A method and device for suppressing inter-mode crosstalk between multi-core optical fiber super-Nyquist signals

Abstract

The present invention relates to the field of optical fiber technology, and more particularly to a method, device, apparatus, and computer storage medium for suppressing inter-mode crosstalk between super-Nyquist signals in a multi-core optical fiber. The method utilizes a nonlinear decision-feedback neural network equalizer to effectively improve equalization performance, suppress inter-mode crosstalk between super-Nyquist signals in a multi-core optical fiber, and reduce bit error rates.

Description

Technical Field

[0001] The present invention relates to the field of optical fiber technology, and in particular to a method, equipment, device and computer storage medium for suppressing inter-mode crosstalk between multi-core optical fiber super-Nyquist signals. Background Art

[0002] To overcome the nonlinear constraints imposed by fiber nonlinearity on single-mode fiber, space-division multiplexing technology in multi-core fiber can provide an additional dimension based on wavelength-division multiplexing and polarization multiplexing technologies, further increasing the capacity of a single fiber. Multi-core fiber is a new type of fiber that has more cores than single-mode fiber, and the signals transmitted in each core will have crosstalk. However, due to the limitations of current MIMO-DSP deployment and fiber manufacturing methods, further increasing capacity has become very difficult. In addition, to improve spectral efficiency, super-Nyquist technology is used. The interaction between mode coupling and inter-symbol interference caused by this technology exhibits a nonlinear distortion. Since traditional MIMO-DSP has a linear structure, which increases signal noise, its performance is poor in channels with severe crosstalk and low optical signal-to-noise ratio, making it difficult to balance the signal damage exhibited by the inter-mode crosstalk interaction caused by super-Nyquist technology. Summary of the Invention

[0003] Therefore, the technical problem to be solved by the present invention is to overcome the problem in the prior art that the linear structure MIMO-DSP is difficult to equalize the inter-mode crosstalk caused by the super-Nyquist technology.

[0004] In order to solve the above technical problems, the present invention provides a method for suppressing inter-mode crosstalk between multi-core optical fiber super-Nyquist signals, comprising:

[0005] In the digital domain, the randomly generated bit sequence is modulated into a transmission signal through QAM;

[0006] Dividing the transmission signal into multiple channels, filtering each channel through super-Nyquist filtering, and modulating the multiple channels into optical signals;

[0007] The multi-channel optical signals are transmitted through the transmitting ends of different cores of a multi-core optical fiber, and spatial demultiplexing and polarization demultiplexing are performed at the receiving end to obtain multi-channel analog signals;

[0008] Converting the multi-channel analog signals into multi-channel digital signals;

[0009] The multi-channel digital signals are input in parallel into a trained decision feedback neural network equalizer for equalization, thereby suppressing inter-mode crosstalk between multi-core optical fiber super-Nyquist signals.

[0010] Preferably, the training process of the decision feedback neural network equalizer is:

[0011] Step 1: Initialize feedback input is a vector of all 0s;

[0012] Step 2: Take any set of normalized multi-channel digital signals X and feedback input in the training set At the same time, the input is input into the decision feedback neural network equalizer, and after being processed by the first fully connected layer, it is processed by the activation function, and the output is processed by the second fully connected layer to obtain the network output Y;

[0013] Step 3: Based on the network output Y and the corresponding reference signal Calculating losses Update the network through loss value and backpropagation algorithm;

[0014] Step 4: Send the network output Y to the decision device for decision, and assign the decision signal to the feedback input

[0015] Step 5: Determine whether the loss has converged. If not, repeat steps 2-5 until the loss converges and the training is completed.

[0016] Preferably, the step of inputting the multiple digital signals in parallel into a trained decision feedback neural network equalizer for equalization to suppress inter-mode crosstalk between multi-core optical fiber super-Nyquist signals comprises:

[0017] Step 1: Initialize feedback input is a vector of all 0s;

[0018] Step 2: The normalized first set of multi-channel digital signals X and feedback input At the same time, the input is input into the decision feedback neural network equalizer, and after being processed by the first fully connected layer, it is processed by the activation function, and the output is processed by the second fully connected layer to obtain the network output Y;

[0019] Step 3: Send the network output Y to the decision maker for decision, and update the feedback input according to the decision signal.

[0020] Step 4: Use the next set of multi-channel digital signals and the updated feedback input as the input of the neural network, and repeat the above steps 2-4 until all the data are processed.

[0021] Preferably, the normalization of the multiple digital signals includes:

[0022] Normalize the mean and variance of the multi-channel digital signal X so that the mean of the data is 0 and the variance is 1. The formula is as follows:

[0023]

[0024] Among them, X mean and σ represent the mean and variance of the multi-channel digital signal X respectively.

[0025] Preferably, the normalized multi-channel digital signals X and At the same time, the decision feedback neural network equalizer is input, and after being processed by the first fully connected layer, it is processed by the activation function, and the output is processed by the second fully connected layer to obtain the network output.

[0026] Among them, the signal in the i-th fiber core is x i ∈X, the feedback decision signal of the i-th fiber core is n means that the multi-core fiber has n modes of signal, W k and b k represents the weight matrix and bias matrix, k = 1, 2, and f(·) represents the activation function.

[0027] Preferably, the sending the network output Y to a decision device for decision-making includes:

[0028] Each signal in the network output is judged in turn. If it is greater than the preset threshold, the judgment is 1; if it is less than the preset threshold, the judgment is 0.

[0029] Preferably, the preset threshold is 0.5.

[0030] The present invention also provides a multi-core optical fiber super-Nyquist signal inter-mode crosstalk suppression device, comprising:

[0031] A transmission signal acquisition module is used to modulate a randomly generated bit sequence into a transmission signal through QAM in the digital domain;

[0032] An optical signal acquisition module is used to divide the transmission signal into multiple channels, and respectively pass super-Nyquist filtering to modulate them into multiple optical signals;

[0033] An analog signal acquisition module is used to transmit the multi-channel optical signals through the transmitting ends of different cores of the multi-core optical fiber, and perform spatial demultiplexing and polarization demultiplexing at the receiving end to obtain multi-channel analog signals;

[0034] an analog-to-digital conversion module, configured to convert the multiple analog signals into multiple digital signals;

[0035] The crosstalk suppression module is used to input the multi-channel digital signals in parallel into a trained decision feedback neural network equalizer for equalization, thereby suppressing inter-mode crosstalk between multi-core optical fiber super-Nyquist signals.

[0036] The present invention also provides a multi-core optical fiber super-Nyquist signal inter-mode crosstalk suppression device, comprising:

[0037] The memory is used to store computer programs; the processor is used to implement the above-mentioned method steps for suppressing inter-mode crosstalk between multi-core optical fiber super-Nyquist signals when executing the computer program.

[0038] The present invention also provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of the above-mentioned method for suppressing inter-mode crosstalk between multi-core optical fiber super-Nyquist signals are implemented.

[0039] The above technical solution of the present invention has the following advantages over the prior art:

[0040] The method for suppressing inter-mode crosstalk between super-Nyquist signals in multi-core optical fibers described in the present invention utilizes a decision feedback neural network equalizer with a nonlinear structure to effectively improve equalization performance, suppress inter-mode crosstalk between super-Nyquist signals in multi-core optical fibers, and reduce the bit error rate. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to make the content of the present invention more clearly understood, the present invention is further described in detail below based on specific embodiments of the present invention in conjunction with the accompanying drawings, wherein:

[0042] Figure 1 This is a flow chart of a method for suppressing inter-mode crosstalk between multi-core optical fiber super-Nyquist signals according to the present invention;

[0043] Figure 2 It is a schematic diagram of the decision feedback neural network structure;

[0044] Figure 3 It is a schematic diagram of Q factor under different cutoff frequencies and balanced conditions;

[0045] Figure 4 This is a structural block diagram of a multi-core optical fiber super-Nyquist signal inter-mode crosstalk suppression device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0046] The core of the present invention is to provide a method, device, equipment and computer storage medium for suppressing inter-mode crosstalk between super-Nyquist signals in multi-core optical fibers, which effectively suppresses inter-mode crosstalk between super-Nyquist signals in multi-core optical fibers.

[0047] In order to enable those skilled in the art to better understand the present invention, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0048] Please refer to Figure 1 , Figure 1 This is a flowchart of a method for suppressing inter-mode crosstalk between multi-core optical fiber super-Nyquist signals provided by the present invention; the specific operation steps are as follows:

[0049] S101: In the digital domain, a randomly generated bit sequence is modulated into a transmission signal through QAM.

[0050] S102: Split the transmission signal into multiple channels, and perform super-Nyquist filtering on each channel to modulate the signals into multiple optical signals;

[0051] S103: transmitting the multiple optical signals through transmitting ends of different cores of a multi-core optical fiber, and performing spatial demultiplexing and polarization demultiplexing at a receiving end to obtain multiple analog signals;

[0052] Since it is a multi-core optical fiber, there are multiple cores in one cladding, so space demultiplexing is required at the receiving end. The purpose is to separate the data in each core; the demodulated multiple independent data signals are transmitted simultaneously by different polarization states of light.

[0053] S104: Converting the multiple analog signals into multiple digital signals;

[0054] Since the decision feedback neural network equalizer is a digital equalizer, it is necessary to convert the input analog signal into a digital signal.

[0055] S105: Inputting the multiple digital signals in parallel into a trained decision feedback neural network equalizer for equalization to suppress inter-mode crosstalk between multi-core optical fiber super-Nyquist signals.

[0056] The method for suppressing inter-mode crosstalk between super-Nyquist signals in multi-core optical fibers described in the present invention utilizes a decision feedback neural network equalizer with a nonlinear structure to effectively improve equalization performance, suppress inter-mode crosstalk between super-Nyquist signals in multi-core optical fibers, and reduce the bit error rate.

[0057] like Figure 2 Based on the above embodiments, this embodiment describes the decision feedback neural network equalizer in detail, as follows:

[0058] The training process of the decision feedback network equalizer is as follows:

[0059] Step 1: Initialize feedback input is a vector of all 0s;

[0060] Step 2: Take any set of normalized multi-channel digital signals X and feedback input in the training set At the same time, the input is input into the decision feedback neural network equalizer, and after being processed by the first fully connected layer, it is processed by the activation function, and the output is processed by the second fully connected layer to obtain the network output Y;

[0061] Step 3: Based on the network output Y and the corresponding reference signal Calculating losses Update the network through loss value and backpropagation algorithm;

[0062] Step 4: Send the network output Y to the decision device for decision, and assign the decision signal to the feedback input

[0063] Step 5: Determine whether the loss has converged. If not, repeat steps 2-5 until the loss converges and the training is completed.

[0064] The step of inputting the multiple digital signals in parallel into a trained decision feedback network equalizer for equalization to suppress inter-mode crosstalk between multi-core optical fiber super-Nyquist signals comprises:

[0065] Step 1: Initialize feedback input is a vector of all 0s;

[0066] Step 2: The normalized first set of multi-channel digital signals X and feedback input At the same time, the input is input into the decision feedback neural network equalizer, and after being processed by the first fully connected layer, it is processed by the activation function, and the output is processed by the second fully connected layer to obtain the network output Y;

[0067] Step 3: Send the network output Y to the decision maker for decision, and update the feedback input according to the decision signal.

[0068] Step 4: Use the next set of multi-channel digital signals and the updated feedback input as the input of the neural network, and repeat the above steps 2-4 until all the data are processed.

[0069] The normalization of the multi-channel digital signals includes:

[0070] Normalize the mean and variance of the multi-channel digital signal X so that the mean of the data is 0 and the variance is 1. The formula is as follows:

[0071]

[0072] Among them, X mean and σ represent the mean and variance of the multi-channel digital signal X. This can improve the model accuracy and the convergence speed of the equalizer.

[0073] The normalized multi-channel digital signals X and At the same time, the input is input into the neural network, processed by the first layer of fully connected layer, processed by the activation function, and the output is processed by the second layer of fully connected layer to obtain the network output

[0074] Among them, the signal in the i-th fiber core is x i ∈X, the feedback decision signal of the i-th fiber core is n means that the multi-core optical fiber has n modes of signal, W k and b k represents the weight matrix and bias matrix, k = 1, 2, and f(·) represents the activation function.

[0075] The sending of the network output Y into the decision device for decision-making comprises:

[0076] Each signal in the network output is judged in turn. If it is greater than a preset threshold, the judgment is 1, and if it is less than the preset threshold, the judgment is 0. The preset threshold is 0.5.

[0077] Based on the above embodiments, this embodiment verifies the effectiveness of the present invention through specific experiments:

[0078] Four-core, 100km transmission and super-Nyquist signaling setup: The baud rate is set to 100 Gbaud. Therefore, using the QAM-16 format, the total capacity across the eight spatial modes of the multi-core few-mode fiber is 3.2 Tbps. The fiber's attenuation coefficient is 0.18 dB / km, the dispersion coefficient is 18 ps / nm / km, and the nonlinear coefficient is 0.81 W / m. When the transmit power is 0 dBm, the OSNR is 20 dB. The coupling length of the four-core few-mode fiber is 100 m.

[0079] By adjusting the cutoff frequency of the super-Nyquist filter to 35GHz, 37.5GHz, and 40GHz respectively. A classic 8×8 MIMO-DSP is used, and 30,000 known symbols are used to update 11 forward taps during training. Similarly, a training sequence of the same length is applied to a DFNNE containing 11 feedforward inputs and 5 feedback inputs. The number of neurons in the hidden layer and output layer is 64 and 16 respectively. By changing the transmit power conditions, the Q factor under different cutoff frequencies and equalization conditions is recorded, such as Figure 3As shown in Figure 2, when FTN filtering is used, processing the coupled signal using classic MIMO-DSP significantly degrades signal transmission performance. Typically, at a transmit power of 7dBm, the Q factor drops by nearly 10dB at a cutoff frequency of 40GHz. When the cutoff frequencies are 37.5GHz and 35GHz, respectively, this algorithm achieves 9dB and 6dB Q gains for the coupled 8-mode signal compared to traditional MIMO-DSP. This is because classic MIMO-DSP has limitations in handling the crosstalk caused by FTN filtering.

[0080] Please refer to Figure 4 , Figure 4 A structural block diagram of a multi-core optical fiber super-Nyquist signal inter-mode crosstalk suppression device provided by an embodiment of the present invention; the specific device may include:

[0081] The transmission signal acquisition module 100 is used to modulate the randomly generated bit sequence into a transmission signal through QAM in the digital domain;

[0082] The optical signal acquisition module 200 is used to divide the transmission signal into multiple channels, and respectively pass super-Nyquist filtering to modulate them into multiple optical signals;

[0083] The analog signal acquisition module 300 is used to transmit the multi-channel optical signals through the transmitting ends of different cores of the multi-core optical fiber, and perform spatial demultiplexing and polarization demultiplexing at the receiving end to obtain multi-channel analog signals;

[0084] The analog-to-digital conversion module 400 is configured to convert the multiple analog signals into multiple digital signals;

[0085] The crosstalk suppression module 500 is used to input the multi-channel digital signals in parallel into a trained decision feedback neural network equalizer for equalization, thereby suppressing inter-mode crosstalk between multi-core optical fiber super-Nyquist signals.

[0086] The machine vision-based surface defect detection device of this embodiment is used to implement the aforementioned multi-core optical fiber super-Nyquist signal inter-mode crosstalk suppression method. Therefore, the specific implementation method of the multi-core optical fiber super-Nyquist signal inter-mode crosstalk suppression device can be seen in the embodiment part of the multi-core optical fiber super-Nyquist signal inter-mode crosstalk suppression method mentioned above. For example, the sending signal acquisition module 100, the optical signal acquisition module 200, the analog signal acquisition module 300, the analog-to-digital conversion module 400, and the crosstalk suppression module 500 are respectively used to implement steps S101, S102, S103, S104 and S105 in the above-mentioned multi-core optical fiber super-Nyquist signal inter-mode crosstalk suppression method. Therefore, its specific implementation method can refer to the description of the corresponding each part of the embodiment, and will not be repeated here.

[0087] A specific embodiment of the present invention further provides a device for suppressing inter-mode crosstalk between multi-core optical fiber super-Nyquist signals, comprising: a memory for storing a computer program; and a processor for implementing the steps of the above-mentioned method for suppressing inter-mode crosstalk between multi-core optical fiber super-Nyquist signals when executing the computer program.

[0088] A specific embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of the above-mentioned method for suppressing inter-mode crosstalk between multi-core optical fiber super-Nyquist signals are implemented.

[0089] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0090] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the steps in the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0091] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0092] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1a process or multiple processes and / or boxes Figure 1 The steps for the function specified in one or more boxes.

[0093] Obviously, the above embodiments are merely examples for clarity of explanation and are not intended to limit the implementation methods. Those skilled in the art will appreciate that other variations or modifications can be made based on the above description. It is not necessary and impossible to enumerate all implementation methods here. Obvious variations or modifications arising therefrom remain within the scope of protection of the present invention.

Claims

1. A method for suppressing inter-mode crosstalk between multi-core optical fiber super-Nyquist signals, characterized in that: include: In the digital domain, the randomly generated bit sequence is modulated into a transmission signal through QAM; Dividing the transmission signal into multiple channels, filtering each channel through super-Nyquist filtering, and modulating the multiple channels into optical signals; The multi-channel optical signals are transmitted through the transmitting ends of different cores of a multi-core optical fiber, and spatial demultiplexing and polarization demultiplexing are performed at the receiving end to obtain multi-channel analog signals; Converting the multi-channel analog signals into multi-channel digital signals; The multi-channel digital signals are input in parallel into a trained decision feedback neural network equalizer for equalization, thereby suppressing inter-mode crosstalk between multi-core optical fiber super-Nyquist signals.

2. The method for suppressing inter-mode crosstalk between multi-core optical fiber super-Nyquist signals according to claim 1, characterized in that: The training process of the decision feedback neural network equalizer is: Step 1: Initialize feedback input is a vector of all 0s; Step 2: Take any set of normalized multi-channel digital signals X and feedback input in the training set At the same time, the input is input into the decision feedback neural network equalizer, and after being processed by the first fully connected layer, it is processed by the activation function, and the output is processed by the second fully connected layer to obtain the network output Y; Step 3: Based on the network output Y and the corresponding reference signal Calculating losses Update the network through loss value and backpropagation algorithm; Step 4: Send the network output Y to the decision device for decision, and assign the decision signal to the feedback input Step 5: Determine whether the loss has converged. If not, repeat steps 2-5 until the loss converges and the training is completed.

3. The method for suppressing inter-mode crosstalk between multi-core optical fiber super-Nyquist signals according to claim 1, characterized in that: The step of inputting the multiple digital signals in parallel into a trained decision feedback neural network equalizer for equalization to suppress inter-mode crosstalk between multi-core optical fiber super-Nyquist signals comprises: Step 1: Initialize feedback input is a vector of all 0s; Step 2: The normalized first set of multi-channel digital signals X and feedback input At the same time, the input is input into the decision feedback neural network equalizer, and after being processed by the first fully connected layer, it is processed by the activation function, and the output is processed by the second fully connected layer to obtain the network output Y; Step 3: Send the network output Y to the decision maker for decision, and update the feedback input according to the decision signal. Step 4: Use the next set of multi-channel digital signals and the updated feedback input as the input of the neural network, and repeat the above steps 2-4 until all the data are processed.

4. The method for suppressing inter-mode crosstalk between multi-core optical fiber super-Nyquist signals according to claim 2 or 3, characterized in that: The normalization of the multi-channel digital signals includes: Normalize the mean and variance of the multi-channel digital signal X so that the mean of the data is 0 and the variance is 1. The formula is as follows: Among them, X mean and σ represent the mean and variance of the multi-channel digital signal X respectively.

5. The method for suppressing inter-mode crosstalk between multi-core optical fiber super-Nyquist signals according to claim 2 or 3, characterized in that: The normalized multi-channel digital signals X and At the same time, the decision feedback neural network equalizer is input, and after being processed by the first fully connected layer, it is processed by the activation function, and the output is processed by the second fully connected layer to obtain the network output. Among them, the signal in the i-th fiber core is x i ∈X, the feedback decision signal of the i-th fiber core is n means that the multi-core optical fiber has n modes of signal, W k and b k represents the weight matrix and bias matrix, k = 1, 2, and f(·) represents the activation function.

6. The method for suppressing inter-mode crosstalk between multi-core optical fiber super-Nyquist signals according to claim 2 or 3, characterized in that: The sending of the network output Y into the decision device for decision-making comprises: Each signal in the network output is judged in turn. If it is greater than the preset threshold, the judgment is 1; if it is less than the preset threshold, the judgment is 0.

7. The method for suppressing inter-mode crosstalk between multi-core optical fiber super-Nyquist signals according to claim 6, characterized in that: The preset threshold is 0.

5.

8. A multi-core optical fiber super-Nyquist signal inter-mode crosstalk suppression device, characterized in that: include: A transmission signal acquisition module is used to modulate a randomly generated bit sequence into a transmission signal through QAM in the digital domain; An optical signal acquisition module is used to divide the transmission signal into multiple channels, and respectively pass super-Nyquist filtering to modulate them into multiple optical signals; An analog signal acquisition module is used to transmit the multi-channel optical signals through the transmitting ends of different cores of the multi-core optical fiber, and perform spatial demultiplexing and polarization demultiplexing at the receiving end to obtain multi-channel analog signals; an analog-to-digital conversion module, configured to convert the multiple analog signals into multiple digital signals; The crosstalk suppression module is used to input the multi-channel digital signals in parallel into a trained decision feedback neural network equalizer for equalization, thereby suppressing inter-mode crosstalk between multi-core optical fiber super-Nyquist signals.

9. A multi-core optical fiber super-Nyquist signal inter-mode crosstalk suppression device, characterized in that: include: Memory for storing computer programs; A processor is configured to implement the steps of a method for suppressing inter-mode crosstalk between multi-core optical fiber super-Nyquist signals as claimed in any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the method for suppressing inter-mode crosstalk between multi-core optical fiber super-Nyquist signals as claimed in any one of claims 1 to 7 are implemented.

Images