Optical module data acquisition and analysis method and system

Abstract

The invention relates to the field of intelligent detection, and provides an optical module data acquisition and analysis method and system, which adopt a time series data analysis and processing technology based on deep learning. And whether the optical module has a fault at the current time point is judged by performing time sequence analysis on the transmitting optical power values, the receiving optical power values and the error rate data of the optical module at the plurality of time points. Therefore, the operation state of the optical module can be known more comprehensively, the trend of performance change can be captured, and the accuracy and reliability of fault detection of the optical module can be improved.

Description

Technical Field

[0001] The present application relates to the field of intelligent detection, and more specifically, to an optical module data acquisition and analysis method and system. Background Art

[0002] An optical module is an optoelectronic device that performs photoelectric and electro-optical conversion. Its main function is to realize the mutual conversion between optical signals and electrical signals. At the transmitting end, it converts electrical signals into optical signals and then transmits them through optical fibers; at the receiving end, it converts the received optical signals back into electrical signals for subsequent processing by electronic devices.

[0003] In traditional optical module data collection and analysis, people often only focus on the optical module operation data at a single time point, which easily leads to a lack of comprehensive understanding of the optical module operation status. In addition, in traditional methods, the thresholds of various optical module data are often simply judged. Once the data exceeds a certain fixed range, it is judged as a fault. This method is too simple and crude, and it is easy to make misjudgments. For example, in actual network communications, optical modules will be affected by environmental factors, equipment aging, etc., and their performance may gradually decline. Relying on unchanging thresholds to judge will affect the accuracy of the detection results.

[0004] Therefore, an optimized optical module data collection and analysis solution is needed. Summary of the invention

[0005] In view of the shortcomings of the prior art, this application provides a method and system for collecting and analyzing optical module data, which adopts time series data analysis and processing technology based on deep learning, and judges whether the optical module is faulty at the current time point by performing time series analysis on the optical module's transmitted optical power value, received optical power value, and bit error rate data at multiple time points. In this way, the operating status of the optical module can be more comprehensively understood, the trend of performance changes can be captured, and the accuracy and reliability of optical module fault detection can be improved.

[0006] A method for collecting and analyzing optical module data, comprising: Obtaining the transmit optical power value, the receive optical power value and the bit error rate of the optical module at multiple predetermined time points within a predetermined time period; Performing a fusion timing analysis on the transmitted optical power values ​​and the received optical power values ​​at the plurality of predetermined time points to obtain an optical power timing feature vector; Performing time series analysis on the bit error rates at the plurality of predetermined time points to obtain a bit error rate time series feature vector; Performing attribute association expression optimization based on feature space compatibility adjustment on the optical power time series feature vector and the bit error rate time series feature vector to obtain an optical module detection feature vector; A detection result of the optical module is obtained based on the optical module detection feature vector.

[0007] In the optical module data acquisition and analysis method provided above, the transmitted optical power values ​​and the received optical power values ​​at the multiple predetermined time points are subjected to a fused timing analysis to obtain an optical power timing feature vector, including: arranging the transmitted optical power values ​​at the multiple predetermined time points into a transmitted optical power timing input vector according to the time dimension and then passing through an optical power timing encoder to obtain a transmitted optical power timing feature vector; arranging the received optical power values ​​at the multiple predetermined time points into a received optical power timing input vector according to the time dimension and then passing through the optical power timing encoder to obtain a received optical power timing feature vector; fusing the transmitted optical power timing feature vector and the received optical power timing feature vector to obtain the optical power timing feature vector.

[0008] In the optical module data acquisition and analysis method provided above, the bit error rates of the multiple predetermined time points are subjected to timing analysis to obtain a bit error rate timing feature vector, including: arranging the bit error rates of the multiple predetermined time points into a bit error rate timing input vector according to the time dimension and then passing it through a bit error rate timing encoder to obtain the bit error rate timing feature vector.

[0009] In the optical module data acquisition and analysis method provided above, the bit error rate timing encoder is a convolutional neural network model including a fully connected layer and a one-dimensional convolutional layer, and the optical power timing encoder is a convolutional neural network model including a first convolutional layer and a second convolutional layer, the first convolutional layer uses a convolutional neural network model with a one-dimensional convolutional kernel of a first scale, and the second convolutional layer uses a convolutional neural network model with a one-dimensional convolutional kernel of a second scale, wherein the first scale is different from the second scale.

[0010] In the optical module data acquisition and analysis method provided above, the optical power timing feature vector and the bit error rate timing feature vector are optimized by attribute association expression based on feature space compatibility adjustment to obtain an optical module detection feature vector, including: constructing an optical module detection fusion distance topological matrix between the optical power timing feature vector and the bit error rate timing feature vector; constructing an optical module detection fusion feature value granularity association matrix between the optical power timing feature vector and the bit error rate timing feature vector; performing topological association modulation on the optical module detection fusion feature value granularity association matrix based on the optical module detection fusion distance topological matrix to obtain an optical module detection fusion topological modulation association matrix; using the optical module detection fusion topological modulation association matrix as the perspective co-projection space, projecting the optical power timing feature vector and the bit error rate timing feature vector into the perspective co-projection space to obtain a perspective modulation optical power timing feature vector and a perspective modulation bit error rate timing feature vector; fusing the perspective modulation optical power timing feature vector and the perspective modulation bit error rate timing feature vector to obtain the optical module detection feature vector.

[0011] In the optical module data collection and analysis method provided above, the optical module detection fusion eigenvalue granularity association matrix is ​​topologically associated modulated based on the optical module detection fusion distance topological matrix to obtain the optical module detection fusion topological modulation association matrix, including: creating a new Spring Boot project and adding the dependencies required by the Spring Boot project in the pom.xml file; creating a model class, the model class is used to represent the optical module detection fusion distance topological matrix, the optical module detection fusion eigenvalue granularity association matrix and the optical module detection fusion topological modulation association matrix; creating a service class, the service class is used to implement the logic of topological association modulation; creating a controller class, the controller class is used to process HTTP requests; starting the Spring Boot project, and packaging and deploying the Spring Boot project to the server.

[0012] In the optical module data collection and analysis method provided above, the detection result of the optical module is obtained based on the optical module detection feature vector, including: passing the optical module detection feature vector through a detection classifier to obtain the detection result, and the detection result is used to indicate whether the optical module is faulty at the current time point.

[0013] An optical module data acquisition and analysis system, comprising: The optical module data acquisition module is used to obtain the transmitted optical power value, received optical power value and bit error rate of the optical module at multiple predetermined time points within a predetermined time period; An optical module optical power data encoding module, used for performing a fusion timing analysis on the transmitted optical power values ​​and the received optical power values ​​at the plurality of predetermined time points to obtain an optical power timing feature vector; An optical module bit error rate data encoding module, used for performing a time series analysis on the bit error rates at the plurality of predetermined time points to obtain a bit error rate time series feature vector; An optical module related feature fusion module, used for optimizing the attribute association expression of the optical power timing feature vector and the bit error rate timing feature vector based on feature space compatibility adjustment to obtain an optical module detection feature vector; The optical module detection result generating module is used to obtain the detection result of the optical module based on the optical module detection feature vector.

[0014] In the optical module data acquisition and analysis system provided above, the optical module optical power data encoding module includes: a transmission optical power timing feature extraction unit, which is used to arrange the transmission optical power values ​​of the multiple predetermined time points into a transmission optical power timing input vector according to the time dimension and then pass it through an optical power timing encoder to obtain a transmission optical power timing feature vector; a reception optical power timing feature extraction unit, which is used to arrange the reception optical power values ​​of the multiple predetermined time points into a reception optical power timing input vector according to the time dimension and then pass it through the optical power timing encoder to obtain a reception optical power timing feature vector; an optical power timing feature fusion unit, which is used to fuse the transmission optical power timing feature vector and the reception optical power timing feature vector to obtain the optical power timing feature vector.

[0015] In the optical module data collection and analysis system provided above, the optical module detection result generation module is used to: pass the optical module detection feature vector through a detection classifier to obtain the detection result, and the detection result is used to indicate whether the optical module is faulty at the current time point.

[0016] This application has significant technical effects due to the adoption of the above technical solutions: The optical module data collection and analysis method and system provided by the present application adopts the time series data analysis and processing technology based on deep learning, and judges whether the optical module is faulty at the current time point by performing time series analysis on the optical module's transmitted optical power value, received optical power value and bit error rate data at multiple time points. In this way, the operating status of the optical module can be more comprehensively understood, the trend of performance changes can be captured, and the accuracy and reliability of optical module fault detection can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present application will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0018] Figure 1 The figure is a flow chart of a method for collecting and analyzing optical module data according to an embodiment of the present application.

[0019] Figure 2 Schematic diagram of data flow of the optical module data collection and analysis method according to an embodiment of the present application.

[0020] Figure 3This is a flow chart of performing a fusion timing analysis on the transmitted optical power values ​​and the received optical power values ​​at the plurality of predetermined time points in the optical module data collection and analysis method according to an embodiment of the present application to obtain an optical power timing feature vector.

[0021] Figure 4 4 is a system block diagram of an optical module data acquisition and analysis system according to an embodiment of the present application. DETAILED DESCRIPTION

[0022] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0023] As an optoelectronic device, the optical module plays a vital role in the field of modern communications. It ensures that information can be accurately transmitted in the optical fiber network by realizing the mutual conversion between optical signals and electrical signals. Specifically, at the transmitting end, the optical module can convert the electrical signals generated by electronic devices (such as computers, servers, etc.) into optical signals with the help of the internal photoelectric conversion mechanism. These optical signals are then transmitted over long distances along the optical fiber. At the receiving end, the optical module can accurately convert the optical signals transmitted through the optical fiber back into electrical signals, so that subsequent electronic devices can process these signals and realize data communication and interaction.

[0024] In the traditional practice of optical module data collection and analysis, the focus is often only on data acquisition at a single time point. This single-time point data collection method cannot reflect the changes in the operating status of the optical module in different time periods, which leads to a lack of comprehensiveness and systematic understanding of the operating status of the optical module. In most cases, fixed thresholds are simply set for various data of the optical module to make judgments. For example, a fixed upper and lower threshold is set for the key indicator of the optical module's transmitted optical power. When the collected data exceeds this range, it is directly determined that the optical module is faulty. However, in the actual network communication environment, the operation of the optical module is affected by many complex factors. On the one hand, environmental factors have a significant impact on the performance of the optical module. On the other hand, equipment aging is also a factor that cannot be ignored. As the optical module runs for a long time, the internal photoelectric conversion element will gradually age. In this case, simply judging based on fixed thresholds is prone to misjudgment.

[0025] In recent years, deep learning and neural networks have been widely used in computer vision, natural language processing, text signal processing and other fields. In addition, deep learning and neural networks have also shown a level close to or even beyond that of humans in image classification, object detection, semantic segmentation, text translation and other fields. The development of deep learning and neural networks provides new solutions and solutions for optical module data collection and analysis.

[0026] Figure 1 The figure is a flow chart of a method for collecting and analyzing optical module data according to an embodiment of the present application. Figure 2 Schematic diagram of data flow of the optical module data collection and analysis method according to an embodiment of the present application. Figure 1 and Figure 2 As shown, the optical module data collection and analysis method according to the embodiment of the present application includes: S110, obtaining the transmitted optical power value, received optical power value and bit error rate of the optical module at multiple predetermined time points within a predetermined time period; S120, performing a fusion timing analysis on the transmitted optical power value and the received optical power value at the multiple predetermined time points to obtain an optical power timing feature vector; S130, performing a timing analysis on the bit error rates at the multiple predetermined time points to obtain a bit error rate timing feature vector; S140, performing attribute association expression optimization based on feature space compatibility adjustment on the optical power timing feature vector and the bit error rate timing feature vector to obtain an optical module detection feature vector; S150, obtaining the detection result of the optical module based on the optical module detection feature vector.

[0027] In step S110, the emission optical power value, the received optical power value and the bit error rate of the optical module at multiple predetermined time points within a predetermined time period are obtained. It should be understood that the emission optical power refers to the output optical power of the light source at the transmitting end of the optical module, which can be understood as the intensity of light. It reflects the ability of the optical module to send signals. During the normal operation of the optical module, the emission optical power needs to be kept within a certain range to ensure that the signal can be accurately and stably transmitted to the receiving end. The received optical power refers to the minimum optical power that the optical module can receive at the receiving end. During the communication process, the size of the received optical power will directly affect the transmission quality and bit error rate of the signal. If the received optical power is too low, the signal may not be received correctly, resulting in bit errors. The bit error rate refers to the ratio of misplaced bits generated between the transmitting end and the receiving end in the number of bits transmitted during the communication process. It reflects the accuracy of the optical module transmission signal. In general, the three parameters of emission optical power, received optical power and bit error rate are interrelated, and they jointly reflect the working state of the optical module. By comprehensively evaluating the changes in these three parameters, the performance and health of the optical module can be more comprehensively understood.

[0028] In step S120, the transmitted optical power values ​​and the received optical power values ​​at the plurality of predetermined time points are subjected to a fusion time series analysis to obtain an optical power time series feature vector. Specifically, Figure 3 The present invention is a flowchart of performing a fusion time series analysis on the transmitted optical power values ​​and received optical power values ​​at the plurality of predetermined time points in the optical module data collection and analysis method according to the embodiment of the present application to obtain an optical power time series feature vector. Figure 3 As shown, the transmitted optical power values ​​and the received optical power values ​​at the multiple predetermined time points are subjected to a fused timing analysis to obtain an optical power timing characteristic vector, including: S121, arranging the transmitted optical power values ​​at the multiple predetermined time points into a transmitted optical power timing input vector according to the time dimension and then passing through an optical power timing encoder to obtain a transmitted optical power timing characteristic vector; S122, arranging the received optical power values ​​at the multiple predetermined time points into a received optical power timing input vector according to the time dimension and then passing through the optical power timing encoder to obtain a received optical power timing characteristic vector; S123, fusing the transmitted optical power timing characteristic vector and the received optical power timing characteristic vector to obtain the optical power timing characteristic vector.

[0029] In step S121, the emission optical power values ​​at the multiple predetermined time points are arranged into an emission optical power timing input vector according to the time dimension and then passed through an optical power timing encoder to obtain an emission optical power timing feature vector. In the optical module data acquisition process, a series of emission optical power values ​​arranged in chronological order are usually obtained, and these values ​​reflect the emission performance of the optical module at different time points. In order to use the timing analysis method to capture the variation law of emission power over time, it is necessary to organize the emission optical power at multiple time points into an ordered structure, i.e., a vector. Through vectorized data representation, we can more intuitively observe the variation trend of the emission optical power, so as to more accurately judge the performance status of the optical module. Next, the arranged vector is input into the optical power timing encoder for feature encoding processing. The essence of the optical power timing encoder in the present application is a convolutional neural network model comprising a first convolutional layer and a second convolutional layer, wherein the first convolutional layer uses a convolutional neural network model with a first-scale one-dimensional convolutional kernel, and the second convolutional layer uses a convolutional neural network model with a second-scale one-dimensional convolutional kernel, wherein the first scale is different from the second scale. It should be understandable that by using one-dimensional convolution kernels of different scales, the optical power time series encoder can more comprehensively capture the short-term and long-term characteristics in the transmitted optical power time series data, which helps the model to more accurately understand the inherent laws of the data.

[0030] In step S122, the received optical power values ​​at the plurality of predetermined time points are arranged into a received optical power timing input vector according to the time dimension and then passed through the optical power timing encoder to obtain a received optical power timing feature vector. It should be understood that the received optical power will also change over time and have a timing characteristic. In order to retain this timing information and provide a basis for subsequent timing analysis, it is first necessary to arrange the received optical power values ​​at a plurality of time points into a received optical power timing input vector according to the time dimension. The arranged vector contains the rising and falling trend information of the received optical power changing over time, and also contains the periodic change information of the received optical power. In addition, it also specifically contains the information that the received optical power fluctuates abnormally at a certain time point or within a short period of time. In order to reflect the inherent laws and potential patterns of the collected received optical power timing data and provide strong support for subsequent analysis, it is necessary to input the received optical power timing input vector into the optical power timing encoder to extract the key information in the optical power timing data.

[0031] In step S123, the emission optical power timing characteristic vector and the reception optical power timing characteristic vector are fused to obtain the optical power timing characteristic vector. It should be understood that the emission optical power and the reception optical power are two important indicators of the performance of the optical module. They respectively reflect the ability of the optical module in sending and receiving signals. In order to comprehensively consider the performance of the optical module in both sending and receiving, so as to more comprehensively evaluate the overall performance status of the optical module, it is necessary to fuse the emission optical power timing characteristic vector and the reception optical power timing characteristic vector in the technical solution of the present application to obtain the optical power timing characteristic vector. By fusing the two characteristic vectors, the errors or deviations that may exist in a single indicator can be eliminated, thereby improving the accuracy of the evaluation.

[0032] In step S130, the bit error rates at the plurality of predetermined time points are subjected to a time series analysis to obtain a bit error rate time series feature vector. Specifically, in an embodiment of the present application, the bit error rates at the plurality of predetermined time points are subjected to a time series analysis to obtain a bit error rate time series feature vector, including: arranging the bit error rates at the plurality of predetermined time points into a bit error rate time series input vector according to the time dimension and then passing through a bit error rate time series encoder to obtain the bit error rate time series feature vector. It should be understood that in order to observe the changing trend of the bit error rate time series data and thus more easily detect abnormal values ​​or abnormal patterns, the bit error rate data at the plurality of time points first need to be arranged into a bit error rate time series input vector according to the time dimension. Then the arranged bit error rate time series input vector is input into the bit error rate time series encoder. The bit error rate time series encoder described in the present application is a convolutional neural network model comprising a fully connected layer and a one-dimensional convolutional layer. The one-dimensional convolutional layer can automatically extract useful features from the input time series data, and the fully connected layer can further integrate the features extracted by the convolutional layer and use them for the final output. That is, the convolutional neural network can capture local anomalies and global trends in the bit error rate time series data by combining local perception and global understanding, thereby improving the accuracy of anomaly detection. It is also worth mentioning that the parameter sharing mechanism in the convolutional neural network can reduce the number of parameters of the model, thereby reducing the computational complexity and improving the training speed and generalization ability of the model.

[0033] In step S140, the optical power timing feature vector and the bit error rate timing feature vector are optimized by attribute association expression based on feature space compatibility adjustment to obtain an optical module detection feature vector. It should be understood that the optical power timing feature vector includes the optical power value of the optical module when sending and receiving signals and the change trend of these values ​​over time, which can reflect the stability and fluctuation of the optical power of the optical module during long-term operation, and the bit error rate timing feature vector includes the bit error rate value of the optical module at different time points, and the change trend of these values ​​over time, which reflects the performance stability and error rate change of the optical module during transmission. In order to combine the performance of the optical module in terms of optical signal strength and transmission error rate, so as to more comprehensively evaluate the overall performance status of the optical module, it is necessary to merge the optical power timing feature vector and the bit error rate timing feature vector.

[0034] In particular, considering that the optical power time series feature vector and the bit error rate time series feature vector represent information of two different dimensions, namely, the intensity and accuracy of optical signal transmission, respectively, they have significant differences in physical sense; at the same time, the information change in the optical power time series feature vector may be relatively smooth, which is usually related to factors such as physical aging and temperature change of the optical module. The changes caused by these factors are often gradual. In contrast, the information change in the bit error rate time series feature vector may show more drastic fluctuations. Some factors (such as signal interference, multipath effect, etc.) may cause the bit error rate to change significantly in a short period of time. The combined influence of these two aspects will lead to a more significant feature heterogeneity between the optical power time series feature vector and the bit error rate time series feature vector. Due to the feature heterogeneity between the optical power time series feature vector and the bit error rate time series feature vector, the two vectors will be distributed in different feature areas in the feature space, forming different feature flows. When the two feature vectors are fused, if they are simply spliced ​​or merged, the feature manifold may be discontinuous and non-smooth, thereby affecting the distribution of the feature vector in the feature space. At the same time, since the feature manifold is not smooth, the fused feature vector may not accurately express the true state of the optical module, thereby affecting the accuracy and certainty of the detection result. Therefore, in the technical solution of the present application, the optical power timing feature vector and the bit error rate timing feature vector are optimized based on the attribute association expression of feature space compatibility adjustment to obtain the optical module detection feature vector.

[0035] Specifically, in an embodiment of the present application, the optical power timing feature vector and the bit error rate timing feature vector are optimized by attribute association expression based on feature space compatibility adjustment to obtain an optical module detection feature vector, including: constructing an optical module detection fusion distance topological matrix between the optical power timing feature vector and the bit error rate timing feature vector; constructing an optical module detection fusion feature value granularity association matrix between the optical power timing feature vector and the bit error rate timing feature vector; performing topological association modulation on the optical module detection fusion feature value granularity association matrix based on the optical module detection fusion distance topological matrix to obtain an optical module detection fusion topological modulation association matrix; using the optical module detection fusion topological modulation association matrix as a perspective co-projection space, projecting the optical power timing feature vector and the bit error rate timing feature vector into the perspective co-projection space to obtain a perspective modulation optical power timing feature vector and a perspective modulation bit error rate timing feature vector; fusing the perspective modulation optical power timing feature vector and the perspective modulation bit error rate timing feature vector to obtain the optical module detection feature vector.

[0036] In an embodiment of the present application, another expression method of optimizing the attribute association expression of the optical power timing feature vector and the bit error rate timing feature vector based on feature space compatibility adjustment to obtain the optical module detection feature vector may be: the optical power timing feature vector and the bit error rate timing feature vector are processed by the following formula to obtain the optical module detection feature vector; wherein the formula is:

[0037]

[0038]

[0039]

[0040]

[0041]

[0042] in, represents the optical power time series characteristic vector, represents the bit error rate timing characteristic vector, is the optical module detection fusion distance topology matrix between the optical power timing characteristic vector and the bit error rate timing characteristic vector, that is, , The optical module detects the fusion distance topology matrix The eigenvalues ​​of the positions, represents the calculation of the optical power time series characteristic vector The characteristic value of the position and the first characteristic vector of the bit error rate timing The distance of the eigenvalue of the position, and and are column vectors, Indicates the optical module detection fusion distance topology matrix, represents the transpose of a vector, represents matrix multiplication, represents the optical module detection fusion feature value granularity correlation matrix, represents the natural exponential function, represents the optical module detection fusion topology modulation association matrix, represents the time series characteristic vector of the viewing angle modulation optical power, represents the timing characteristic vector of the bit error rate modulation of the viewing angle, and represents the weighted hyperparameter, Represents the optical module detection feature vector.

[0043] More specifically, in an embodiment of the present application, the optical module detection fusion eigenvalue granularity association matrix is ​​topologically associated modulated based on the optical module detection fusion distance topology matrix to obtain the optical module detection fusion topology modulation association matrix, including: creating a new Spring Boot project and adding the dependencies required by the Spring Boot project in the pom.xml file; creating a model class, the model class is used to represent the optical module detection fusion distance topology matrix, the optical module detection fusion eigenvalue granularity association matrix and the optical module detection fusion topology modulation association matrix; creating a service class, the service class is used to implement the logic of topology association modulation; creating a controller class, the controller class is used to process HTTP requests; starting the Spring Boot project, and packaging and deploying the Spring Boot project to the server.

[0044] In an embodiment of the present application, part of the implementable example code for performing topological correlation modulation on the optical module detection fusion feature value granularity correlation matrix based on the optical module detection fusion distance topological matrix to obtain the optical module detection fusion topological modulation correlation matrix is: / / Model class package com.example.opticalmodule.model; public class OpticalModule { private double[][] distanceMatrix; private double[][] featureMatrix; / / Getters and Setters } / / Service class package com.example.opticalmodule.service; import org.springframework.stereotype.Service; @Service public class MatrixService { public double[][] modulateMatrices(double[][] distanceMatrix,double[][] featureMatrix) { / / Implement the logic of topological correlation modulation here / / Returns the optical module detection fusion topology modulation correlation matrix return new double[][]{{}}; / / Example return } } / / Controller class package com.example.opticalmodule.controller; import com.example.opticalmodule.service.MatrixService; import org.springframework.web.bind.annotation.*; @RestController @RequestMapping(" / api / optical-module") public class OpticalModuleController { private final MatrixService matrixService; public OpticalModuleController(MatrixService matrixService) { this.matrixService = matrixService; } @PostMapping(" / modulate") public double[][] modulate(@RequestBody double[][] distanceMatrix,@RequestBody double[][] featureMatrix) { return matrixService.modulateMatrices(distanceMatrix, featureMatrix); } } Here, the advantages of using the Spring mechanism to deploy and implement "topological correlation modulation of the optical module detection fusion feature value granularity correlation matrix based on the optical module detection fusion distance topological matrix to obtain the optical module detection fusion topological modulation correlation matrix" include: a. The Spring framework supports modular development, which is convenient for separating different functions into independent modules and enhancing the maintainability and scalability of the code; b. Spring provides a variety of configuration methods (such as Java configuration, XML configuration), so that developers can flexibly choose according to project requirements, and facilitate environment switching and parameter adjustment; c. Spring has a large developer community and rich documentation resources, and can quickly find solutions or get support when encountering problems; d. The automatic configuration feature of Spring Boot can quickly start applications, reduce development and deployment time, while maintaining good performance; e. Spring MVC provides powerful RESTful API support, which is convenient for building and maintaining API interfaces, suitable for the needs of modern microservice architecture; f. Spring Security provides a comprehensive security solution that can protect applications from various security threats and ensure data security.

[0045] That is, in the technical solution of the present application, the optical power timing feature vector and the bit error rate timing feature vector are optimized for attribute association expression based on feature space compatibility adjustment, and firstly, an optical module detection fusion distance topological matrix between the optical power timing feature vector and the bit error rate timing feature vector is constructed. This step utilizes the idea of ​​graph theory, takes the eigenvalues ​​of each position in the optical power timing feature vector and the bit error rate timing feature vector as nodes, and uses the distance metric function to determine the low-dimensional embedding expression of the edges between nodes, so as to construct a feature expression that can quantify the topological association relationship between the optical power timing feature vector and the bit error rate timing feature vector. This step is crucial for understanding the spatial distribution of data points and their mutual connections, and also provides basic structural information for subsequent steps.

[0046] Next, construct an optical module detection fusion eigenvalue granularity association matrix between the optical power timing feature vector and the bit error rate timing feature vector. This step focuses on exploring the correlation between the internal attributes of the feature vectors. In a specific example, the product between the transposed vectors of the optical power timing feature vector and the bit error rate timing feature vector is calculated to obtain the optical module detection fusion eigenvalue granularity association matrix. Furthermore, the optical module detection fusion eigenvalue granularity association matrix is ​​topologically associated modulated based on the optical module detection fusion distance topology matrix to obtain the optical module detection fusion topology modulation association matrix. That is, the eigenvalue granularity association information of the optical power timing feature vector and the bit error rate timing feature vector is driven to wander on the distance topology map to construct a low-dimensional modulation perspective co-projection space domain for mapping the optical power timing feature vector and the bit error rate timing feature vector to a continuous high-dimensional regression space attribute.

[0047] Next, the optical module detection fusion topology modulation correlation matrix is ​​used as the perspective co-projection space, and the optical power timing feature vector and the bit error rate timing feature vector are projected into the perspective co-projection space to obtain the perspective modulation optical power timing feature vector and the perspective modulation bit error rate timing feature vector. In other words, the key to this technical solution is to find an implicit third space with modulation capability, in which as much information of the original high-dimensional data as possible is retained, so that the converted new feature vector not only reduces the dimension, but also maintains the relative position relationship between the original features.

[0048] Finally, the perspective modulation optical power timing feature vector and the perspective modulation bit error rate timing feature vector are fused to obtain the optical module detection feature vector. Fusion can be achieved in a variety of forms, ranging from simple arithmetic average to complex ensemble learning algorithms, depending on the application scenario and the goals pursued. The final optical module detection feature vector combines the main features of the two sets of input data and provides a more comprehensive data representation, which is not only conducive to improving the performance of machine learning models, but also promotes the ability of cross-modal or multi-source data analysis, thereby significantly enhancing the understanding and utilization efficiency of complex data structures.

[0049] In step S150, based on the optical module detection feature vector, the detection result of the optical module is obtained. Specifically, in an embodiment of the present application, based on the optical module detection feature vector, the detection result of the optical module is obtained, including: passing the optical module detection feature vector through a detection classifier to obtain the detection result, and the detection result is used to indicate whether the optical module is faulty at the current time point. It should be understood that the optical module detection feature vector obtained by fusion contains feature information in multiple aspects such as optical power and bit error rate. In order to convert these feature information into a detection result with clear meaning, that is, to determine whether the optical module is faulty, it is necessary to further process and analyze these feature vectors through a detection classifier. The detection classifier is a deep learning model that can learn the fault characteristics and laws of the optical module from a large amount of training data, thereby realizing accurate classification and judgment of the newly input feature vector. The detection result can reflect the performance status of the optical module in real time. When the optical module fails, an early warning can be issued in time to remind the operation and maintenance personnel to handle it, which helps to avoid further expansion of the fault and ensure the stable operation of the optical communication system.

[0050] In summary, the optical module data collection and analysis method based on the embodiment of the present application is explained, which adopts the time series data analysis and processing technology based on deep learning, and judges whether the optical module is faulty at the current time point by performing time series analysis on the optical module's transmitted optical power value, received optical power value and bit error rate data at multiple time points. In this way, the operating status of the optical module can be more comprehensively understood, the trend of performance changes can be captured, and the accuracy and reliability of optical module fault detection can be improved.

[0051] Figure 4 FIG. 1 is a system block diagram of an optical module data acquisition and analysis system according to an embodiment of the present application. Figure 4 As shown, according to the optical module data acquisition and analysis system 100 of the embodiment of the present application, it includes: an optical module data acquisition module 110, which is used to obtain the transmitted optical power value, the received optical power value and the bit error rate of the optical module at multiple predetermined time points within a predetermined time period; an optical module optical power data encoding module 120, which is used to perform a fusion timing analysis on the transmitted optical power value and the received optical power value at the multiple predetermined time points to obtain an optical power timing feature vector; an optical module bit error rate data encoding module 130, which is used to perform a timing analysis on the bit error rate at the multiple predetermined time points to obtain a bit error rate timing feature vector; an optical module related feature fusion module 140, which is used to perform attribute association expression optimization based on feature space compatibility adjustment on the optical power timing feature vector and the bit error rate timing feature vector to obtain an optical module detection feature vector; an optical module detection result generation module 150, which is used to obtain the detection result of the optical module based on the optical module detection feature vector.

[0052] Here, those skilled in the art can understand that the specific functions and operations of the various units and modules in the optical module data acquisition and analysis system 100 have been described in the above reference. Figures 1 to 3 The optical module data collection and analysis method has been described in detail, and therefore, its repeated description will be omitted.

[0053] In summary, the optical module data acquisition and analysis system 100 based on the embodiment of the present application is explained, which adopts the time series data analysis and processing technology based on deep learning, and judges whether the optical module is faulty at the current time point by performing time series analysis on the transmitted optical power value, received optical power value and bit error rate data of the optical module at multiple time points. In this way, the operating status of the optical module can be more comprehensively understood, the trend of performance changes can be captured, and the accuracy and reliability of optical module fault detection can be improved.

[0054] As described above, the optical module data acquisition and analysis system 100 according to the embodiment of the present application can be implemented in various wireless terminals, such as a server for optical module data acquisition and analysis. In one example, the optical module data acquisition and analysis system 100 according to the embodiment of the present application can be integrated into the wireless terminal as a software module and / or a hardware module. For example, the optical module data acquisition and analysis system 100 can be a software module in the operating system of the wireless terminal, or can be an application developed for the wireless terminal; of course, the optical module data acquisition and analysis system 100 can also be one of the many hardware modules of the wireless terminal.

[0055] Alternatively, in another example, the optical module data acquisition and analysis system 100 and the wireless terminal may also be separate devices, and the optical module data acquisition and analysis system 100 may be connected to the wireless terminal via a wired and / or wireless network and transmit interactive information in accordance with an agreed data format.

Claims

1. A method for collecting and analyzing optical module data, characterized in that: include: Obtaining the transmit optical power value, the receive optical power value and the bit error rate of the optical module at multiple predetermined time points within a predetermined time period; Performing a fusion timing analysis on the transmitted optical power values ​​and the received optical power values ​​at the plurality of predetermined time points to obtain an optical power timing feature vector; Performing time series analysis on the bit error rates at the plurality of predetermined time points to obtain a bit error rate time series feature vector; Performing attribute association expression optimization based on feature space compatibility adjustment on the optical power time series feature vector and the bit error rate time series feature vector to obtain an optical module detection feature vector; A detection result of the optical module is obtained based on the optical module detection feature vector.

2. The optical module data collection and analysis method according to claim 1, characterized in that: Performing a fusion timing analysis on the transmitted optical power values ​​and the received optical power values ​​at the plurality of predetermined time points to obtain an optical power timing feature vector includes: Arranging the transmit optical power values ​​of the plurality of predetermined time points into a transmit optical power time series input vector according to the time dimension and then passing the vector through an optical power time series encoder to obtain a transmit optical power time series feature vector; Arranging the received optical power values ​​of the plurality of predetermined time points into a received optical power timing input vector according to the time dimension and then passing the received optical power timing encoder to obtain a received optical power timing characteristic vector; The transmitted optical power timing characteristic vector and the received optical power timing characteristic vector are fused to obtain the optical power timing characteristic vector.

3. The optical module data collection and analysis method according to claim 2, characterized in that: Performing timing analysis on the bit error rates at the multiple predetermined time points to obtain a bit error rate timing feature vector includes: arranging the bit error rates at the multiple predetermined time points into a bit error rate timing input vector according to the time dimension and then passing it through a bit error rate timing encoder to obtain the bit error rate timing feature vector.

4. The optical module data collection and analysis method according to claim 3, characterized in that: The bit error rate timing encoder is a convolutional neural network model including a fully connected layer and a one-dimensional convolutional layer, and the optical power timing encoder is a convolutional neural network model including a first convolutional layer and a second convolutional layer, wherein the first convolutional layer uses a convolutional neural network model with a one-dimensional convolutional kernel of a first scale, and the second convolutional layer uses a convolutional neural network model with a one-dimensional convolutional kernel of a second scale, wherein the first scale is different from the second scale.

5. The optical module data collection and analysis method according to claim 4, characterized in that: The optical power time series feature vector and the bit error rate time series feature vector are optimized based on feature space compatibility adjustment to obtain an optical module detection feature vector, including: Constructing an optical module detection fusion distance topology matrix between the optical power timing characteristic vector and the bit error rate timing characteristic vector; Constructing an optical module detection fusion eigenvalue granularity correlation matrix between the optical power timing eigenvector and the bit error rate timing eigenvector; Based on the optical module detection fusion distance topology matrix, the optical module detection fusion feature value granularity correlation matrix is ​​topologically modulated to obtain an optical module detection fusion topology modulation correlation matrix; Taking the optical module detection fusion topology modulation correlation matrix as the perspective co-projection space, projecting the optical power timing feature vector and the bit error rate timing feature vector into the perspective co-projection space to obtain the perspective modulation optical power timing feature vector and the perspective modulation bit error rate timing feature vector; The optical module detection feature vector is obtained by fusing the viewing angle modulation optical power timing feature vector and the viewing angle modulation bit error rate timing feature vector.

6. The optical module data collection and analysis method according to claim 5, characterized in that: The optical module detection fusion feature value granularity association matrix is ​​topologically associated modulated based on the optical module detection fusion distance topological matrix to obtain an optical module detection fusion topological modulation association matrix, including: Create a new Spring Boot project and add the dependencies required by the Spring Boot project in the pom.xml file; Create a model class, the model class is used to represent the optical module detection fusion distance topology matrix, the optical module detection fusion eigenvalue granularity association matrix and the optical module detection fusion topology modulation association matrix; Creating a service class, wherein the service class is used to implement the logic of topology-associated modulation; Create a controller class, which is used to process HTTP requests; Start the Spring Boot project, package the Spring Boot project, and deploy it on the server.

7. The optical module data collection and analysis method according to claim 6, characterized in that: Obtaining a detection result of the optical module based on the optical module detection feature vector includes: passing the optical module detection feature vector through a detection classifier to obtain the detection result, wherein the detection result is used to indicate whether the optical module is faulty at a current time point.

8. An optical module data acquisition and analysis system, characterized in that: include: The optical module data acquisition module is used to obtain the transmitted optical power value, received optical power value and bit error rate of the optical module at multiple predetermined time points within a predetermined time period; An optical module optical power data encoding module, used for performing a fusion timing analysis on the transmitted optical power values ​​and the received optical power values ​​at the plurality of predetermined time points to obtain an optical power timing feature vector; An optical module bit error rate data encoding module, used for performing a time series analysis on the bit error rates at the plurality of predetermined time points to obtain a bit error rate time series feature vector; An optical module related feature fusion module, used for optimizing the attribute association expression of the optical power timing feature vector and the bit error rate timing feature vector based on feature space compatibility adjustment to obtain an optical module detection feature vector; The optical module detection result generating module is used to obtain the detection result of the optical module based on the optical module detection feature vector.

9. The optical module data acquisition and analysis system according to claim 8, characterized in that: The optical module optical power data encoding module comprises: An emission optical power timing feature extraction unit, configured to arrange the emission optical power values ​​of the plurality of predetermined time points into an emission optical power timing input vector according to the time dimension and then pass the vector through an optical power timing encoder to obtain an emission optical power timing feature vector; A received optical power timing feature extraction unit, configured to arrange the received optical power values ​​at the plurality of predetermined time points into a received optical power timing input vector according to the time dimension and then pass the received optical power timing encoder to obtain a received optical power timing feature vector; The optical power timing characteristic fusion unit is used to fuse the transmitted optical power timing characteristic vector and the received optical power timing characteristic vector to obtain the optical power timing characteristic vector.

10. The optical module data acquisition and analysis system according to claim 9, characterized in that: The optical module detection result generating module is used to: pass the optical module detection feature vector through a detection classifier to obtain the detection result, and the detection result is used to indicate whether the optical module is faulty at a current time point.

Images