Network quality prediction method, apparatus, device, and storage medium

Abstract

The application discloses a network quality prediction method and device, equipment and a storage medium, relates to the technical field of communication, and is used for improving the accuracy of predicting the network quality of a communication link, and comprises the following steps: acquiring a plurality of basic parameters of a target network at a current time, wherein the plurality of basic parameters affect the network quality of the target network; performing normalization processing on the plurality of basic parameters to obtain target parameters, wherein the normalization processing is used for unifying the dimension and value range of each basic parameter in the plurality of basic parameters; inputting the target parameters into a preset network quality prediction model to determine a quality prediction result of the target network, wherein the preset network quality prediction model is obtained by training based on a plurality of training data and an elastic network model, and the quality prediction result is indicated by at least one of the following: throughput, delay and packet loss rate. The application is applied to the scene of predicting network communication quality.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of communication, and particularly relates to a network quality prediction method and device, equipment and a storage medium. BACKGROUND

[0002] With the continuous development of the 5th Generation Mobile Communication Technology (5G), the 5G network is widely used in the fields of environmental monitoring, intelligent transportation and smart furniture. However, the 5G network will be disturbed by noise in the environment during data transmission, resulting in poor network communication quality, data loss, and further retransmission of messages, which will further reduce network performance. Therefore, it is necessary to predict the network communication quality to select a high-quality communication link for data transmission, to ensure normal data transmission, reduce data loss and data retransmission, and improve network performance and throughput.

[0003] At present, the network quality of a communication link is determined based on link characteristics, which depends on network equipment. However, network equipment has calibration errors, and the network quality of a communication link cannot be accurately predicted. The network quality of a communication link is determined based on probability estimation, which requires sending a large amount of probe packet data, resulting in additional communication data of the network communication link and affecting the prediction of the network quality of the communication link. The network quality of a communication link is determined based on machine learning, which is determined by evaluating one influencing factor, and cannot evaluate multiple influencing factors, which also affects the prediction of the network quality of the communication link. Therefore, the accuracy of predicting the network quality of a communication link is low. SUMMARY

[0004] The present application provides a network quality prediction method, device, equipment and storage medium for improving the accuracy of predicting the network quality of a communication link.

[0005] To achieve the above object, the present application adopts the following technical solution:

[0006] In a first aspect, a network quality prediction method is provided, comprising: obtaining a plurality of basic parameters of a target network at a current time, the plurality of basic parameters affecting the network quality of the target network; performing normalization processing on the plurality of basic parameters to obtain target parameters, the normalization processing being used to unify the dimension and value range of each basic parameter in the plurality of basic parameters; inputting the target parameters into a preset network quality prediction model to determine a quality prediction result of the target network, the preset network quality prediction model being obtained by training based on a plurality of training data and an elastic network model, and the quality prediction result being indicated by at least one of the following: throughput, latency and packet loss rate.

[0007] In a possible implementation, the method further includes: obtaining a plurality of groups of first parameters corresponding to the target network, the plurality of groups of first parameters being network parameters corresponding to the target network at a plurality of historical time points, and each group of first parameters including a plurality of basic parameters and a plurality of performance parameters corresponding to a historical time point; for any group of first parameters in the plurality of groups of first parameters, performing normalization processing on the plurality of basic parameters included in the any group of first parameters to obtain a plurality of second parameters corresponding to the historical time point of the any group of first parameters, and each second parameter being obtained by performing normalization processing on a basic parameter; determining at least one third parameter from the plurality of second parameters as a group of third parameters, the at least one third parameter including a second parameter with an influence coefficient greater than a first preset threshold, and the influence coefficient being used to indicate a correlation between the second parameter and the plurality of performance parameters; and training the elastic network model based on the plurality of groups of third parameters determined based on the plurality of groups of first parameters to obtain the preset network quality prediction model.

[0008] In a possible implementation, the determining, from the plurality of second parameters, at least one third parameter as a group of third parameters includes: for any second parameter in the plurality of second parameters, determining, based on a first preset algorithm, an influence coefficient corresponding to the any second parameter, the first preset algorithm being used to determine a correlation between any two parameters; and determining, from the plurality of second parameters, at least one third parameter with an influence coefficient greater than a first preset threshold as a group of third parameters.

[0009] In a possible implementation, the training, based on the plurality of groups of third parameters determined based on the plurality of groups of first parameters, of the elastic network model to obtain the preset network quality prediction model includes: determining, based on the plurality of groups of third parameters determined based on the plurality of groups of first parameters, a comprehensive influence coefficient between any third parameter in the at least one third parameter and any performance parameter in the plurality of performance parameters by using a second preset algorithm, the second preset algorithm being used to determine a correlation between any two parameters in a time dimension; determining, from the at least one third parameter, a fourth parameter with a comprehensive influence coefficient less than a second preset threshold, and inputting the fourth parameter into the elastic network model for training to obtain the preset network quality prediction model.

[0010] In a possible implementation, the determining, for any second parameter in the plurality of second parameters, based on a first preset algorithm, of an influence coefficient corresponding to the any second parameter includes: for any second parameter in the plurality of second parameters, determining, based on a first preset algorithm, a correlation between the any second parameter and any performance parameter in the plurality of performance parameters; and summing the correlation between the any second parameter and each performance parameter in the plurality of performance parameters as the influence coefficient corresponding to the any second parameter.

[0011] In a possible implementation, the method further includes: training, by the target network, the preset verification model based on the plurality of second parameters and the plurality of performance parameters corresponding to each of the plurality of historical moments, to obtain a trained verification model, the trained verification model being used to verify accuracy of the preset network quality prediction model; inputting the target parameter corresponding to the target network at the current moment into the trained verification model to obtain a quality verification result; and determining the accuracy of the preset network quality prediction model based on the quality verification result and the quality prediction result.

[0012] In a second aspect, a network quality prediction apparatus is provided, which includes an acquisition unit and a processing unit. The acquisition unit is configured to acquire a plurality of basic parameters of a target network at a current moment, the plurality of basic parameters affecting network quality of the target network. The processing unit is configured to perform normalization processing on the plurality of basic parameters to obtain a target parameter, the normalization processing being configured to unify a dimension and a value range of each of the plurality of basic parameters. The processing unit is further configured to input the target parameter into a preset network quality prediction model to determine a quality prediction result of the target network, the preset network quality prediction model being obtained by training based on a plurality of training data and an elastic network model, and the quality prediction result being indicated by at least one of throughput, latency, and packet loss rate.

[0013] In a possible implementation, the acquisition unit is further configured to acquire a plurality of groups of first parameters corresponding to the target network, the plurality of groups of first parameters being network parameters of the target network at a plurality of historical moments, and each group of first parameters including a plurality of basic parameters and a plurality of performance parameters corresponding to a historical moment. The processing unit is further configured to, for any one of the plurality of groups of first parameters, perform normalization processing on the plurality of basic parameters included in the any one group of first parameters to obtain a plurality of second parameters corresponding to the historical moment of the any one group of first parameters, and each second parameter being obtained by performing normalization processing on one basic parameter. The processing unit is further configured to determine at least one third parameter from the plurality of second parameters as a group of third parameters, the at least one third parameter including a second parameter with an influence coefficient greater than a first preset threshold, the influence coefficient being used to indicate a correlation between the second parameter and the plurality of performance parameters. The processing unit is further configured to train the elastic network model based on the plurality of groups of third parameters determined based on the plurality of groups of first parameters to obtain the preset network quality prediction model.

[0014] In a possible implementation, the processing unit is specifically configured to, for any one of the plurality of second parameters, determine an influence coefficient corresponding to the any one second parameter based on a first preset algorithm, the first preset algorithm being used to determine a correlation between any two parameters. The processing unit is specifically configured to determine at least one third parameter with an influence coefficient greater than a first preset threshold from the plurality of second parameters as a group of third parameters.

[0015] In a possible implementation, the processing unit is specifically configured to determine, based on the plurality of sets of third parameters determined based on the plurality of sets of first parameters, a comprehensive influence coefficient between any third parameter in the at least one set of third parameters and any performance parameter in the plurality of performance parameters by using a second preset algorithm, the second preset algorithm being used to determine a correlation between any two parameters in a time dimension; and the processing unit is specifically configured to determine, from the at least one set of third parameters, a fourth parameter with a comprehensive influence coefficient less than a second preset threshold, and input the fourth parameter into the elastic network model for training to obtain the preset network quality prediction model.

[0016] In a possible implementation, the processing unit is specifically configured to determine, for any second parameter in the plurality of second parameters, a correlation between the second parameter and any performance parameter in the plurality of performance parameters based on a first preset algorithm; and the processing unit is specifically configured to sum the correlations between the second parameter and each performance parameter in the plurality of performance parameters as an influence coefficient corresponding to the second parameter.

[0017] In a possible implementation, the processing unit is further configured to train, by using the target network, the preset verification model based on the plurality of second parameters and the plurality of performance parameters corresponding to each historical moment in the plurality of historical moments, to obtain a trained verification model, the trained verification model being used to verify an accuracy of the preset network quality prediction model; the processing unit is further configured to input the target parameter corresponding to the target network at the current moment into the trained verification model to obtain a quality verification result; and the processing unit is further configured to determine the accuracy of the preset network quality prediction model based on the quality verification result and the quality prediction result.

[0018] In a third aspect, an electronic device is provided, including a processor and a memory; the memory is configured to store one or more programs including computer execution instructions; when the electronic device is running, the processor executes the computer execution instructions stored in the memory, so that the electronic device executes the network quality prediction method in the first aspect.

[0019] In a fourth aspect, a computer readable storage medium storing one or more programs is provided, the one or more programs including instructions that, when executed by a computer, cause the computer to perform the network quality prediction method in the first aspect.

[0020] This application provides a network quality prediction method, apparatus, device, and storage medium, applied to scenarios involving the prediction of network communication quality. First, multiple fundamental parameters affecting network quality in the target network at the current moment are obtained. Then, the dimensions and value ranges of each fundamental parameter are standardized to obtain normalized target parameters. These target parameters are then input into a pre-set network quality prediction model to obtain quality prediction results for the target network, including throughput, latency, and packet loss rate. Throughput, latency, and packet loss rate indicate the quality prediction results. By standardizing the dimensions and value ranges of each fundamental parameter based on a pre-trained network quality prediction model, the model's data processing efficiency and the accuracy of predicting the network quality of the target network can be improved. This effectively enhances the efficiency and accuracy of predicting the network quality of communication links. Attached Figure Description

[0021] Figure 1 A schematic diagram of a network quality prediction system structure provided for an embodiment of this application;

[0022] Figure 2 A schematic flowchart of a network quality prediction method provided for embodiments of this application. Figure 1 ;

[0023] Figure 3 A schematic flowchart of a network quality prediction method provided for embodiments of this application. Figure 2 ;

[0024] Figure 4 A schematic flowchart of a network quality prediction method provided for embodiments of this application. Figure 3 ;

[0025] Figure 5 A schematic flowchart of a network quality prediction method provided for embodiments of this application. Figure 4 ;

[0026] Figure 6 A method for determining network quality prediction model parameters provided in embodiments of this application. Figure 1 ;

[0027] Figure 7 A method for determining network quality prediction model parameters provided in embodiments of this application. Figure 2 ;

[0028] Figure 8 A schematic flowchart of a network quality prediction method provided for embodiments of this application. Figure 5 ;

[0029] Figure 9 A schematic flowchart of a network quality prediction method provided for embodiments of this application.Figure 6 ;

[0030] Figure 10 A network quality prediction device structure schematic diagram provided for an embodiment of the present application;

[0031] Figure 11 An electronic device structure schematic diagram provided for an embodiment of the present application. DETAILED DESCRIPTION

[0032] The technical solutions in the embodiments of the present application will be described below with reference to the drawings in the embodiments of the present application.

[0033] In the description of the present application, unless otherwise specified, " / " means "or", for example, A / B can mean A or B. "And / or" in this document only describes the association relationship of the associated objects, which means that there can be three relationships, for example, A and / or B, which can mean that A exists alone, A and B exist together, and B exists alone. In addition, "at least one" "multiple" means two or more. "First", "second", etc. do not limit the quantity and execution order, and "first", "second", etc. do not necessarily mean different.

[0034] At present, the network quality of the network communication link is predicted, and then the network communication link with higher network quality is selected, so as to ensure the high-quality transmission of data becomes the top priority of network communication.

[0035] However, using the method based on link characteristics to predict network quality depends on network equipment, but network equipment has calibration error, and the network quality prediction of the communication link is not accurate enough.

[0036] Using the method based on probability estimation to predict network quality needs to send a large number of probe packets, which not only increases the additional communication overhead, but also the probe packets are not sensitive enough to the network link condition.

[0037] Using the link quality prediction method based on machine learning to predict network quality is to transform the network link quality prediction problem into a time series prediction problem, use the method of machine learning or deep learning to mine the characteristics of the training sample, deeply learn the potential relationship between the input and output of the model, and use the obtained model to predict the network link quality at the next moment. Although the method of machine learning can effectively mine the characteristics of the training sample, it has higher accuracy and stability, but the algorithm based on machine learning can only output the quality prediction level of a single influencing factor, and cannot associate multiple communication quality influencing factors, that is, the input and output data are quality data or only use the classification model to classify the quality data, cannot consider multiple influencing factors, and cannot fully investigate the communication quality.

[0038] The network quality prediction method provided by the embodiments of the present application can be applied to a network quality prediction system. Figure 1 A structural schematic diagram of the network quality prediction system is shown. As shown in Figure 1 The network quality prediction system 20 includes an electronic device 21 and a network device 22.

[0039] The network device 22 can be a network device corresponding to an operator, such as a base station, a mobile base station, etc., and is configured to provide network services for users, such as 5G network, wireless sensor network, etc.

[0040] The electronic device 21 can acquire a plurality of basic parameters of the network, normalize the plurality of basic parameters, obtain target parameters with unified dimensions and value ranges, and then input the target parameters into a preset network quality prediction model, so as to predict the network quality of the network and obtain a network quality prediction result.

[0041] A network quality prediction method provided by the embodiments of the present application is described below with reference to the accompanying drawings. As shown in Figure 2 The network quality prediction method provided by the embodiments of the present application includes S201-S203.

[0042] S201, acquiring a plurality of basic parameters of a target network at a current time.

[0043] The plurality of basic parameters affect the network quality of the target network.

[0044] Optionally, the plurality of basic parameters can include communication distance, transmission power, networking mode, traffic volume, load volume, motion communication, whether it is a time-sensitive network, receiving sensitivity, etc.

[0045] It can be understood that by collecting the plurality of basic parameters affecting the network quality, the network quality during network transmission can be predicted, so as to filter out network links with high quality and improve the transmission efficiency of data.

[0046] Specifically, the topology of the network in the target network, the network device constituting the network, and the link information connecting the network can be collected, so as to acquire the plurality of basic parameters of the target network at the current time.

[0047] S202, normalizing the plurality of basic parameters to obtain target parameters.

[0048] The normalization processing is configured to unify the dimensions and value ranges of each of the plurality of basic parameters.

[0049] It should be noted that the dimension refers to the basic attribute of a physical quantity, different physical quantities have different dimensions, and the units used are also different, therefore, the units of multiple basic parameters need to be unified, thereby providing a standard for further processing of multiple basic parameters.

[0050] Optionally, the International System of Units can be used to process multiple basic parameters. For example, length (meter), mass (kilogram), time (second), electric current (ampere), thermodynamic temperature (kelvin), amount of substance (mole), and luminous intensity (candela).

[0051] S203, input the target parameter into the preset network quality prediction model to determine the quality prediction result of the target network.

[0052] Among them, the preset network quality prediction model is obtained by training based on multiple training data and an elastic network model, and the quality prediction result is indicated by at least one of the following: throughput, delay, and packet loss rate.

[0053] Specifically, the throughput refers to the amount of data effectively transmitted in the network channel per unit time. If the number of successfully transmitted data frames per unit time is M and the frame length of the data frame is L, then the current throughput S is shown in the following formula one:

[0054] S = M x L Formula one

[0055] Specifically, the delay can be divided into propagation delay and transmission delay. The propagation delay refers to the time required for data to propagate in the channel medium, which is mainly related to the distance; the transmission delay refers to the time required to push all bits in the data packet to the link. The size of the delay determines whether the nodes in the network can receive the information sent by other nodes in time.

[0056] Specifically, the packet loss rate refers to the ratio of the number of lost data packets to the total number of sent data packets during data transmission. The packet loss rate is related to the length of the transmitted data packet and the sending frequency of the data packet. The packet loss rate can be calculated by the following formula two:

[0057]

[0058] It can be understood that the throughput, delay, and packet loss rate of the target network can be predicted according to the target parameter and the preset network quality prediction model, thereby realizing the prediction of the network quality of the target network.

[0059] In the embodiment of the present application, first, a plurality of basic parameters affecting network quality in the target network at the current time are acquired, then the dimension and value range of each basic parameter in the plurality of basic parameters are uniformly processed to obtain normalized target parameters, and then the target parameters are input into a preset network quality prediction model to obtain quality prediction results of the target network including throughput, delay and packet loss rate, so as to indicate the quality prediction results through the throughput, delay and packet loss rate. Through the above method, on the basis of the preset network quality prediction model obtained by pre-training, the dimension and value range corresponding to each basic parameter are uniformly processed, which can improve the processing efficiency of the model on data and improve the accuracy of predicting the network quality of the target network. Thus, the efficiency and accuracy of predicting the network quality of the communication link are effectively improved.

[0060] In one design, as shown in FIG. 1, the embodiment of the present application provides a network quality prediction method, which can further include steps S301-S304: Figure 3

[0061] S301, a plurality of groups of first parameters corresponding to the target network are acquired.

[0062] Among them, the plurality of groups of first parameters are network parameters corresponding to the target network at a plurality of historical moments, and a group of first parameters includes a plurality of basic parameters and a plurality of performance parameters corresponding to a historical moment.

[0063] Specifically, the network parameters of the target network at a plurality of historical moments can be collected in combination with the communication scenario of the target network, and a sample set is constructed in chronological order, so as to acquire a plurality of groups of first parameters of the target network.

[0064] For example, eight basic parameters of the target network at a plurality of historical moments are acquired, including communication distance, transmission power, networking mode, traffic volume, load volume, motion communication, whether it is a time-sensitive network, and receiving sensitivity, denoted as x i (i=1, 2, ··· 8); and three performance parameters of the target network at a plurality of historical moments are acquired, including throughput, delay, and packet loss rate, denoted as y j (j=1, 2, 3). The eight basic parameters and the three performance parameters included in a group of first parameters are stored in a data set {x i , y j}.

[0065] S302, for any one group of first parameters in the plurality of groups of first parameters, the plurality of basic parameters included in any one group of first parameters are normalized to obtain a plurality of second parameters corresponding to the historical moment of any one group of first parameters.

[0066] Among them, one second parameter is obtained by normalizing one basic parameter. ​

[0067] Specifically, the plurality of basic parameters included in any one of the first parameters is normalized to ensure that each of the plurality of basic parameters included in any one of the first parameters corresponds to a unified dimension and value range, thereby obtaining a plurality of second parameters corresponding to the historical time of any one of the first parameters.

[0068] For example, the eight basic parameters x i (i = 1, 2, ··· 8) included in any one of the first parameters obtained are normalized to obtain a plurality of second parameters x i ’ (i = 1, 2, ··· 8).

[0069] S303, at least one third parameter is determined from the plurality of second parameters as a group of third parameters.

[0070] Among them, at least one third parameter includes a second parameter with an influence coefficient greater than a first preset threshold, and the influence coefficient is used to indicate the correlation between the second parameter and the plurality of performance parameters.

[0071] Specifically, the correlation between each of the plurality of second parameters and each of the plurality of performance parameters can be determined by a first preset algorithm, and then the influence coefficient corresponding to each second parameter is determined, and further at least one third parameter with an influence coefficient greater than a first preset threshold is selected from the plurality of second parameters as a group of third parameters.

[0072] It can be understood that by screening the plurality of second parameters, at least one third parameter with an influence coefficient greater than a first preset threshold between the performance parameters and the network quality is determined, and then the network is predicted by the at least one third parameter, thereby improving the prediction accuracy of the network quality.

[0073] S304, based on the plurality of third parameters determined by the plurality of first parameters, the elastic network model is trained to obtain a preset network quality prediction model.

[0074] Specifically, the comprehensive influence coefficient between any one of the at least one third parameter and any one of the plurality of performance parameters can be determined from the plurality of third parameters by a second preset algorithm based on the plurality of third parameters determined by the plurality of first parameters, and a fourth parameter less than a second preset threshold is determined from the plurality of third parameters, and then the fourth parameter is input into the elastic network model for training to obtain a preset network quality prediction model.

[0075] It can be understood that by screening at least one third parameter, a fourth parameter with a greater correlation between the third parameter and the performance parameter in the time dimension is further determined, thereby improving the accuracy of the prediction result of the network quality.

[0076] In the embodiments of the present application, the plurality of basic parameters included in any one of the plurality of groups of first parameters of the target network are filtered to determine parameters having a large correlation with the performance parameters. The preset network quality prediction model is obtained by training the elastic network model. Then, the network quality can be predicted by the plurality of basic parameters of the target network, and the accuracy of the prediction result is ensured.

[0077] In one design, as shown in FIG. 3, the network quality prediction method provided by the embodiments of the present application can include the following steps S301-S303: Figure 4 The method in step S303 can specifically include steps S401-S402.

[0078] S401, for any one of the plurality of second parameters, determining the influence coefficient corresponding to the any one second parameter based on a first preset algorithm.

[0079] The first preset algorithm is used to determine the correlation between any two parameters.

[0080] Optionally, the first preset algorithm can be a Pearson correlation coefficient algorithm, a Granger causality algorithm, or other algorithms that can determine the correlation between any two parameters.

[0081] Specifically, the correlation between any one of the plurality of second parameters and each of the plurality of performance parameters can be determined according to the first preset algorithm, and then the correlation between any one second parameter and the plurality of performance parameters is added to obtain the influence coefficient corresponding to the any one second parameter.

[0082] For example, according to the Pearson correlation coefficient algorithm, the correlation between any one of the eight second parameters x i ’ (i=1, 2, ··· 8) and any one of the three performance parameters y j (j=1, 2, 3) can be determined, and then the influence coefficient corresponding to any one second parameter is determined.

[0083]

[0084] S402, at least one third parameter having an influence coefficient greater than a first preset threshold is determined from the plurality of second parameters as a group of third parameters.

[0085] Optionally, in addition to determining at least one third parameter based on the influence coefficient being greater than the first preset threshold, the influence coefficients corresponding to multiple second parameters can be sorted in descending order, and the second parameter with the higher ranking can be selected as a set of third parameters.

[0086] Understandably, an influence coefficient greater than the first preset threshold indicates a strong correlation between the corresponding second parameter and the performance parameter. In other words, the second parameter has a significant impact on the performance parameter and also on the network quality. Therefore, filtering it out allows for more accurate prediction of the target network and reduces computational load.

[0087] For example, the influence coefficients corresponding to multiple second parameters Sort the data in descending order and select the second parameter corresponding to the top n influence coefficients as the third parameter x. i ’ (i = 1, 2, ..., n), where n is less than 8.

[0088] In this embodiment, a first preset algorithm is used to perform a horizontal analysis and comparison of the correlation between the second parameter and the performance parameter to determine at least one third parameter, thereby ensuring the accuracy of network quality prediction and reducing the computational load of irrelevant factors, thus improving the computational efficiency of the preset network quality prediction model.

[0089] In a design, such as Figure 5 As shown, in the network quality prediction method provided in this application embodiment, the method in step S304 above may specifically include steps S501-S502:

[0090] S501. Based on multiple sets of third parameters determined by multiple sets of first parameters, a second preset algorithm is used to determine the comprehensive influence coefficient between any one of the at least one third parameter and any one of the multiple performance parameters.

[0091] The second preset algorithm is used to determine the correlation between any two parameters in the time dimension.

[0092] It should be noted that because multiple third parameters are correlated and coupled, selecting parameters only horizontally can lead to an unstable model, thus affecting the model's predictive performance. Therefore, it is also necessary to select multiple parameters vertically along the time dimension to effectively extract parameters, reduce data dimensionality, improve model computation speed, and enhance model prediction accuracy.

[0093] Optionally, the second preset algorithm can use Granger causality test. Statistically, the prediction error is usually represented by the sum of squared residuals. Thus, a regression equation can be established, and the coefficient can be tested for zero using the hypothesis testing method (F test).

[0094] Exemplarily, the regression equation at the time t can be determined based on the Granger causality test, a plurality of groups of third parameters determined according to a plurality of groups of first parameters at a plurality of historical moments, and a plurality of performance parameters of the first parameter obtained, as shown in Formula Three:

[0095] y j,t = a0+ a1y j,t-1 + ··· + a p y j,t-p + b0+ b1x i,t-1 ’ + ··· + b p x i,t-p ’ + ε t Formula Three

[0096] wherein y j,t is a performance parameter at the time t, x i,t-1 ’ (i = 1, 2, ··· n) is a third parameter at the time t-1, a and b are constant coefficients, b1 to b p is a comprehensive influence coefficient between any third parameter in the at least one third parameter and any performance parameter in the plurality of performance parameters, and ε t is an error term at the time t.

[0097] It should be noted that there is a group of comprehensive influence coefficients between each performance parameter y j (j = 1, 2, 3) and each third parameter x i ’ (i = 1, 2, ··· n).

[0098] S502, determining a fourth parameter from the at least one third parameter, wherein a comprehensive influence coefficient of the fourth parameter is less than a second preset threshold value, and inputting the fourth parameter into the elastic network model for training to obtain a preset network quality prediction model.

[0099] It should be noted that in the case where the comprehensive influence coefficient of any third parameter is less than the second preset threshold value, it indicates that there is an obvious causal relationship between the third parameter and the plurality of performance parameters, and therefore the third parameter can be determined as the fourth parameter and input into the elastic network model for further training to obtain the preset network quality prediction model.

[0100] Specifically, the elastic network model is a linear regression model similar to the least absolute shrinkage and selection operator (Lasso) and ridge regression algorithm, which combines the characteristics of Lasso regression algorithm and ridge regression algorithm. The main difference between Lasso regression algorithm and ridge regression algorithm is the penalty term. Lasso regression algorithm takes the sum of squares of coefficients as the penalty term, that is, the L1 regularization term, while ridge regression algorithm takes the sum of absolute values of coefficients as the penalty term, that is, the L2 regularization term. The advantage of ridge regression algorithm is to improve the prediction accuracy, but it cannot discard any feature to make the regression coefficient 0, and it is relatively stable compared with Lasso regression algorithm. While Lasso regression algorithm can shrink some associated feature coefficients to 0, so it is beneficial to feature selection and greatly improves the interpretability of the model. Elastic network is a compromise between the two methods, which automatically performs feature selection while continuously shrinking the coefficients, can select group-related features, and can select more than the number of samples to make the feature saturated, suitable for models with multiple features related to each other, and achieves better stability, higher accuracy, and better generalization and interpretability. Elastic network algorithm has been applied and has good effect in the fields of wind speed prediction, stock prediction and load prediction.

[0101] For example, in combination with Equation Three, if the third parameter x i ’ is not significantly 0, the comprehensive influence coefficient of each performance parameter y j (j = 1, 2, 3) is not significantly 0, and the fourth parameter x i ’ (i = 1, 2, ··· n) is the selected fourth parameter x i ’ (i = 1, 2, ··· m), where m is less than or equal to n.

[0102] Specifically, the selected fourth parameter is taken as the input quantity and the performance parameter is taken as the output quantity, and input into the loss function of the elastic network model as shown in Equation Four below:

[0103]

[0104] Where, α is a mixing parameter, γ is a complexity parameter, x i ′ is the independent variable (fourth parameter), ‖β i ‖1 is the Lasso term, is the ridge regression term.

[0105] It should be noted that α and γ are two hyperparameters. The mixture parameter α (0≤α≤1) characterizes the degree of Lasso regression and ridge regression algorithms; for example, α=0 represents ridge regression and α=1 represents Lasso regression. The complexity parameter γ characterizes the degree of compression (penalty).

[0106] Optional, such as Figure 6 As shown, this represents the fit of the elastic network model with the hybrid parameter α ranging from 0 to 1. This allows us to determine the accurate value of the hyperparameter (hybrid parameter) α in the elastic network model, and further, obtain the preset network quality prediction model.

[0107] Optional, such as Figure 7 As shown, this represents the fit of the elastic network model to the complex parameter γ within the range of 0 to 4. This allows us to determine the accurate value of the complex parameter γ in the elastic network model and, further, obtain the preset network quality prediction model.

[0108] In this embodiment, by filtering multiple sets of third parameters in the time dimension, a fourth parameter that has a causal relationship with the performance parameters is obtained and input into the elastic network model for training, thereby obtaining the hyperparameters of the model and further obtaining the preset network quality prediction model.

[0109] In a design, such as Figure 8 As shown, in the network quality prediction method provided in this application embodiment, the method in step S401 above may specifically include steps S601-S602:

[0110] S601. For any one of the multiple second parameters, based on the first preset algorithm, determine the correlation between any one of the second parameters and any one of the multiple performance parameters.

[0111] For example, based on the Pearson correlation coefficient algorithm, with eight second parameters x i ’ (i = 1, 2, ..., 8) are used as variables, and three performance parameters y j Using (j = 1, 2, 3) as the objective variable, the correlation between any second parameter and any performance parameter among multiple performance parameters can be determined, i.e., the second parameter x i ’ With three performance parameters y j The corresponding relevance is r ij For example, the second parameter x1 ’ The correlation between the three performance parameters y1, y2, and y3 is r. 11 r 12 r 13 .

[0112] S602, sum the correlation between any second parameter and each performance parameter in the plurality of performance parameters as an influence coefficient corresponding to the any second parameter.

[0113] For example, the correlation r between any second parameter x i ’ and each performance parameter in the plurality of performance parameters is summed to obtain an influence coefficient corresponding to the any second parameter i1 , r i2 , r i3 .

[0114]

[0115] It can be understood that, according to the obtained influence coefficient corresponding to any second parameter, comparison is made with the first preset threshold, and in a case where the influence coefficient of a certain second parameter is greater than the first preset threshold, the second parameter is determined as a third parameter, and then a plurality of groups of third parameters are determined, and the elastic network model is trained.

[0116] In the embodiment of the application, the correlation between any second parameter and any performance parameter in the plurality of performance parameters is determined by the first preset algorithm, and then the correlation is added to obtain an influence coefficient corresponding to the any second parameter, and the influence coefficient can be further distinguished to select a third parameter from the second parameters.

[0117] In one design, as shown in Figure 9 , the network quality prediction method provided by the embodiment of the application can further include steps S701-S703:

[0118] S701, training a preset verification model by using the plurality of second parameters and the plurality of performance parameters corresponding to each historical moment in the plurality of historical moments of the target network, to obtain a trained verification model.

[0119] The trained verification model is used to verify the accuracy of the preset network quality prediction model.

[0120] Optionally, the preset verification model can be a back propagation (BP) neural network model.

[0121] Exemplarily, using the BP neural network model as the preset verification model, the mapping relationship between the input and the output mode is obtained by learning the plurality of second parameters and the plurality of performance parameters corresponding to each of the plurality of historical moments, and the model is trained. This process is not learning a mathematical equation describing the mapping relationship, but through the total error of the network, using the gradient descent algorithm, constantly adjusting the threshold value of each node in the network and the weight value between the nodes, thereby determining the trained verification model. Generally, the BP neural network is composed of an input layer, a hidden layer and an output layer. The input layer corresponds to several attributes of each data; the hidden layer can be one or more layers, each layer having several nodes; the output layer can be classified by different results output by several nodes. Each node itself has a threshold value, and has a weight value with each node of the next layer. Through multiple iterations of the input sample data, these parameters are finally determined, so that new data can be predicted.

[0122] S702, inputting the target parameter corresponding to the target network of the current moment into the trained verification model to obtain a quality verification result.

[0123] It can be understood that, through the trained verification model, according to the target parameter corresponding to the target network of the current moment, the trained verification model can realize the prediction of the network quality of the target network, and obtain the quality verification result, so that the quality prediction result obtained by the preset network quality prediction model can be further compared, thereby realizing the verification of the accuracy of the preset network quality prediction model.

[0124] S703, determining the accuracy of the preset network quality prediction model based on the quality verification result and the quality prediction result.

[0125] Optionally, the quality verification result and the quality prediction result can be judged by the mean square error, the average error and the average relative error, so as to determine the accuracy of the preset network quality prediction model.

[0126] Optionally, the preset network quality prediction model can be corrected according to the quality verification result, so that the preset network quality prediction model is more convergent, and the network quality prediction result is more accurate.

[0127] In the embodiment of the application, the preset verification model is trained to obtain the quality verification result, which is further compared with the quality prediction result of the preset network quality prediction model to determine the accuracy of the preset network quality prediction model, so that the preset network quality prediction model can be continuously corrected to ensure the accuracy of the network quality prediction result.

[0128] The application provides a network quality prediction method. First, a plurality of basic parameters affecting network quality in a target network at a current time are acquired, then a dimension and a value range of each basic parameter in the plurality of basic parameters are uniformly processed to obtain normalized target parameters, then the target parameters are input into a preset network quality prediction model to obtain a quality prediction result of the target network including throughput, delay and packet loss rate, so as to indicate the quality prediction result by the throughput, the delay and the packet loss rate. Through the above method, on the basis of the preset network quality prediction model obtained by pre-training, the dimension and the value range corresponding to each basic parameter are uniformly processed, the processing efficiency of the model on data can be improved, and the accuracy of predicting the network quality of the target network is improved. Thus, the efficiency and the accuracy of predicting the network quality of the communication link are effectively improved.

[0129] The above mainly introduces the scheme provided by the embodiments of the application from the perspective of the method. To implement the above functions, the hardware structure and / or software module corresponding to each function are included. Those skilled in the art should easily realize that, in combination with the units and algorithm steps of the examples described in the embodiments disclosed herein, the embodiments of the application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a certain function is implemented in hardware or computer software driven hardware depends on the specific application of the technical solution and the design constraint conditions. The skilled person can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the application.

[0130] The embodiments of the application can divide a network quality prediction device into functional modules according to the above method examples. For example, each functional module can be divided according to each function, or two or more functions can be integrated into one processing module. The integrated module can be implemented in the form of hardware or software functional module. Optionally, the division of the modules in the embodiments of the application is illustrative, and is only a logical functional division. In actual implementation, another division mode can be used.

[0131] Figure 10 A structural schematic diagram of a network quality prediction device provided by the embodiments of the application is shown in FIG. 1. As shown in FIG. 1, a network quality prediction device 100 is used to improve the accuracy of predicting the network quality of a communication link, for example, to implement the network quality prediction method shown in FIG. 2. Figure 10 As shown in FIG. 2, the network quality prediction method includes the following steps. Figure 2 The network quality prediction device 100 includes an acquisition unit 1001 and a processing unit 1002.

[0132] The acquisition unit 1001 is configured to acquire a plurality of basic parameters of a target network at a current time, and the plurality of basic parameters affect the network quality of the target network.

[0133] The processing unit 1002 is configured to normalize the plurality of basic parameters to obtain target parameters, and the normalization is configured to unify the dimension and value range of each basic parameter in the plurality of basic parameters.

[0134] The processing unit 1002 is further configured to input the target parameters into a preset network quality prediction model to determine a quality prediction result of the target network, and the preset network quality prediction model is obtained by training based on a plurality of training data and an elastic network model, and the quality prediction result is indicated by at least one of the following: throughput, delay, and packet loss rate.

[0135] In a possible implementation, in the network quality prediction apparatus 100 provided by the embodiment of the present application, the acquisition unit 1001 is further configured to acquire a plurality of groups of first parameters corresponding to the target network, the plurality of groups of first parameters are network parameters corresponding to the target network at a plurality of historical moments, and one group of first parameters includes a plurality of basic parameters and a plurality of performance parameters corresponding to one historical moment.

[0136] The processing unit 1002 is further configured to, for any one group of first parameters in the plurality of groups of first parameters, normalize the plurality of basic parameters included in the any one group of first parameters to obtain a plurality of second parameters corresponding to the historical moment of the any one group of first parameters, and one second parameter is obtained by normalizing one basic parameter.

[0137] The processing unit 1002 is further configured to determine at least one third parameter from the plurality of second parameters as one group of third parameters, the at least one third parameter includes a second parameter with an influence coefficient greater than a first preset threshold, and the influence coefficient is used to indicate the correlation between the second parameter and the plurality of performance parameters.

[0138] The processing unit 1002 is further configured to train the elastic network model based on the plurality of groups of third parameters determined based on the plurality of groups of first parameters to obtain the preset network quality prediction model.

[0139] In a possible implementation, in the network quality prediction apparatus 100 provided by the embodiment of the present application, the processing unit 1002 is specifically configured to, for any one second parameter in the plurality of second parameters, determine an influence coefficient corresponding to the any one second parameter based on a first preset algorithm, and the first preset algorithm is used to determine the correlation between any two parameters.

[0140] The processing unit 1002 is specifically configured to determine at least one third parameter with an influence coefficient greater than a first preset threshold from the plurality of second parameters as one group of third parameters.

[0141] In a possible implementation, in the network quality prediction apparatus 100 provided by the embodiment of the present application, the processing unit 1002 is specifically configured to determine, by a second preset algorithm, a comprehensive influence coefficient between any third parameter in the at least one third parameter and any performance parameter in the plurality of performance parameters, the second preset algorithm being used to determine a correlation between any two parameters in a time dimension;

[0142] The processing unit 1002 is specifically configured to determine, from the at least one third parameter, a fourth parameter with a comprehensive influence coefficient less than a second preset threshold, and input the fourth parameter into the elastic network model for training to obtain the preset network quality prediction model.

[0143] In a possible implementation, in the network quality prediction apparatus 100 provided by the embodiment of the present application, the processing unit 1002 is specifically configured to determine, for any second parameter in the plurality of second parameters, a correlation between the any second parameter and any performance parameter in the plurality of performance parameters based on a first preset algorithm.

[0144] The processing unit 1002 is specifically configured to sum the correlation between the any second parameter and each performance parameter in the plurality of performance parameters as an influence coefficient corresponding to the any second parameter.

[0145] In a possible implementation, in the network quality prediction apparatus 100 provided by the embodiment of the present application, the processing unit 1002 is further configured to train the preset verification model by the plurality of second parameters and the plurality of performance parameters corresponding to each historical moment in the plurality of historical moments of the target network, to obtain a trained verification model, the trained verification model being used to verify the accuracy of the preset network quality prediction model.

[0146] The processing unit 1002 is further configured to input the target parameter corresponding to the target network at the current moment into the trained verification model to obtain a quality verification result.

[0147] The processing unit 1002 is further configured to determine the accuracy of the preset network quality prediction model based on the quality verification result and the quality prediction result.

[0148] In the case of implementing the functions of the above integrated modules in the form of hardware, the embodiment of the present application provides another possible structural diagram of the electronic device involved in the above embodiments. As shown in Figure 11 An electronic device 110 is used to improve the accuracy of predicting the network quality of a communication link, for example, to perform Figure 2An electronic device 110 is shown. The electronic device 110 includes a processor 1101, a memory 1102 and a bus 1103. The processor 1101 and the memory 1102 can be connected through the bus 1103.

[0149] The processor 1101 is a control center of the communication device, which can be one processor or a general term of multiple processing elements. For example, the processor 1101 can be a general central processing unit (CPU), or other general-purpose processors, etc. The general-purpose processor can be a microprocessor or any conventional processor, etc.

[0150] As an embodiment, the processor 1101 can include one or more CPUs, such as the CPU 0 and the CPU 1 shown in FIG. 1. Figure 11

[0151] The memory 1102 can be a read-only memory (ROM) or other type of static storage device that can store static information and instructions, a random access memory (RAM) or other type of dynamic storage device that can store information and instructions, an electrically erasable programmable read-only memory (EEPROM), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by a computer, but is not limited to this.

[0152] As a possible implementation, the memory 1102 can exist independently of the processor 1101. The memory 1102 can be connected to the processor 1101 through the bus 1103, and used to store instructions or program codes. When the processor 1101 invokes and executes the instructions or program codes stored in the memory 1102, the method for predicting network quality provided in the embodiments of the present application can be implemented.

[0153] In another possible implementation, the memory 1102 can also be integrated with the processor 1101.

[0154] ​The bus 1103 can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or other types of bus. The bus can be divided into an address bus, a data bus, a control bus, and the like. For ease of representation, Figure 11 Only one thick line is used to represent the bus in the figure, but it does not mean that there is only one bus or only one type of bus.

[0155] It should be noted that, Figure 11 The structure shown does not constitute a limitation on the electronic device 110. In addition to Figure 11 the components shown, the electronic device 110 can include more or fewer components than shown, or combine certain components, or different component arrangements.

[0156] As an example, in combination with Figure 10 , the functions implemented by the acquisition unit 1001 and the processing unit 1002 in the electronic device are the same as the functions of the processor 1101 in Figure 11 .

[0157] Optionally, as Figure 11 shown, the electronic device 110 provided by the embodiments of the present application can further include a communication interface 1104.

[0158] The communication interface 1104 is used to connect with other devices through a communication network. The communication network can be an Ethernet, a wireless access network, a wireless local area network (WLAN), and the like. The communication interface 1104 can include a receiving unit for receiving data, and a sending unit for sending data.

[0159] In one design, in the electronic device provided by the embodiments of the present application, the communication interface can also be integrated in the processor.

[0160] Through the above description of the embodiments, those skilled in the art can clearly understand that, for the convenience and brevity of description, only the above division of functional units is taken as an example. In actual application, the above functions can be completed by different functional units according to needs, that is, the internal structure of the device is divided into different functional units to complete all or part of the functions described above. The specific working process of the system, device and unit described above can refer to the corresponding process in the foregoing method embodiments, which will not be described here.

[0161] The embodiment of the present application further provides a computer readable storage medium, and the computer readable storage medium stores instructions. When a computer executes the instructions, the computer executes each step in the method flow shown in the method embodiment.

[0162] The embodiment of the present application provides a computer program product containing instructions, which, when executed on a computer, cause the computer to perform a network quality prediction method in the method embodiment.

[0163] The computer readable storage medium may, for example, be, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or apparatus, or any combination of the above. More specific examples (a non-exhaustive list) of the computer readable storage medium include an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a register, a hard disk, an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any other suitable combination of the above, or any other medium from which the program can be derived. Thus, the computer readable storage medium can be embodied by a computer program product, a memory, a volatile memory, a non-volatile memory, a floppy disk device, a CD-ROM, a DVD, a Blu-ray Disc, a hard disk, a removable media repository, or any other medium that can be used to store the desired program code in a non-transitory form so that it can be executed by the computer.

[0164] An exemplary storage medium is coupled to the processor such that the processor can read information from, and write information to, the storage medium. Of course, the storage medium can be a part of the processor. The processor and the storage medium can be located in an Application Specific Integrated Circuit (ASIC).

[0165] In the embodiment of the present application, the computer readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus or device.

[0166] The electronic device, the computer readable storage medium and the computer program product in the embodiment of the present application can be applied to the above method, and the technical effects that can be obtained are also referable to the above method embodiment, which will not be described here in the embodiment of the present application.

[0167] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto, and any change or replacement within the technical scope disclosed in the present application should be covered in the protection scope of the present application.

Claims

1. A network quality prediction method characterized by, The method comprises: obtaining a plurality of basic parameters of a target network at a current time, the plurality of basic parameters affecting network quality of the target network; normalizing the plurality of basic parameters to obtain target parameters, the normalization being used to unify the dimension and value range of each basic parameter in the plurality of basic parameters; inputting the target parameters into a preset network quality prediction model to determine a quality prediction result of the target network, the preset network quality prediction model being obtained by training based on a plurality of training data and an elastic network model, the quality prediction result being indicated by at least one of throughput, latency and packet loss rate; The method further comprises: obtaining a plurality of groups of first parameters corresponding to the target network, the plurality of groups of first parameters being network parameters corresponding to the target network at a plurality of historical times, and a group of first parameters comprising a plurality of basic parameters and a plurality of performance parameters corresponding to a historical time; for any group of first parameters in the plurality of groups of first parameters, normalizing the plurality of basic parameters included in the any group of first parameters to obtain a plurality of second parameters corresponding to the historical time of the any group of first parameters, one second parameter being obtained by normalizing one basic parameter; determining at least one third parameter from the plurality of second parameters as a group of third parameters, the at least one third parameter comprising a second parameter with an influence coefficient greater than a first preset threshold, the influence coefficient being used to indicate the correlation between the second parameter and the plurality of performance parameters; based on the plurality of groups of third parameters determined based on the plurality of groups of first parameters, determining a comprehensive influence coefficient between any third parameter in the at least one third parameter and any performance parameter in the plurality of performance parameters by a second preset algorithm, the second preset algorithm being used to determine the correlation between any two parameters in the time dimension; determining a fourth parameter with a comprehensive influence coefficient less than a second preset threshold from the at least one third parameter, and inputting the fourth parameter into the elastic network model for training to obtain the preset network quality prediction model.

2. The method of claim 1, wherein, The determination of the at least one third parameter from the plurality of second parameters as a group of third parameters comprises: for any second parameter in the plurality of second parameters, determining an influence coefficient corresponding to the any second parameter based on a first preset algorithm, the first preset algorithm being used to determine the correlation between any two parameters; determining at least one third parameter with an influence coefficient greater than the first preset threshold from the plurality of second parameters as a group of third parameters.

3. The method of claim 2, wherein, The determination of the influence coefficient corresponding to the any second parameter based on the first preset algorithm comprises: for any second parameter in the plurality of second parameters, determining the correlation between the any second parameter and any performance parameter in the plurality of performance parameters based on a first preset algorithm; summing the correlation between the any second parameter and each performance parameter in the plurality of performance parameters as the influence coefficient corresponding to the any second parameter.

4. The method according to any one of claims 1 to 3, characterized in that, The method further comprises: training, by the target network, a preset verification model based on a plurality of second parameters and a plurality of performance parameters corresponding to each of a plurality of historical time points, to obtain a trained verification model, the trained verification model being used to verify accuracy of the preset network quality prediction model; inputting the target parameter corresponding to the target network at a current time point into the trained verification model to obtain a quality verification result; determining accuracy of the preset network quality prediction model based on the quality verification result and the quality prediction result.

5. A network quality prediction apparatus characterized by comprising: The network quality prediction device comprises an acquisition unit and a processing unit. The acquisition unit is configured to acquire a plurality of basic parameters of a target network at a current time point, the plurality of basic parameters affecting network quality of the target network. The processing unit is configured to perform normalization processing on the plurality of basic parameters to obtain a target parameter, the normalization processing being used to unify a dimension and a value range of each of the plurality of basic parameters. The processing unit is further configured to input the target parameter into a preset network quality prediction model to determine a quality prediction result of the target network, the preset network quality prediction model being obtained by training based on a plurality of training data and an elastic network model, the quality prediction result being indicated by at least one of throughput, latency, and packet loss rate. The acquisition unit is further configured to acquire a plurality of groups of first parameters corresponding to the target network, the plurality of groups of first parameters being network parameters corresponding to a plurality of historical time points of the target network, and one group of first parameters comprising a plurality of basic parameters and a plurality of performance parameters corresponding to one historical time point. The processing unit is further configured to, for any one group of first parameters in the plurality of groups of first parameters, perform normalization processing on a plurality of basic parameters included in the any one group of first parameters to obtain a plurality of second parameters corresponding to a historical time point of the any one group of first parameters, one second parameter being obtained by performing normalization processing on one basic parameter. The processing unit is further configured to determine at least one third parameter from the plurality of second parameters as one group of third parameters, the at least one third parameter comprising a second parameter with an influence coefficient greater than a first preset threshold, the influence coefficient being used to indicate a correlation degree between the second parameter and the plurality of performance parameters. The processing unit is further configured to determine, based on a plurality of groups of third parameters determined based on the plurality of groups of first parameters, a comprehensive influence coefficient between any one third parameter in the at least one third parameter and any one performance parameter in the plurality of performance parameters by using a second preset algorithm, the second preset algorithm being used to determine a correlation degree between any two parameters in a time dimension. The processing unit is further configured to determine a fourth parameter with a comprehensive influence coefficient less than a second preset threshold from the at least one third parameter, and input the fourth parameter into the elastic network model to obtain the preset network quality prediction model.

6. The network quality prediction apparatus according to claim 5, wherein The processing unit is specifically configured to determine, for any second parameter in the plurality of second parameters, an influence coefficient corresponding to the any second parameter based on a first preset algorithm, the first preset algorithm being used to determine the correlation between any two parameters. The processing unit is specifically configured to determine, from the plurality of second parameters, at least one third parameter with an influence coefficient greater than the first preset threshold as a group of third parameters.

7. The network quality prediction apparatus according to claim 6, wherein The processing unit is specifically configured to determine, for any second parameter in the plurality of second parameters, the correlation between the any second parameter and any performance parameter in the plurality of performance parameters based on a first preset algorithm. The processing unit is specifically configured to sum the correlation between the any second parameter and each performance parameter in the plurality of performance parameters as the influence coefficient corresponding to the any second parameter.

8. The network quality prediction apparatus according to any one of claims 5-7, wherein, The processing unit is further configured to train a preset verification model by the plurality of second parameters and the plurality of performance parameters corresponding to each historical time of the plurality of historical times through the target network, to obtain a trained verification model, and the trained verification model is used to verify the accuracy of the preset network quality prediction model. The processing unit is further configured to input the target parameter corresponding to the target network at the current time into the trained verification model to obtain a quality verification result. The processing unit is further configured to determine the accuracy of the preset network quality prediction model based on the quality verification result and the quality prediction result.

9. An electronic device, comprising: Comprising: A processor and a memory; wherein the memory is used to store one or more programs, the one or more programs including computer execution instructions, when the electronic device is running, the processor executes the computer execution instructions stored in the memory, so that the electronic device executes the network quality prediction method in any one of claims 1-4.

10. A computer-readable storage medium storing one or more programs, the one or more programs comprising instructions for: The one or more programs include instructions that, when executed by a computer, cause the computer to perform a network quality prediction method as claimed in any one of claims 1-4.

Images