FTTR network stability analysis method, system, device and medium

Abstract

The invention discloses an FTTR network stability analysis method, system and device and a medium, and relates to the technical field of networks, and the method comprises the steps that a sending device group S sends a test data packet to a receiving device R based on a preset sending parameter; obtaining each performance parameter of each network node in each sending period, and generating a performance matrix of each network node according to each performance parameter; determining a longitudinal feature vector corresponding to the network node according to the performance matrix, and determining a transverse feature vector corresponding to the network node according to the performance matrix; and based on a network stability analysis model, generating a network stability analysis result according to the longitudinal feature vector and the transverse feature vector. The FTTR network stability evaluation method and device have the effect of improving the accuracy of FTTR network stability evaluation.

Description

Technical Field

[0001] The present application relates to the technical field of networks, and in particular, to a method, system, device, and medium for analyzing the stability of an FTTR network. Background Art

[0002] In the current FTTR (Fiber To The Room) network stability analysis technology, common methods mainly rely on static performance tests and the measurement of the capacity of a single data packet. These technologies usually evaluate the performance of the network by monitoring basic indicators such as the transmission rate, packet loss rate, and latency of the network. However, this method has multiple defects.

[0003] First of all, static tests often cannot accurately reflect the actual performance of the network under different usage scenarios or load conditions. Since the usage environment of the FTTR network may be very complex, the needs of users and the data traffic will fluctuate significantly at different times, and static evaluation may ignore the impacts brought by these dynamic changes, resulting in misjudgment of the network stability.

[0004] Secondly, the measurement of the capacity of a single data packet limits the comprehensive understanding of the network performance. In actual applications, different data traffic will generate different network responses, and relying only on fixed test conditions cannot reveal various problems that the network may encounter in actual use.

[0005] In addition, the existing technologies often lack in-depth analysis of the performance differences between different links. Usually, they can only provide global network performance indicators, rather than being detailed to the specific performance of each node. In an FTTR network environment where multiple optical network terminals work together, this is likely to lead to inaccurate problem positioning, thus delaying the time for troubleshooting and repair.

[0006] In summary, although the current FTTR network stability evaluation technology provides certain performance monitoring capabilities, due to its limitations mainly relying on static tests and the measurement of the capacity of a single data packet, the accuracy of the evaluation is relatively low, and it cannot comprehensively reflect the actual stability of the network under complex application scenarios. Therefore, there is an urgent need to develop more detailed and dynamic evaluation methods to improve the accuracy of FTTR network stability analysis. Summary of the Invention

[0007] In order to improve the accuracy of FTTR network stability evaluation, the present application provides a method, system, device, and medium for analyzing the stability of an FTTR network.

[0008] In the first aspect, the present application provides a method for analyzing the stability of an FTTR network, adopting the following technical solution:

[0009] A method for analyzing the stability of an FTTR network includes:

[0010] Based on the preset sending parameters, the sending device group S sends test data packets to the receiving device R. Among them, the sending devices in the sending device group {S1, S2, ……, S n} have a one-to-one correspondence communication relationship with the slave optical modems in the FFTR network, and the receiving device communicates with the master optical modem. The preset sending parameters include a preset sending frequency, a preset sending number, and a data packet capacity setting set P = {P1, P2, ……, P n}, P1, P2, ……, P n show an increasing trend, and the data packet capacities sent by the same sending device are the same, while the data packet capacities sent by different sending devices are different;

[0011] Obtain various performance parameters of each network node in each sending cycle, and generate a performance matrix of each network node according to the various performance parameters. Among them, the network nodes include the master optical modem and multiple slave optical modems, and the rows or columns of the performance matrix represent the various performance parameters of the network nodes in the same sending cycle;

[0012] Determine the longitudinal eigenvector corresponding to the network node according to the performance matrix, and determine the transverse eigenvector corresponding to the network node according to the performance matrix. Among them, the longitudinal eigenvector is used to represent the performance fluctuation condition of the network nodes on the same data transmission link, and the transverse eigenvector is used to represent the performance difference condition between the network nodes on different data transmission links;

[0013] Based on the network stability analysis model, generate a network stability analysis result according to the longitudinal eigenvector and the transverse eigenvector. Among them, the network stability analysis result is used to represent the stability of the FTTR network under different data packet capacities.

[0014] By adopting the above technical solution, first, based on the preset sending parameters, the sending device group S sends test data packets to the receiving device R. Among them, the sending devices in the sending device group {S1, S2, ……, S n} have a one-to-one correspondence communication relationship with the slave optical modems in the FFTR network, and the receiving device communicates with the master optical modem. The preset sending parameters include a preset sending frequency, a preset sending number, and a data packet capacity setting set P = {P1, P2, ……, P n}, P1, P2, ……, P nIt shows an increasing trend. The data packet capacities sent by the same sending device are all the same, and the data packet capacities sent by different sending devices are all different. Then, various performance parameters of each network node in each sending cycle are obtained, and a performance matrix of each network node is generated according to the various performance parameters. Among them, the network nodes include a main optical network terminal (ONT) and multiple slave ONTs. The rows or columns of the performance matrix represent the various performance parameters of the network nodes in the same sending cycle. Then, the longitudinal eigenvector corresponding to the network nodes is determined according to the performance matrix, and the transverse eigenvector corresponding to the network nodes is determined according to the performance matrix. Among them, the longitudinal eigenvector is used to represent the performance fluctuation condition of the network nodes on the same data transmission link, and the transverse eigenvector is used to represent the performance difference condition between the network nodes on different data transmission links. Finally, based on the network stability analysis model, and according to the longitudinal eigenvector and the transverse eigenvector, a network stability analysis result is generated. Among them, the network stability analysis result is used to represent the stability of the Fiber to the Room (FTTR) network under different data packet capacities; the dual analysis of the FTTR network in the longitudinal and transverse directions of this application enables the quantification of the performance fluctuations of network nodes and the differences between links. This multi-dimensional evaluation method not only improves the reliability and accuracy of data, but also provides operators with deeper insights, enabling them to make timely adjustments during peak network traffic periods and reducing the possibility of service interruptions.

[0015] Optionally, the performance matrix includes the first performance matrix of the main ONT and the second performance matrices corresponding to each of the slave ONTs. The step of determining the longitudinal eigenvector corresponding to the network nodes according to the performance matrix includes:

[0016] Perform row difference processing on the first performance matrix A to obtain the first performance fluctuation matrix M, and perform row difference processing on the second performance matrix B to obtain the second performance fluctuation matrices N corresponding to each of the slave ONTs;

[0017] For each data transmission link, determine the performance fluctuation ratio matrix O according to the first performance fluctuation matrix M and the second performance fluctuation matrix N corresponding to the slave ONT in the data transmission link. Among them, for each element O[i][j] in the performance fluctuation ratio matrix O, The performance fluctuation ratio matrix O is used to represent the performance relationship between the main ONT and the slave ONTs;

[0018] Construct a performance fluctuation ratio vector A according to the performance fluctuation ratio matrix O, where, m is the number of rows of the performance fluctuation ratio matrix O;

[0019] Normalize the elements in the performance fluctuation ratio vector A to obtain the normalized vector A0;

[0020] Obtain the importance coefficients corresponding to each performance parameter, and determine the link fluctuation vector corresponding to each data transmission link according to the normalized vector A0 and the importance coefficients;

[0021] Determine the first weight of each data transmission link according to the data packet capacity setting set P, and splice the link fluctuation vectors based on the first weight to generate a link feature vector.

[0022] By adopting the above technical solution, in order to generate a link feature vector, perform row difference processing on the first performance matrix A to obtain the first performance fluctuation matrix M, and perform row difference processing on the second performance matrix B to obtain the second performance fluctuation matrix N corresponding to each slave optical network unit. Then, for each data transmission link, in order to determine the longitudinal feature vector corresponding to the network node, determine the performance fluctuation ratio matrix O according to the first performance fluctuation matrix M and the second performance fluctuation matrix N corresponding to the slave optical network unit in the data transmission link. Among them, for each element O[i][j] in the performance fluctuation ratio matrix O, The performance fluctuation ratio matrix O is used to represent the performance relationship between the master optical network unit and the slave optical network unit. Then, construct the performance fluctuation ratio vector A according to the performance fluctuation ratio matrix O, where, m is the number of rows of the performance fluctuation ratio matrix O. Then, normalize the elements in the performance fluctuation ratio vector A to obtain the normalized vector A0. Then, obtain the importance coefficients corresponding to each performance parameter, and determine the link fluctuation vector corresponding to each data transmission link according to the normalized matrix O0 and the importance coefficients. Finally, determine the first weight of each data transmission link according to the data packet capacity setting set P, and splice the link fluctuation vectors based on the first weight to generate a link feature vector.

[0023] Optionally, the step of determining the first weight of each data transmission link according to the data packet capacity setting set P includes:

[0024] Determine the minimum value of the elements in the data packet capacity setting set P according to the data packet capacity setting set P;

[0025] Calculate the element ratio corresponding to each element in the data packet capacity setting set P with the minimum value of the elements, and use the element ratio as the first weight of the corresponding data transmission link.

[0026] By adopting the above technical solution, in order to determine the first weight of each data transmission link, determine the minimum value of the elements in the data packet capacity setting set P according to the data packet capacity setting set P, then calculate the element ratio corresponding to each element in the data packet capacity setting set P with the minimum value of the elements, and use the element ratio as the first weight of the corresponding data transmission link.

[0027] Optionally, the steps of determining the horizontal feature vector corresponding to the node according to the performance matrix include:

[0028] Generate the corresponding set of slave optical network units F according to the data packet capacity setting set P, where F = {F1, F2,..., F n}, and the elements in the data packet capacity setting set P correspond one-to-one with the elements in the set of slave optical network units F;

[0029] Generate n - 1 node performance difference matrices C according to the second performance matrices of each slave optical network unit, where C = {C1, C2,..., C n-1}, C k [i][j] = F k+1 [i][j] - F k [i][j], k = 1, 2,..., n - 1;

[0030] For each node performance difference matrix C, construct n - 1 corresponding node performance difference vectors Q according to the node performance difference matrix C, where m represents the number of rows of the node performance difference matrix C;

[0031] Determine the respective second weights corresponding to the n - 1 node performance difference matrices according to the data packet capacity setting set P, and splice the n - 1 node performance difference matrices based on the second weights to generate a horizontal feature vector.

[0032] By adopting the above technical solution, in order to generate a horizontal feature vector, first generate the corresponding set of slave optical network units F according to the data packet capacity setting set P, where F = {F1, F2,..., F n}, and the elements in the data packet capacity setting set P correspond one-to-one with the elements in the set of slave optical network units F, then generate n - 1 node performance difference matrices C according to the second performance matrices of each slave optical network unit, where C = {C1, C2,..., C n-1}, C k [i][j] = F k+1 [i][j] - F k [i][j], k = 1, 2,..., n - 1, then for each node performance difference matrix C, construct n - 1 corresponding node performance difference vectors Q according to the node performance difference matrix C, where m represents the number of rows of the node performance difference matrix C, and finally determine the respective second weights corresponding to the n - 1 node performance difference matrices according to the data packet capacity setting set P, and splice the n - 1 node performance difference matrices based on the second weights to generate a horizontal feature vector.

[0033] Optionally, the step of determining the respective second weights corresponding to the n - 1 node performance difference matrices according to the data packet capacity setting set P includes:

[0034] Determine the data packet capacity difference set R according to the data packet capacity setting set P, where R = {R1, R2,..., R n-1}, R j = P j+1 - P j , j = 1, 2,..., n - 1;

[0035] Determine the sum of the elements of the data packet capacity difference set R according to the data packet capacity difference set R, and determine the second weight corresponding to each of the n - 1 node performance difference matrices according to the element values and the sum of the elements in the data packet capacity difference set R.

[0036] By adopting the above technical solution, in order to determine the second weight, first determine the data packet capacity difference set R according to the data packet capacity setting set P, where R = {R1, R2,..., R n-1}, R j = P j+1 - P j , j = 1, 2,..., n - 1, and then determine the sum of the elements of the data packet capacity difference set R according to the data packet capacity difference set R, and determine the second weight corresponding to each of the n - 1 node performance difference matrices according to the element values and the sum of the elements in the data packet capacity difference set R.

[0037] Optionally, the network stability analysis model based on includes an input layer, a hidden layer, and an output layer. The steps of generating a network stability analysis result based on the network stability analysis model and according to the longitudinal eigenvector and the transverse eigenvector include:

[0038] Through the input layer, generate an input feature vector according to the longitudinal eigenvector and the transverse eigenvector;

[0039] Through the hidden layer, generate a feature extraction vector according to the input feature vector;

[0040] Through the output layer, generate an output vector according to the feature extraction vector, where the output vector includes a first element and a second element. The first element is used for the acceptable data packet capacity range of the FTTR network, and the second element is used to characterize the stability score of the FTTR network within different data packet capacity ranges;

[0041] Generate a network stability analysis result based on the output vector.

[0042] By adopting the above technical solution, in order to generate the network stability analysis result, through the input layer, an input feature vector is generated according to the longitudinal feature vector and the transverse feature vector, and then through the hidden layer, a feature extraction vector is generated according to the input feature vector, and then through the output layer, an output vector is generated according to the feature extraction vector. Among them, the output vector includes a first element and a second element. The first element is used for the packet capacity range acceptable by the FTTR network, and the second element is used to characterize the stability score of the FTTR network within different packet capacity ranges. Finally, based on the output vector, the network stability analysis result is generated.

[0043] Optionally, the hidden layer includes a first hidden layer, a second hidden layer, and a third hidden layer. The neuron ratio between the first hidden layer, the second hidden layer, and the third hidden layer is 6:3:1. The step of generating the feature extraction vector according to the input feature vector through the hidden layer includes:

[0044] Based on the first hidden layer and according to the input feature vector, a first intermediate feature vector is generated;

[0045] Based on the second hidden layer and according to the first intermediate feature vector, a second intermediate feature vector is generated;

[0046] Based on the third hidden layer and according to the second intermediate feature vector, a feature extraction vector is generated.

[0047] By adopting the above technical solution, in order to generate the feature extraction vector, first, based on the first hidden layer and according to the input feature vector, a first intermediate feature vector is generated, then, based on the second hidden layer and according to the first intermediate feature vector, a second intermediate feature vector is generated, and finally, based on the third hidden layer and according to the second intermediate feature vector, a feature extraction vector is generated.

[0048] In a second aspect, the present application also provides an FTTR network stability analysis system, adopting the following technical solution:

[0049] An FTTR network stability analysis system includes:

[0050] A test module is used to send test packets from the sending device group S to the receiving device R based on preset sending parameters. Among them, the sending devices in the sending device group {S1, S2,..., Sn} have a one-to-one correspondence communication relationship with the slave optical modems in the FFTR network, and the receiving device communicates with the master optical modem. The preset sending parameters include a preset sending frequency, a preset sending number, and a packet capacity setting set P = {P1, P2,..., Pn}, where P1, P2,..., Pn show an increasing trend, the packet capacities sent by the same sending device are the same, and the packet capacities sent by different sending devices are different;

[0051] A performance matrix generation module, configured to obtain various performance parameters of each network node in each transmission period, and generate a performance matrix for each network node according to the various performance parameters, where the network nodes include a main optical network terminal (ONT) and multiple slave ONTs, and the rows or columns of the performance matrix represent the various performance parameters of the network nodes in the same transmission period;

[0052] An eigenvector determination module, configured to determine a longitudinal eigenvector corresponding to a network node according to the performance matrix, and determine a transverse eigenvector corresponding to the network node according to the performance matrix, where the longitudinal eigenvector is used to represent the performance fluctuation condition of the network nodes on the same data transmission link, and the transverse eigenvector is used to represent the performance difference condition between the network nodes on different data transmission links;

[0053] A stability analysis module, configured to generate a network stability analysis result based on a network stability analysis model and according to the longitudinal eigenvector and the transverse eigenvector, where the network stability analysis result is used to represent the stability of the Fiber to the Room (FTTR) network under different data packet capacities.

[0054] In a third aspect, the present application further provides a computer device, adopting the following technical solution:

[0055] A computer device includes a memory and a processor, and a computer program that can run on the processor is stored on the memory. When the processor executes the computer program, the method described in the first aspect is implemented.

[0056] In a fourth aspect, the present application further provides a computer-readable storage medium, adopting the following technical solution:

[0057] A computer-readable storage medium stores a computer program that can be loaded and executed by a processor to implement the method described in the first aspect.

[0058] In summary, the present application at least includes the following beneficial technical effects: First, based on preset transmission parameters, a sending device group S sends test data packets to a receiving device R, where the sending devices in the sending device group {S1, S2,..., S n} have a one-to-one correspondence communication relationship with the slave ONTs in the FFTR network, the receiving device communicates with the main ONT, and the preset transmission parameters include a preset transmission frequency, a preset number of transmissions, and a data packet capacity set P = {P1, P2,..., P n}, P1, P2,..., P nIt shows an increasing trend. The data packet capacities sent by the same sending device are all the same, and the data packet capacities sent by different sending devices are all different. Then, various performance parameters of each network node in each sending cycle are obtained, and a performance matrix of each network node is generated according to the various performance parameters. Among them, the network nodes include a main optical modem and multiple slave optical modems. The rows or columns of the performance matrix represent the various performance parameters of the network nodes in the same sending cycle. Then, the longitudinal eigenvector corresponding to the network node is determined according to the performance matrix, and the transverse eigenvector corresponding to the network node is determined according to the performance matrix. Among them, the longitudinal eigenvector is used to represent the performance fluctuation condition of the network nodes on the same data transmission link, and the transverse eigenvector is used to represent the performance difference condition between the network nodes on different data transmission links. Finally, based on the network stability analysis model, the network stability analysis result is generated according to the longitudinal eigenvector and the transverse eigenvector. Among them, the network stability analysis result is used to represent the stability of the FTTR network under different data packet capacities; the dual analysis of the FTTR network in the longitudinal and transverse directions in this application quantifies the fluctuations of the network node performance and the differences between the links. This multi-dimensional evaluation method not only improves the reliability and accuracy of the data, but also provides deeper insights for the operator, enabling it to make timely adjustments during peak network traffic periods and reducing the possibility of service interruption. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 is the overall process schematic diagram of the embodiment of the present application.

[0060] Figure 2 is the structural schematic diagram of the system of the present application.

[0061] Figure 3 is the structural block diagram of the computer device of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0062] In order to make the objectives, technical solutions and advantages of the present application clearer, the following further describes the present application in detail with reference to the attached Figures 1 - 3 drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0063] The embodiment of the present application discloses a method for analyzing the stability of an FTTR network.

[0064] Referring to Figure 1 , a method for analyzing the stability of an FTTR network, characterized in that it includes:

[0065] Step S11, based on preset sending parameters, a sending device group S sends test data packets to a receiving device R.

[0066] Among them, the sending device group {S1, S2,..., Sn The sending device in {} and the slave optical network terminal in the FFTR network have a one-to-one correspondence communication relationship. The receiving device communicates with the master optical network terminal. The preset sending parameters include a preset sending frequency, a preset sending number of times, and a data packet capacity setting set P = {P1, P2, ……, P n}, P1, P2, ……, P n shows an increasing trend. The data packet capacities sent by the same sending device are the same, and the data packet capacities sent by different sending devices are different.

[0067] It should be noted that in step S11, the sending device group {S1, S2, ……, Sn} will send test data packets to the receiving device R according to the preset sending parameters. Each sending device maintains a one-to-one correspondence communication relationship with the slave optical network terminal, and the receiving device communicates with the master optical network terminal. According to the preset sending frequency, the preset sending number of times, and the data packet capacity set P = {P1, P2, ……, Pn}, each sending device will select a specified capacity to send data packets. Specifically, the data packet capacities sent by the same sending device remain consistent, while the capacities of different devices are set according to the increasing trend of the values in P. Therefore, the sending device S1 sends data packets with a capacity of P1, S2 sends data packets with a capacity of P2, and so on until Sn. During the transmission process, the system will ensure that each device accurately executes the data packet sending operation according to the established frequency and number of times, thereby realizing effective test data transmission.

[0068] Step S12, obtain various performance parameters of each network node in each sending cycle, and generate a performance matrix of each network node according to the various performance parameters.

[0069] Among them, the network nodes include the master optical network terminal and multiple slave optical network terminals. The rows or columns of the performance matrix represent the various performance parameters of the network nodes in the same sending cycle.

[0070] It should be noted that in step S12, the system will automatically obtain the performance parameters of each network node by monitoring the network activities in each sending cycle. These performance parameters may include key indicators such as latency, packet loss rate, and bandwidth utilization rate. After obtaining these data, the system will summarize and organize them into a performance matrix for subsequent analysis. The performance matrix of each network node will present the various performance indicators of the node during the test, thereby providing a comprehensive view of the network status. Finally, these performance matrices not only help to evaluate the overall performance of the network but also provide an important basis for subsequent optimization and troubleshooting.

[0071] Step S13, determine the longitudinal eigenvector corresponding to the network node according to the performance matrix, and determine the transverse eigenvector corresponding to the network node according to the performance matrix.

[0072] Among them, the vertical feature vector is used to represent the performance fluctuation status of network nodes on the same data transmission link, and the horizontal feature vector is used to represent the performance difference status between network nodes on different data transmission links.

[0073] It should be noted that according to the performance matrix, the system can respectively extract the vertical feature vector and the horizontal feature vector of network nodes in the FTTR network. The vertical feature vector is obtained by analyzing the changes in performance parameters of network nodes (the main optical network unit and the slave optical network unit in this application) in different transmission cycles, reflecting the performance stability and dynamic characteristics of the main optical network unit and the slave optical network unit in the data transmission link. Relatively speaking, the horizontal feature vector is obtained by comparative analysis of the performance parameters of the slave optical network unit, aiming to reveal the performance differences between different network nodes. By aggregating the performance indicators of all nodes to form a unified feature vector, it can be used for global performance analysis to evaluate the overall performance of the network and identify potential bottlenecks or problem areas. The combination of the two provides a basic basis for network optimization, enabling more targeted adjustment and optimization under the guidance of the feature vector.

[0074] Step S14, based on the network stability analysis model, and generate a network stability analysis result according to the vertical feature vector and the horizontal feature vector.

[0075] Among them, the network stability analysis result is used to represent the stability of the FTTR network under different data packet capacities.

[0076] In the above embodiment, first, based on the preset transmission parameters, the sending device group S sends test data packets to the receiving device R. Among them, the sending devices in the sending device group {S1, S2,..., S n} have a one-to-one correspondence communication relationship with the slave optical network units in the FFTR network, the receiving device communicates with the main optical network unit, and the preset transmission parameters include a preset transmission frequency, a preset number of transmissions, and a data packet capacity setting set P = {P1, P2,..., P n}, P1, P2,..., P nIt shows an increasing trend. The packet capacities sent by the same sending device are all the same, and the packet capacities sent by different sending devices are all different. Then, various performance parameters of each network node in each sending cycle are obtained, and a performance matrix of each network node is generated according to the various performance parameters. Among them, the network nodes include a main optical network terminal (ONT) and multiple slave ONTs. The rows or columns of the performance matrix represent the various performance parameters of the network nodes in the same sending cycle. Then, the longitudinal eigenvector corresponding to the network node is determined according to the performance matrix, and the transverse eigenvector corresponding to the network node is determined according to the performance matrix. Among them, the longitudinal eigenvector is used to represent the performance fluctuation condition of the network nodes on the same data transmission link, and the transverse eigenvector is used to represent the performance difference condition between the network nodes on different data transmission links. Finally, based on the network stability analysis model, the network stability analysis result is generated according to the longitudinal eigenvector and the transverse eigenvector. Among them, the network stability analysis result is used to represent the stability of the Fiber to the Room (FTTR) network under different packet capacities; the dual analysis of the FTTR network in the longitudinal and transverse directions in this application enables the quantification of the performance fluctuations of network nodes and the differences between links. This multi-dimensional evaluation method not only improves the reliability and accuracy of data but also provides operators with deeper insights, enabling them to make timely adjustments during peak network traffic periods and reducing the possibility of service interruptions.

[0077] As a further implementation manner of the method, the performance matrix includes the first performance matrix of the main ONT and the second performance matrix corresponding to each slave ONT. The step of determining the longitudinal eigenvector corresponding to the network node according to the performance matrix includes:

[0078] Step S21, perform row difference processing on the first performance matrix A to obtain the first performance fluctuation matrix M, and perform row difference processing on the second performance matrix B to obtain the second performance fluctuation matrix N corresponding to each slave ONT.

[0079] Step S22, for each data transmission link, determine the performance fluctuation ratio matrix O according to the first performance fluctuation matrix M and the second performance fluctuation matrix N corresponding to the slave ONT in the data transmission link.

[0080] Among them, for each element O[i][j] in the performance fluctuation ratio matrix O, The performance fluctuation ratio matrix O is used to represent the performance relationship between the main ONT and the slave ONT.

[0081] It can be understood that in step S22, each data transmission link includes at least the main ONT and the slave ONT. Therefore, each data transmission link (or each slave ONT) corresponds to a performance fluctuation ratio matrix O, that is, O = {O1, O2, ……, On}, and n represents the number of data transmission links (that is, the number of slave ONTs). n}.

[0082] Step S23: Construct a performance fluctuation ratio vector A based on the performance fluctuation ratio matrix O.

[0083] Among them, m is the number of rows of the performance fluctuation ratio matrix O.

[0084] It can be understood that there are n performance fluctuation ratio vectors A, that is, A = {A1, A2, ……, A n}.

[0085] Step S24: Normalize the elements in the performance fluctuation ratio vector A to obtain a normalized vector A0.

[0086] Step S25: Obtain the importance coefficients corresponding to each performance parameter, and determine the link fluctuation vectors corresponding to each data transmission link based on the normalized vector A0 and the importance coefficients.

[0087] It should be noted that the importance coefficient is used to quantify the importance of certain performance parameters in the overall performance evaluation. Through the importance coefficient, it helps to highlight more critical parameters, so that when calculating the link fluctuation vector, parameters with higher importance can have a greater impact on the final result.

[0088] Step S26: Determine the first weight of each data transmission link according to the data packet capacity setting set P, and splice the link fluctuation vectors based on the first weight to generate a link feature vector.

[0089] It should be noted that in the FTTR network stability analysis method, the main role of the first weight is to eliminate the differences in data packet capacity between different links, ensuring a more fair and accurate performance evaluation. Specifically, the first weight can eliminate link capacity differences. By combining the performance fluctuations of each link with the corresponding first weight, the performance of different links can be evaluated according to a unified standard. Then, links with larger capacities will not have a disproportionate impact on the analysis results due to their higher data volumes. At the same time, the first weight makes the performance comparison between different links more reasonable, and can better reflect the performance fluctuations of each link in the network on the same basis, avoiding unfair evaluations caused by capacity differences. In addition, the first weight can also strengthen the consistency of the link feature vector. Different links eliminate the influence brought by the data packet capacity through the first weight. Thus, when generating the link feature vector, the fluctuation characteristics of each link can fairly reflect its actual performance in the network.

[0090] In the above embodiment, in order to generate a link feature vector, a row difference process is performed on the first performance matrix A to obtain a first performance fluctuation matrix M, and a row difference process is performed on the second performance matrix B to obtain a corresponding second performance fluctuation matrix N for each slave optical network unit. Then, for each data transmission link, in order to determine the longitudinal feature vector corresponding to the network node, a performance fluctuation ratio matrix O is determined according to the first performance fluctuation matrix M and the second performance fluctuation matrix N corresponding to the slave optical network unit in the data transmission link. Among them, for each element O[i][j] in the performance fluctuation ratio matrix O, The performance fluctuation ratio matrix O is used to represent the performance relationship between the master optical network unit and the slave optical network unit. Then, a performance fluctuation ratio vector A is constructed according to the performance fluctuation ratio matrix O. Among them, m is the number of rows of the performance fluctuation ratio matrix O. Then, the elements in the performance fluctuation ratio vector A are normalized to obtain a normalized vector A0. Then, the importance coefficients corresponding to each performance parameter are obtained, and the link fluctuation vectors corresponding to each data transmission link are determined according to the normalized matrix O0 and the importance coefficients. Finally, according to the data packet capacity setting set P, the first weight of each data transmission link is determined, and the link fluctuation vectors are concatenated based on the first weight to generate a link feature vector.

[0091] As a further embodiment of the method, the step of determining the first weight of each data transmission link according to the data packet capacity setting set P includes:

[0092] Step S31, determining the minimum value of the elements in the data packet capacity setting set P according to the data packet capacity setting set P.

[0093] Step S32, calculating the element ratio corresponding to the minimum value of the element and each element in the data packet capacity setting set P, and using the element ratio as the first weight of the corresponding data transmission link.

[0094] In the above embodiment, in order to determine the first weight of each data transmission link, the minimum value of the elements in the data packet capacity setting set P is determined according to the data packet capacity setting set P. Then, the element ratio corresponding to the minimum value of the element and each element in the data packet capacity setting set P is calculated, and the element ratio is used as the first weight of the corresponding data transmission link.

[0095] As a further embodiment of the method, the step of determining the horizontal feature vector corresponding to the node according to the performance matrix includes:

[0096] Step S41, generating a corresponding set F of slave optical network units according to the data packet capacity setting set P.

[0097] Among them, F = {F1, F2,..., F n}, and the elements in the data packet capacity setting set P and the elements in the set F of slave optical network units correspond one by one.

[0098] Step S42: Generate n - 1 node performance difference matrices C according to the second performance matrices of each slave optical network terminal (ONT).

[0099] Where C = {C1, C2,..., C n-1}, C k [i][j] = F k+1 [i][j] - F k [i][j], k = 1, 2,..., n - 1.

[0100] Step S43: For each node performance difference matrix C, construct n - 1 corresponding node performance difference vectors Q according to the node performance difference matrix C.

[0101] Where m represents the number of rows of the node performance difference matrix C.

[0102] Step S44: Determine the respective second weights of the n - 1 node performance difference matrices according to the data packet capacity setting set P, and splice the n - 1 node performance difference matrices based on the second weights to generate a horizontal feature vector.

[0103] It should be noted that the second weight reflects the differences between different data packet capacities, providing a dynamic comparison standard for performance analysis. It helps analyze the performance difference degree among the slave ONT nodes under different data packet capacity conditions; by applying the second weight to the node performance difference matrix, the generated horizontal feature vector will more accurately express the performance change of each node under different data packet capacities. This feature vector not only makes the performance difference more obvious but also enhances the sensitivity of the model to the change of data packet capacity; the introduction of the second weight makes the analysis result not only depend on the performance value itself but also takes into account the change trend of the data packet capacity, thus further improving the reliability and scientificity of network performance evaluation.

[0104] In the above embodiment, in order to generate a horizontal feature vector, first generate the corresponding slave ONT set F according to the data packet capacity setting set P, where F = {F1, F2,..., F n}, the elements in the data packet capacity setting set P and the elements in the slave ONT set F are in one - to - one correspondence, and then generate n - 1 node performance difference matrices C according to the second performance matrices of each slave ONT, where C = {C1, C2,..., C n-1}, C k [i][j] = F k+1 [i][j] - F k[i][j], where k = 1, 2, ……, n - 1, and then for each node performance difference matrix C, n - 1 corresponding node performance difference vectors Q are constructed based on the node performance difference matrix C. Among them, m represents the number of rows of the node performance difference matrix C. Finally, according to the data packet capacity setting set P, the second weight corresponding to each of the n - 1 node performance difference matrices is determined, and based on the second weight, the n - 1 node performance difference matrices are concatenated to generate a horizontal feature vector.

[0105] As a further implementation of the method, the step of determining the second weight corresponding to each of the n - 1 node performance difference matrices according to the data packet capacity setting set P includes:

[0106] Step S51, determining a data packet capacity difference set R according to the data packet capacity setting set P.

[0107] Among them, R = {R1, R2, …… R n-1}, R j = P j+1 - P j , where j = 1, 2, ……, n - 1.

[0108] Step S52, determining the sum of the elements of the data packet capacity difference set R according to the data packet capacity difference set R, and determining the second weight corresponding to each of the n - 1 node performance difference matrices according to the element values and the sum of the elements in the data packet capacity difference set R.

[0109] In the above implementation, in order to determine the second weight, first a data packet capacity difference set R is determined according to the data packet capacity setting set P. Among them, R = {R1, R2, …… R n-1}, R j = P j+1 - P j , where j = 1, 2, ……, n - 1. Then, the sum of the elements of the data packet capacity difference set R is determined according to the data packet capacity difference set R, and the second weight corresponding to each of the n - 1 node performance difference matrices is determined according to the element values and the sum of the elements in the data packet capacity difference set R.

[0110] As a further implementation of the method, based on the network stability analysis model including an input layer, a hidden layer, and an output layer, the step of generating a network stability analysis result based on the network stability analysis model and according to the longitudinal feature vector and the horizontal feature vector includes:

[0111] Step S61, generating an input feature vector through the input layer according to the longitudinal feature vector and the horizontal feature vector.

[0112] Step S62, generating a feature extraction vector through the hidden layer according to the input feature vector.

[0113] Step S63: Through the output layer, generate an output vector based on the feature extraction vector.

[0114] Among them, the output vector includes a first element and a second element. The first element is used for the range of data packet capacities acceptable by the FTTR network, and the second element is used to characterize the stability score of the FTTR network within different data packet capacity ranges.

[0115] Step S64: Generate a network stability analysis result based on the output vector.

[0116] In the above embodiment, in order to generate a network stability analysis result, through the input layer, an input feature vector is generated according to the longitudinal feature vector and the transverse feature vector, then through the hidden layer, a feature extraction vector is generated according to the input feature vector, and then through the output layer, an output vector is generated according to the feature extraction vector. Among them, the output vector includes a first element and a second element. The first element is used for the range of data packet capacities acceptable by the FTTR network, and the second element is used to characterize the stability score of the FTTR network within different data packet capacity ranges. Finally, a network stability analysis result is generated based on the output vector.

[0117] As a further embodiment of the method, the hidden layer includes a first hidden layer, a second hidden layer, and a third hidden layer. The neuron ratio between the first hidden layer, the second hidden layer, and the third hidden layer is 6:3:1. The step of generating a feature extraction vector according to the input feature vector through the hidden layer includes:

[0118] Step S71: Based on the first hidden layer and according to the input feature vector, generate a first intermediate feature vector.

[0119] Step S72: Based on the second hidden layer and according to the first intermediate feature vector, generate a second intermediate feature vector.

[0120] Step S73: Based on the third hidden layer and according to the second intermediate feature vector, generate a feature extraction vector.

[0121] In the above embodiment, in order to generate a feature extraction vector, first, based on the first hidden layer and according to the input feature vector, a first intermediate feature vector is generated, then based on the second hidden layer and according to the first intermediate feature vector, a second intermediate feature vector is generated, and finally based on the third hidden layer and according to the second intermediate feature vector, a feature extraction vector is generated.

[0122] The embodiment of the present application also discloses an FTTR network stability analysis system.

[0123] Refer to Figure 2 An FTTR network stability analysis system includes:

[0124] A test module, configured to send test data packets from a sending device group S to a receiving device R based on preset sending parameters, where the sending devices in the sending device group {S1, S2, ……, S n} have a one-to-one correspondence communication relationship with the slave optical modems in the FFTR network, the receiving device communicates with the master optical modem, and the preset sending parameters include a preset sending frequency, a preset number of sending times, and a data packet capacity setting set P = {P1, P2, ……, P n}, P1, P2, ……, P n shows an increasing trend, and the data packet capacities sent by the same sending device are the same, while the data packet capacities sent by different sending devices are different;

[0125] A performance matrix generation module, configured to obtain various performance parameters of each network node in each sending cycle, and generate a performance matrix for each network node according to the various performance parameters, where the network nodes include the master optical modem and multiple slave optical modems, and the rows or columns of the performance matrix represent the various performance parameters of the network nodes in the same sending cycle;

[0126] A feature vector determination module, configured to determine the longitudinal feature vector corresponding to the network node according to the performance matrix, and determine the transverse feature vector corresponding to the network node according to the performance matrix, where the longitudinal feature vector is used to represent the performance fluctuation condition of the network nodes on the same data transmission link, and the transverse feature vector is used to represent the performance difference condition between the network nodes on different data transmission links;

[0127] A stability analysis module, configured to generate a network stability analysis result based on a network stability analysis model and according to the longitudinal feature vector and the transverse feature vector, where the network stability analysis result is used to represent the stability of the FTTR network under different data packet capacities.

[0128] A FTTR network stability analysis system of the present invention can implement any one of the methods in a FTTR network stability analysis method, and the specific working process of a FTTR network stability analysis system of the present invention can refer to the corresponding process in the above-mentioned FTTR network stability analysis method.

[0129] The embodiments of the present application also disclose a computer device.

[0130] Refer to Figure 3 , a computer device, including a memory and a processor, where a computer program that can run on the processor is stored on the memory, and when the processor executes the computer program, it implements any one of the methods in the above-mentioned FTTR network stability analysis method.

[0131] The embodiments of the present application also disclose a computer-readable storage medium.

[0132] A computer-readable storage medium stores a computer program that can be loaded and executed by a processor to perform any of the above-described FTTR network stability analysis methods.

[0133] Among them, the computer-readable storage medium can be any tangible medium that contains or stores a program, and this program can be used by or in combination with an instruction execution system, apparatus, or device; the program code contained on the computer-readable medium can be transmitted by any suitable medium, including but not limited to wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0134] The above are all preferred embodiments of the present application. The protection scope of the present application is not limited thereby. Any feature disclosed in this specification (including the abstract and drawings), unless specifically described, can be replaced by other equivalent or similar-purpose alternative features. That is, unless specifically described, each feature is only an example of a series of equivalent or similar features.

Claims

1. A method for analyzing the stability of an FTTR network, characterized in that, Including: Based on preset sending parameters, the sending device group S sends test data packets to the receiving device R, where the sending devices in the sending device group {S1, S2, ……, S n} have a one-to-one correspondence communication relationship with the slave optical modems in the FFTR network, the receiving device communicates with the master optical modem, and the preset sending parameters include a preset sending frequency, a preset sending number of times, and a data packet capacity setting set P = {P1, P2, ……, P n}, P1, P2, ……, P n show an increasing trend, the data packet capacities sent by the same sending device are the same, and the data packet capacities sent by different sending devices are different; Obtain various performance parameters of each network node in each transmission cycle, and generate a performance matrix for each of the network nodes according to the various performance parameters, where the network nodes include the main optical network terminal and multiple slave optical network terminals, and the rows or columns of the performance matrix represent the various performance parameters of the network nodes in the same transmission cycle; Determine the longitudinal eigenvector corresponding to the network node according to the performance matrix, and determine the transverse eigenvector corresponding to the network node according to the performance matrix, where the longitudinal eigenvector is used to represent the performance fluctuation condition of the network nodes on the same data transmission link, and the transverse eigenvector is used to represent the performance difference condition between the network nodes on different data transmission links; Based on the network stability analysis model, generate a network stability analysis result according to the longitudinal eigenvector and the transverse eigenvector, where the network stability analysis result is used to represent the stability of the FTTR network under different data packet capacities.

2. The method for analyzing the stability of an FTTR network according to claim 1, wherein The performance matrix includes the first performance matrix of the main optical network terminal and the second performance matrix corresponding to each of the slave optical network terminals. The step of determining the longitudinal eigenvector corresponding to the network node according to the performance matrix includes: Perform row difference processing on the first performance matrix A to obtain a first performance fluctuation matrix M, and perform row difference processing on the second performance matrix B to obtain a second performance fluctuation matrix N corresponding to each of the slave optical network terminals; For each data transmission link, a performance fluctuation ratio matrix O is determined according to the first performance fluctuation matrix M and the second performance fluctuation matrix N corresponding to the slave optical network unit in the data transmission link. For each element O[i][j] in the performance fluctuation ratio matrix O, the performance fluctuation ratio matrix O is used to represent the performance relationship between the master optical network unit and the slave optical network unit; Construct a performance fluctuation ratio vector A based on the performance fluctuation ratio matrix O, where m is the number of rows of the performance fluctuation ratio matrix O; Normalize the elements in the performance fluctuation ratio vector A to obtain a normalized vector A0; Obtain the importance coefficients corresponding to the various performance parameters, and determine the link fluctuation vectors corresponding to each data transmission link according to the normalized vector A0 and the importance coefficients; Determine the first weight of each data transmission link according to the data packet capacity setting set P, and splice the link fluctuation vectors based on the first weight to generate a link eigenvector.

3. The method for analyzing the stability of an FTTR network according to claim 2, wherein The step of determining the first weight of each data transmission link according to the data packet capacity setting set P includes: Determine the minimum value of the elements in the data packet capacity setting set P according to the data packet capacity setting set P; Calculate the element ratio corresponding to each element in the data packet capacity setting set P with respect to the minimum value of the elements, and use the element ratio as the first weight of the corresponding data transmission link.

4. The method for analyzing the stability of an FTTR network according to claim 2, wherein, The step of determining the transverse eigenvector corresponding to the node according to the performance matrix includes: Generate the corresponding set F of slave optical network units according to the data packet capacity setting set P, where F = {F1, F2,..., F n}, and the elements in the data packet capacity setting set P correspond one-to-one with the elements in the set F of slave optical network units; Generate n - 1 node performance difference matrices C according to the second performance matrix of each of the slave optical network units, where C = {C1, C2, …… C n-1}, C k [i][j] = F k+1 [i][j] - F k [i][j], k = 1, 2, ……, n - 1; For each of the node performance difference matrices C, construct n - 1 corresponding node performance difference vectors Q according to the node performance difference matrix C, where m represents the number of rows of the node performance difference matrix C; Determine the second weight corresponding to each of the n - 1 node performance difference matrices according to the data packet capacity setting set P, and splice the n - 1 node performance difference matrices based on the second weight to generate a transverse eigenvector.

5. The method for analyzing the stability of an FTTR network according to claim 4, wherein The step of determining the second weight corresponding to each of the n - 1 node performance difference matrices according to the data packet capacity setting set P includes: Determine the data packet capacity difference set R according to the data packet capacity setting set P, where R = {R1, R2,... R n-1}, R j = P j+1 - P j , j = 1, 2,..., n - 1; Determine the sum of the elements of the data packet capacity difference set R according to the data packet capacity difference set R, and determine the second weight corresponding to each of the n - 1 node performance difference matrices according to the element values in the data packet capacity difference set R and the sum of the elements.

6. The method for analyzing the stability of an FTTR network according to claim 1, wherein The network stability analysis model includes an input layer, a hidden layer, and an output layer. The steps of generating a network stability analysis result based on the longitudinal feature vector and the transverse feature vector by the network stability analysis model include: Through the input layer, an input feature vector is generated according to the longitudinal feature vector and the transverse feature vector. Through the hidden layer, a feature extraction vector is generated according to the input feature vector. Through the output layer, an output vector is generated according to the feature extraction vector. The output vector includes a first element and a second element. The first element is used for the range of packet capacities acceptable for the FTTR network, and the second element is used to characterize the stability score of the FTTR network within different packet capacity ranges. Based on the output vector, a network stability analysis result is generated.

7. The method for analyzing the stability of an FTTR network according to claim 6, wherein The hidden layer includes a first hidden layer, a second hidden layer, and a third hidden layer. The neuron ratio between the first hidden layer, the second hidden layer, and the third hidden layer is 6:3:

1. The steps of generating a feature extraction vector according to the input feature vector through the hidden layer include: Based on the first hidden layer, a first intermediate feature vector is generated according to the input feature vector. Based on the second hidden layer, a second intermediate feature vector is generated according to the first intermediate feature vector. Based on the third hidden layer, a feature extraction vector is generated according to the second intermediate feature vector.

8. An FTTR network stability analysis system, characterized in that, Includes: A test module is configured to send test data packets from a sending device group S to a receiving device R based on preset sending parameters. Among them, the sending devices in the sending device group {S1, S2, ……, S n} have a one-to-one correspondence communication relationship with the slave optical modems in the FFTR network. The receiving device communicates with the master optical modem. The preset sending parameters include a preset sending frequency, a preset number of sending times, and a data packet capacity setting set P = {P1, P2, ……, P n}, and P1, P2, ……, P n shows an increasing trend. The data packet capacities sent by the same sending device are the same, and the data packet capacities sent by different sending devices are different; A performance matrix generation module for obtaining various performance parameters of each network node in each transmission cycle and generating a performance matrix for each network node according to the various performance parameters. The network nodes include the main optical modem and multiple slave optical modems. The rows or columns of the performance matrix represent the various performance parameters of the network nodes in the same transmission cycle. A feature vector determination module for determining the longitudinal feature vector corresponding to the network node according to the performance matrix and determining the transverse feature vector corresponding to the network node according to the performance matrix. The longitudinal feature vector is used to represent the performance fluctuation condition of the network nodes on the same data transmission link, and the transverse feature vector is used to represent the performance difference condition between the network nodes on different data transmission links. A stability analysis module for generating a network stability analysis result based on the network stability analysis model and according to the longitudinal feature vector and the transverse feature vector. The network stability analysis result is used to represent the stability of the FTTR network under different packet capacities.

9. A computer device, characterized in that, Includes a memory and a processor. A computer program that can run on the processor is stored on the memory. When the processor executes the computer program, the method according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium, characterized in that, Stores a computer program that can be loaded and executed by the processor to perform the method according to any one of claims 1 to 7.

Images