Sample generation method and device, computer equipment, storage medium and product

Abstract

The invention relates to the technical field of communication, in particular to a sample generation method and device, computer equipment, a storage medium and a product. Comprising the following steps: receiving sample light path data of a light path simulated by a generator; determining a target light path attenuation coefficient of the light path according to the sample light path data; performing confidence analysis on the sample light path data according to the target light path attenuation coefficient to obtain a confidence probability corresponding to the sample light path data; and sending the sample light path data and the confidence probability corresponding to the sample light path data to a generator, so that the generator adjusts the sample light path data of the light path according to the sample light path data and the confidence probability, and the confidence probability of the adjusted sample light path data is close to a standard value, according to the invention, it is ensured that the generator adjusts the sample light path data of the light path according to the sample light path data and the confidence probability, so that the confidence probability of the adjusted sample light path data is close to a standard value, and then it is ensured that the subsequent generator can be used for model training when outputting the sample light path data.

Description

Technical Field

[0001] This application relates to the field of communication technology, and in particular to a sample generation method, apparatus, computer equipment, storage medium, and product. Background Technology

[0002] With the continuous evolution of communication technologies, optical access networks are gradually developing from static to dynamic. To make fuller use of physical layer resources and improve network resource utilization efficiency, precise awareness of the network's real-time operational status is necessary. In modern optical networks, the application of coherent transmission technology provides a large number of flexibly configurable parameters for optical path deployment. To evaluate the effectiveness of a given optical path configuration in real time and optimize the overall network design, it is crucial to quickly and accurately estimate the Quality of Transmission (QoT) of candidate optical paths before deployment.

[0003] However, in actual optical network deployments, due to the limited number of optical paths deployed in the early networks and the lack of monitoring equipment at network nodes, it is difficult to collect sufficient data for model training. Summary of the Invention

[0004] Therefore, it is necessary to provide a sample generation method, apparatus, computer equipment, storage medium, and product that can collect sufficient data for model training to address the aforementioned technical problems.

[0005] Firstly, this application provides a sample generation method. The method includes:

[0006] Receive sample optical path data of the optical path simulated by the generator;

[0007] Based on the sample optical path data, determine the target optical path attenuation coefficient of the optical path;

[0008] Based on the target optical path attenuation coefficient, a confidence analysis is performed on the sample optical path data to obtain the confidence probability corresponding to the sample optical path data.

[0009] The sample optical path data and the corresponding confidence probability are sent to the generator so that the generator adjusts the sample optical path data of the optical path according to the sample optical path data and the confidence probability, so that the confidence probability of the adjusted sample optical path data is close to the standard value.

[0010] In one embodiment, determining the target optical path attenuation coefficient of the optical path based on the sample optical path data includes:

[0011] Obtain the target transmission quality data of the optical path at the time when the sample optical path data is generated;

[0012] Based on the target transmission quality data, optical path attenuation analysis is performed on the optical path to obtain the target optical path attenuation coefficient corresponding to the sample optical path data.

[0013] In one embodiment, the step of performing optical path attenuation analysis on the optical path based on the target transmission quality data to obtain the target optical path attenuation coefficient corresponding to the sample optical path data includes:

[0014] Based on the correspondence between different transmission quality data and optical path attenuation coefficients, a target optical path attenuation coefficient that matches the target transmission quality data is determined.

[0015] In one embodiment, the step of performing confidence analysis on the sample optical path data based on the target optical path attenuation coefficient to obtain the confidence probability corresponding to the sample optical path data includes:

[0016] Determine the actual optical path attenuation coefficient and the tolerance error corresponding to the sample optical path data;

[0017] Confidence analysis is performed on the sample optical path data based on the target optical path attenuation coefficient, the actual optical path attenuation coefficient, and the tolerance error to obtain the confidence probability corresponding to the sample optical path data.

[0018] In one embodiment, the step of performing confidence analysis on the sample optical path data based on the target optical path attenuation coefficient, the actual optical path attenuation coefficient, and the tolerance error to obtain the confidence probability corresponding to the sample optical path data includes:

[0019] Determine the probability decay rate corresponding to the tolerance error;

[0020] The result of the calculation is to determine the difference between the target optical path attenuation coefficient and the actual optical path attenuation coefficient;

[0021] The confidence probability corresponding to the sample optical path data is determined based on the ratio between the difference calculation result and the attenuation rate.

[0022] In one embodiment, the sample optical path data is data output by the generator based on random noise.

[0023] Secondly, this application also provides a sample generation apparatus. The apparatus includes:

[0024] The receiving module is used to receive sample optical path data of the optical path simulated by the generator;

[0025] The determining module is used to determine the target optical path attenuation coefficient of the optical path based on the sample optical path data.

[0026] The analysis module is used to perform confidence analysis on the sample optical path data based on the target optical path attenuation coefficient to obtain the confidence probability corresponding to the sample optical path data.

[0027] The sending module is used to send the sample optical path data and the confidence probability corresponding to the sample optical path data to the generator, so that the generator adjusts the sample optical path data of the optical path according to the sample optical path data and the confidence probability, so that the confidence probability of the adjusted sample optical path data is close to the standard value.

[0028] Thirdly, this application also provides a computer device. The computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to perform the following steps:

[0029] Receive sample optical path data of the optical path simulated by the generator;

[0030] Based on the sample optical path data, determine the target optical path attenuation coefficient of the optical path;

[0031] Based on the target optical path attenuation coefficient, a confidence analysis is performed on the sample optical path data to obtain the confidence probability corresponding to the sample optical path data.

[0032] The sample optical path data and the corresponding confidence probability are sent to the generator so that the generator adjusts the sample optical path data of the optical path according to the sample optical path data and the confidence probability, so that the confidence probability of the adjusted sample optical path data is close to the standard value.

[0033] Fourthly, this application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program thereon, which, when executed by a processor, performs the following steps:

[0034] Receive sample optical path data of the optical path simulated by the generator;

[0035] Based on the sample optical path data, determine the target optical path attenuation coefficient of the optical path;

[0036] Based on the target optical path attenuation coefficient, a confidence analysis is performed on the sample optical path data to obtain the confidence probability corresponding to the sample optical path data.

[0037] The sample optical path data and the corresponding confidence probability are sent to the generator so that the generator adjusts the sample optical path data of the optical path according to the sample optical path data and the confidence probability, so that the confidence probability of the adjusted sample optical path data is close to the standard value.

[0038] Fifthly, this application also provides a computer program product. The computer program product includes a computer program that, when executed by a processor, performs the following steps:

[0039] Receive sample optical path data of the optical path simulated by the generator;

[0040] Based on the sample optical path data, determine the target optical path attenuation coefficient of the optical path;

[0041] Based on the target optical path attenuation coefficient, a confidence analysis is performed on the sample optical path data to obtain the confidence probability corresponding to the sample optical path data.

[0042] The sample optical path data and the corresponding confidence probability are sent to the generator so that the generator adjusts the sample optical path data of the optical path according to the sample optical path data and the confidence probability, so that the confidence probability of the adjusted sample optical path data is close to the standard value.

[0043] The aforementioned sample generation method, apparatus, computer equipment, storage medium, and product receive sample optical path data of the simulated optical path by the generator; determine the target optical path attenuation coefficient of the optical path based on the sample optical path data; then perform confidence analysis on the sample optical path data based on the target optical path attenuation coefficient to obtain the confidence probability corresponding to the sample optical path data; finally, send the sample optical path data and the corresponding confidence probability to the generator so that the generator adjusts the sample optical path data of the optical path based on the sample optical path data and the confidence probability, so that the confidence probability of the adjusted sample optical path data is close to the standard value. As described above, in the process of sample generation, this application pre-acquires sample optical path data of the generator's simulated optical path. Then, it uses the target optical path attenuation coefficient of the optical path corresponding to the sample optical path data to determine the confidence probability corresponding to the sample optical path data. Furthermore, the confidence probability characterizes the data generation validity of the sample optical path data. By sending the sample optical path data and the corresponding confidence probability to the generator, it ensures that the generator adjusts the sample optical path data of the optical path according to the sample optical path data and the confidence probability, so that the confidence probability of the adjusted sample optical path data is close to the standard value. Thus, it ensures that the generator can use the output sample optical path data for model training. Attached Figure Description

[0044] Figure 1 An application environment diagram of a sample generation method provided in this application embodiment;

[0045] Figure 2 A flowchart illustrating the first sample generation method provided in this application embodiment;

[0046] Figure 3A flowchart illustrating the second sample generation method provided in this application embodiment;

[0047] Figure 4 A flowchart illustrating the third sample generation method provided in this application embodiment;

[0048] Figure 5 A flowchart illustrating the fourth sample generation method provided in this application embodiment;

[0049] Figure 6 A structural block diagram of a sample generation device provided in an embodiment of this application;

[0050] Figure 7 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0051] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0052] With the continuous evolution of communication technology, optical access networks are gradually developing from static to dynamic. To make fuller use of physical layer resources and improve network resource utilization efficiency, accurate real-time perception of the network's operational status is necessary. In modern optical networks, the application of coherent transmission technology provides a large number of flexibly configurable parameters for optical path deployment. To evaluate the effectiveness of a given optical path configuration scheme in real time and optimize the overall network design, it is crucial to quickly and accurately estimate the Quality of Transmission (QoT) of candidate optical paths before deployment. In recent years, machine learning methods have been increasingly applied to QoT estimation due to their flexibility, efficiency, speed, and accuracy. However, regardless of the machine learning method, a large number of training samples are required to train the model. In actual optical network deployments, due to the small number of optical paths deployed in early networks and the lack of monitoring equipment at network nodes, it is difficult to collect sufficient data for model training. This lack of training samples leads to insufficient QoT estimation accuracy in practical applications.

[0053] To ensure sufficient sample data can be collected for model training, data-driven deep neural networks (DNNs) can be used to enhance the optical path dataset by combining physical analytical formulas, i.e., Gaussian noise models. This increases the number of samples without introducing other data costs when the number of QoT optical path samples is limited.

[0054] Furthermore, Deep Neural Networks (DNNs) are machine learning models that mimic the connection patterns of neurons in the human brain. Composed of multiple hidden layers, they automatically learn complex feature representations from data. They compute predictions through forward propagation and then adjust network parameters using backpropagation and gradient descent optimization algorithms (such as Adam) to minimize loss functions (such as cross-entropy or mean squared error). The power of deep neural networks lies in their hierarchical structure: shallow layers learn basic features (such as edges and textures), while deeper layers combine these features to form high-level semantics (such as objects and scenes). Common variants include Convolutional Neural Networks (CNNs) for processing grid data, Recurrent Neural Networks (RNNs) for modeling sequences, and Transformers based on self-attention. Through its powerful non-linear modeling capabilities and hierarchical feature learning, DNNs have become a core technology in the field of data generation, and their development is evolving towards greater efficiency, interpretability, and generality.

[0055] In optical communication networks, the optical signal-to-noise ratio (OSNR) is generally used to estimate the performance of coherent optical systems. The OSNR of channel i is specifically defined as:

[0056] ;

[0057] in, This represents the transmission power of the optical signal at the transmitting end. This represents the noise power generated by the amplified spontaneous emission (ASE) of an optical amplifier, often referring to the linear noise experienced by the signal. This represents the nonlinear noise power caused by inter-channel interference affecting the signal.

[0058] In optical communication networks, the Gaussian noise model (GN-model) is a relatively simple and reliable tool for simulating random interference encountered by optical signals during transmission. The GN-model assumes that the noise follows a Gaussian distribution (also known as a normal distribution), with its probability density function having a bell-shaped curve symmetrical to the mean.

[0059] According to the GN model, the ASE noise power generated by N amplifiers is defined as:

[0060] ;

[0061] Where G represents the gain generated by the EDFA; B represents the fixed bandwidth of each channel; represents the spontaneous emissivity of the EDFA, which is proportional to the noise figure; h represents the Planck coefficient; and v represents the center frequency of the optical signal during transmission in the optical fiber. Therefore, in optical communication systems, The noise power is relatively readily available and fixed, depending only on the number and parameters of the amplifiers. Therefore, when estimating the uncertainty of the channel OSNR, the main focus is on the uncertainty caused by its nonlinear interference.

[0062] According to the GN model, the nonlinear noise of the i-th channel can be defined as:

[0063] ;

[0064] in, It is the total number of channels. It is the channel phase modulation factor, which can be specifically expressed as:

[0065] ;

[0066] ;

[0067] Where γ represents the fiber nonlinearity coefficient, α represents the fiber attenuation coefficient, and β² represents the fiber second-order dispersion coefficient. This represents the frequency interval between channel i and channel j.

[0068] QoT estimation algorithm based on artificial neural network (ANN)

[0069] In the problem of QoT regression prediction in optical networks, many studies employ classic ANNs to construct QoT regression prediction models. This model collects transmission configuration parameters of deployed optical paths, such as the total length of the optical path, the length of the longest link in the path, the number of fiber segments traversed, the total number of links, and transmission power, to construct an input vector x representing the optical path information. The QoT values ​​of the optical paths measured by OPMs deployed in the network, such as OSNR measurements, are used as sample labels y. Assuming that the relevant transmission parameters x and corresponding QoT measurements y of N deployed optical paths in the network are collected, a training dataset is constructed to train the model.

[0070] The sample generation method provided in this application embodiment can be applied to, for example... Figure 1In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be integrated onto server 104 or placed on a cloud or other network server. By receiving sample optical path data simulated by the generator, the system determines the target optical path attenuation coefficient based on the sample optical path data. Then, it performs confidence analysis on the sample optical path data based on the target optical path attenuation coefficient to obtain the confidence probability corresponding to the sample optical path data. Finally, it sends the sample optical path data and the corresponding confidence probability to the generator, so that the generator adjusts the sample optical path data according to the sample optical path data and the confidence probability, making the confidence probability of the adjusted sample optical path data closer to the standard value. Terminal 102 can be, but is not limited to, various personal computers, laptops, smartphones, tablets, IoT devices, and portable wearable devices. IoT devices can be smart speakers, smart TVs, smart air conditioners, smart in-vehicle devices, etc. Portable wearable devices can be smartwatches, smart bracelets, head-mounted devices, etc. Server 104 can be implemented using a standalone server or a server cluster consisting of multiple servers.

[0071] In one embodiment, such as Figure 2 As shown, a sample generation method is provided, which can be applied to... Figure 1 Taking server 104 as an example, the following steps are included:

[0072] S201 receives sample optical path data of the optical path simulated by the generator.

[0073] It should be noted that the sample optical path data is the data output by the generator based on random noise.

[0074] Therefore, the generator in this application, when outputting sample optical path data, may include the following: the generator receives a random noise z as input and generates sample optical path data; wherein, the loss function corresponding to the generator... The goal of generator G is to enable discriminator D to determine the confidence probability of the output sample optical path data of the generated sample optical path data. The generator should be as close as possible to a standard value (e.g., the standard value could be 1). By minimizing this loss function, the generator obtains the highest probability that the discriminator classifies it as true.

[0075] S202, Determine the target optical path attenuation coefficient based on the sample optical path data.

[0076] Among them, the target optical path attenuation coefficient is a key parameter for measuring the degree of power loss when an optical signal is transmitted in an optical path.

[0077] In one embodiment of this application, when it is necessary to determine the target optical path attenuation coefficient of an optical path based on sample optical path data, the following may be included: obtaining the target transmission quality data of the optical path at the time when the sample optical path data is generated; performing optical path attenuation analysis on the optical path based on the target transmission quality data to obtain the target optical path attenuation coefficient corresponding to the sample optical path data.

[0078] In another embodiment of this application, when it is necessary to determine the target optical path attenuation coefficient of the optical path based on the sample optical path data, the following may also be included: inputting the sample optical path data into a pre-trained attenuation coefficient analysis model and obtaining the output result of the attenuation coefficient analysis model, which is the target optical path attenuation coefficient of the optical path.

[0079] The training process of the attenuation coefficient analysis model includes: acquiring at least one reference optical path data, determining the reference optical path attenuation coefficient corresponding to the reference optical path data through manual annotation, and then training the initial model based on the reference optical path data annotated with the reference optical path attenuation coefficient to obtain the trained attenuation coefficient analysis model.

[0080] S203, perform confidence analysis on the sample optical path data based on the target optical path attenuation coefficient to obtain the confidence probability corresponding to the sample optical path data.

[0081] It should be noted that when it is necessary to perform confidence analysis on the sample optical path data based on the target optical path attenuation coefficient to obtain the confidence probability corresponding to the sample optical path data, the following can be included: determining the true optical path attenuation coefficient and the tolerance error corresponding to the sample optical path data; performing confidence analysis on the sample optical path data based on the target optical path attenuation coefficient, the true optical path attenuation coefficient, and the tolerance error to obtain the confidence probability corresponding to the sample optical path data.

[0082] S204, send the sample optical path data and the corresponding confidence probability to the generator so that the generator can adjust the sample optical path data of the optical path according to the sample optical path data and the confidence probability, so that the confidence probability of the adjusted sample optical path data is close to the standard value.

[0083] In one embodiment of this application, it is continuously verified whether the confidence probability of the sample optical path data output by the generator is greater than a preset threshold. If the confidence probability of the sample optical path data is greater than the preset threshold, it indicates that the sample optical path data output by the generator at this time has met the conditions for being used as training samples for the model. Therefore, by generating samples through the generator, the ultimate goal of expanding the optical path dataset is achieved, avoiding overfitting of the machine learning model due to insufficient training data.

[0084] In one embodiment of this application, the number of times sample optical path data and the confidence probability corresponding to the sample optical path data are sent to the generator can be counted. Then, when the number of times reaches the threshold, it is determined that the sample optical path data output by the generator at this time meets the conditions for being used as training samples for the model. Therefore, the generator generates samples to achieve the ultimate goal of expanding the optical path dataset and avoids overfitting of the machine learning model due to insufficient training data.

[0085] The above-described sample generation method receives sample optical path data from the simulated optical path by the generator; determines the target optical path attenuation coefficient based on the sample optical path data; then performs confidence analysis on the sample optical path data based on the target optical path attenuation coefficient to obtain the confidence probability corresponding to the sample optical path data; finally, sends the sample optical path data and the corresponding confidence probability to the generator so that the generator adjusts the sample optical path data based on the sample optical path data and the confidence probability, making the confidence probability of the adjusted sample optical path data closer to the standard value. As described above, in the process of sample generation, this application pre-acquires sample optical path data of the generator's simulated optical path. Then, it uses the target optical path attenuation coefficient of the optical path corresponding to the sample optical path data to determine the confidence probability corresponding to the sample optical path data. Furthermore, the confidence probability characterizes the data generation validity of the sample optical path data. By sending the sample optical path data and the corresponding confidence probability to the generator, it ensures that the generator adjusts the sample optical path data of the optical path according to the sample optical path data and the confidence probability, so that the confidence probability of the adjusted sample optical path data is close to the standard value. Thus, it ensures that the generator can use the output sample optical path data for model training.

[0086] In one embodiment, such as Figure 3 As shown, when it is necessary to determine the target optical path attenuation coefficient based on sample optical path data, the following can be included:

[0087] S301, acquire the target transmission quality data of the optical path at the moment the sample optical path data is generated.

[0088] It should be noted that when it is necessary to determine the target transmission quality data, the following calculation formula can be used:

[0089] ;

[0090] in, This refers to the target transmission quality data. For noise power, This represents the transmission power of the optical signal at the transmitting end. It is the channel phase modulation factor.

[0091] S302, perform optical path attenuation analysis on the optical path based on the target transmission quality data to obtain the target optical path attenuation coefficient corresponding to the sample optical path data.

[0092] It should be noted that when it is necessary to perform optical path attenuation analysis on the optical path based on the target transmission quality data to obtain the target optical path attenuation coefficient corresponding to the sample optical path data, the following can be included: based on the correspondence between different transmission quality data and optical path attenuation coefficients, determine the target optical path attenuation coefficient that matches the target transmission quality data.

[0093] In one embodiment of this application, the correspondence between different transmission quality data and optical path attenuation coefficients can be recorded, with different transmission quality data corresponding to different optical path attenuation coefficients. Therefore, after determining the target transmission quality data, a reference transmission quality parameter that is the same as the target transmission quality data is selected from at least one transmission quality parameter included in the correspondence, and the optical path attenuation coefficient corresponding to the reference transmission quality parameter in the correspondence is used as the target optical path attenuation coefficient that matches the target transmission quality data.

[0094] In another embodiment of this application, the correspondence between different transmission quality data and optical path attenuation coefficient can be characterized by the following calculation formula:

[0095] ;

[0096] Here, α refers to the target optical path attenuation coefficient; therefore, by substituting the relevant parameters into the above calculation formula, the target optical path attenuation coefficient that matches the target transmission quality data is obtained.

[0097] The above sample generation method uses the target transmission quality data of the optical path at the moment of sample optical path data generation to determine the order of the target optical path attenuation coefficient, which provides a data basis for obtaining the confidence probability corresponding to the sample optical path data and ensures the smooth progress of subsequent processes.

[0098] In one embodiment, such as Figure 4 As shown, when it is necessary to perform confidence analysis on sample optical path data based on the target optical path attenuation coefficient to obtain the confidence probability corresponding to the sample optical path data, the following can be included:

[0099] S401, determine the actual optical path attenuation coefficient and the tolerance error corresponding to the sample optical path data.

[0100] Among them, the real optical path attenuation coefficient refers to the optical path attenuation coefficient under real conditions; the tolerance error refers to the degree of difference between the sample optical path data and the real optical path data that is allowed to exist.

[0101] S402, perform confidence analysis on the sample optical path data based on the target optical path attenuation coefficient, the actual optical path attenuation coefficient, and the tolerance error to obtain the confidence probability corresponding to the sample optical path data.

[0102] It should be noted that when it is necessary to perform confidence analysis on sample optical path data based on the target optical path attenuation coefficient, the actual optical path attenuation coefficient, and the tolerance error to obtain the confidence probability corresponding to the sample optical path data, the following can be included: determining the probability attenuation rate corresponding to the tolerance error; determining the result of the difference calculation between the target optical path attenuation coefficient and the actual optical path attenuation coefficient; and determining the confidence probability corresponding to the sample optical path data based on the ratio between the difference calculation result and the attenuation rate.

[0103] Based on the above, the formula for calculating the confidence probability is as follows:

[0104] ;

[0105] in, This refers to the confidence probability. This refers to the actual optical path attenuation coefficient. δ refers to the target optical path attenuation coefficient; δ refers to the probability attenuation rate corresponding to the tolerance error.

[0106] The above sample generation method ensures that the generator adjusts the sample optical path data according to the sample optical path data and the confidence probability by using the confidence probability corresponding to the sample optical path data. This makes the confidence probability of the adjusted sample optical path data close to the standard value, thereby ensuring that the generator can use the output sample optical path data for model training.

[0107] In one embodiment, such as Figure 5 As shown, when it is necessary to send the sample optical path data and the corresponding confidence probability to the generator, the following can be included:

[0108] S501 receives sample optical path data of the optical path simulated by the generator.

[0109] S502, acquire the target transmission quality data of the optical path at the moment the sample optical path data is generated.

[0110] S503 determines the target optical path attenuation coefficient that matches the target transmission quality data based on the correspondence between different transmission quality data and optical path attenuation coefficient.

[0111] S504 determines the actual optical path attenuation coefficient and the tolerance error corresponding to the sample optical path data.

[0112] S505 determines the probability decay rate corresponding to the tolerance error.

[0113] S506, determine the result of the calculation of the difference between the target optical path attenuation coefficient and the actual optical path attenuation coefficient.

[0114] S507 determines the confidence probability corresponding to the sample optical path data based on the ratio between the difference calculation result and the attenuation rate.

[0115] S508 sends the sample optical path data and the corresponding confidence probability to the generator so that the generator adjusts the sample optical path data of the optical path according to the sample optical path data and the confidence probability, so that the confidence probability of the adjusted sample optical path data is close to the standard value.

[0116] In one embodiment of this application, the optical path sample data enhancement process based on digital-analog co-processing may include the following: a generator receives random noise z as input; the generator generates data G(z) of the corresponding dimension; a discriminator receives the generated data G(z) as input; the discriminator references a physical model, substitutes the sample data into the physical formula, determines whether the data is true based on whether the calculated parameter values ​​are within the confidence range, and outputs the confidence probability D(G(z)) that the data is true. The generator uses... The loss function is used to "trick" the discriminator as much as possible to obtain a higher discrimination probability; the final generator produces usable samples to achieve the ultimate goal of expanding the optical path dataset.

[0117] The above-described sample generation method receives sample optical path data from the simulated optical path by the generator; determines the target optical path attenuation coefficient based on the sample optical path data; then performs confidence analysis on the sample optical path data based on the target optical path attenuation coefficient to obtain the confidence probability corresponding to the sample optical path data; finally, sends the sample optical path data and the corresponding confidence probability to the generator so that the generator adjusts the sample optical path data based on the sample optical path data and the confidence probability, making the confidence probability of the adjusted sample optical path data closer to the standard value. As described above, in the process of sample generation, this application pre-acquires sample optical path data of the generator's simulated optical path. Then, it uses the target optical path attenuation coefficient of the optical path corresponding to the sample optical path data to determine the confidence probability corresponding to the sample optical path data. Furthermore, the confidence probability characterizes the data generation validity of the sample optical path data. By sending the sample optical path data and the corresponding confidence probability to the generator, it ensures that the generator adjusts the sample optical path data of the optical path according to the sample optical path data and the confidence probability, so that the confidence probability of the adjusted sample optical path data is close to the standard value. Thus, it ensures that the generator can use the output sample optical path data for model training.

[0118] It should be understood that although the steps in the flowcharts of the above embodiments are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the above embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0119] Based on the same inventive concept, this application also provides a sample generation apparatus for implementing the sample generation method described above. The solution provided by this apparatus is similar to the implementation described in the above method; therefore, the specific limitations in one or more sample generation apparatus embodiments provided below can be found in the limitations of the sample generation method described above, and will not be repeated here.

[0120] In one embodiment, such as Figure 6 As shown, a sample generation device is provided, including: a receiving module 10, a determining module 20, an analyzing module 30, and a sending module 40, wherein:

[0121] The receiving module 10 is used to receive sample optical path data of the optical path simulated by the generator.

[0122] The determination module 20 is used to determine the target optical path attenuation coefficient of the optical path based on the sample optical path data.

[0123] The analysis module 30 is used to perform confidence analysis on the sample optical path data based on the target optical path attenuation coefficient to obtain the confidence probability corresponding to the sample optical path data.

[0124] The sending module 40 is used to send the sample optical path data and the confidence probability corresponding to the sample optical path data to the generator, so that the generator adjusts the sample optical path data of the optical path according to the sample optical path data and the confidence probability, so that the confidence probability of the adjusted sample optical path data is close to the standard value.

[0125] In one embodiment, target transmission quality data of the optical path is acquired at the moment when the sample optical path data is generated;

[0126] Based on the target transmission quality data, optical path attenuation analysis is performed on the optical path to obtain the target optical path attenuation coefficient corresponding to the sample optical path data.

[0127] In one embodiment, a target optical path attenuation coefficient that matches the target transmission quality data is determined based on the correspondence between different transmission quality data and optical path attenuation coefficients.

[0128] In one embodiment, the actual optical path attenuation coefficient and the tolerance error corresponding to the sample optical path data are determined.

[0129] Confidence analysis is performed on the sample optical path data based on the target optical path attenuation coefficient, the actual optical path attenuation coefficient, and the tolerance error to obtain the confidence probability corresponding to the sample optical path data.

[0130] In one embodiment, the probability decay rate corresponding to the tolerance error is determined;

[0131] The result of calculating the difference between the target optical path attenuation coefficient and the actual optical path attenuation coefficient;

[0132] The confidence probability corresponding to the sample optical path data is determined based on the ratio between the difference calculation result and the attenuation rate.

[0133] In one embodiment, the sample optical path data is data output by the generator based on random noise.

[0134] The aforementioned sample generation device receives sample optical path data from the simulated optical path of the generator; determines the target optical path attenuation coefficient based on the sample optical path data; then performs confidence analysis on the sample optical path data based on the target optical path attenuation coefficient to obtain the confidence probability corresponding to the sample optical path data; finally, sends the sample optical path data and the corresponding confidence probability to the generator, so that the generator adjusts the sample optical path data of the optical path based on the sample optical path data and the confidence probability, making the confidence probability of the adjusted sample optical path data closer to the standard value. As described above, in the process of sample generation, this application pre-acquires sample optical path data of the generator's simulated optical path. Then, it uses the target optical path attenuation coefficient of the optical path corresponding to the sample optical path data to determine the confidence probability corresponding to the sample optical path data. Furthermore, the confidence probability characterizes the data generation validity of the sample optical path data. By sending the sample optical path data and the corresponding confidence probability to the generator, it ensures that the generator adjusts the sample optical path data of the optical path according to the sample optical path data and the confidence probability, so that the confidence probability of the adjusted sample optical path data is close to the standard value. Thus, it ensures that the generator can use the output sample optical path data for model training.

[0135] Each module in the aforementioned sample generation device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the operations corresponding to each module.

[0136] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 7 As shown, the computer device includes a processor, memory, input / output interface, communication interface, display unit, and input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interface. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The input / output interface is used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements a sample generation method. The display unit is used to form a visually visible image and can be a display screen, projection device, or virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0137] Those skilled in the art will understand that Figure 7 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0138] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0139] Receive sample optical path data of the optical path simulated by the generator;

[0140] Based on the sample optical path data, determine the target optical path attenuation coefficient of the optical path;

[0141] Confidence analysis is performed on the sample optical path data based on the target optical path attenuation coefficient to obtain the confidence probability corresponding to the sample optical path data.

[0142] The sample optical path data and the corresponding confidence probability are sent to the generator so that the generator can adjust the sample optical path data according to the sample optical path data and the confidence probability, so that the confidence probability of the adjusted sample optical path data is close to the standard value.

[0143] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0144] Acquire the target transmission quality data of the optical path at the moment the sample optical path data is generated;

[0145] Based on the target transmission quality data, optical path attenuation analysis is performed on the optical path to obtain the target optical path attenuation coefficient corresponding to the sample optical path data.

[0146] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0147] Based on the correspondence between different transmission quality data and optical path attenuation coefficients, the target optical path attenuation coefficient that matches the target transmission quality data is determined.

[0148] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0149] Determine the actual optical path attenuation coefficient and the tolerance error corresponding to the sample optical path data;

[0150] Confidence analysis is performed on the sample optical path data based on the target optical path attenuation coefficient, the actual optical path attenuation coefficient, and the tolerance error to obtain the confidence probability corresponding to the sample optical path data.

[0151] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0152] Determine the probability decay rate corresponding to the tolerance error;

[0153] The result of calculating the difference between the target optical path attenuation coefficient and the actual optical path attenuation coefficient;

[0154] The confidence probability corresponding to the sample optical path data is determined based on the ratio between the difference calculation result and the attenuation rate.

[0155] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0156] The sample optical path data is the data output by the generator based on random noise.

[0157] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, the computer program performing the following steps when executed by a processor:

[0158] Receive sample optical path data of the optical path simulated by the generator;

[0159] Based on the sample optical path data, determine the target optical path attenuation coefficient of the optical path;

[0160] Confidence analysis is performed on the sample optical path data based on the target optical path attenuation coefficient to obtain the confidence probability corresponding to the sample optical path data.

[0161] The sample optical path data and the corresponding confidence probability are sent to the generator so that the generator can adjust the sample optical path data according to the sample optical path data and the confidence probability, so that the confidence probability of the adjusted sample optical path data is close to the standard value.

[0162] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0163] Acquire the target transmission quality data of the optical path at the moment the sample optical path data is generated;

[0164] Based on the target transmission quality data, optical path attenuation analysis is performed on the optical path to obtain the target optical path attenuation coefficient corresponding to the sample optical path data.

[0165] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0166] Based on the correspondence between different transmission quality data and optical path attenuation coefficients, the target optical path attenuation coefficient that matches the target transmission quality data is determined.

[0167] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0168] Determine the actual optical path attenuation coefficient and the tolerance error corresponding to the sample optical path data;

[0169] Confidence analysis is performed on the sample optical path data based on the target optical path attenuation coefficient, the actual optical path attenuation coefficient, and the tolerance error to obtain the confidence probability corresponding to the sample optical path data.

[0170] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0171] Determine the probability decay rate corresponding to the tolerance error;

[0172] The result of calculating the difference between the target optical path attenuation coefficient and the actual optical path attenuation coefficient;

[0173] The confidence probability corresponding to the sample optical path data is determined based on the ratio between the difference calculation result and the attenuation rate.

[0174] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0175] The sample optical path data is the data output by the generator based on random noise.

[0176] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, performs the following steps:

[0177] Receive sample optical path data of the optical path simulated by the generator;

[0178] Based on the sample optical path data, determine the target optical path attenuation coefficient of the optical path;

[0179] Confidence analysis is performed on the sample optical path data based on the target optical path attenuation coefficient to obtain the confidence probability corresponding to the sample optical path data.

[0180] The sample optical path data and the corresponding confidence probability are sent to the generator so that the generator can adjust the sample optical path data according to the sample optical path data and the confidence probability, so that the confidence probability of the adjusted sample optical path data is close to the standard value.

[0181] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0182] Acquire the target transmission quality data of the optical path at the moment the sample optical path data is generated;

[0183] Based on the target transmission quality data, optical path attenuation analysis is performed on the optical path to obtain the target optical path attenuation coefficient corresponding to the sample optical path data.

[0184] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0185] Based on the correspondence between different transmission quality data and optical path attenuation coefficients, the target optical path attenuation coefficient that matches the target transmission quality data is determined.

[0186] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0187] Determine the actual optical path attenuation coefficient and the tolerance error corresponding to the sample optical path data;

[0188] Confidence analysis is performed on the sample optical path data based on the target optical path attenuation coefficient, the actual optical path attenuation coefficient, and the tolerance error to obtain the confidence probability corresponding to the sample optical path data.

[0189] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0190] Determine the probability decay rate corresponding to the tolerance error;

[0191] The result of calculating the difference between the target optical path attenuation coefficient and the actual optical path attenuation coefficient;

[0192] The confidence probability corresponding to the sample optical path data is determined based on the ratio between the difference calculation result and the attenuation rate.

[0193] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0194] The sample optical path data is the data output by the generator based on random noise.

[0195] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of related data must comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0196] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0197] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0198] The above embodiments are merely illustrative of several implementation methods of this application, and their descriptions are relatively specific and detailed. However, they should not be construed as limiting the scope of this application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A sample generation method, characterized in that, The method includes: Receive sample optical path data of the optical path simulated by the generator; Based on the sample optical path data, determine the target optical path attenuation coefficient of the optical path; Based on the target optical path attenuation coefficient, a confidence analysis is performed on the sample optical path data to obtain the confidence probability corresponding to the sample optical path data. The sample optical path data and the corresponding confidence probability are sent to the generator so that the generator adjusts the sample optical path data of the optical path according to the sample optical path data and the confidence probability, so that the confidence probability of the adjusted sample optical path data is close to the standard value.

2. The method according to claim 1, characterized in that, The step of determining the target optical path attenuation coefficient of the optical path based on the sample optical path data includes: Obtain the target transmission quality data of the optical path at the time when the sample optical path data is generated; Based on the target transmission quality data, optical path attenuation analysis is performed on the optical path to obtain the target optical path attenuation coefficient corresponding to the sample optical path data.

3. The method according to claim 2, characterized in that, The step of performing optical path attenuation analysis on the optical path based on the target transmission quality data to obtain the target optical path attenuation coefficient corresponding to the sample optical path data includes: Based on the correspondence between different transmission quality data and optical path attenuation coefficients, a target optical path attenuation coefficient that matches the target transmission quality data is determined.

4. The method according to claim 1, characterized in that, The step of performing confidence analysis on the sample optical path data based on the target optical path attenuation coefficient to obtain the confidence probability corresponding to the sample optical path data includes: Determine the actual optical path attenuation coefficient and the tolerance error corresponding to the sample optical path data; Confidence analysis is performed on the sample optical path data based on the target optical path attenuation coefficient, the actual optical path attenuation coefficient, and the tolerance error to obtain the confidence probability corresponding to the sample optical path data.

5. The method according to claim 4, characterized in that, The step of performing confidence analysis on the sample optical path data based on the target optical path attenuation coefficient, the actual optical path attenuation coefficient, and the tolerance error to obtain the confidence probability corresponding to the sample optical path data includes: Determine the probability decay rate corresponding to the tolerance error; The result of the calculation is to determine the difference between the target optical path attenuation coefficient and the actual optical path attenuation coefficient; The confidence probability corresponding to the sample optical path data is determined based on the ratio between the difference calculation result and the attenuation rate.

6. The method according to any one of claims 1-5, characterized in that, The sample optical path data is the data output by the generator based on random noise.

7. A sample generation device, characterized in that, The device includes: The receiving module is used to receive sample optical path data of the optical path simulated by the generator; The determining module is used to determine the target optical path attenuation coefficient of the optical path based on the sample optical path data. The analysis module is used to perform confidence analysis on the sample optical path data based on the target optical path attenuation coefficient to obtain the confidence probability corresponding to the sample optical path data. The sending module is used to send the sample optical path data and the confidence probability corresponding to the sample optical path data to the generator, so that the generator adjusts the sample optical path data of the optical path according to the sample optical path data and the confidence probability, so that the confidence probability of the adjusted sample optical path data is close to the standard value.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images