With modulus The ratios are concatenated into an OAM parameter vector and then mapped through a fully connected layer to generate group numbers. Order of control vector and Modular Control Vector The order of the control vector and Modular Control Vector The group number obtained by performing the Hadamard product is Input control vector The order of the control vector Absolute value and modulus control vector Summing yields the group number as Forgetting control vector , initialize the previous first The group number is Hidden state vector and the previous The group number is Historical dispersion of the amplified intrinsic signal spliced ​​together as the previous number The group number is joint input vector , This is the total number for historical reference.

[0023] Group number is Input control vector Forgetting control vector Compared with the previous The group number is joint input vector By employing a cooperative input modulation timing model and executing a cooperative loop mechanism, the group number for this experiment was derived as follows: Hidden state vector And the group number is obtained by mapping it to the fully connected layer through a self-attention mechanism. Predicted dispersion , and group number Amplified intrinsic signal Dispersion After accumulation, dispersion pre-compensation is performed to obtain group number as Compensated intrinsic signals .

[0024] Furthermore, the cooperative loop mechanism of the modulation timing model includes the following steps:

[0025] The previous number The group number is joint input vector With group number Input control vector The previous nonlinear mapping was performed separately to obtain the first... The group number is Input gate vector ;

[0026] The previous number The group number is joint input vector With group number Forgetting control vector The previous nonlinear mapping was performed separately to obtain the first... The group number is Forgetting Gated Vector ;

[0027] The previous number The group number is Forgetting Gated Vector Compared with the previous The group number is Hidden state vector Hadamaji plus the previous number The group number is Latest learning vector Compared with the previous The group number is Input gate vector The Hadama accumulation, obtained from the previous [number] The group number is Hidden state vector Among them, the previous number The group number is Latest learning vector Equal to the previous number The group number is joint input vector Hyperbolic tangent modulation mapping results;

[0028] By executing the sequence cyclically and deducing step by step, the group number for this iteration is obtained. Hidden state vector .

[0029] Furthermore, the total wavelength bandwidth of the OAM mode... With the upper limit wavelength bandwidth of the pump core light source Adding 1 to the integer part of the ratio gives the total number of pump wavelengths. ,collection The group compensates for the transmission loss of the intrinsic signal, and the group number is... Target Raman gain Set the group number as Transmission loss With Raman fiber loss coefficient The product of superposition of redundant gains Summation The target Raman gain of the group is obtained as the total target Raman gain. The group number is Target Raman gain Total Raman gain of the target The ratio multiplied by the total pump power The group number is obtained as Pump group power Based on the total number of injection points Uniformly segmented transmission fiber and determination of single segment axis length , will the Injection point number of each injection point Total number of injection points The ratio multiplied by the group number is Pump group power The group number is obtained as Compensated intrinsic signals In the Matching power at each injection point , .

[0030] Furthermore, the group number is of Matching pause time for each injection point The same, both equal to the basic Raman establishment time. With order and modulus The product of linear recurrence relations, order and modulus The linear recurrence relation is equal to 1 plus the order influence coefficient. Multiply by the order Modulus influence coefficient Multiply by the modulus .

[0031] Furthermore, the group number is Compensated intrinsic signals In the Signal light output power at each injection point equals group number Compensated intrinsic signals In the Signal light output power at each injection point Multiply by 10 as the base, the first From the injection point to the... The group number between each injection point is Single-segment target Raman gain One-tenth of the power value of the exponent, the th From the injection point to the... The group number between each injection point is Single-segment target Raman gain Equal to target Raman gain Multiply by the injection point number Total number of injection points The ratio of .

[0032] Furthermore, the first The group number of each injection point is Pump light output power equal to the The group number of each injection point is Pump light output power With group number Compensated intrinsic signals In the Signal light output power at each injection point The sum multiplied by the order Modulus exponential decay rate order Modulus exponential decay rate Equal to the natural constant as the base and the order as the resultant. Modulus decay rate With single-segment axis length The negative of the product is the power of the exponent.

[0033] Furthermore, the group number is Compensated intrinsic signals In the FWM noise power at each injection point equals group number Compensated intrinsic signals In the Signal light output power at each injection point The square of the first The group number of each injection point is Pump light output power Multiply by a constant coefficient Multiply by 1 and subtract the order Modulus exponential decay rate According to the group number Compensated intrinsic signals In the FWM noise power at each injection point With FWM threshold The magnitude relationship will affect the phase modulation depth Set as or Combining PRBS random sequences with phase modulation depth Generate random phase perturbation The natural constant will be used as the base and the phase will be randomly perturbed. The power value of the exponent and the group number are Compensated intrinsic signals The product is used to break the nonlinear phase matching condition of FWM.

[0034] Compared with the prior art, this invention proposes an order and modulus separation and order-path gain algorithm for OAM mode. It uses the spiral phase characteristics and radial distribution characteristics of OAM mode to separate the intrinsic signal, adjusts the corresponding gain coefficient according to the cooperative attenuation characteristics of OAM mode order and modulus, and pre-amplifies each group of intrinsic signals after separation to accurately match the attenuation requirements of different mode signals and effectively compensate for signal loss.

[0035] A modulation-type timing model is designed. The order and modulus of each group of intrinsic signals are used to generate control vectors through feature mapping. The predicted dispersion is derived, and the actual dispersion is superimposed to perform dispersion pre-compensation for each group of amplified intrinsic signals. The matching power and matching pause time of each group of compensated intrinsic signals at each injection point are allocated according to the preset total number of injection points to achieve distributed Raman amplification. The FWM clutter power at each injection point is collected, and random phase modulation is used to destroy the nonlinear phase matching condition of FWM, suppress FWM clutter interference, and solve the dispersion distortion problem. Attached Figure Description

[0036] Figure 1 This is a flowchart of a forward Raman nonlinear suppression and amplification method for OAM optical communication.

[0037] Figure 2 Flowchart for separating intrinsic signals from OAM composite signals;

[0038] Figure 3 Here is a flowchart of the dispersion pre-compensation process;

[0039] Figure 4 This is a modulation timing model diagram. Detailed Implementation

[0040] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments.

[0041] like Figure 1 As shown, a specific embodiment of the present invention discloses a forward Raman nonlinearity suppression and amplification method for OAM optical communication, comprising the following steps:

[0042] Obtain the OAM composite signal at the transmitting end. Separation is based on the spiral phase characteristics and radial distribution characteristics of the OAM mode. The intrinsic signals are grouped, and the corresponding gain coefficients are dynamically adjusted based on the synergistic attenuation characteristics of the order and modulus of the OAM mode using the step-path gain algorithm. The intrinsic signals of each group are pre-amplified to obtain The group amplifies the intrinsic signal, among which, Here, is the radial coordinate, and is the perpendicular distance from any point within the transmission fiber to the central axis of the fiber in the polar coordinate system. is an angular coordinate, where represents the angle of rotation of the reference direction within the fiber cross-section around the origin of the cross-section. The origin of the cross-section is the intersection of the fiber's central axis and the fiber's cross-section. For axial coordinates, it is equal to the axial distance from the fiber's origin along the fiber's central axis to any point within the transmission fiber's cross-section. The total number of eigensets equals the total order. Multiply by the total modulus ;

[0043] Collect and calculate The dispersion of the amplified intrinsic signals is analyzed. Based on a modulation-based time-series model, the order and modulus of each intrinsic signal are used to generate control vectors through feature mapping, which are then used as modulation constraints for the gating mechanism in the Long Short-Term Memory (LSM) network. This allows for differentiated extrapolation based on trends consistent with OAM patterns. The amplified intrinsic signal is subjected to predicted dispersion over a specific time period, and the corresponding dispersion is superimposed to respectively... The amplified intrinsic signal is pre-compensated for dispersion to obtain... The group compensates for the intrinsic signals, where a specific duration is the OAM composite signal. The acquisition interval duration;

[0044] Total wavelength bandwidth based on OAM mode Determine the total number of pump wavelengths And collect The transmission loss of the intrinsic signal is compensated by grouping and calculated separately. The target Raman gain of the group compensates for the intrinsic signal, combined with the preset total number of injection points. Adaptive allocation of each group of compensated intrinsic signals in The matching power and matching pause duration at each injection point are adjusted to achieve effective distributed Raman amplification and suppress backscattering. (The text then abruptly shifts to a seemingly unrelated topic: acquiring...) The FWM clutter power at each injection point is used to disrupt the FWM nonlinear phase matching condition through random phase modulation. Distributed Raman amplification at each injection point yields... Raman amplification of intrinsic signals;

[0045] Using a photodetector at the receiving end The Raman amplified intrinsic signal is then converted into photoelectric signal to obtain... The intrinsic electrical signal is sequentially amplified, filtered, and equalized to remove noise and compensate for signal distortion. Combined with the modulation rules at the transmitting end, the signal is demodulated. The original data carried by the intrinsic electrical signal is used to realize OAM optical communication with forward Raman nonlinear suppression and amplification.

[0046] Furthermore, the OAM composite signal transmitted in the transmission fiber is captured by a transmission fiber mode coupler deployed at the transmitting end. for The composition of the intrinsic signals follows the Laguerre-Gaussian function. The details are as follows:

[0047] ,

[0048] in, For group number The intrinsic signal, For group number, and The group numbers are respectively The amplitude and propagation coefficient of the intrinsic signal, the propagation coefficient The order of the OAM pattern Modulus The waveguide parameters of the transmission optical fiber determine the overall structure of the fiber. order Modulus The radial field distribution function is determined by the waveguide structure of the transmission fiber and is a solution to the Helmholtz equation for the transmission fiber. order Angular spiral phase With order Modulus Axial propagation phase The product of and These are the exponential function and the imaginary number, respectively, with the natural constant as the base.

[0049] like Figure 2 As shown, further, the OAM composite signal Separate into The process of generating intrinsic signals includes the following steps:

[0050] Using the electric domain spiral phase filtering algorithm, at order The range of values Select the target order in sequence And construct a digital spiral phase filter core ;

[0051] OAM composite signal in polar coordinate system With angular spiral phase filter kernel Multiply and diagonally to coordinates By performing integration, based on the orthogonality of the angular spiral phases, the product of angular spiral phases of different orders is zero. After integration, only the target order is retained. intermediate signal The details are as follows:

[0052] ,

[0053] Among them, the diagonal coordinates The maximum value is Target order intermediate signal Only the radial and axial distributions are retained;

[0054] In modulus The range of values Select the target modulus in sequence Based on the target order Target Modulus radial field distribution function Constructing radial conjugate matched filter kernels in conjugate form ;

[0055] The target order in polar coordinates intermediate signal With digital radial matched filter kernel Multiply and relative to radial coordinates Perform integration, using the target order of The radial orthogonality and conjugate between radial field distribution functions of each modulus are considered, and only the target order is retained after integration. Target Modulus intermediate single-mode signal The details are as follows:

[0056] ,

[0057] Wherein, radial coordinates The maximum value is the core radius. Target order Target Modulus intermediate single-mode signal equal to the Amplitude of the intrinsic signal of the group With the target order Target Modulus Axial propagation phase The product;

[0058] target order Target Modulus intermediate single-mode signal Multiplied by the same order of angular spiral phase To obtain the target order Target Modulus intrinsic signals When the target order With target modulus When all values ​​have been traversed within their respective ranges, the OAM composite signal will be... Separate into Group intrinsic signals.

[0059] Furthermore, the path gain algorithm is used for... Each intrinsic signal is adaptively pre-amplified to compensate for losses introduced during subsequent signal processing and transmission, generating... Group amplification of intrinsic signals, among which, group number is intrinsic signals The path gain algorithm is executed, and the order is fitted according to the pre-calibrated attenuation characteristic table. Modulus decay rate The details are as follows:

[0060] ,

[0061] in, This is the reference decay rate in OAM mode with order 1 and modulus 1. , and These are the order factor, module factor, and cross factor, used to describe the order. Modulus Basic attenuation contribution and order With modulus The cooperative decay contribution, since the OAM mode follows the Laguerre-Gaussian function, decay rate With order and modulus The attenuation characteristics exhibit an exponential correlation. The attenuation characteristic table is obtained by testing the attenuation of intrinsic signals of different orders and moduli over a fixed transmission distance, and then fitting the data using the least squares method, based on the order of the intrinsic signal. Modulus decay rate With average decay rate The ratio of the base gain of the preamplifier Perform an adaptive linear correction with an upper limit constraint to determine the order. Modulus signal gain The details are as follows:

[0062] ,

[0063] Among them, order Modulus decay rate The higher the order, the better. Modulus signal gain The larger the gain, the less than or equal to the upper limit gain of the preamplifier. , order Modulus signal gain With intrinsic signal Multiply to perform pre-amplification, resulting in group number . Amplified intrinsic signal , , .

[0064] like Figure 3 As shown, further, dispersion pre-compensation includes the following steps:

[0065] Data is collected and calculated using a coherent detector. The dispersion of the amplified intrinsic signal is increased by the group number. Amplified intrinsic signal The coherent detector acquires and amplifies the intrinsic signal. Pulse delay difference Based on the total wavelength bandwidth of OAM mode and the total axial length of the transmission fiber Calculate the amplified intrinsic signal Dispersion Among them, the total axial length of the transmission optical fiber Essentially an OAM composite signal The transmission distance;

[0066] Group number is Amplified intrinsic signal Dispersion Defined as the dispersion amount in this case. Calling the previous The group number obtained after acquisition, separation, and pre-amplification is: The historical dispersion measures of the historical amplified intrinsic signal are concatenated into a historical dispersion measure sequence according to the acquisition order. The details are as follows:

[0067] ,

[0068] in, For the previous first The group number is Historical amplification of intrinsic signals This is the total number for historical reference.

[0069] order absolute value, modulus and order With modulus The ratios are concatenated column-wise into an OAM parameter vector, and then mapped through two fully connected branches to generate group numbers. Order of control vector and Modular Control Vector The order of the control vector and Modular Control Vector The group number obtained by performing the Hadamard product is Input control vector The order of the control vector Absolute value and modulus control vector Summing yields the group number as Forgetting control vector , will be the previous The group number is Hidden state vector Initialize as a vector of all zeros in the corresponding dimension and then merge it with the previous zero vector. The group number is Historical dispersion of the amplified intrinsic signal Arrange them column by column to form the previous [number]. The group number is joint input vector ;

[0070] Based on group number Input control vector Forgetting control vector Compared with the previous The group number is joint input vector By utilizing the coordinated loop mechanism of the input gate, forget gate, and output gate in the modulation timing model, the group number for this operation is obtained through step-by-step deduction along the time sequence. Hidden state vector It also utilizes a self-attention mechanism to capture the interdependencies of hidden states, and obtains the group number through mapping via a fully connected layer. Predicted dispersion , and group number Amplified intrinsic signal Dispersion The cumulative sum obtained is group number The dispersion pre-compensation amount was determined, and the programmable dispersion compensation component was adjusted to perform dispersion pre-compensation, resulting in group number [number missing]. Compensated intrinsic signals Since dispersion changes dynamically over time, dispersion pre-compensation overcomes the lag problem caused by the processing delay of real-time compensation, meets the stringent requirements of phase coherence for dispersion compensation accuracy, and ensures the stability and coherence of OAM optical communication. In this embodiment, the programmable dispersion compensation component is an existing dispersion compensation device, and the principle of dispersion pre-compensation based on the dispersion pre-compensation amount will not be described in detail.

[0071] like Figure 4 As shown, the cooperative loop mechanism of the modulation timing model further includes the following steps:

[0072] The previous number The group number is joint input vector With group number Input control vector The signal is transmitted to the input gate, where it undergoes joint input modulation and control input modulation, and then summed to obtain the previously obtained signal. The group number is Input gate vector ,in, and These are the joint input weight matrix and joint input bias vector for joint input modulation, respectively. and These are the control input weight matrix and the control input bias vector, respectively, for control input modulation. Represents the Sigmoid function;

[0073] The previous number The group number is joint input vector With group number Forgetting control vector The data is transmitted to the forget gate, where joint forgetting modulation and controlled forgetting modulation are performed and summed to obtain the previously obtained result. The group number is Forgetting Gated Vector ,in, and These are the joint forgetting weight matrix and joint forgetting bias vector, respectively, for joint forgetting modulation. and These are the control forgetting weight matrix and the control forgetting bias vector, respectively, which control forgetting modulation.

[0074] The previous number The group number is joint input vector Input gate vector Forgetting Gated Vector Transmitted to the output gate, the previous [number] [number] [unit] [is transmitted]. The group number is Forgetting Gated Vector Compared with the previous The group number is Hidden state vector The Hadamard product is used to reflect the degree to which the modulation-type time series model preserves historical hidden states, thus obtaining the previous... The group number is Preserving hidden state vectors , will be the previous The group number is joint input vector The hyperbolic tangent modulation mapping is used sequentially to obtain the previous (i)th (j ... The group number is Latest learning vector , will be the previous The group number is Preserving hidden state vectors In addition to the previous number The group number is Latest learning vector Compared with the previous The group number is Input gate vector The Hadama accumulation, obtained from the previous [number] The group number is Hidden state vector ,in, and These are the hyperbolic tangent weight matrix and hyperbolic tangent bias vector for hyperbolic tangent modulation, respectively. Represents the hyperbolic tangent function. It represents the Hadamardi (or Hadama) stack;

[0075] Based on the coordinated cyclic operation of the input gate, forget gate, and hide gate in the modulation-type timing model, the group number for this test is obtained by step-by-step deduction along the time sequence. Hidden state vector The group number for this event is... joint input vector The group number for the first time was Hidden state vector The group number for this event is The dispersion of the amplified intrinsic signal Obtained by piecing together columns.

[0076] Furthermore, using a coherent DFB laser array as the core pump source requires that the pump wavelength range cover the total wavelength bandwidth of the OAM mode. Upper wavelength bandwidth based on coherent DFB lasers Total wavelength bandwidth With upper limit wavelength bandwidth Adding 1 to the integer part of the ratio gives the total number of pump wavelengths. Data collected using an optical loss meter The group compensates for the transmission loss of the intrinsic signal, where the group number is... Compensated intrinsic signals The transmission loss is The core of Raman amplification lies in compensating for transmission loss. The linear loss induced by Raman fiber, therefore, group number is Compensated intrinsic signals Target Raman gain Equal to transmission loss With Raman fiber loss coefficient The product of superposition of redundant gains That is, the target Raman gain Redundancy gain As a buffer, to avoid insufficient gain due to equipment fluctuations, summation The target Raman gain of the group is obtained as the total target Raman gain. According to the group number Target Raman gain Total Raman gain of the target The ratio from the total pump power The group number was assigned as follows Pump group power Based on the preset total number of injection points Injection points are selected at equal intervals along the central axis of the transmission fiber to achieve uniform segmentation of the transmission fiber and determine the axial length of a single segment. , will the Injection point number of each injection point Total number of injection points The ratio multiplied by the group number is Pump group power The group number is obtained as Compensated intrinsic signals In the Matching power at each injection point , .

[0077] Furthermore, since Raman gain is not instantaneous, it requires a certain amount of time for the pump light and signal light to complete energy transfer through the Raman effect. During the Raman gain build-up period, the intensity of backscattering reaches its peak. If pump light is continuously injected during this time, the backscattering will carry more energy, severely interfering with normal signal transmission. Therefore, temporarily cutting off the pump during the gain build-up period can reduce the pump energy input during this period, thereby avoiding the backscattering peak and reducing interference with the transmitted signal. An optical coupler is deployed at each injection point to inject pump light, and a high-speed switch is connected in series beforehand. During the Raman settling period, pump light injection is temporarily cut off to avoid reflection interference. The order of the OAM mode... and modulus The larger the value, the more complex the Raman gain establishment process becomes, and The distribution of injection points does not affect the energy transfer efficiency or order. and modulus The inherent connection, based on the duration of the basic Raman establishment. Establish about order and modulus The linear recurrence relation yields the group number as of Matching pause time for each injection point ,in, and These are the order influence coefficient and the modulus influence coefficient, respectively, which are determined by measuring the Raman settling time of the intrinsic signals of different orders and moduli during the experimental stage and fitting them using the least squares method.

[0078] Furthermore, the group number is Compensated intrinsic signals In the Signal light output power at each injection point equals group number Compensated intrinsic signals In the Signal light output power at each injection point With base 10 and number 10 as the base, From the injection point to the... The group number between each injection point is Single-segment target Raman gain One-tenth of the product of the exponent's power value is... , No. From the injection point to the... The group number between each injection point is Single-segment target Raman gain Equal to target Raman gain Multiply by the injection point number Total number of injection points The ratio of the two values, the 0th injection point is the fiber optic start point. .

[0079] Furthermore, the group number is Compensated intrinsic signals Signal light output power at injection point 0 equals group number Compensated intrinsic signals The signal power, the first The group number of each injection point is Pump light output power equal to the The group number of each injection point is Pump light output power With group number Compensated intrinsic signals In the Signal light output power at each injection point The sum multiplied by the order Modulus exponential decay rate order Modulus exponential decay rate Equal to the natural constant as the base and the order as the resultant. Modulus decay rate With single-segment axis length The negative of the product is the power of the exponent. .

[0080] Furthermore, the group number is Compensated intrinsic signals In the FWM noise power at each injection point With group number Compensated intrinsic signals In the Signal light output power at each injection point The square and the first The injection point group number is Pump light output power The product is related to the rest and is used to reflect the rest. The compensated intrinsic signals of each group are in the first... The injection point and the first The injection point group number is The coupling crosstalk between the pump beams is as follows:

[0081] ,

[0082] in, For fixed constant coefficients, the nonlinear refractive index of the transmission fiber, the cross-sectional area of ​​the fiber, and the OAM composite signal are considered. The center wavelength is related, and the comparison group number is... Compensated intrinsic signals In the FWM noise power at each injection point With FWM threshold If it is greater than the FWM threshold Set the phase modulation depth for If less than or equal to the FWM threshold Set the phase modulation depth for Based on PRBS random sequences, based on phase modulation depth With a signal much higher than OAM composite signal Modulation frequency of transmission frequency Generate random phase perturbation The natural constant will be used as the base and the phase will be randomly perturbed. The power value of the exponent and the group number are Compensated intrinsic signals The product is used to apply a phase perturbation, disrupting the nonlinear phase matching condition of the FWM, while the group number is... Compensated intrinsic signals It operates in the linear or weakly nonlinear region, and the signal carrying method is independent of nonlinear phase matching, so it is not affected by random phase disturbances. The influence of this, thereby suppressing the rest The compensated intrinsic signals of each group are in the first... The injection point and the first The injection point group number is The coupling crosstalk between the pump lights.

[0083] This invention discloses a forward Raman nonlinear suppression amplification method for OAM optical communication. It acquires the OAM composite signal and separates multiple intrinsic signals based on spiral phase and radial distribution characteristics. Using a step-path gain algorithm, it adjusts the corresponding gain coefficients according to the synergistic attenuation characteristics of the OAM mode's order and modulus. Each intrinsic signal is pre-amplified to obtain a corresponding amplified intrinsic signal, accurately matching the attenuation requirements of different OAM mode signals and effectively compensating for signal loss. The method also collects and calculates the dispersion of each amplified intrinsic signal. Based on a modulation-type timing model, it generates a control vector from the order and modulus of each intrinsic signal through feature mapping and uses it as a modulation constraint. This yields a predicted dispersion of a set of amplified intrinsic signals. The corresponding dispersion is then superimposed, and dispersion pre-compensation is performed on each set of amplified intrinsic signals to obtain a corresponding compensated intrinsic signal, achieving accurate dispersion pre-compensation. To avoid signal distortion due to dispersion and ensure signal phase coherence, the total number of pump wavelengths is determined and the transmission loss of each set of compensated intrinsic signals is collected. The target Raman gain of each set of compensated intrinsic signals is calculated, and the matching power and matching pause time of each set of compensated intrinsic signals at each injection point are allocated based on the preset total number of injection points. The FWM clutter power at each injection point is collected, and random phase modulation is used to destroy the FWM nonlinear phase matching condition. After distributed Raman amplification at each injection point, each set of Raman amplified intrinsic signals is obtained, achieving efficient distributed Raman amplification while suppressing FWM clutter interference and reducing the impact of backscattering on signal transmission. At the receiving end, each set of Raman amplified intrinsic signals is photoelectrically converted and then sequentially amplified, filtered, equalized, and demodulated to obtain the corresponding raw data, completing OAM optical communication and ensuring the overall quality and stability of OAM optical communication.

[0084] The above description is merely a preferred embodiment of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principles of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. A forward Raman nonlinearity suppression and amplification method for OAM optical communication, characterized in that, Includes the following steps: Based on the spiral phase characteristics and radial distribution characteristics, intrinsic signals of different orders and moduli are separated from the OAM composite signal, and the corresponding gain coefficients are adjusted using the step-radius gain algorithm for pre-amplification to obtain the corresponding amplified intrinsic signals. The dispersion of each group of amplified intrinsic signals is collected and calculated. Based on the modulation-type timing model, the order and modulus of each group of intrinsic signals are used to generate control vectors through feature mapping and used as modulation constraints. The predicted dispersion of each group of amplified intrinsic signals is derived. The corresponding dispersion is superimposed and dispersion pre-compensation is performed on each group of amplified intrinsic signals to obtain the compensated intrinsic signals of all groups. The total number of pump wavelengths is determined and the transmission loss of each group of compensated intrinsic signals is collected. The target Raman gain of each group of compensated intrinsic signals is calculated. Based on the total number of injection points, the matching power and matching pause time of each group of compensated intrinsic signals at all injection points are allocated. The FWM clutter power of each injection point is collected and the FWM nonlinear phase matching condition is destroyed by random phase modulation. Through distributed Raman amplification at all injection points, the Raman amplified intrinsic signals of all groups are obtained and sent to the receiver.

2. The forward Raman nonlinear suppression and amplification method for OAM optical communication as described in claim 1, characterized in that, The intrinsic signals of different orders and moduli can be separated from the OAM composite signal, including: Select the target order within the range of order values ​​to construct a digital spiral phase filter kernel; The intermediate signal of the target order is obtained by diagonally integrating the product of the OAM composite signal and the digital spiral phase filter kernel. Within the range of modulus values, select the target modulus and construct a radial conjugate matched filter kernel based on the conjugate form of the radial field distribution function of the target order and the modulus. The intermediate single-mode signal is obtained by integrating the product of the intermediate signal and the radial conjugate matched filter kernel over the radial coordinates; Multiply the intermediate single-mode signal by the angular spiral phase of the same order to obtain the eigensignals of the target order and modulus. Iterate through the order and modulus to obtain all eigensignals.

3. The forward Raman nonlinear suppression and amplification method for OAM optical communication as described in claim 1 or 2, characterized in that, The step-path gain algorithm for a single set of intrinsic signals includes: The attenuation rate corresponding to the order and modulus of a single intrinsic signal is determined based on the attenuation characteristic table. The attenuation rate is equal to the product of the reference attenuation rate with order 1 and modulus 1 and the exponential correction term of the order and modulus of the single intrinsic signal. The exponential correction term is equal to the absolute value of the order factor multiplied by the order minus 1, the modulus factor multiplied by the modulus minus 1, the cross factor, and the product of the absolute value of the order minus 1 and the modulus minus 1. The attenuation characteristic table is determined by fitting the attenuation of intrinsic signals with different orders and modulus over the same transmission distance. The smaller value between the product of the ratio of the attenuation rate to the average attenuation rate plus 1 and the base gain and the upper limit gain is used as the signal gain, and pre-amplification is performed on the single intrinsic signal.

4. The forward Raman nonlinearity suppression and amplification method for OAM optical communication as described in claim 1, characterized in that, Dispersion pre-compensation includes: acquiring the pulse delay difference of a single amplified intrinsic signal and dividing it by the total wavelength bandwidth and total axial length to obtain the dispersion amount, which is defined as the dispersion amount in this case; concatenating the absolute value of the order, the modulus, and the ratio of order to modulus of the single amplified intrinsic signal into an OAM parameter vector, and generating the order control vector and the modulus control vector through mapping by a fully connected layer; performing a Hadamard product between the order control vector and the modulus control vector to obtain the input control vector, and summing the absolute value of the order control vector with the modulus control vector to obtain the forgetting control vector; initializing the previous... The hidden state vector of the [number]th time, compared with the previous [number]th time... The historical dispersion values ​​of the same group number are spliced ​​together to form the previous number. The joint input vector of the second time; combining the input control vector, the forgotten control vector, and the previous first-order joint input vector. The joint input vector is input into the modulation timing model and a cooperative loop mechanism is executed to deduce the hidden state vector for this step. The predicted dispersion is obtained by mapping it to a fully connected layer through a self-attention mechanism. After summing the dispersion of the single-set amplified intrinsic signal, dispersion pre-compensation is performed to obtain a single-set compensated intrinsic signal. This is the total number for historical reference.

5. The forward Raman nonlinear suppression and amplification method for OAM optical communication as described in claim 4, characterized in that, The cooperative loop mechanism of the modulation timing model includes: […]. The joint input vector and the input control vector are nonlinearly mapped to obtain the previous [number]th [number] [unit]. The input gating vector of the previous th time; The joint input vector and the forgetting control vector of the previous step are nonlinearly mapped to obtain the result of the previous step. The forgetting gating vector of the previous time; The forgetting gating vector of the second time is the same as the previous one. Perform the Hadamard product on the hidden state vector of the nth time, and add the previous nth time... The Hadamard product of the latest learning vector and the input gated vector is obtained from the previous [number]th [number]. The hidden state vector of the th iteration; the latest learned vector is equal to the previously learned th iteration. The hyperbolic tangent modulation mapping result of the joint input vector is used to cyclically derive the hidden state vector for this time.

6. The forward Raman nonlinearity suppression and amplification method for OAM optical communication as described in claim 1, characterized in that, Add 1 to the integer part of the ratio of the total wavelength bandwidth of the OAM mode to the upper limit wavelength bandwidth of the pump core light source to obtain the total number of pump wavelengths. The transmission loss of a single set of compensated intrinsic signals is collected and the corresponding target Raman gain is set as the superposition of the product of the transmission loss and the Raman fiber loss coefficient, resulting in redundant gain. Summing up the target Raman gains of all groups yields the total target Raman gain. Multiplying the ratio of the target Raman gain of a single compensated intrinsic signal to the total target Raman gain by the total pump power yields the corresponding pump group power. The axial length of a single segment is determined by evenly dividing the transmission fiber according to the total number of injection points. The ratio of the injection point number to the total number of injection points is multiplied by the corresponding pump group power to obtain the matching power of a single set of compensated intrinsic signals at the injection point corresponding to the injection point number.

7. The forward Raman nonlinear suppression and amplification method for OAM optical communication as described in claim 1, characterized in that, Suppressing FWM clutter includes: using the square of the signal light output power of the single-group compensated intrinsic signal at a single injection point, the pump light output power of the same group number at the single injection point, a fixed constant coefficient, and the product of 1 minus the exponential attenuation rate of the same group number as the FWM clutter power of the single-group compensated intrinsic signal at a single injection point; comparing the FWM clutter power with the FWM threshold, if it is greater than the threshold, the phase modulation depth is set to the first value, and if it is less than or equal to the threshold, it is set to the second value; combining the PRBS random sequence and the phase modulation depth to generate random phase perturbation, multiplying the power value of the random perturbation phase as the exponent with the single-group compensated intrinsic signal to destroy the nonlinear phase matching condition of FWM.

8. The forward Raman nonlinearity suppression and amplification method for OAM optical communication as described in claim 6, characterized in that, The matching pause time of a single-group compensated intrinsic signal is the same at all injection points; the matching pause time is equal to the product of the basic Raman setup time and the linear recursive formula of the order and the modulus; the linear recursive formula is equal to 1 plus the product of the order influence coefficient and the order plus the product of the modulus influence coefficient and the modulus.

9. The forward Raman nonlinearity suppression and amplification method for OAM optical communication as described in claim 6, characterized in that, The signal light output power of a single injection point of a single compensated intrinsic signal is equal to the product of the signal light output power of the single compensated intrinsic signal at the previous injection point and the product of the power of the single-segment target Raman gain of the single compensated intrinsic signal between the previous injection point and the single injection point, with 10 as the base and one-tenth as the exponent; the single-segment target Raman gain is equal to the target Raman gain of the same group number multiplied by the ratio of the injection point number to the total number of injection points; The pump light output power of a single injection point in the same group number is equal to the sum of the pump light output power of the previous injection point in the same group number and the signal light output power of a single group of compensated intrinsic signals in a single injection point, multiplied by the exponential decay rate of the corresponding order and modulus. The exponential decay rate is a power of the product of the decay rate of the same group number and the single segment axis length, with the natural constant as the base.

10. The forward Raman nonlinear suppression and amplification method for OAM optical communication as described in claim 1 or 2, characterized in that, The OAM composite signal is the sum of multiple intrinsic signals and follows the Laguerre-Gaussian function; a single intrinsic signal is equal to the product of the amplitude, the radial field distribution function of the corresponding order and modulus, the angular spiral phase of the corresponding order, and the axial propagation phase of the corresponding order and modulus.

Images